74911 PNG Health Workforce Crisis: A Call to Action October, 2011 PNG Health Workforce Crisis: A Call to Action October 2011 1305747 PNG HR Report 11-27-12.indb 1 11/27/12 11:38 AM Disclaimer: The findings, interpretations and conclusions expressed in this paper are entirely those of the authors, and do not necessarily represent the views of the World Bank, its Executive Directors, or the countries they represent. 1305747 PNG HR Report 11-27-12.indb 2 11/27/12 11:38 AM Contents Foreword ....................................................................................................................................................................ix Acknowledgments .....................................................................................................................................................xi Abbreviations and Acronyms ................................................................................................................................. xiii Executive Summary ..................................................................................................................................................xv Chapter 1:  Introduction and Background ................................................................................................................1 1.1 Introduction ......................................................................................................................................................................... 1 1.2 Health System Structural Changes: Emergence of an Almost Unmanagable System ................................................ 1 1.3 Health Outcomes Remain a Serious Challenge.............................................................................................................. 2 1.4 Government Health Expenditures Substantial But Inefficient and Fragmented ........................................................ 3 1.5 The Emerging Human Resource Crisis: The Reason for the Study .............................................................................. 3 1.6 Outline of the Report .......................................................................................................................................................... 4 Chapter 2:  Characteristics of the Current Health Public Sector Workforce ...........................................................7 2.1 Introduction......................................................................................................................................................................... 7 2.2 Size and Deployment of the Health Workforce ............................................................................................................... 7 Annex 2.1  Notes on Data Sources and Data Constraints. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Comments on the Current Status of the Key Human Resource Information Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Methodology Used to Estimate the Health Workforce and its Key Characteristics.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Annex 2.2  Publicly Financed Health Staff by Province and Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Chapter 3:   Health Workforce Training Capacity and Issues .................................................................................27 3.1 Introduction....................................................................................................................................................................... 27 3.2 A Brief Discussion of the Historic Context of the National Health Workforce Training System ................................................................................................................................................................. 27 3.3 A Brief Description of the Health-Related Training System Institutions and Programs ........................................ 29 3.3.1 Universities .......................................................................................................................................................... 29 3.3.2 Schools of Nursing ............................................................................................................................................... 30 3.3.3 Community Health Worker Training Schools (All Mission Owned) ................................................................ 30 iii 1305747 PNG HR Report 11-27-12.indb 3 11/27/12 11:38 AM 3.4 Preservice Enrollments and Outputs of Health Training Institutions ....................................................................... 30 3.4.1 Medical, Health Extension Officer, Nursing and Community Health Worker Training Programs ............ 31 3.5 Postgraduate Training Programs ..................................................................................................................................... 36 3.5.1 Nurse Midwifery Programs ............................................................................................................................... 36 3.6 A Note on Subsequent Chapters of the Report ............................................................................................................. 36 Annex 3.1  Real Per Capita Recurrent and Development Expenditure (1975–2009). . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Annex 3.2  Health Training Institutions, Programs, Enrollments and Graduates (2009). . . . . . . . . . . . . . . . . . . . . . 41 Chapter 4:  Characteristics of Schools of Nursing and CHW Training Institutions ..............................................45 4.1 Introduction ....................................................................................................................................................................... 45 4.2 Nurse and CHW Training Institution Expenditures 1995–2009 ................................................................................ 45 4.2.1 Schools of Nursing ............................................................................................................................................. 45 4.2.2 CHW Training Schools ....................................................................................................................................... 46 4.3 Unit Costs of Nurse Schools and CHW Training Institutions .................................................................................... 50 4.4 Staffing Numbers and Other Characteristics of Nurse and CHW Training Schools ............................................... 52 4.4.1 Nursing Schools ................................................................................................................................................. 52 4.4.2 CHW Training Schools ...................................................................................................................................... 54 4.5 Key Indicators for Nurse and CHW Schools: A Summary ......................................................................................... 57 4.6 A Qualitative and Quantitative Assessment of Nurse and CHW Training Schools: A Summary Report of the Survey of Principal’s Views ...................................................................................................................... 57 4.6.1 Nursing School Student Entry Requirements, Quality of Curriculum and of Graduates ........................... 58 4.6.2 Nursing School Assets and Infrastructure Quality ......................................................................................... 61 4.6.3 CHW School Student Entry Requirements, Quality of Curriculum and of Graduates.............................. 62 4.6.4 CHW School Assets and Infrastructure Quality .............................................................................................. 64 4.7 A Brief Conclusion to Chapter 4 ..................................................................................................................................... 65 Chapter 5:  Future Staffing Demand Scenarios for PNG’s National Health System ...............................................67 5.1 Introduction ....................................................................................................................................................................... 67 5.2 Key Aspects of Demand for Direct Health Service Providers ..................................................................................... 68 5.2.1 Historic Government Recurrent Resource Constraints .................................................................................. 68 5.2.2 Health Outcomes and Revealed Demand for Health Services ...................................................................... 68 5.2.3 Population Growth Notes: Projections, Rural-Urban Distribution and Mobility ....................................... 70 5.2.4 Resources Available to Health: the Future ...................................................................................................... 72 5.3 Projecting the Demand for Health Cadres: Five Scenarios ......................................................................................... 74 5.4 Future Health Human Resource Demand Scenarios 2010–2030 ............................................................................... 78 5.4.1 Scenario 1: The No-Change-in-Supply Scenario ............................................................................................ 79 5.4.2 Scenario 2: The PNGDSP-Posited Aspirational Scenario ............................................................................ 80 5.4.3 Scenario 3: Maintaining Current Population-to-Staff Ratios Scenario ....................................................... 83 5.4.4 Scenario 4: The WHO-Recommended “Threshold” Service-Delivery Staff Scenario ................................ 83 5.4.5 Scenario 5: The Recommended Scenario ........................................................................................................ 86 5.5 Conclusions ........................................................................................................................................................................ 88 Chapter 6:  Supply and Demand: Key Health Cadre Supply Gaps ..........................................................................91 6.1 Introduction....................................................................................................................................................................... 91 6.2 The Range and Scale of Supply and Demand Gaps by Key Health Worker Cadre ................................................... 19 6.3 Detailed Scenario Analysis .............................................................................................................................................. 92 6.3.1 Scenario 1: No Change in Human Resource Supply Capacity 2010–2030 ................................................. 92 iv  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 4 11/27/12 11:38 AM 6.3.2 Scenario 2: Aspirational Health Workforce Targets Envisioned in the PNGDSP 2010–2030................ 102 6.3.3 Scenario 3: Maintaining Existing Population-to-Service Delivery Staff Cadre Ratios 2010 to 2030 ..................................................................................................................................................... 108 6.3.4 Scenario 4: WHO “Threshold” Service Delivery Staff Density Targets by 2030 ...................................... 116 6.3.5 Scenario 5: A Recommended Preservice Training Scenario to Meet Key Health Human Resource Needs .............................................................................................................. 125 Chapter 7:  Meeting Human Resource Needs: Options and Recommendations to Enhance Health Human Resource Supply Responses to meet Needs ..................................................................................137 7.1 Introduction ..................................................................................................................................................................... 137 7.2 Critical Need for Improved Data on Health Human Resources ............................................................................... 137 7.2.1 The Stock of Health Human Resources: Public and Private ....................................................................... 137 7.2.2 Documentation of Current National Training Capacity of Health Workers .............................................. 139 7.3 Health Training Program Curriculum Issues .............................................................................................................. 140 7.4 Health Service-Delivery Cadre Issues .......................................................................................................................... 140 7.5 Health Service-Delivery Staffing (Demand) and Training (Supply) Scenarios: The Implications .............................................................................................................................................................. 140 7.6 Implementing the Human Resource Plan: Need for “Whole-of-Government” Approach ................................... 144 7.7 The Role of Development Partners ............................................................................................................................... 144 List of Figures Figure 3‑1: Real Per Capita Expenditure on Nursing Colleges as Percentage of 2009 Expenditure................................ 29 List of Tables Table 2‑1: Composition and Growth of the Public Sector Health Workforce 1988–2009................................................ 8 Table 2‑2: Total Number of Publicly Financed Health Facilities.......................................................................................... 9 Table 2‑3: Total Public Sector-Financed Health Employees Urban and Rural (2009).................................................... 10 Table 2‑4: Distribution of Publicly-Financed Health Staff by Province and Population (2009).................................... 12 Table 2‑5: Total Publicly Financed Service-Delivery Staff by Gender and Occupation.................................................. 13 Table 2‑6: Total Publicly Financed Service-Delivery Staff by Occupation and Age Group (2009)............................... 15 Table 2‑7: Total Publicly Financed Service-Delivery Staff by Occupation and Age Group (2009) (%)........................ 16 Table 3‑1: Summary of Key Preservice Training Enrollments and Graduates (2009).................................................... 31 Table 3‑2: Nurse Preservice Training Enrollments and Graduates (2009)....................................................................... 33 Table 3‑3: Community Health Worker Training Enrollments and Graduates (2009).................................................... 35 Table 3‑4: Postgraduate Training Program Graduates (2009)............................................................................................ 37 Table 4‑1: Recurrent Expenditures of All Schools of Nursing 2005–2009 (Kina)........................................................... 46 Table 4‑2: Total Expenditures on All SoNs by Expenditure Categories 2005–2009 (Kina)........................................... 47 Table 4‑3: Total Expenditures on All SoNs by Expenditure Categories 2005–2009 (%)................................................. 47 Table 4‑4: Total Recurrent Costs CHW Schools 2005–2009 (Kina).................................................................................. 48 Table 4‑5: Total Expenditures of All CHW Schools by Expenditure Categories 2005–2009 (Kina)............................. 49 Table 4‑6: Total Expenditures of All CHW Schools by Expenditure Categories 2005–2009 (Percent)........................ 49 Table 4‑7: Recurrent Unit Costs per Student for Schools of Nursing 2006–2009 (Kina)............................................... 50 Table 4‑8: Average Recurrent Unit Costs CHW Schools 2006–2009 (Kina).................................................................... 51 Table 4‑9: Total Full-Time Nursing School Teaching Staff by Sex (2009)......................................................................... 52 Table 4‑10: Total Full-Time Nursing School Teaching Staff by Age (2009)........................................................................ 53 Table 4‑11: Education Qualifications of Full-Time SoN Teaching Staff (2009)(Percent)................................................. 53 Contents   //  v 1305747 PNG HR Report 11-27-12.indb 5 11/27/12 11:38 AM Table 4‑12: Teaching Qualifications of Full-Time Nursing School Teaching Staff (2009) (Percent)............................... 54 Table 4‑13: Average Years of Experience of Full-Time Nursing School Teaching Staff (2009)........................................ 54 Table 4‑14: Total Full-Time CHW School Teaching Staff by Gender (2009)..................................................................... 55 Table 4‑15: Total Full-Time CHW School Teaching Staff by Age (2009)........................................................................... 55 Table 4‑16: Education Qualifications of Full-Time CHW Schools Teaching Staff (2009)(Percent)................................ 56 Table 4‑17: Teaching Qualification of Full-Time CHW Schools Teaching Staff (2009)(Percent)................................... 56 Table 4‑18: Average Years of Experience of Full-Time CHW School Teaching Staff (2009)............................................ 57 Table 4‑19: Key Indicators for SoNs (2009)............................................................................................................................ 58 Table 4‑20: Key Indicators for CHW Schools (2009)............................................................................................................ 58 Table 5‑1: Official Population Estimates and Projections 2000–2030 (Selected Years ‘000).......................................... 70 Table 5‑2: Potential Resource Scenarios for Public Expenditure on Health (Real 2009 Prices in ‘000 Kina).............. 72 Table 5‑3: Costing of the NHP 2010–2020 (Real Average per Year)(Millions of Kina 2010 Prices)............................. 74 Table 5‑4: Technical Options for Interventions to Deliver the MDGs by Level of Service in PNG.............................. 76 Table 5‑5: Relative Costs of Core Health Service-Delivery Cadres (2009)....................................................................... 78 Table 5‑6: Scenario 1: Public Sector Health Workforce Envisioned by PNGDSP 2010–2030....................................... 79 Table 5‑7: Scenario 1: Public Sector Health Workforce Annual Growth Rates Envisioned by PNGDSP 2010–2030............................................................................................................................................................... 80 Table 5‑8: Scenario 1: Costs of Public Sector Health Workforce Envisioned by PNGDSP 2009–2030........................ 80 Table 5‑9: Scenario 2: Public Sector Health Workforce Envisioned by PNGDSP 2010–2030....................................... 81 Table 5‑10: Scenario 2: Annual Public Sector Health Workforce Growth Rates Envisioned by PNGDSP 2010–2030.............................................................................................................................................. 81 Table 5‑11: Scenario 2: Costs of Public Sector Health Workforce Envisioned by PNGDSP 2009–2030........................ 82 Table 5‑12: Scenario 3: Health Sector Service-Delivery Workforce Maintaining Current (2009) Population-to-Staff Ratios 2009–2030................................................................................................................. 83 Table 5‑13: Scenario 3: Costs of Health Service-Delivery Staff Workforce When Maintaining Current (2009) Population Staff Ratios (2009–2030)....................................................................................................... 84 Table 5‑14: Scenario 4: Achieving WHO “Threshold” Health Service-Delivery Staff Density (2009–2030)................. 85 Table 5‑15: Scenario 4: Service-Delivery Staff Growth Rates Required to Achieve WHO “Threshold” Service-Delivery Staff Density 2009–2030 ......................................................................................................... 85 Table 5‑16: Scenario 4: Costs of Achieving WHO “Threshold” Health Service-Delivery Staff Density 2009–2030............................................................................................................................................................... 86 Table 5‑17: Scenario 5: Recommended Scenario for Direct Service-Delivery Health Staff 2010–2030......................... 87 Table 5‑18: Scenario 5: Direct Service-Delivery Health Workforce Growth Rates For Recommended Scenario 2010–2030............................................................................................................................................... 87 Table 5‑19: Scenario 5: Costs of Recommended Scenario for Direct Service-Delivery Health Staff 2010–2030............................................................................................................................................................... 88 Table 6‑1: Scenario 1: Medical Officers (MO): No Change in Supply Capacity (2010–2030)....................................... 94 Table 6‑2: Scenario 1: Nursing Officers: No Change in Supply Scenario (2010–2030)................................................... 95 Table 6‑3: Scenario 1: Community Health Workers (CHWs): No Change in Supply Capacity Scenario (2010–2030)............................................................................................................................................................ 97 Table 6‑4: Scenario 1: Health Extension Officers (HEOs): No Change in Supply Capacity Scenario (2010–2030)............................................................................................................................................................ 99 Table 6‑5: Scenario 1: Total Service-Delivery Staff: No Change in Supply Capacity Scenario (2010–2030)............. 100 Table 6‑6: Scenario 2: Medical Officers (MOs): Aspirational Targets Envisioned by PNGDSP (2010–2030)........... 103 Table 6‑7: Scenario 2: Nursing Officers (NOs): Aspirational Targets Envisioned by PNGDSP (2010–2030)............ 105 Table 6‑8: Scenario 2: Community Health Workers (CHWs): Aspirational Targets Envisioned by PNGDSP (2010–2030)......................................................................................................................................... 107 vi  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 6 11/27/12 11:38 AM Table 6‑9: Scenario 2: Total Service-Delivery Staff: Aspirational Targets Envisioned by PNGDSP (2010–2030)......................................................................................................................................... 109 Table 6‑10: Scenario 3: Medical Officers (MOs): Maintaining Existing Population to Staff Ratio (2010–2030)......... 111 Table 6‑11: Scenario 3: Nursing Officers (NOs): Maintaining Existing Population to Staff Ratios (2010–2030).......................................................................................................................................................... 112 Table 6‑12: Scenario 3: Community Health Workers (CHW): Maintaining Existing Population to Staff Ratios (2010–2030)................................................................................................................................. 113 Table 6‑13: Scenario 3: Health Extension Officers (HEOs): Maintaining Existing Population-to-Staff Ratios (2010–2030)............................................................................................................ 115 Table 6‑14: Scenario 3: Total Service-Delivery Staff: Maintaining Existing Staff to Population Ratios for all Cadres (2010–2030).................................................................................................. 117 Table 6‑15: Scenario 4: Medical Officers (MO): WHO “Threshold” Service Delivery Staff Density Target (2010–2030)................................................................................................................................ 119 Table 6‑16: Scenario 4: Nursing Officers (NOs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030).............................................................................................................................. 121 Table 6‑17: Scenario 4: Community Health Workers (CHWs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030)..................................................................................................................... 123 Table 6‑18: Scenario 4: Health Extension Officers (HEOs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030)..................................................................................................................... 124 Table 6‑19: Scenario 4: Total Service-Delivery Staff: WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030).............................................................................................................................. 126 Table 6‑20: Scenario 5: Medical Officers (MOs): A Suggested Preservice Training Scenario (2010–2030)................. 128 Table 6‑21: Scenario 5: Nursing Officers (NOs): A Suggested Preservice Training Scenario (2010–2030)................. 129 Table 6‑22: Scenario 5: Community Health Workers (CHWs): A Suggested Preservice Training Scenario (2010–2030).......................................................................................................................................... 131 Table 6‑23: Scenario 5: Health Extension Officers (HEOs): A Suggested Preservice Training Scenario (2010–2030).......................................................................................................................................... 134 Table 6‑24: Scenario 5: Total Service-Delivery Staff: A Suggested Preservice Training Scenario (2010–2030).......................................................................................................................................................... 135 Contents   //  vii 1305747 PNG HR Report 11-27-12.indb 7 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 8 11/27/12 11:38 AM FOREWORD This report comes at a very opportune time. It demonstrates clearly and systematically that Papua New Guinea faces a health work- force supply crisis. If we do not redress the supply and demand imbalances arising from: (a) the current severely constrained training system for new health workforce cadres; (b) the rapid aging of the existing workforce; and (c) the expanding demand for services over the next 10 to 20 years that arises from the sustained increase of the population; we will not be able to achieve our vision for improved health outcomes for our population articulated in the National Health Plan 2011–2020. In fact, without decisive action to immediately expand the number of nurses being trained, the PNG government financed public health sector may well have fewer nurses in 2020 than we have at present. The picture for other service delivery cadres is similar. This, challenge also emerges at a time when there is evidence the private health sector is expanding significantly and will add to the demand for health staff from our training institutions and also result in increased transfers of the workforce the private sector. The Health Plan forthrightly argues that our health system requires a “back to basics” approach to reform of our health system to arrest and reverse the nation’s deteriorating health indicators. We need to systematically strengthen our primary health care approach and ensure that those at the front line of health service delivery are equipped with the necessary facilities, supplies, equipment and training. We know our health sector workforce delivers services under trying conditions. Improving rural health service delivery fundamentally means that there needs to be an adequate supply of quality health human resources and that they need to be strategically and equitably placed across the country in health facilities that also have access to operational funding and medical supplies. Currently our workforce is inequitably distributed across provinces and districts. It will take a whole of government approach to achieve appropriate change. We need to find mechanisms to both increase our pro- vincially based workforce and to deploy it according to workload needs. This will require a concerted effort by all stakeholders. We need to reach consensus with training institutions—universities, government and church managed—on how best to expand sup- ply to meet identified needs, including reaching agreement on appropriate curricula. We need to improve the quality of our work- force through the re-establishment of in-service training programs—focussed initially on efforts to reduce our unacceptably high Maternal Mortality Rate. The Health Plan also acknowledges that it is imperative that we cultivate strong, cooperative, and innovative partnerships to assist us in meeting our health objectives. Nowhere is this more important than with respect to training of our workforce. We are com- mitted to strengthening our extremely important relationship with the Christian Health Services and with the University sector which undertakes most of our new health workforce training. ix 1305747 PNG HR Report 11-27-12.indb 9 11/27/12 11:38 AM The report canvasses a range of scenarios for the health workforce and documents the costs of each scenario. The recommended scenario is one which is both affordable given our nation’s likely development and fiscal path and is technically appropriate given our health needs. This is a well-timed call to arms. I wish to thank the World Bank for the report. Mr Pasco Kase Secretary of Health 1305747 PNG HR Report 11-27-12.indb 10 11/27/12 11:38 AM Acknowledgments This report was prepared by Mr. Ian Morris, Human base Manager and Researcher, reconstructed the Human Resource Economist for the World Bank under the direc- Resource Information Base of the Human Resource Divi- tion of Ms. Aparnaa Somanathan, Task Team Leader. Mr. sion of NDoH and analyzed the database for the tables Emmanuel Jimenez, Sector Director, Human Develop- on staffing presented in this report. Ms. Ellen Kulumbu ment and Mr. Juan Pablo Uribe, Sector Manager, Health, of the Port Moresby Office of the World Bank provided Nutrition and Population of the East Asia Region of the very important operational support and coordinated the World Bank provided overall management guidance for dialogue with the PNG Government. the report. The Bank team would like to thank the contribution Mr. Pius Kalambe, Consultant Researcher, under- made to the report by many individuals within the NDoH took a significant part of the work analyzing the Health and from the principals of the health-related training insti- Training Institution Survey. This survey was undertaken tutions who worked to design and complete the training jointly with the Human Resources Division of the National institution survey. Department of Health (NDoH) and the Secretariat for The report has benefited from the peer review pro- the PNG Universities Review 2010 (undertaken by Pro- cess within the Bank, including comments from Mr. Too- fessor Ross Garnaut and the Rt. Honourable Sir Rab- mas Palu and Mr. Tim Bulman, as well as those of Mr. bie Namaliu) which was established in the Commission Robert Christie of AusAID, and Mr. Jim Buchan, a con- of Higher Education. Ms. Zillar Miro, Consultant Data- sultant to AusAID. Cover photography by Gregg Maxwell/World Bank. xi 1305747 PNG HR Report 11-27-12.indb 11 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 12 11/27/12 11:38 AM Abbreviations and Acronyms ARI Acute Respiratory Infections HMTEF Health Medium-Term Expenditure AusAID Australian Agency for International Framework Development HRD Human Resource Development BAR Bougainville Autonomous Region HRIS Health Human Resource Information BoM Board of Management System CACC Central Agencies Coordinating IMF International Monetary Fund Committee IMR Infant Mortality Rate Central Central Province LIC Low-Income Country CHE Commission of Higher Education LLG Local Level Government Chimbu Chimbu Province Madang Madang Province CHS Church Health Services MBP Milne Bay Province CHW Community Health Workers M&E Monitoring and Evaluation CMC Church Medical Council MDGs Millennium Development Goals CMR Child Mortality Rate MHERST Ministry of Higher Education, Research, DHP District Health Post Science and Technology DHS Demographic and Household Survey MHS Minimum Health Standards 2002 DNP&M Department of National Planning and MIC Middle-Income Country Monitoring MMR Maternal Mortality Rate DoF Department of Finance MoF Ministry of Finance DoT Department of Treasury MoH Ministry of Health DP Development Partners Morobe Morobe Province DPM Department of Personnel Management MTDP Medium-Term Development Plan DWU Divine Word University 2011–2015 EHP Eastern Highlands Province MTEF Medium-Term Expenditure Framework ENBP East New Britain Province 2011–2015 (of Health) ESP East Sepik Province NCD National Capital District GDP Gross Domestic Product NC of PNG Nursing Council of Papua New Guinea GoPNG Government of Papua New Guinea NDoH National Department of Health Gulf Gulf Province NEFC National Economic and Fiscal HC Health Center Commission HEO Health Extension Officer NGO Nongovernmental Organization HIV/AIDS Human Immunodeficiency Virus/ NHAA National Health Administration Act Acquired Immune Deficiency Syndrome NHB National Health Board xiii 1305747 PNG HR Report 11-27-12.indb 13 11/27/12 11:38 AM NHCS National Headcount Survey SHP Southern Highlands Province NHEP National Higher Education Plan SoN School of Nursing NHP National Health Plan 2011–2020 STI Sexually Transmitted Infection NIP New Ireland Province STR Student Teacher Ratio NOL New Organic Law on Provincial and SWAP Sector Wide Approach Local Level Government, 1995 TB Tuberculosis OHE Office of Higher Education TBA Traditional Birth Attendant OP Oro Province TESAS Tertiary Education Study Assistance PAU Pacific Adventist University Scheme PHA Provincial Health Advisor TFR Total Fertility Rate PHC Primary Health Care UoG University of Goroka PMGH Port Moresby General Hospital UPNG University of Papua New Guinea PNG Papua New Guinea WB World Bank PNG DSP Papua New Guinea Development WHO World Health Organization Strategic Plan 2010–2030 WHP Western Highlands Province POM Port Moresby WNB West New Britain Province PSRMU Public Sector Reform Management Unit WP Western Province Sandaun Sandaun Province xiv  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 14 11/27/12 11:38 AM Executive Summary A P AR apua New Guinea’s health sector is facing a series of major challenges–including Aus an emerging workforce crisis–that must be dealt with if it hopes to deliver better health care. The sector’s shortcomings are manifesting themselves in a worrying BA BoM health picture. Over the past 35 years, there has been little improvement, and evidence CA from the past decade indicates extremely fragile outcomes. Rates of maternal and infant mortality, and traditional communicable diseases—which together account Cen CH for about 60 percent of the total disease burden—remain unacceptably high (Box 1). Ch Making matters worse is the emergence of new diseases, including the HIV and AIDS CH epidemics, and lifestyle-related diseases. CH CM CM The National Health Plan (NHP) 2011–20 sets out • structural changes in the sector’s governance, includ- DH the strategic directions for the development of PNG’s ing flawed provincial governance and financing DH health sector over the next decade. It recognizes that arrangements; DN these challenges have their roots in: Do Do Box 1: A Snapshot of Health Outcomes DP DP Preliminary results from the just-completed 2006 Demographic Health Survey reinforce the conclusion of a crisis in health outcomes and the DW hurdles PNG faces in achieving its national Millennium Development Goals (MDGs): EH • The national maternal mortality rate (MMR) is reported to have almost doubled since 2006 to 733 per 100,000 (UNICEF estimates that the average rate for developing countries is 450), with the infant mortality rate (IMR) at 57 per 1,000 live births. Pneumonia and diarrhea, EN together with underlying malnutrition, are the key causes of post-neonatal death in young children. ESP • The disease burden among adults is still dominated by infectious and vector-borne diseases, especially tuberculosis and malaria. GD • HIV is now well established as a generalized and accelerating heterosexually transmitted epidemic—one of the region’s most serious. The Go World Bank estimated in 2005/6 that HIV prevalence among sexually active adults exceeds 1 percent in rural areas, 2 percent in many urban/ enclave areas, and 3 percent in the capital, Port Moresby. Gu • Life expectancy at birth remains low at 57 years. HC • Only 32 percent of the rural population has access to safe water and 42 percent to sanitation. While the situation in urban areas is HE better—88 percent have access to safe water and 67 percent to sanitation facilities—antedotal evidence indicates that the situation in HIV urban squatter settlements is deteriorating and urban settlements are growing faster than official records indicate. xv 1305747 PNG HR Report 11-27-12.indb 15 11/27/12 11:38 AM • poor governance and administrative capacity across service-delivery health staff over the next two decades. the health system, including sound information sys- It also draws out the supply and demand gap implica- tems to facilitate decision making; and tions of the scenarios. • significantly declining real recurrent resources per This exercise is set within the context of important capita—including for health—since independence strategic work by the government on health. The NHP in 1975. Recurrent health outlays fell 9.4 percent in 2011–20 lays out the strategic directions for the develop- real terms from 1996–2004 (the latest year for which ment of the health sector over the next decade—framed there is a full statement of public health expenditures). within the Papua New Guinea Vision 2050, the Papua New Guinea Development Strategic Plan 2010–2030 (PNG Against this backdrop, the National Department DSP), the Medium-Term Development Plan 2011–2015 of Health (NDoH) has sponsored legislative changes (MTDP), and the NDoH’s Medium-Term Expenditure which were recently passed by Parliament that enable Framework 2011–2015 (MTEF). provincial governments to establish Provincial Health The report argues that the government’s response Authorities (PHA). PHAs are to be responsible for both needs to deal with: (i) the immediate supply-side cri- primary and secondary health care (hospitals) in the sis (quantity); (ii) the qualitative side, including preser- province. The NDoH has also initiated a major organi- vice and in-service training (especially for emergency zational restructure that will enable it to better provide obstetric care for existing staff); and (iii) incentives to technical support and guide priority provincial pro- ensure staff are able to be deployed where needed, par- grams rather than implement them; monitor and eval- ticularly in rural areas. To that end, its recommendations uate overall sector performance; and support efforts to focus on ameliorating the information problem, improv- ensure that human resources, logistics support, and infra- ing the training curriculum, tinkering with the compo- structure planning do not remain key constraints to ser- sition of the health-delivery staff to boost the number of vice delivery capacity. doctors and nurses, and finding a viable delivery staff sce- This report was commissioned because NDoH and nario to close the supply gap. development partners supporting the health sector One possible path forward is Scenario 5—the rec- increasingly recognize that the health sector is facing ommended scenario—which is not only affordable but an emerging health sector workforce crisis. The triggers also responds to the demand requirements for staff include: (i) an aging workforce; (ii) limited preservice from the health system while leaving space in the recur- training capacity to replenish the workforce; (iii) weak- rent health budget to boost quality. The bottom line is nesses in the curriculum of training programs supplying that drastic short- and long-term steps must be taken to new entrants to the direct service-delivery health work- remove health human resources as a major long-term force; and (iv)an almost total lack of systematic in-service constraint on the health sector’s capacity to deliver bet- training, especially for rural health. Moreover, data on the ter health services, both public and private, over the next health workforce is woefully inadequate for health human decade and more. resource planning and management purposes. In response, this report documents for the first time in over a decade the current stock, age, and gen- A Profile of the Health Sector der structure of the publicly financed health workforce, The Supply Side along with the capacity of the health-related train- ing institutions. It presents the results of an important Given the woefully inadequate data on the size, charac- 2009 PNG survey of health training institutions, which teristics, and deployment of the current publicly financed enable unit costs, staffing, and other aspects of the insti- workforce—and the need for better data to form a basis tutions to be analyzed, together with an assessment of for the new NHP 2010–2020—the Human Resources the quality of students and facilities by training school Division of NDoH undertook a special National Head- principals. It uses the data gathered to present a set of count Survey (NHCS) in 2009. The survey shows that the five alternative demand and supply scenarios for direct size of the health workforce financed by the public sec- xvi  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 16 11/27/12 11:38 AM Table 1: Composition and Growth of the Public Sector Health Workforce 1988–2009 Change Change Category 1988 1998 2004 2009 1998–2009 (%) 2004–2009 (%) Doctors and Dentists 384 316 524 570 80.4 8.8 Health Extension Officers 357 233 575 486 108.6 –15.5 Nurses 2,917 2,920 3,980 3,618 23.9 –9.1 Allied Health 283 372 440 318 –14.5 –27.7 Med Lab. Technical 159 150 254 258 72.0 1.6 Community Health Workers 4,982 3,926 5,358 4,419 12.6 –17.5 Other/Administration – 2,874 1,224 3,394 18.1 177.3 Total 9,082 10,791 12,355 13,063 21.1 5.7 Source: Data for 1988, 1998 and 2004 as presented in and documented in Chapter 2 of Strategic Directions for Human Development (World Bank, 2007) and 2009 from the National Head Count Survey 2009 (NDoH, 2009). tor has grown from 10,791 in 1998 to 13,063 in 2009— ing serious lack of information on the output of training an overall rise of 21.1 percent in the past 11 years, or a institutions. While PNG’s population has grown, spend- growth of 1.9 percent per annum (Table 1). Since 2004, ing on health training has fallen sharply. The question is however, the pace of growth has slowed down markedly. whether the capacity to train all health cadres has been Of this total workforce, direct service-delivery staff with reduced so much over the past 15 years or more that it is direct service delivery occupations make up 8,844, with 62 now producing newly qualified staff well below historic percent female and 38 percent male, although the share of attrition rates from the workforce. Hence, without dras- males in the rural areas rises to 47 percent. This group— tic action on the supply side—which will take at least the the focus of this report—includes doctors and dentists, best part of the next decade with concerted efforts begin- nurses, midwives, community health workers (CHWs), ning immediately—both short- and long-term human and health extension officers (HEOs). To date, the NHP resource supply gaps are to be expected. has not indicated that it believes a major adjustment needs The public sector finances the operation of some to be made in the structure of delivery cadres. 2,746 health facilities, of which 94 are urban and 2,652 As for age structure—vital for many reasons includ- are rural (Table 2). A striking fact is that in rural areas— ing experience “on the job” and planning for replacement which includes 80 percent of PNG’s population—most of staff owing to retirement—the survey validates wor- facilities are open aid posts, which offer simple cura- ries about an aging workforce. Key observations include: tive and preventive care. Moreover, there is an extremely (i) almost 16 percent of the service delivery workforce of uneven spread among provinces of not only facilities but 8,844 in 2009 (1,381) was aged 55 years or more; (ii) a fur- also health delivery staff. ther 37.7 percent (3,338) are currently in the 45–55 year age group and will reach retirement age over the next decade The Demand Side with a further one third (3,033) reaching retirement age in the subsequent decade; and (iii) only 12.3 percent of staff One important indicator of the effective demand for in 2009 (1,090) are less than 35 years of age (Table 2.6). health services is outpatient visits per capita per annum. On the training front, the survey shows that over This is an indicator of the overall use of the health sys- time, two important gaps have emerged: the diffusion tem given the state of the health system (funding, staff- of responsibility for training oversight and a continu- ing, pharmaceuticals, and other medical supplies) and the Executive Summary   //  xvii 1305747 PNG HR Report 11-27-12.indb 17 11/27/12 11:38 AM Table 2: Total Number of Publicly Financed Health Facilities Facility Type Government Mission Other Total Urban Hospitals 21 0 0 21 Urban Clinics 45 14 14 73 Total Urban 66 14 14 94 Rural Health Center 143 43 6 192 Health Subcenter 162 278 7 447 Rural Hospitals 4 6 2 12 District Hospitals 1 1 0 2 Clinics 1 0 0 1 Open Aid Posts 1,998 0 0 1,998 Total Rural 2,309 328 15 2,652 Total Facilities 2,375 342 29 2,746 Source: National Head Count Survey (NDoH, 2009). disease burden of the population. Available data leave lit- diture on service-delivery staff. This will help ensure tle doubt that ambulatory care visits per capita have been that demand for health services—as expressed by outpa- decreasing while health outcomes have been deteriating tient visits per capita per annum, including natal care and and that this trend, without a reversal, will further hurt immunizations—also increases. Given the demand for ser- health outcomes. The number of outpatient visits per per- vices as documented in the 2009 Monash Report, exist- son per annum declined from 1.54 in 1999 to 1.37 in 2008, ing staff numbers are some 40 percent over the required and outpatient visits per capita per annum in rural areas numbers—at least in rural areas. There is, therefore, con- on average for 2007–8 are only 0.88. siderable scope to increase rural services with existing Another demand factor is resources with the gov- staff. On the other hand, existing service demand is well ernment estimating that resources for health will rise below what should be demanded if the health system were significantly over the next five years as well as in the responding to the population’s disease burden. longer-term to 2030—in part because of planned LNG projects. For this report’s supply-demand scenarios, it is Key Supply Gaps assumed that the health budget will grow at about 5 per- cent in real terms per annum, about twice the growth in Armed with this supply and demand information, per capita GDP by 2030. The NHP indicates that popula- the report runs five scenarios to draw out the impli- tion will increase at about 2.8 percent per annum over the cations for each health cadre and for all service deliv- period 2010 to 2020, then fall to about 2.5 percent over ery staff, including affordability. The five scenarios the period 2020 to 2025 and to 2.1 percent between 2025 vary from no change in existing supply capacity to four and 2030. These projections are probably ambitious, how- alternative scenarios with supply adjusted to meet the ever, unless decisive action is taken on the family planning postulated demand. These scenarios are summarized in front and on education, particularly of girls. Table 3 below together with the 2009 baseline and then The reality is that the final effective demand for discussed in more detail. health workers will depend in no small part on the health system’s efforts to increase the quality enhanc- Scenario 1: No change in human resource supply capacity. ing items of the nonsalary budget, which needs to rise This scenario highlights the implications of a “Do Noth- faster than total expenditure on health and of expen- ing” strategy on the supply side from 2010 to 2030—that xviii  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 18 11/27/12 11:38 AM Table 3: Summary of the Five Demand Scenarios Scenario 4 Scenario 5 (WHO (Preservice Threshold Training for Scenario 1 Scenario 2 Scenario 3 Service Service Delivery Baseline (No Change in (Aspirational (Maintain 2009 Delivery Staff Workforce (2009) Supply) PNGDSP) Pop/Staff Ratio) Ratios) -Recommended) Category Staff Numbers 2030 Doctors 379 656 4,184 647 6,231 1,535 Nurses 3,252 2,869 19,526 5,551 8,497 8,012 CHWs 4,398 3,537 18,795 7,507 10,310 8,256 HEOs 411 607 200 702 702 604 Total 8,440 7,669 42,705 14,407 25,739 18,406 Population to Service Delivery Staff Ratio 2030* Doctors 6,637:1 17,277:1 2,707:1 17,512:1 1,818:1 7,380:1 Nurses 17,512:1 3,949:1 580:1 2,041:1 1,333:1 1,414:1 CHWs 2,041:1 3,203:1 603:1 1,509:1 1,099:1 1,372:1 HEOs 1,509:1 18,663:1 56,645:1 16,138:1 16,138:1 18,756:1 Total Staff 16,148:1 1,477:1 265:1 786:1 440:1 616:1 Staff Costs 2030 (Millions of Kina at 2009 Prices) Doctors 34 59 377 58 561 138 Nurses 74 65 443 126 193 182 CHWs 73 59 312 125 171 137 HEOs 11 16 5 19 19 16 Total 192 199 1,137 328 944 473 Costs Kina Mn 2009 Prices Expected 513.0 1,460.5 1,460.5 1,460.5 1,460.5 1,460.5 Recurrent Budget Service Staff 37.3 13.6 77.7 22.4 64.5 32.3 Costs as % of Budget Nurse & CHW 5.7 6.5 96.4 17.5 40.5 28.6 Training Costs Quality-Enhanced 8.5 9.7 149.1 26.9 70.3 43.5 Training Costs Training Costs as 1.1 0.4 6.6 1.2 2.8 2.0 % of Recurrent Budget Quality-Enhanced 1.6 0.7 10.2 1.8 4.8 3.0 Training Costs as % of Recurrent Budget Note: * Population ratios based on high population estimates. Executive Summary   //  xix 1305747 PNG HR Report 11-27-12.indb 19 11/27/12 11:38 AM is, there is no change in the current preservice training The share of the health budget allocated to staff would capacity for doctors, nurses, CHWs and HEOs. increase from 37.3 percent to 77.7 percent by 2030, an The scenario shows an impending crisis that will unsustainably high share. The recurrent costs of train- result in a fast shrinking service-delivery workforce. ing nurses and CHWs would increase from K5.7 mil- The total number of direct service-delivery staff will fall lion in 2009 to K96.4 million by 2030. The real costs of a from 8,440 in 2009 to 7,669 in 2030, and the population “quality-enhanced” training package would increase from to staff ratio more than double from 786 to one staff to K8.3 million in 2009 to almost K150 million in 2030. This 16,418 to one staff over the same period. Most significantly, would represent 8.5 percent of the budget in 2030 for cur- there would be a large decline in CHWs and nurses—the rent-level quality training and around 10 percent with the backbone of rural service delivery. Although the number “quality- enhanced” training package. of doctors and HEOs would expand slightly, the popula- tion per doctor and HEO would decline. Total staff costs Scenario 2 demonstrates that the extremely ambitious aspi- would only increase slightly from K191.3 million now to rational targets envisioned by PNGDSP 2010–2030 are not K198.3 million in 2030, while the share of the health bud- only unaffordable but fail to adequately reflect the likely needed composition of cadres in the future health get allocated to staff would decline sharply from 37.3 per- workforce. cent to 13.6 percent over the period. Scenario 1 underscores the crisis facing the human resource Scenario 3: Maintaining existing population to service- requirements of PNG’s health sector over the next two delivery ratios. This scenario is driven by population decades if there is no change in human resource supply growth. It assumes that the core direct service-deliv- capacity. There will be a major staff supply crisis, a major ery health cadres maintain their current share of the decline in staff relative to the population, and a huge decline workforce and the current (2009) population-to-staff in CHWs and nurses, the backbone of rural service ratios over the period 2009–30 which would be in keep- delivery. ing with the NHP’s thinking that no major adjustment needs to be made in the structure of delivery cadres. Scenario 2: Aspirational targets envisioned by PNGDSP The scenario suggests that additional demand for ser- 2010–2030.This scenario is driven by the PNGDSP’s pro- vices can be achieved by using the existing staff more posed plan for sharply expanding human resources for efficiently and letting the workforce grow at the rate of health and achieving ambitious health outcome targets. population growth. It envisages the total service delivery staff increasing from Under this scenario, the number of doctors would a base of 8,440 in 2009 to 42,705 by 2030, an increase of increase from 379 in 2009 to 647 in 2030, sustaining a over 400 percent. The population to service delivery staff population to doctor ratio of 17,511 to one. The number ratio would improve from 786 per staff in 2009 to only of nurses would rise from 3,252 in 2009 to 5,551 in 2030, 265 per staff by 2030—far below what even the WHO pro- sustaining a population to nurse ratio of 2,041 to one. The poses for a country at PNG’s epidemiological stage. The number of CHWs would grow from 4,398 in 2009 to 7,507 plan calls for an increase in: in 2030, sustaining a population to CHW ratio of 1,509 • doctors from 379 in 2009 to 4,184 by 2030. to one. The number of HEOs would increase from 411 • nurses from 3,252 in 2009 to 19,526 in 2030. in 2009 to 702 in 2030, sustaining a population to HEO • CHWs from 4,398 in 2009 to 18,795 in 2030. ratio of 16,148 to one. As for maintaining the population to service-delivery staff ratio of 786 to one, total service This scenario is not affordable given the expected delivery staff numbers would need to increase from 8,440 growth of the economy and the health budget. Specif- in 2009 to 14,407 in 2030. ically, staff remuneration, assuming all staff are financed This scenario is probably affordable but unlikely to by government, would increase from K191.3 million in result in the right mix of cadres required for the health 2009 to K1,135 million in 2030—a real increase of about workforce. The real remuneration costs would grow 500 percent over 21 years (or over 20 percent per year). from K191.3 million in 2009 to K326.5 million in 2030– xx  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 20 11/27/12 11:38 AM an increase of 2.8 percent per annum (the estimated rate • The number of HEOs would grow at the same rate as of population increase). This would be more than afford- the population because they represent a small propor- able given the expected recurrent health budget growth tion of the total (411 in 2009 to 702 in 2030). of 5 percent per annum in real terms. In this scenario the share of staff costs in the health budget would decline from This scenario, however, is not affordable given 37.3 percent in 2009 to 22.4 percent by 2030. expected economic growth and the health budget. To begin with, staff remuneration would increase from K191.3 Scenario 3 shows that it is affordable to maintain the exist- million in 2009 to K942.2 million in 2030—a real 390 per- ing population to service-delivery ratios but probably would cent budgetary increase over 21 years (or about 18 percent not result in the right mix of cadres required for the health per year). As a percentage of the recurrent health bud- workforce. get, staff remuneration would increase from 37.3 percent in 2009 to 64.5 percent by 2030, an unsustainably high Scenario 4: WHO recommended “threshold” service- share. The recurrent costs of training nurses and CHWs delivery staff-density targets. This scenario is driven by would jump from K5.7 million in 2009 to K40.5 million the WHO “threshold” density of 2.28 per 1,000 popu- by 2030. The real costs of a “quality-enhanced” training lation (or population-to-staff ratio of 439 to 1) of doc- package would rise from K8.5 million in 2009 to about tors, nurses (registered and enrolled), and midwives, K70 million in 2030. below which, according to WHO, coverage of essen- tial interventions—including those necessary to reach Scenario 4 is not affordable, particularly in the outer years, the health-related Millennium Development Goals and it recommends a doctor-to-population ratio that is (MDGs)—is not likely. The breakdown would be a doc- probably not feasible from a supply constraint per tor density of 0.55 doctors per 1,000 and a nondoctor spective and is lower than is needed to meet the popula- tion’s health needs—raising questions about cost- staff density of 1.73. effectiveness. This would mean an overall increase in total staff from 8,440 in 2009 to almost 26,000 in 2030—an increase of over 200 percent over 21 years. This is a significant Scenario 5: A suggested preservice training scenario for expansion of staff but one that is substantially lower than direct service-delivery staff. This scenario—the broadly the almost 43,000 proposed in Scenario 2. Similarly, this recommended one—envisages a new mix of direct ser- scenario projects a population-to-staff ratio by 2030 of 440 vice-delivery staff. It is driven by: (i) the growth in the to one, a big improvement from the current 786 to one, resource envelope likely to be available for health and ser- but not as much as in Scenario 2 (265 to one). By cadre, vice-delivery staff; and (ii) the feasibility and speed with Scenario 4 projects that: which preservice training can be ramped up to meet the • The number of doctors would rise from 379 in 2009 demands of workforce attrition and the needs of a grow- to 6,231 by 2030, with the population-to-doctor ratio ing population. improving from the current 17,512 to one to 1,818 to Specifically, it calls for a reasonable expansion of one by 2030. The proportion of doctors in the direct the number of doctors (to be targeted for rural facilities) service-delivery workforce would rise from 4.5 per- and an expansion of general nursing graduates relative cent in 2009 to about 25 percent in 2030. to CHWs. It also assumes that the existing capacity for • The number of nurses would increase from 3,252 in producing HEOs is sustained given their value as a vital 2009 to 8,497 in 2030, with the population-to-nurse management and supervisory cadre, especially for rural ratio improving from 2,041 per nurse in 2009 to 1,333 health. Underlying this scenario is a firm suggestion that per nurse in 2030. there needs to be a significant expansion of recurrent (and • The number of CHWs would grow from 4,398 in capital) resources to support: (i) the expansion of pretrain- 2009 to 10,310 in 2030, with the population-to- ing and in-service training; (ii) additional staff for sup- CHW ratio improving from 1,509 to one in 2009 to port services; and (iii) more quality-enhancing nonsalary 1,099 by 2030. budget expenditures. Executive Summary   //  xxi 1305747 PNG HR Report 11-27-12.indb 21 11/27/12 11:38 AM In Scenario 5, staff numbers would rise from 8,440 Scenario 5 is affordable, responds to the demand require- in 2009 to 18,406 in 2030—an increase of 118 per- ments for staff from the health system, and leaves space cent. This would sustain an improvement in the popula- for recurrent health resources to be allocated to a signifi- tion to direct service-delivery staff ratio from 786 to one cant expansion of training (preservice and in-service)— in 2009 to 616 to one in 2030. By cadre the mix would while also leaving space for increased allocations to both change as follows: support staff and quality- enhancing nonsalary budgets. • The number of doctors would rise from 379 in 2009 to 1,535 in 2030. The population-to-doctor ratio would improve from 17,512 per doctor in 2009 to Recommendations and Options 7,380 per doctor in 2030. By 2030 doctors would represent 8.4 percent of the workforce, up from 4.5 This report identifies a number of issues that require percent in 2009. decisive action by the government. They fall into five • The number of nurses would increase from 3,252 in groups: (i) information deficits; (ii) training curriculum; 2009 to 8,012 by 2030. This will enable the popula- (iii) service-delivery cadres; (iv) service-delivery staffing; tion-to-nurse ratio to improve from around 2,041 to and (v) the role of development partners. one nurse in 2009 to 1,414 to one by 2030. Nurses would represent 44 percent of the workforce in 2030, Information Deficits up from 38.5 percent in 2009. • The number of CHWs would grow from 4,398 in Issue: A serious information constraint on the health 2009 to 8,256 in 2030. The population-to-CHW ratio workforce and its trends completely undermines NDoH would improve from around 1,500 to one CHW in capacity to monitor the health workforce. 2009 to about 1,372 per CHW in 2030. CHWs would Recommendation 1. The NDoH should create a man- represent about 45 percent of the workforce in 2030, agement committee with appropriate support from quali- slightly down from 52 percent in 2009. fied technical NDoH staff to review human resource data requirements for management and planning purposes Scenario 5 is affordable. The cost of employing all and to decide how best to rationalize current data system staff would increase from K191.3 million in 2009, or about arrangements (within NDoH’s control). 37 percent of the total health recurrent budget, to K472 million in 2030, or about 32 percent. Recommendation 2. The NDoH should immediately re- Space would be left for training costs and other establish—and make operational—the Health Profession- quality-enhancing efforts to improve health care deliv- als Database(s). This is critical for further insight into the ery. The recurrent costs of training nurses and CHWs scale and operational trends of the private sector. would increase from K5.7 million in 2009 to almost K29 Issue: There is highly inadequate information available million by 2030. The real costs of a “quality-enhanced” on the capacity and operations of health-related train- training package would rise from K8.5 million in 2009 ing institutions. to almost K43.5 million in 2030. This would represent 2 percent of the heath recurrent budget in 2030 for cur- Recommendation 3. The NDoH and the Office of Higher rent level quality training of nurses and CHWs and 3 Education (OHE) should form a joint ad hoc management percent with the “quality-enhanced” nurse and CHW committee to determine how best to generate the key infor- training package. The costs of employing this number of mation required on health training program enrollment doctors in real terms would increase from K34 million policies (including all universities), institution through- in 2009 to K138 million in 2030. The costs of employ- puts by program, and costs. This needs to be agreed at a ing nurses would rise in real terms from about K74 mil- high level (perhaps with an “all of government approach”) lion in 2009 to K182 million in 2030 while the costs of and involve the mission training facilities (currently mis- employing CHWs would rise from K73 million in 2009 sion nurse training is under partial OHE’s oversight and to K137 million by 2030. CHW training is under NDoH oversight). xxii  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 22 11/27/12 11:38 AM Training Curriculum the “emergency response” to human resources for the health sector. As discussed, “no change” on the training Issue: A range of critical curriculum issues must be supply side is not a feasible option. A specific set of deci- dealt with, including (i) the appropriate level of train- sions need to be made on: (i) the mix of cadres; (ii) the ing on birth complications and emergency obstetric future of HEOs (and the use of specialty nurses or nurse care; (ii) how and whether to monitor and ensure that practitioners); (iii) the relative balance of pre- and in-ser- nurses and CHWs are taught the full agreed curricu- vice training of nurses and CHWs; and (iv) how to deal lum on family planning, STIs, and HIV prevention; with emergency obstetric care knowledge in the health (iii) how to boost universities’ accountability to NDoH workforce. and other major employers to meet nationally set cur- riculum standards; and (iv) defining in-service train- Service-delivery Staffing ing program requirements (including special modules Issue: One important implication of the analyses (in on emergency obstetric care for nurses and CHWs). all five scenarios) is that there will be a serious drop Recommendation 4. The NDoH Executive should weigh in the number of nurses and CHWs over the period how to deal with the “all of government” issues that arise of the current NHP and extending into the next NHP because the evolving institutional structures for health period because the attrition rate of the health work- worker training have not been kept synchronized. It should force will exceed the numbers being trained and enter- do this in consultation with OHE, universities, represen- ing the workforce. tatives of other health-related training institutions, and Recommendation 6. NDoH management should annually the central government agencies (including the Treasury match supply and demand or build scenarios for human and its Budgets Division and the Department of National resource development—from the top down and the bot- Planning and Monitoring (DNP&M)). tom up. This could be linked to the MTEF process and the process of expanding and redeveloping the health system Service-delivery Cadres province by province. Issue: The report suggests: (i) the expansion of the Recommendation 7. The PNG Government should estab- health workforce with modest changes in the compo- lish a Whole-of-Government Taskforce—comprising at sition of the workforce mix by cadres (Scenario 5)— least NDoH, church health agencies, OHE, relevant univer- more doctors and nurses relative to CHWs; (ii) further sities, DNP&M, Budgets Division of Treasury, Economic work on nondirect service-delivery staff and postgrad- Policy Division of Treasury, National Department of Edu- uate training requirements; (iii) further consideration cation, Prime Minister’s Department, and the Department of how best to deal (in staff training terms) with the of Personnel Management. The taskforce should imme- major health problems (including emergency obstet- diately review options to: ric care); and (iv) a review of hospital needs—espe- • develop a costed plan to expand training capacity as cially the role of CHWs in hospitals, given that they agreed that should also explore short–term options were trained for rural areas, and whether a new cadre to expand supply capacity with training institutions; in hospitals is needed. • explore options and incentives to encourage exist- ing staff retention through incentives to reduce early Recommendation 5. NDoH management needs to retirement and postpone retirement; make decisions on the issues raised in this report follow- • explore options to encourage redeploying staff to rural ing extensive consultation with key stakeholders. These areas and deploying new graduates to rural areas, par- include: (i) the immediate steps needed to respond to the ticularly those with staff shortages; and impending crisis in direct service-delivery staff (such as • significantly refurbish existing training facilities. expanding general nurse and CHW preservice training); and (ii) which broad scenario discussed in this report (or This taskforce should also be responsible for address- another arising from further dialogue) should underpin ing the set of “Whole-of-Government” implementation Executive Summary   //  xxiii 1305747 PNG HR Report 11-27-12.indb 23 11/27/12 11:38 AM issues that arises because NDoH lacks responsibility for ernment, and church agencies. The World Bank could their implementation. give further presentations to: (i) the National Executive of NDoH; (ii) core related agencies (including DNP&M, Role of Development Partners Treasury, and OHE); (iii) a special meeting of church agen- cies; and (iv) development partners accompanied by senior Issue: Development partners have recognized the need NDoH staff who will be responsible for driving the imple- for this study and have been concerned for some time mentation of the report’s recommendations. This could about the impending crisis in human resources for be followed by a one to two day conference to discuss the health based on the fragmentary evidence that was pre- results, leading to the development of a strategic plan in viously available. response to the issues identified in the report. Recommendation 8. This report should be widely dis- cussed and disseminated throughout NDoH, the gov- xxiv  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 24 11/27/12 11:38 AM CHAPTER 1 INTRODUCTION AND BACKGROUND 1.1. Introduction fronting any viable strategy designed to implement the objectives it has set for the health sector. This report helps This report documents for the first time in over a decade document the nature of this emerging crisis and makes the current stock of the publicly financed health work- specific recommendations on the way forward. force in PNG and their characteristics and deployment across the country by type of health facility and health cadre. It also documents the capacity of the health-related 1.2. Health System Structural Changes: training institutions and presents the results of an impor- Emergence of an Almost Unmanagable tant survey of health training institutions which enable System unit costs, staffing and other aspects of the institutions to be analyzed together with a qualitative assessment of By the time of independence (in 1975), a formal gov- the quality of students and of facilities by training school ernment-funded health system provided basic primary principals. The report presents a set of five demand sce- health care in most parts of the country through front- narios and draws out the implications for the health train- line workers with minimal amounts of training and lim- ing system and of the health budget for these scenarios. ited basic drugs. Various Christian missions supplemented The report is set within the context of important government efforts through government-subsidized health work done by the government in the recent past on services. The system was managed by a centralized health health. The National Department of Health (NDoH) has department which managed the whole system, including produced a new National Health Plan 2011–2020 (NHP) hospitals, and delegated powers to regional, district, and which sets out the strategic directions for the develop- line staff and facilities. Since independence, however, there ment of the health sector over the next decade. The NHP have been successive attempts to decentralize the provi- has been framed within: (i) a new, more strategic, 40–50 sion of services to provincial and district governments and year framework launched by the Government of Papua to allow provinces (and local-level governments) to have New Guinea (GoPNG)–Papua New Guinea Vision 2050 enhanced control over health sector resources. (sponsored by the Department of National Planning and In 1995 the New Organic Law on Provincial and Monitoring (DNP&M) and the subsequent Papua New Local Level Government (NOL) devolved primary Guinea Development Strategic Plan 2010–2030 (PNGDSP) health care services to provincial level. The provinces which maps out “how to get PNG to where our Papua New are responsible for managing primary health care services Guinea Vision 2050 wants us to be”; (ii) the Medium-Term (with local-level governments and communities responsi- Development Plan 2011–2015 (MTDP) which sets out the ble for maintaining health facilities) managed by a provin- specific inter-sectoral targets to be achieved over the next cial health advisor reporting to a provincial administrator five years; and (iii) the NDoH’s Medium-Term Expendi- as the chief accountability officer for all health staff and ture Framework 2011–2015 (MTEF). primary health services. For budget and personnel man- The NHP recognizes that the emerging crisis in agement purposes, each province deals directly with the the health human resources area is a critical issue con- central agencies responsible for budget, as well as personnel 1 1305747 PNG HR Report 11-27-12.indb 1 11/27/12 11:38 AM management and planning–independently of the NDoH. (ii) monitor and evaluate overall sector performance; and The NDoH has retained responsibility for: (i) policy over- (iii) support efforts to ensure human resources, logistics sight, albeit with very limited capacity for enforcement; support and infrastructure planning do not continue as (ii) hospitals, which are now managed by autonomous key constraints to service-delivery capacity. boards able to make budget and human resource deci- sions including management of their own payrolls inde- pendently of the NDoH; (iii) pharmaceutical purchases; 1.3. Health Outcomes Remain a Serious and (iv) coordination of most external resources to the Challenge health sector, as agreed with the central agencies and indi- vidual development partners. Health outcomes have stalled over the last quarter cen- The 19 provinces and 20 hospitals and a largely tury and have even declined in the decade to 2000, with publicly funded church health system operate in a very maternal (MMR) and infant mortality rates (IMR) and diffuse and largely uncoordinated manner. This system communicable diseases remaining unacceptably high. accounts for about one-half of all ambulatory care–a not The high maternal and infant death rates and communi- insignificant share of inpatient care days. For the purposes cable diseases dominate the burden of disease–accounting of this study it meant that the workforce is scattered across for about 60 percent of the total disease burden. Women are the 40–50 distinct entities with their own independent pay- particularly disadvantaged, as evidenced by poor maternal rolls and Human Resource (HR) Management Informa- health and lack of access to family planning. By 2000, infant tion Systems (HRMIS) with the consequence that NDoH and maternal mortality had declined to 64 per 1,000 live has no ability to centrally monitor HR trends and effec- births (72 in 1980) and 370 per 100,000 live births (400 in tively fulfill its HR planning and oversight roles. 1980) respectively. In the Highlands region, however, the The health system employed about 13,000 staff in MMR had increased to 625 per 100,000 around 2000. Total 2009 and its infrastructure primarily comprises 19 pro- fertility remained high at 4.6 children per woman (albeit vincial hospitals, 73 urban clinics, 192 health centers, down from 5.4 in 1980) and contraceptive prevalence rates and 447 health subcenters. There were about 2,000 health remained low at 26 percent. Preliminary results from the posts–but many more were previously operational. One just-completed 2006 Demographic Health Survey reinforce study showed that at least 300 health posts closed between the crisis in health outcomes and the real challenges PNG 1995 and 2000 mainly affecting those in lower-asset quin- faces in achieving its national MDG goals. The national tiles and those living in remote areas. NDoH cannot verify MMR is reported to have almost doubled since 2006 to be how many aid posts are now operating with any surety. Fur- 733 per 100,000 while there have been modest improve- ther, over the past decade or more many staff have retreated ments in the aggregate IMR since 2000 which is reported from peripheral health facilities and work at more centrally to have fallen to 57 per 1,000 live births.2 located facilities even though they are formally recorded Pneumonia and diarrhea, together with under- as working at the more peripheral facilities. lying malnutrition, are the most important causes In response to these identified problems, and to redress some of the key issues arising from the NOL, 1 This refers to the Provincial Health Authority Act (2007), the NDoH has recently initiated legislative and orga- which enables streamlining of provincial health services, bring- nizational changes. The legislative changes, which have ing together the provincial health departments, hospitals and dis- now been passed by parliament,1 enable provincial gov- trict health services under one management board thus unifying the structure of the previously fragmented public health services ernments to establish Provincial Health Authorities (PHA) at the subnational level. Critically, this legislation enables prov- to be responsible for both primary and secondary health inces to decide to opt in to this framework–it is not compulsory. care (hospitals) in the province. These are currently split One province, Eastern Highlands, has formally agreed to opt in between national and provincial governments. NDoH has to this framework. A number of others are seriously consider- ing it.The national government cannot make this compulsory also initiated a major organizational restructure which will without an ”Organic law” or constitutional change to the pow- enable it to better: (i) provide technical support and guide ers held by the national government. priority provincial programs rather than implement them; 2  Disaggregated data by region is not yet available. 2  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 2 11/27/12 11:38 AM of post-neonatal death in young children. The recent fell 77 percent. At the same time expenditures on salaries DHS reports a slight improvement in neonatal deaths increased 10 percent in real terms. Significantly, at least compared to the 1996 DHS results with 29 per 1,000 300 aid posts were closed between 1995 and 2000, antena- compared to 32 in 1996 and in post neonatal deaths and tal coverage declined from 80 percent in 1991 to 58 per- improvement from 38 per 1,000 in 2006 to 28 in 2006. cent in 2004 and there were frequent shortages of basic Among adults, the disease burden is still dominated by drugs in most rural and urban health facilities. infectious and vector-borne diseases, especially tubercu- losis and malaria respectively. This situation is seriously compounded by HIV which is now well established in 1.5. The Emerging Human Resource the form of a generalized and accelerating heterosexu- Crisis: The Reason for the Study ally transmitted epidemic–one of the most serious in the region. Bank estimates in 2005/06 indicate HIV prev- The last few national health conferences and NDoH alence among sexually active adults exceeds 1 percent management have clearly recognized that health human in rural areas, 2 percent in many urban/enclave areas resources are becoming the major long-term constraint and 3 percent in the capital, Port Moresby. More recent on the capacity of the health sector–public and private– consensus workshop estimates suggest HIV incidence to deliver improved health services over the next decade may have fallen. The World Bank, in cooperation with and more. Partial evidence showed the health workforce government and other development partners is under- was aging quickly–particularly those cadres responsible for taking a national HIV Bio-behavioral survey–a crucial delivering front-line health services to the rural population step–to better understand the scale and drivers of the and urban poor (Community Health Workers or CHWs, epidemic in PNG in parallel with this proposed piece nurses and midwives)—and large swathes of the peripheral of sector work. health services have collapsed. At the same time, there was concern that the capacity to train all cadres of health staff had also been significantly reduced over the past 15 years 1.4. Government Health Expenditures or more to the point that it is now producing new qualified Substantial But Inefficient and staff well below historic attrition rates from the workforce. Fragmented There was concern that attrition rates could be increasing significantly because of the age of the workforce. Without In 2004 public health expenditures represented 11.9 drastic short- and longer-term action on the supply side– percent of total public expenditures and 3.8 percent of which will take at least the best part of the next decade with GDP representing a substantial commitment to health. concerted efforts beginning immediately—both short- and Nevertheless, the poor health status of the population has long-term human resource supply gaps are to be expected. also been accompanied by a decline in health system per- The report shows that these concerns were well founded. formance with decreased coverage and quality of services While previous health conferences as long as a despite a substantial increase in public spending on the decade ago–including the Mount Hagen Health Con- health sector of 35 percent in real terms between 1996 and ference in 2002–recognized that health human resource 2004 (the latest year for which there is a full statement of constraints were looming, other structural challenges public expenditures on health).3 facing the health system have also loomed large and Over the period 1996 to 2004 there were very sig- received more attention. These include problems with nificant changes in the composition of health expen- decentralization and establishment of coherent health ditures which have threatened the quality of health expenditures. Recurrent health expenditures declined 3  The fragmentation of health authorities as noted above also by 9.4 percent in real terms while development expen- means that there is no central point for consolidation of either ditures–mostly donor financed–increased 110 percent. government-financed or donor-financed health expenditures. No effort has been made since 2004 by either government or Government expenditures on goods and services fell 27 development partners to sustain the efforts of the initial health percent in real terms while expenditures on capital items SWAP team to generate this data. INTRODUCTION AND BACKGROUND   //  3 1305747 PNG HR Report 11-27-12.indb 3 11/27/12 11:38 AM systems at the provincial level. On the human resource distribution compared to population. Finally, details are side NDoH also recognized that there were problems provided on the distribution between hospitals and rural with the numbers of people on the various and dispersed service delivery of the key service-delivery occupations by health payrolls and that many were not legitimate. Con- sex and age group. The chapter concludes with a summary siderable efforts over the past years have been devoted to of the numbers and proportions of each service-delivery installing a new payroll system and trying to confirm all occupation currently at retirement age and those that will health employees on the payroll were legitimate. In 2005 reach retirement age over the next decade. it was estimated that perhaps K20 million could be saved Chapter 3 outlines the current capacity of the health by cleansing the payroll of inappropriate and ghost work- workforce-related training system in PNG. It partic- ers. Significant efforts have been made by NDoH man- ularly focuses on the current size of the training effort agement to complete these reforms. for the major service-delivery occupations as outlined Efforts to validate the existing stock of qualified in Chapter 2–doctors, HEOs, nurses and midwives and health workers actually working within the health sys- CHWs. It starts with an outline of the current institutional tem has met with only partial success. The process of arrangements for the training of the health-related work- validating the age, qualification, experience and other force, including the formal responsibilities and configu- characteristics of the existing health workforce on the ration of the current health-related training system. The payroll system has all but collapsed. This meant that the chapter also notes how responsibilities have evolved and original plan for the study to use the payroll system was significant gaps emerged in the coherence of the training not possible. The base line headcount for staff was estab- system over the past two decades. In this context it notes lished through a special survey undertaken by NDoH for the emergence of two important gaps in the current insti- the NHP. This study used this data together with other tutional relationships–the diffusion of responsibility for data sets to estimate the characteristics of the workforce the oversight of the training and a continuing serious lack as outlined in Chapter 2 Annex 1. of information on the need for training institutions out- Thus there was an emerging consensus and recog- puts. The chapter concludes with a summary documenta- nition of an evolving HR crisis and a belief that: (i) HR– tion of the health-related postgraduate training program rather than resources–was emerging as a major binding enrollments and graduates for 2009. constraint on achieving health outcome objectives; and Chapter 4 reviews in considerable detail the char- (ii) relieving the HR crisis needed to become a major acteristics of two categories of health training institu- theme of both the NHP and the health MTEF. tions–the Schools of Nursing (SoNs) and Community Health Worker Training Schools.4 These institutions are responsible for supplying the core of the health-delivery 1.6. Outline of the Report system in PNG. To this end Chapter 4 documents and dis- cusses: (i) expenditures of the training schools; (ii) staff- Chapter 2 documents the size of the publicly financed student ratios and the unit costs of students in each of health workforce for the latest year for which data is the nurse and CHW training institutions; and (iii) char- available (2009) and trends in the total size of the health acteristics of nurse and CHW trainers–qualifications and workforce since 1988. As background to a discussion of teacher-training qualification, as well as the age, sex and the deployment of publicly financed health staff, the cur- years of experience of the teaching workforce in each of rent number and location of health facilities by type and the two types of training schools. Chapter 4 also reports management arrangement is documented. The compo- the details of a qualitative and quantitative assessment of sition of the total publicly financed health workforce is then analyzed by rural-urban (overwhelmingly hospital) 4  The information reported in this chapter is derived from the deployment, occupation and whether in service deliv- Health Training Facilities Survey of 2009. This Survey was con- ducted jointly with the Secretariat of the Garnaut–Namaliu ery, service-delivery support or administration. Subse- Review of Higher Education. By and large the universities did quent analysis focuses on the distribution of health staff not respond to the survey—as a consequence only data on the by province and the provincial equity of provincial staff SoNs and CHW training schools can be reported. 4  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 4 11/27/12 11:38 AM the conditions under which these training schools oper- jections discussed in Chapter 5) in population to staff and ate (as reported by principals). This includes the quality staff per 1,000 population ratios; (ii) the expected attrition of teaching support and teaching processes and the qual- from the workforce; (iii) the expected outputs from the ity of buildings and equipment. training schools in the initial years of the NHP (prior to Chapter 5 explores factors which may affect the implementing any expansion plans and allowing for new demand for health service-delivery staff over the com- trainees to graduate) as well as how fast graduations from ing two decades. It notes that there is an expectation training schools will need to ramp upwards to the employ- that more resources will be available to health than in ment targets implied by each of the scenarios; (iv) the the immediate past, however, it is still too early to be sure first year intakes to respective training schools required what the full revenue implications to government of the to ensure–given expected dropout rates for each school– LNG project and other developments will be. Neverthe- the needed graduates to reach the employment targets for less, as discussed in Chapter 5, the report does assume the specific scenario; and (v) the costs in 2009 prices of economic growth rates will be positive and significantly employing all the graduates. The recurrent costs of training above those recorded in the past decade. The NHP projects both nurses and CHWs based on unit costs of 2009 and that the population will increase at about 2.8 percent per “quality-enhanced” unit costs of training are also shown. annum over the next decade while the PNGDSP expects Each set of scenario tables is summarized in a sum- annual population growth to fall to about 2.5 percent over mary table looking at the total service-delivery staff. the period 2020 to 2025 and to 2.1 percent between 2025 This includes: (i) staff to be employed; (ii) population to and 2030. As discussed in more detail in Chapter 5, this service-delivery staff ratios and total service-delivery staff report has used the NHP population projections for the per 1,000 population ratios; and critically (iii) it shows period through 2020 and has subsequently assumed for the direct service-delivery staff salary costs of each sce- the period 2020–2030 the population growth rate will nario, the expected total recurrent budget (as discussed begin a modest decline. in Chapter 5) and the share of the total service-delivery The NHP proposes to turn the current decline in staff costs of the expected total recurrent budget. Finally, outpatient services per capita in rural areas around–fol- the implications for policy on the expansion of direct ser- lowing a substantial decline in recent years. This will be vice-delivery staff training programs are discussed for each achieved by a variety of actions but, strategically, the pro- scenario together with the total recurrent costs of train- portion of the budget available for service-delivery staff ing of nurses and CHWs. will decline while that for nonsalary quality-enhancing Chapter 7 presents the core recommendations of expenditures will increase–albeit of an expanding budget the report. These relate to: (i) future priorities for data in real terms per capita. Chapter 5 summarizes five sce- system development for health human resources; (ii) insti- narios–one of which assumes that there is no change in tutionalization of the documentation of the existing sup- existing supply capacity for direct health service-delivery ply capacity for health human resources; (iii) curriculum staff and demonstrates the human resource crisis facing issues for core health service-delivery staff (including for the health sector and four alternative demand scenar- emergency obstetric care); (iv) summarizes the core train- ios with supply adjusted to meet the postulated demand. ing system expansion recommendations arising from the Chapter 6 presents the detailed demand and sup- recommended scenario for the future development of ply projections for the five scenarios described in Chap- health service-delivery staffing; (v) the establishment of ter 5 and, within each scenario, the implications for a “Whole-of-Government” Taskforce to manage imple- each cadre. In addition to showing the detailed (year- mentation of the key agreed recommendations from the by-year) growth of the direct service-delivery staff (by report; and (vi) an extensive consultation process on the cadre) implied for each scenario, the detailed tables also key results of the report in order to establish a national show: (i) the expected trend (given the population pro- consensus on the way forward. INTRODUCTION AND BACKGROUND   //  5 1305747 PNG HR Report 11-27-12.indb 5 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 6 11/27/12 11:38 AM CHAPTER 2 Characteristics of the Current Health Public Sector Workforce 2.1. Introduction data systems which purport or aim to provide insights on the size and distribution of the workforce—both public This chapter documents the size of the publicly financed and private. For a range of reasons these are only partial health workforce for the latest year for which data is and often duplicative. The problems with the existing data available–2009—and trends in the total size of the health sources are discussed in Annex 2.1 to Chapter 2. workforce since 1988. As background to a discussion of In recognition of this problem the Human Resources the deployment of publicly financed health staff, the cur- Division of NDoH undertook a special National Head- rent number and location of health facilities by type and count Survey (NHCS) in 2009 to establish a reasonable management arrangement is documented. The compo- estimate of the current size of the publicly financed sition of the total publicly financed health workforce is health workforce in order to form a basis for the new then analyzed by the rural-urban deployment, occupa- National Health Plan 2010–2020. This is the best sin- tion and whether they are in service delivery, service- gle estimate of the total size of the current health work- delivery support or administration. Subsequent analysis force employed by government and missions and thus of focuses on the distribution of health staff by province and the health workforce financed by the public budget. This the equity of the distribution of provincial staff compared survey collected information on only the major occupa- to population. Finally, details are provided on the distri- tions and sex of health staff. Other information on the bution between hospitals (overwhelmingly urban) and workforce had to be derived from other, sometimes par- rural service delivery of the key service-delivery occupa- tial databases. The method by which characteristics of the tions by sex and age group. The chapter concludes with health workforce, particularly age, have been estimated is a summary of the numbers and proportions of each ser- discussed in Annex 2.1 to Chapter 2. vice-delivery occupation currently at retirement age and In the last quarter of 2009 it is estimated by the those who will reach retirement age over the next decade. NHCS that there were a total of 12,608 health staff employed—although this survey did not cover the staff employed by NDoH in Port Moresby. It is estimated 2.2. Size and Deployment of the Health by the Human Resource Information System (HRIS) of Workforce the NDoH that there were 455 staff employed by NDoH in 2009. Thus total staff financed by government in late Data on the size and characteristics of the publicly 2009 is estimated at 13,063 (Table 2-1). The overall size financed health workforce are woefully inadequate of the health workforce financed by the public sector has while data on the workforce in the private sector is grown from 10,791 in 1998 to 13,063 in 2009—an overall almost nonexistent. Nevertheless, there are a number of increase of 21.1 percent in the last 11 years or a growth 7 1305747 PNG HR Report 11-27-12.indb 7 11/27/12 11:38 AM Table 2‑1: Composition and Growth of the Public Sector Health Workforce 1988–2009 Change Change Category 1988 1998 2004 2009 1998–2009 (%) 2004–2009 (%) Doctors and Dentists 384 316 524 570 80.4 8.8 HEOs 357 233 575 486 108.6 – 15.3 Nurses 2,917 2,920 3,980 3,618 23.9 – 9.1 Allied Health 283 372 440 318 – 14.5 – 27.7 Med Lab. Technical 159 150 254 258 72.0 1.6 CHWs 4,982 3,926 5,358 4,419 12.6 – 17.5 Other/Administration – 2,874 1,224 3,394 18.1 177.3 Total 9,082 10,791 12,355 13,063 21.1 5.7 Source: Data for 1988, 1998 and 2004 as presented in and documented in Chapter 2 of Strategic Directions for Human Development (World Bank, 2007) and 2009 from the NHCS 2009 (NDoH, 2009). of 1.9 percent per annum. During this time the popu- the number of nurses by 1,000 and CHWs by 1,700 due lation is estimated to have increased at about 2.8 per- mainly to significant ghost workers on the payroll and cent per annum. Between 2004 and 2009, the workforce inadequate payroll cleansing. Nevertheless, the 2004 data increased by only 5.7 percent—from 12,355 in 2004 to does show the staff numbers the health sector was paying 13,063 in 2009. This represents an increase of 1.1 percent for in 2004. It is estimated that the number of doctors was per annum and is somewhat slower than the estimated approximately correct.5 If these numbers are correct then rate of population increase and significantly lower than the number of nurses may well have been steady over the that recorded over the period 1998–2004 of 14.5 percent period 1988 to 2004 but have increased over 20 percent or about 2.4 percent per annum. in the last five years. The picture is similar for CHWs— It is interesting to look at the trends in the num- the numbers will have been steady over the period 1988 bers for specific occupations within the health sector. to 2009 but have shown some increase over the past five The number of doctors (and dentists) has increased sig- years. Data for Other/Administration in 2004 were also nificantly over the past 11 years (80 percent) and five years problematic and probably underestimated—thus the (9 percent) respectively. The number of HEOs apparently apparent increase 2004 to 2009 is probably overestimated. increased quite fast over the last 11 years (109 percent) Health outcomes and the cost effectiveness of those but the numbers appear to have declined over the past five outcomes are strongly influenced by the size, composi- years. The number of nurses increased 24 percent over the tion and deployment of the health staff. The deployment last 11 years but recorded numbers over the last five years of staff is strongly influenced by the number and location indicate that there has been a 5 percent decrease in num- of facilities. Table 2-2 summarizes the rural-urban loca- bers. A similar story is apparent for CHWs—over the last tion, type of facilities and who manages facilities financed 11 years numbers increased by nearly 13 percent and in by the public sector in PNG. The public sector finances the last five years declined 17 percent. The data from the the operation of some 2,746 health facilities. NHCS indicate that the numbers allocated to Other/Admin- The most striking feature of the structure of facil- istration have increased dramatically (177 percent) over ities is the overall importance of Aid Posts—notwith- the past five years and 18 percent over the past 11 years. standing the closure of many over the past 10–15 years As noted above, and discussed more fully Annex 2.1 to Chapter 2, data on staff in the health sector is far 5  This is discussed extensively in Chapter 2 and Annex 2A of from robust. The data for 2004 sourced from the national Strategic Directions for Human Development in Papua New payroll has major constraints and probably overestimated Guinea (World Bank, 2007). 8  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 8 11/27/12 11:38 AM Table 2‑2: Total Number of Publicly Financed Health Facilities Facility Type Government Mission Other Total Urban Hospitals 21 0 0 21 Urban Clinics 45 14 14 73 Total Urban 66 14 14 94 Rural Health Center 143 43 6 192 Health Subcenter 162 278 7 447 Rural Hospitals 4 6 2 12 District Hospitals 1 1 0 2 Clinics 1 0 0 1 Open Aid Posts 1,998 0 0 1,998 Total Rural 2,309 328 15 2,652 Total Facilities 2,375 342 29 2,746 Source: National Headcount Survey (NDoH, 2009). or more. There are almost 2,000 Aid Posts (73 percent sector-owned facilities whose operating costs are substan- of all facilities)—they are government owned and man- tially financed by public funding. Mission facilities over- aged and are essentially one-person facilities—which all account for over 45 percent of ambulatory care visits are rural based and offer simple curative and preventive and government facilities account for 55 percent of visits care and support. Staffing in Aid Postaare almost exclu- (see Annex 2.11; Table 2-1). sively CHWs and are, notionally at least, supported by The current health workforce, its composition and either government health centers or health subcenters. distribution rural-urban is summarized in Table 2-3. Key Aid Posts are designed to serve small population groups observations about the current deployment of staff include: of around 1,000 to 1,500. There are 94 facilities located in • Medical officers and HEOs each constitute around urban areas, including 66 owned and managed by govern- 5 percent of the service-delivery workforce, nurses ment. This includes the 19 provincial hospitals, the Port (including midwives) constitute about 40 percent of Moresby General Hospital and one specialist (psychiat- the workforce and almost 50 percent of the work- ric) hospital. The government owns 45 urban clinics while force comprises CHWs. missions run 14 urban clinics. • Overall, 52 percent of staff (6,801) are engaged in urban The core rural health services, which offer a range areas (including NDoH, hospitals and urban clinics) of services above the basic services provided by Aid and 48 percent (6,262) are deployed to rural areas. Posts, are essentially provided from 639 health cen- • Direct service-delivery staff comprise 69 percent ters and health subcenters which comprise 98 percent (8,954) of total staff and an additional 28 percent of health facilities in rural areas—excluding Aid Posts. are involved in service-delivery support or admin- These facilities are almost evenly split between government istration. and mission management—although it is noteworthy that • Staff providing technical support for service delivery missions run proportionately smaller health subcenters (allied and ancillary health workers) total 520 (4 per- (86 percent of the total for mission centers and subcen- cent of the total workforce) or almost 6 percent of the ters) than health centers while the split between govern- total involved in direct service delivery. ment health centers and subcenters is 47 percent to 53 • Administrative support, including provincial health percent. Other, includes government owned or private offices, totals 3,134 and accounts for 24 percent of the Characteristics of the Current Health Public Sector Workforce   //  9 1305747 PNG HR Report 11-27-12.indb 9 11/27/12 11:38 AM Table 2‑3: Total Public Sector-Financed Health Employees Urban and Rural (2009) Staff Category Urban Rural Total % Urban % Rural 1. Service Delivery Medical Officers 332 51 383 86.7 13.3 HEOs 175 285 460 38.0 62.0 Nursing Officers 1,807 1,472 3,279 55.1 44.9 Midwives 101 192 293 34.5 65.5 CHWs 1,412 3,006 4,418 32.0 68.0 Dentists/Dental Therapists 79 42 121 65.3 34.7 Subtotal Service Delivery * 3,906 5,048 8,954 43.6 56.4 2. Service Delivery Support Medical Laboratory Assistants & Technicians 117 65 182 64.3 35.7 X-Ray Technicians 53 7 60 88.3 11.7 Pharmacists 71 21 92 77.2 22.8 Environmental Health Officers 69 74 143 48.3 51.7 Training Coordinators 33 10 43 76.7 23.3 Subtotal Service Delivery Support 343 177 520 66.0 34.0 3. Administration Administration Support 696 182 878 79.3 20.7 Other Support Staff 419 260 679 61.7 38.3 Cleaners 129 180 309 41.7 58.3 Drivers 121 216 337 35.9 64.1 Casuals 732 199 931 78.6 21.4 Subtotal Administration 2,097 1,037 3,134 66.9 33.1 4. Nat. Dept of Health HQ Medical Officers 62 0 62 100 0 HEOs 26 0 26 100 0 Nursing Officers 41 0 41 100 0 Midwives 5 0 5 100 0 CHWs 1 0 1 100 0 Dental Therapists 4 0 4 100 0 Subtotal NDoH 139 0 139 100 0 HQ Support Service Occupations Med. Lab Assistants & Technicians 13 0 13 100 0 X-Ray Technicians 3 0 3 100 0 Pharmacists 16 0 16 100 0 Environmental Health Officers 24 0 24 100 0 Subtotal HQ Support Service Occupations 56 0 56 100 0 Other Occupations Administration Support 210 0 210 100 0 Other Support Staff 50 0 50 100 0 Subtotal HQ Other Occupations 260 0 260 100 0 Total NDoH Staff 455 0 455 100 0 Total Staff 6,801 6,262 13,063 52.1 47.9 Sources: National Headcount Survey. NDoH staff totals from Human Resource Information System (HRIS), NDoH. Note: * Service delivery staff in this table includes staff with these occupations working in administration at provincial level. 10  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 10 11/27/12 11:38 AM total workforce. In addition the NDoH headquarters There were no substantive explicit mechanisms embodied accounts for an additional 455 staff (3.5 percent of in the organic law on provincial government to equalize the the total workforce). allocation of facilities and/or staffing between provinces. • Only 56 percent of service-delivery staff are deployed The considerable inequalities that existed in 1976 in rural areas (5,048) which accounts for well over 80 have not been systematically addressed since then. A few percent of the population, while 44 percent (3,906) mechanisms through national government budget fund- are deployed in urban areas. ing have attempted to address some of the most egregious • Almost 50 percent of service-delivery staff (4,418) are anomalies with limited success—mainly because they were CHWs who are trained to deliver basic services in never institutionalized including through the National rural-based health posts. Almost one-third of CHWs Public Expenditure Plans of the late 1970s and 1980s. The are deployed in urban areas (mainly hospitals). revision to the Organic Law on Provincial and Local level • Excluding CHWs, there are more service-delivery Government of 1996 did not address inequalities in cap- health staff employed in urban areas (2,494 or 55 per- ital stock between provinces. Simultaneously, foreign aid cent) than deployed in rural areas (2,042 or 45 percent). allocations were increasingly project-based and brought • At an occupational level it is striking that 87 percent off-budget, making it very difficult to address issues as of doctors (332) in service delivery are in urban areas between provinces and between long run recurrent and and only 13 percent (51) are servicing rural popula- capital costs. As a consequence, large differences in the tions. In addition, there are more doctors in the NDoH distribution of staff (and facilities) remain (Table 2-4). headquarters (62) in nonservice delivery roles than There are currently 12,292 publicly financed staff are serving in all rural areas. delivering health services (excluding administrative staff • With respect to nursing officers a majority (1,807 or in provincial headquarters and in the NDoH). The fol- 55 percent) are deployed in urban areas while 1,472 lowing observations are made on the distribution of this (45 percent) are deployed in rural areas. staff as documented (Table 2-4): • Midwives, critical to turning around the crisis in • Government accounts for 71 percent of staff maternal mortality, comprise a fraction over 8 per- (8,732) while missions (including government facil- cent (293) of the total nursing stock. These numbers ities whose management is taken over by missions are clearly small but it is noteworthy that two-thirds and/or private institutions) account for just over one- of midwives are deployed in rural areas and there are quarter of staff (3,560). only five in NDoH headquarters. • The proportion of government staff by province var- • There are 486 HEOs–95 percent (460) of whom are ies very significantly – from a high of 96 percent in involved in direct service delivery. However, over one- Bougainville and 90 percent in the National Capital third (175) are deployed in urban areas rather than District (NCD) to a low of 38.4 percent in East Sepik in rural areas where health centers and health sub- Province and 47.5 percent in Enga province. Con- centers that they are meant to manage are located. versely, the proportion managed by mission is very important in the many provinces (including the lat- Other important dimensions on the distribution ter) and less important in others. Just over half the of health staff include that by province and population. provinces, excluding NCD, have a lower proportion Staff were historically distributed by province largely on of government staff (a higher proportion of mission the basis of the history of the development of the health staff) than the average share. sector—whether financed by government or missions. • There is, on average, one health staff member funded Just after independence in 1976 provincial governments by government for each 561 of the population. How- were established. This froze the recurrent budget in real ever, there are also very significant variations in the terms for provincial functions (including rural health). population-staff ratio between provinces. Exclud- Provinces were made responsible for the creation of new ing the National Capital District and Central Prov- facilities but capital budgets (for maintenance and capi- ince which directly serve overlapping populations, tal works) were also frozen at the 1976 level in real terms. the population per staff ratio varies from a low 270:1 Characteristics of the Current Health Public Sector Workforce   //  11 1305747 PNG HR Report 11-27-12.indb 11 11/27/12 11:38 AM in Manus province and 329:1 in New Ireland prov- provide accessible services. However, the large vari- ince to highs of 964:1, 896:1, 890:1 for Morobe, East ations in population: staff ratios suggest these are not Sepik and Southern Highlands provinces respectively. planned variations. Significantly, four provinces (five • The final column of Table 2-4 indicates the percentage including NCD) would require a 20 percent or more increase or decrease in staff in each province required reduction in staff numbers (Manus 51.9 percent; New to allocate existing staff according to population, i.e., Ireland 41.4 percent; Milne Bay 39 percent; and East assuming an allocation of staff to achieve 561 people per New Britain 24.0 percent) and five provinces would staff member. Some 55 percent of provinces would need require an increase in staff of more than 20 percent to have a reduction in staff numbers while the other (Central 100.5 percent; Morobe 71.9 percent; East 45 percent would have increased staff. In some cases Sepik 59.6 percent; Southern Highlands 58.7 per- the increases or reduction would be very significant. cent; and Madang 33.6 percent). • It is recognized that population is not the only crite- rion which should determine the allocation of staff This report, while recognizing the importance of because the geography (for example mountainous looking at the whole health workforce, is primarily look- and maritime provinces; deficits in road and trans- ing at the structure and features of the service-delivery port infrastructure and its maintenance; and popu- component of the workforce. It will, however, make ref- lation dispersion) will influence the need for staff to erences to the total workforce when detailed breakdowns Table 2‑4: Distribution of Publicly-Financed Health Staff by Province and Population (2009) Gov’t Non-Gov’t Total Gov’t Est. Pop Pop Per Staff if Average % Above or Province Staff Staff Staff Staff % 2009 Health Staff Pop. Per Staff (Below) Average Western 231 176 407 56.8 205,332 505 366 (10.0) Gulf 178 117 295 60.3 157,498 534 281 (4.6) Central 109 97 206 52.9 231,795 1,125 413 100.5 NCD 1,530 171 1,701 90.0 320,206 188 570 (66.5) Milne Bay 529 259 788 67.1 269,779 342 481 (39.0) Oro 276 45 321 86.0 176,677 550 315 (2.0) S. Highlands 621 364 985 63.0 876,938 890 1,563 58.7 Enga 288 319 607 47.5 362,033 596 645 6.2 W. Highlands 599 403 1,002 59.8 559,257 558 996 (0.5) Chimbu 626 92 718 87.2 353,949 493 631 (12.1) E. Highlands 636 198 834 76.3 505,248 606 901 8.0 Morobe 572 83 655 87.3 631,412 964 1,126 71.9 Madang 589 86 675 87.3 506,323 750 902 33.6 East Sepik 192 308 500 38.4 447,773 896 798 59.6 Sandaun 250 220 470 53.2 238,147 507 425 (9.6) Manus 168 38 206 81.6 55,628 270 99 (51.9) New Ireland 246 229 475 51.8 155,775 329 277 (41.4) E. New Britain 390 212 602 64.8 256,197 426 457 (24.0) W. New Britain 370 131 501 73.9 251,466 502 448 (10.6) Bougainville 332 14 346 96.0 196,572 569 350 (1.4) Total 8,732 3,562 12,294 71.0 6,758,583 561 n.a. n.a. Source: Annex Table 2.1 in Annex 2.11. 12  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 12 11/27/12 11:38 AM of the workforce are available. As noted above, the service- areas. The overwhelming majority are working in hos- delivery component of the workforce comprises 8,954 or pitals, while 5,048 (57 percent) work in rural areas. 69 percent of the total public sector workforce. • Almost two-thirds (5,444) of all service-delivery Table 2-5 presents information on the service-deliv- staff are female and 38 percent (3,400) are male. In ery component (direct service-delivery occupations only rural areas, however, males represent 48 percent of and excluding service-delivery staff involved in provin- the workforce compared to only 27 percent in urban cial administration) of the health workforce by hospi- areas and, conversely, females represent 74 percent of tals/urban and rural areas, occupation and gender. Key the workforce in urban areas/hospitals and 53 per- observations about the structure of the workforce from cent in rural areas. these perspectives include: • The service-delivery staff are less concentrated in urban There are significant variations in the gender com- areas as compared to the total workforce although a position and location of the various service-delivery still-significant 3,796 (43 percent) of staff are in urban occupations across the public health sector: Table 2‑5: Total Publicly Financed Service-Delivery Staff by Gender and Occupation1 Category/Function Male Female Total Male (%) Female (%) Hospitals / Urban2 Medical Officers 246 82 328 75.0 25.0 HEOs 67 59 126 53.2 46.8 Nursing Officers 265 1,515 1,780 14.9 85.1 Midwives 12 80 92 13.0 87.0 CHWs 363 1,029 1,392 26.1 73.9 Dentists/Dental Therapists 49 29 78 62.8 37.2 Subtotal – Hospitals 1,002 2,794 3,796 26.40 73.6 Rural Health 3 Medical Officers 40 11 51 78.4 21.6 HEOs 185 100 285 64.9 35.1 Nursing Officers 468 1,004 1,472 31.8 68.2 Midwives 54 138 192 28.1 71.9 CHWs 1,620 1,386 3,006 53.9 46.1 Dental Therapists 31 11 42 73.8 26.2 Subtotal – Rural Health 2,398 2,650 5,048 47.5 52.5 Total Service-Delivery Staff Medical Officers 286 93 379 75.5 24.5 HEOs 252 159 411 61.3 38.7 Nursing Officers 733 2,519 3,252 22.5 77.5 Midwives 66 218 284 23.2 76.8 CHWs 1,983 2,415 4,398 45.1 54.9 Dental Therapists 80 40 120 66.7 33.3 TOTAL 3,400 5,444 8,844 38.4 61.6 Source: National Headcount Survey 2009 (NDoH).    Includes all staff in direct service-delivery provision with these designations. Notes: 1 2   Includes Urban Clinics but excludes provincial administrative staff and NDoH HQ staff with these occupations. 3   Includes administrative staff at district and lower levels estimated at about 415 (or almost 7 percent) of the total rural staff of 6,262 (see Table 2-3). Characteristics of the Current Health Public Sector Workforce   //  13 1305747 PNG HR Report 11-27-12.indb 13 11/27/12 11:38 AM • The vast majority of the 379 doctors in direct service • Significantly, a further 37.7 percent (3,338) are cur- delivery work in hospitals (87 percent) and thus only rently in the 45–54 year age group and will reach 51 (13 percent) are deployed in rural areas. About retirement age over the next decade. A further one- three-quarters of doctors are male and one-quarter third (3,033) will reach retirement age in the subse- female with little difference in the gender composi- quent decade. Thus the need for replacement of staff tion of the doctor workforce between rural health in service delivery will provide a significant demand services and hospitals. for training institutions over the next 20 years. • Two-thirds of HEOs are to be found in rural areas • Only 12.3 percent of staff in 2009 (1,090) are less than and, surprisingly, 126 (one-third) are working in 35 years of age—less than the 16 percent (1,381) aged hospitals—almost two-thirds of HEOs are male, the 55 years or more; confirming the oft-recited state- sex balance in hospitals is 47 percent female and 53 ment that the health workforce is aging. percent male. • Over half (1,872) of nurses (including midwives) work It is important to look at the proportions of each in hospitals/urban areas while 47 percent (1,664) are service-delivery occupation by age grouping to better deployed in rural services. Midwives (total 284) com- understand the breakdown of the aggregate age struc- prise 3 percent of the service-delivery workforce and ture of the workforce. This is also important for plan- 8 percent of the total nurse workforce. Interestingly, ning training requirements and the potential shortfalls two-thirds (192) of midwives are deployed in rural or surpluses in staff given existing supply capacity—see areas—significantly more than for the nursing group Chapters 3, 5 and 6. Table 2-7 shows the proportion of as a whole. each service-delivery occupation by hospitals/urban areas • A significant majority of CHWs (68 percent) are and rural service delivery and for the whole workforce. deployed in rural service delivery, and yet, as noted The following key observations can be made by occu- above, a significant proportion—almost one-third— pational group: are deployed in hospitals/urban areas. The majority Medical Officers of CHWs are female (55 percent) but it is noteworthy • Of the total medical officer workforce (including den- that the vast majority of CHWs in hospitals/urban tists) of 379 just over 10 percent (39) were 55 years areas (almost three-quarters) are female. or more (that is at, or approaching, retirement age). • Dental staff constitute 120 (less than 1.5 percent), of An additional 36.7 percent (139) are aged 45–54 the total service-delivery staff—nearly two-thirds are years of age and will thus reach retirement age over in hospitals/urban areas and one-third are in rural the next decade. areas. Two-thirds of dental staff are male while in rural • Over 40 percent of medical officers (158) are aged areas this proportion rises to nearly three-quarters. 35–44 years of age. • Only 11.3 percent of the medical officers (43) are The age structure of service-delivery staff, including aged 25–34 years of age-only marginally more than of the principal occupations involved in service deliv- the cohort currently at retirement age. ery is important for many reasons including experience • A much higher proportion of medical officers work- “on-the-job” and, most importantly, for planning for ing in rural service delivery are in the older age replacement of staff due to retirement. Table 2-6 pres- groupings compared to the average for all medical ents data on the age structure of each occupation involved officers—almost 14 percent of rural medical officers in direct service delivery by Hospitals/Urban and Rural are at retirement age and a further 60 percent (31) can services. Key observations include: expect to retire over the next decade. • Almost 16 percent of the workforce in 2009 (1,381) was aged 55 years or more and thus due to retire shortly— Health Extension Officers given the age eligibility for formal retirement begins • Of the total number of HEOs in direct service at 55 years of age—with a slightly higher proportion delivery (411), 12.7 percent (52) are aged 55 years (17 percent) in this age group working in hospitals. or more and thus have reached or are approach- 14  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 14 11/27/12 11:38 AM Table 2‑6: Total Publicly Financed Service-Delivery Staff by Occupation and Age Group (2009)1 Category/Function <24 25–34 35–44 45–54 55–64 65+ Total Hospitals/Urban2 Medical Officers 0 40 148 108 27 5 328 HEOs 2 34 54 29 7 0 126 Nursing Officers 5 215 646 588 310 15 1,780 Midwives 0 1 18 38 34 1 92 CHWs 0 100 479 585 216 12 1,392 Dental/Dental Therapists 0 8 27 24 17 1 78 Subtotal Hospitals 7 398 1,372 1,372 611 34 3,796 Percent Total Hospitals 0.2 10.5 36.1 36.1 16.1 0.9 1004 Rural Health3 Medical Officers 0 3 10 31 7 0 51 HEOs 1 22 115 101 40 5 285 Nursing Officers 13 256 476 518 202 7 1,472 Midwives 0 10 55 85 40 2 192 CHWs 13 366 996 1,207 341 83 3,006 Dental Therapists 0 1 9 24 7 2 42 Subtotal Rural Health 27 658 1,661 1,966 637 99 5,048 % Total Rural Health 0.5 13.0 32.9 39.0 12.6 2.0 100 Total Service-Delivery Staff Medical Officers 0 43 158 139 34 5 379 HEOs 3 56 169 130 47 5 411 Nursing Officers 18 471 1,122 1,106 512 22 3,252 Midwives 0 11 73 123 74 3 284 CHWs 13 466 1,475 1,792 557 95 4,398 Dental Therapists 0 9 36 48 24 3 120 TOTAL 342 1,056 3,033 3,338 1,248 133 8,844 Percent of Total 0.4 11.9 34.3 37.7 14.1 1.5 100 Source: National Headcount Survey 2009 (NDoH).    Includes all staff in direct service-delivery provision with these designations. Notes: 1 2   Includes Urban Clinics, excludes provincial administrative staff and NDoH HQ staff with these occupations. 3   Includes direct service-delivery staff working as administrative staff at district and lower levels (relatively small). 4   Errors due to rounding arising from estimation proess see Annex 2. ing retirement age. Another 31.7 percent (130) are • It is interesting to note that a higher proportion of aged 45–54 years and can be expected to retire over the younger cohorts of HEOs are working in hospi- the next decade. tals/urban areas. • Another 41.2 percent (169) are aged 35–44 years of age. • Less than 15 percent of HEOs (59) are aged less than Nursing Officers 35 years—indicating the numbers in this younger • Of the total number of nursing officers (excluding cohort working in service delivery are significantly midwives) of 3,252, over 16 percent (534) working less than for the subsequent two 10 years age cohorts. in direct service delivery are aged 55 years or older Characteristics of the Current Health Public Sector Workforce   //  15 1305747 PNG HR Report 11-27-12.indb 15 11/27/12 11:38 AM and almost one-third more (1,106) are aged 45–54 Midwives years of age and are thus scheduled to retire over • Midwives constitute an important health service occu- the next decade. pation and are presented in this analysis as a separate • Another 35 percent (1,122) are aged 35–44 years of age. occupation—albeit one that requires a nursing qual- • Some 15 percent of nurses (489) are aged less than 35 ification prior to training as a midwife. There are 284 years—about the same proportion of the workforce midwives in service delivery of whom 77 (over 27 per- scheduled to retire imminently. cent) are aged 55 years or more and can be expected • Nursing officers, on average, are younger in rural to retire shortly. Significantly, 38 percent of midwives areas, particularly among the younger age group of working in hospitals are of retirement age compared 25–34 years. to about 22 percent in rural areas. Table 2‑7: Total Publicly Financed Service-Delivery Staff by Occupation and Age Group (2009) (%) Category/Function <24 25–34 35–44 45–54 55–64 65+ Total Hospitals/Urban Medical Officers 0 12.2 45.1 32.9 8.2 1.5 100 HEOs 1.6 27.0 42.9 23.0 5.6 0 100 Nursing Officers 0.3 12.1 36.3 33.0 17.4 0.8 100 Midwives 0 1.1 19.6 41.3 36.9 1.1 100 CHWs 0 7.2 34.4 42.0 15.5 0.9 100 Dentists/Dental Therapists 0 10.4 35.1 31.2 22.1 1.3 100 Subtotal Hospitals 0.2 10.5 36.1 36.1 16.1 0.9 100 Rural Health Medical Officers 0 5.9 19.6 60.8 13.7 0 100 HEOs 0.4 7.7 40.4 35.4 14.0 1.8 100 Nursing Officers 0.9 17.4 32.3 35.2 13.7 0.5 100 Midwives 0 5.2 28.7 44.3 20.8 1.0 100 CHWs 0.4 12.2 33.1 40.2 11.3 2.8 100 Dental Therapists 0 2.4 20.9 55.8 16.3 4.6 100 Subtotal Rural Health 0.5 13.0 32.9 39.0 12.6 2.0 100 Total Service Delivery Staff Medical Officers 0 11.3 41.7 36.7 9.0 1.3 100 HEOs 0.7 13.7 41.2 31.7 11.5 1.2 100 Nursing Officers 0.5 14.5 34.5 34.0 15.7 0.7 100 Midwives 0 3.9 25.7 43.3 26.1 1.1 100 CHWs 0.3 10.6 33.5 40.8 12.7 2.2 100 Dental Therapists 0 7.5 30.0 40.0 20.0 2.5 100 TOTAL (Percent) 0.4 11.9 34.3 37.7 14.1 1.5 100 TOTAL (Numbers) 34 1,056 3,033 3,338 1,248 133 8,844 Source: Calculated from Table 2-6. Note: Errors due to rounding. 16  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 16 11/27/12 11:38 AM • Another 43 percent of midwives (123) are aged • Another one-third of CHWs (1,475) are aged 35–44 45–54 years and can be expected to retire over the years of age and will begin to reach formal retirement next decade. age in another decade. • Only 3.9 percent of midwives (11) are aged below • Around 11 percent (479) of CHWs are aged less than 35 years. 35 years—a proportion lower than that for the cohorts aged 55 years or more. Community Health Workers • CHWs constitute about 50 percent of the total ser- Dental Therapists vice workforce of 4,398 with 15 percent (652) aged • Dental therapists and technicians total 120—the major- 55 years or older and due for retirement. ity in hospitals. The proportion aged 55 years or more • An additional 41 percent of CHWs (1,792) are aged is almost one-quarter (27) and an additional 40 per- 45–54 years and due to reach retirement age over the cent (48) are aged 45–54 years of age. Thus almost next decade. A slightly higher proportion of CHWs two-thirds of dental staff are expected to retire over working in hospitals are aged over 55 years than in the next decade. rural service delivery. • Only 7.5 percent of this occupational group (nine) are aged less than 35 years. Characteristics of the Current Health Public Sector Workforce   //  17 1305747 PNG HR Report 11-27-12.indb 17 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 18 11/27/12 11:38 AM Annex 2.1 Notes on Data Sources and Data Constraints Introduction Comments on the Current Status of the Key Human Resource Information The NDoH), the government more generally, and the Systems Churches Health Council have a number of databases which contain important information on health sec- This section makes comments on the type of infor- tor employees. Many of these systems duplicate infor- mation within each information system (as outlined mation contained in other systems and in other cases above) and includes information on characteristics of the same information is collected but different codes the health workforce, its quality, and ease of access to are used. These data systems include: (i) the Govern- the information. ment Payroll System (Department of Finance); (ii) the (i) Government Payroll System. Over the best part of the Churches Health Council Payroll; (iii) the Health Care last decade the government payroll system has been Practitioner’s Professional Registration System (within transited to the “Concept System” and management of NDoH); (iv) Health Human Resource Management Sys- the payroll has been increasingly delegated to depart- tem (within NDoH); (v) the NHCS 2009 (within NDoH); ments and provinces. This is important for the health and (vi) the Health Management Information System workforce employed by government as all provincial (within NDoH). All these systems have significant con- health staff are, in fact, provincial government employ- straints and data gaps which have degraded over recent ees. As part of the overall reforms of the health sector, years. Even the NHCS 2009 conducted by the NDoH hospitals (while national) are now managed by their to give a basis for total health-related employees in the own boards and are largely and increasingly respon- publicly financed health sector for the NHP has a num- sible for management of their payrolls. Thus there are, ber of constraints as discussed below. Nevertheless, this in effect, over 40 separate subpayrolls for the govern- is the single best count of the publicly financed health ment health sector which have very little central qual- workforce and forms the basis of the estimates of total ity control. As a consequence of both the significant health workforce, location, type of facility, occupation decentralization of payroll management and an inad- and sex used throughout this report and in the NHP. equate central quality control of information on the As described below the data from the Health Human payroll for employees, the data kept on the payroll— Resource Management System ((iv) above) was used to particularly that relating to the characteristics of the estimate the age structure of the workforce—data which employees (sex, age, qualifications, training, and loca- is key to the overall theme of this report. tion)—has deteriorated significantly. This data source 19 1305747 PNG HR Report 11-27-12.indb 19 11/27/12 11:38 AM (used quite successfully for analysis of the character- ing in the private health sector—particularly at this istics of the government health workforce over the stage of PNG’s development with strong private sec- last three decades through 2004), therefore, proved tor development, including anecdotal evidence that to be wholly inadequate for this study. It will take a the private health sector is growing significantly as very significant effort to rectify this situation. At the incomes rise and many employers in rural and rural outset of this study it had been assumed the payroll enclave areas recognize the importance of ensuring would be the principal source of information for the good health care for their employees and their fami- characteristics of the government health workforce. lies and more generally the wider communities within (ii) Churches Health Council Payrolls. These payrolls which the enclaves operate. are operated by the Churches Health Council and/ (iv) Health Human Resource Management System. This or their members to pay the publicly financed health system operates within the Human Resources Divi- workforce employed in mission/church health facili- sion of the NDoH and collects detailed information ties. This would have been the source of information on the publicly financed health workforce. It dupli- for the mission sector had the government payroll cates much of the data which should be available from been a viable source of information for health human the Government (and Church) Payroll Systems. The resource planning. These payrolls should be part of data is, however, inadequately managed and the sys- any review of data sources for health human resource tems are degrading. NDoH’s capacity to continue planning as recommended in Chapter 7 of this report. to collect this information has been significantly The NDoH should insist on a uniform and consistent degraded. Nevertheless, the data for 2009 (proba- supply of information on the mission health work- bly the last year for which data will be usable with- force as part of its “conditionalities” for public financ- out major quality improvements to the systems) was ing of the mission health workforce. found to be useful. As discussed below, it was used (iii) Health Care Practitioner’s Professional Registration to generate information on the age of the workforce System. This is a critical potential source of informa- by location, occupation and sex. Getting the infor- tion on the professional health workforce working in mation from this system required significant techni- PNG (doctors, nurses, dentists and pharmacists). It cal manipulation of the data sources as each province is, in fact, the only current potential source of con- and hospital (over 40 in total) was recorded on indi- sistent information on health professionals working vidual spreadsheets, often using nonstandard codes in the private sector. Under current legislation, all requiring considerable data manipulation to generate the listed health professionals—government, mission a consistent data set able to be analyzed on a national and private sector—are required to register annually basis. It was also evident that the database held infor- and pay an annual fee for the privilege. Without this mation on fewer individuals working in the publicly registration these professionals are not able to prac- financed health sector compared to that recorded in tice as a health professional. According to the NDoH the NHCS. Nevertheless, the information collected most health professionals comply with this require- seemed to be of high quality—notwithstanding the ment. The registration form collects key information lower recorded totals compared to the NHCS. on many other characteristics of the professional, (v) NHCS (2009). This is unambiguously the best single including where they work, age, sex, and highest recent database on the characteristics of the publicly level of training. The database to record this data is, financed health workforce and its totals. As discussed, however, in a state of disrepair and has wholly inade- it was undertaken as a one-off survey because the other quate systems in place to ensure the data is managed systems have significant systemic problems and are with integrity. This would not take long to repair with inadequate for most human resource management adequate systems and training support. It is recom- and planning uses. Unfortunately, it did not collect mended in Chapter 7 that this situation is rectified as information on the age of the workforce—hence the a matter of urgency. It is particularly important that manipulation of the data from the Health Human NDoH has a sound understanding of what is happen- Resource Management System as described below for 20  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 20 11/27/12 11:38 AM this report. Notwithstanding the overall soundness of Care Practitioner’s Professional Registration System this survey, it was not perfect and it took considerable be redeveloped as a matter of immediate priority so effort. If the existing systems were rationalized, as rec- that strategic information on the scale of the private ommended in Chapter 7, the key information should sector can be generated and then subsequently mon- be available on a continuous basis and a national sur- itored over time; (iii) that the feasibility of modify- vey of this kind should not be needed. ing the Health Management Information System (to (vi) Health Management Information System. This include the data outlined above) and making the sys- system is the core management information system tem much more user friendly to use is explored; and of NDoH and collects all the information on ser- (iv) that serious consideration be given to stopping vice delivery, and some chacteristics of health facil- the Health Human Resource Management System ities and staff. The system collects information from on the basis that this system is largely duplicating the ground (facility) up. It was not designed to be a information contained in the above referenced sys- human resource management system yet does collect tems which the NDoH human resource management important information on health staff. A key ques- needs. It is also clear that computer systems resourc- tion is should/could it be adapted or expanded to ing and the technical skills and capacity required to cover key strategic information as collected for the enable them to operate efficiently and effectively is NHCS (expanded to include age). This would mean largely missing from the NDoH. that the type of analysis undertaken in this report could be done on a regular/annual basis alongside the MTEF—as recommended in Chapter 7. It would not, Methodology Used to Estimate the of course, be useful for some other important health Health Workforce and its Key human resource purposes—so this suggestion is not Characteristics a substitute for deciding what other system(s) will be needed for overall health human resource planning. As discussed, the core of the information derived from Another key issue is that this critical system is not user the NHCS forms the basis of the information we have on friendly and in its current form is very difficult and the stock of the publicly financed health workforce and time consuming to integrate. It does not have many its characteristics. The database was adjusted (for each standard output report defined. These are technical province and the 20 hospitals) where there was missing issues that can be resolved with adequate computer information (in each case missing information was quite systems support and should be a matter of priority. small—less than 5 percent in the vast majority of cases) (vii) Overall Conclusion on Current Status of Health in the following manner. For each province and hospital it Human Resource Systems. The overwhelming con- was assumed that a proportion of the known data would clusion is that there is considerable scope to rational- apply to the unknown data—that is the data was scaled ize the human resource information systems within up in the same proportion as the known characteristics the NDoH. This report recommends that NDoH sys- so that the totals were equal to the headcount. tematically reviews its health human resource require- As discussed, the headcount survey did not gener- ments, being careful not to put too many demands ate information on age—an important characteristic of on the system(s). It is probably much better to have the health workforce. To this end it was assumed that the smaller, less complicated, but robust systems meet- age distribution by province/hospital and occupation of ing defined strategic information requirements. Our NHCS (as adjusted) would be the same as that recorded understanding of the current state of the existing sys- in the Health Human Resource Management System data- tems and the needs of NDoH lead us to tentatively base for 2009. There did not seem to be any major system- conclude that: (i) the payroll systems should be fixed atic bias in the missing data (for instance the distribution as a matter of both short- and longer-term priority by occupation, sex and province was generally consistent (including ensuring that the church system payroll between the two data sources). The data was also broadly can generate the required data); (ii) that the Health consistent with published reports of the health workforce Notes on Data Sources and Data Constraints   //  21 1305747 PNG HR Report 11-27-12.indb 21 11/27/12 11:38 AM from around 2000 adjusted for time given changes in sup- the health workforce stock be done annually and moni- ply. Nevertheless, the mission believes the data reported tored carefully to determine if the trends predicted in this in this report is relatively robust—there is probably some report actually eventuate. It also recommends that the pro- difference between the characteristics of the workforce jections are done annually as part of, or in conjunction reported and those actually in place. This simply means with, the annual MTEF updates and that the projections that it is important to improve the databases and the con- are gradually built up from provincial exercises which are sequent information base for this type of analysis. The planned to start under the proposed Asian Development report recommends, in Chapter 7, that the estimates of Bank Project in 2012. 22  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 22 11/27/12 11:38 AM Annex 2.2 Annex Table 2.1: Publicly Financed Health Staff by Province and Function Joint (Government Province / Function Government Mission & Mission) Other Total Western Admin (Urban only) 5 0 0 0 5 Urban Services 173 7 0 0 180 Rural Services 58 164 0 5 227 Subtotal Western 236 171 0 5 412 Gulf Admin (Urban only) 0 0 0 0 0 Urban Services 119 0 0 0 119 Rural Services 59 117 0 0 176 Subtotal Gulf 178 117 0 0 295 Central Admin (Urban only) 29 0 0 0 29 Urban Services 4 0 0 0 4 Rural Services 105 95 0 2 202 Subtotal Central 138 95 0 2 235 National Capital Admin (inc. NDoH HQ) 455 0 0 0 455 Urban Services 1,530 51 0 120 1,701 Rural Services 0 0 0 0 0 Sub-Total NCD 1,985 51 0 120 2,156 Milne Bay Admin (Urban only) 21 0 0 0 21 Urban Services 312 0 0 0 312 Rural Services 217 259 0 0 476 Subtotal Milne Bay 550 259 0 0 809 Oro Admin (Urban only) 10 0 0 0 10 Urban Services 163 0 0 0 163 Rural Services 113 45 0 0 158 Subtotal Oro 286 45 0 0 331 Southern Highlands Admin (Urban only) 32 0 0 0 32 (continued on next page) 23 1305747 PNG HR Report 11-27-12.indb 23 11/27/12 11:38 AM Annex Table 2.1: Publicly Financed Health Staff by Province and Function  (continued) Joint (Government Province / Function Government Mission & Mission) Other Total Urban Services 238 13 0 0 251 Rural Services 383 254 97 0 734 Subtotal SHP 653 267 97 0 1,017 Enga Admin (Urban only) 23 0 0 0 23 Urban Services 163 25 0 0 188 Rural Services 125 203 0 91 419 Subtotal Enga 311 228 0 91 630 Western Highlands Admin (Urban only) 0 0 0 0 0 Urban Services 422 19 0 0 441 Rural Services 177 384 0 0 561 Subtotal WHP 599 403 0 0 1,002 Chimbu Admin (Urban only) 29 0 0 0 29 Urban Services 148 0 0 0 148 Rural Services 478 92 0 0 570 Subtotal Chimbu 655 92 0 0 747 Eastern Highlands Admin (Urban only) 30 0 0 0 30 Urban Services 413 12 0 0 425 Rural Services 223 186 0 0 409 Subtotal EHP 666 198 0 0 864 Morobe Admin (Urban only) 10 0 0 0 10 Urban Services 454 0 0 0 454 Rural Services 118 83 0 0 201 Subtotal Morobe 582 83 0 0 665 Madang Admin (Urban only) 0 0 0 0 0 Urban Services 281 0 0 0 281 Rural Services 308 86 0 0 394 Subtotal Madang 589 86 0 0 675 East Sepik Admin (Urban only) 23 0 0 0 23 Urban Services 175 45 0 0 220 Rural Services 17 232 0 31 280 Subtotal East Sepic 215 277 0 31 523 (continued on next page) 24  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 24 11/27/12 11:38 AM Annex Table 2.1: Publicly Financed Health Staff by Province and Function  (continued) Joint (Government Province / Function Government Mission & Mission) Other Total Sandaun Admin (Urban only) 0 0 0 0 0 Urban Services 189 0 0 0 189 Rural Services 61 220 0 0 281 Subtotal Sandaun 250 220 0 0 470 Manus Admin (Urban only) 21 0 0 0 21 Urban Services 113 0 0 0 113 Rural Services 55 33 0 5 93 Subtotal Manus 189 33 0 5 227 New Ireland Admin (Urban only) 21 0 0 0 21 Urban Services 129 0 0 77 206 Rural Services 117 150 0 0 267 Subtotal Island 267 150 0 77 494 East New Britain Admin (Urban only) 17 0 0 0 17 Urban Services 202 0 0 0 202 Rural Services 188 206 0 6 400 Subtotal ENB 407 206 0 6 619 West New Britain Admin (Urban only) 18 0 0 0 18 Urban Services 246 0 0 0 246 Rural Services 124 131 0 0 255 Subtotal WNB 388 131 0 0 519 Bougainville Admin (Urban only) 27 0 0 0 27 Urban Services 187 0 0 0 187 Rural Services 145 14 0 0 159 Subtotal Bougainville 359 14 0 0 373 TOTAL Admin (Urban only) 771 0 0 0 771 Urban Services 5,661 172 0 197 6,030 Rural Services 3,071 2,954 97 140 6,262 TOTAL STAFFING 9,503 3,126 97 337 13,063 Notes: (1) Rural base does not include aid post staff. (2) Admin Urban (Provincial Health Office’s + NDoH), Urban (Hospitals + Urban Clinics), Rural (DHO’s + RH’s + HC’s + SC’s + Training Institutions). Notes on Data Sources and Data Constraints   //  25 1305747 PNG HR Report 11-27-12.indb 25 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 26 11/27/12 11:38 AM CHAPTER 3 Health Workforce Training Capacity and Issues 3.1. Introduction This capacity, particularly for nurse and CHW training, was strongly supported by a network of mission-managed This chapter outlines the current capacity of the health training institutions which also received direct public sub- workforce-related training system in PNG. It particu- sidies. In this situation, the NDoH had direct decision- larly focuses on the current training effort for the major making authority over numbers entering the vast majority service-delivery occupations as outlined in Chapter 2— of training institutions and it determined the curriculum— doctors, HEOs, nurses and midwives, and CHWs. It starts albeit in collaboration with the church/mission partners. with an outline of the current institutional arrangements There was a national consensus on both the major causes, for the training of the health-related workforce, including and nature, of both the burden of disease in PNG and on the formal responsibilities and configuration of the cur- the roles and functions of the network of different health rent health-related training system. The chapter also notes facilities and their staffing. how responsibilities have evolved and significant gaps Major decisions were made by government in 1999 emerged in the coherence of the training system over the which have had a lasting influence on the structure of past two decades. In this context it notes the emergence of postsecondary education and training in PNG—includ- two important gaps in the current institutional relation- ing for health-related training programs. First, within ships—the diffusion of responsibility for the oversight of the context of a major fiscal adjustment (which required the training and a continuing serious lack of information a 10 percent or more reduction in much of the civil ser- on the needs for the outputs of training institutions. The vice), a number of government nursing schools were closed chapter concludes with a summary documentation of the and the last of the government CHW training schools health-related postgraduate training program enrollments were also closed. Another important government policy and graduates for 2009. decision involved making the Office of Higher Education (OHE) (under the oversight of a relatively new Commis- sion of Higher Education—CHE) responsible for the vast 3.2. A Brief Discussion of the Historic majority of postsecondary education institutions—partic- Context of the National Health ularly those requiring or aspiring to require grade 12 as Workforce Training System a prerequisite for entry to the institution. This included all health-related training (HEO and nursing education) Historically, the NDoH had a major and direct role in but excepted CHW training which formally required the provision of health-related training through a net- grade 10 entry requirements. Part of this reform process work of HEO, nurse and CHW training institutions included a policy decision in 1996 that existing training which it both managed and financed through its budget. institutions (including the largely single program-based 27 1305747 PNG HR Report 11-27-12.indb 27 11/27/12 11:38 AM schools of nursing and the College of Allied Health in 45 percent. There are very few functioning countries, if Madang) either affiliate or amalgamate with a university.6 any, that have been able to deal with this level of resource It also required them to plan to upgrade their programs decline and remain basically functional. Education and to a degree level over time—an issued discussed in more training institutions, including health-related training detail in Chapter 6. institutions, have not been exempt from this fundamen- Under this arrangement, responsibility for the tal resource-scarce picture. financing of government scholarships for students to The data for nursing colleges illustrate this situa- enter degree and diploma programs for all sectors, tion with real recurrent expenditure currently about 50 including health, was centralized in the OHE. This percent of what it was in the mid-1980s. On a real per effectively meant that the OHE set student numbers for capita basis, recurrent budgets are around 25 percent of courses, except at the margin where some institutions began what they were in the 1980s (Figure 3.1). In any discus- accepting private students and a number of aid agencies sion of quality issues and the review of the current state began directly funding some students. In the absence of of health-related training, these stylized facts need to be national or sectoral human resource plans or evidence- kept in mind—in fact, in many senses, it is remarkable that based mechanisms to determine student numbers it was the institutions have been able to maintain any semblance supply-side capacity to train students which was the pri- of professionalism. It is a credit to current management mary determinant of how many were trained. While there that many of the institutions survive at all. It is fortuitous, has been much discussion of human resource planning as discussed in Chapter 5, that the resource picture fac- needs, the last National Human Resource Assessment was ing PNG is less constrained than over the past decade or undertaken in 1987–88 and the OHE still does not have more and that there is scope, assuming appropriate pri- a systematic evidence-based process to determine prior- orities, to redress some of the past resource constraints. ities for training or university programs or for scholar- Annex 3.1 (Figures 3A.1 and 3A.2) presents data on gov- ships financed by the public budget. ernment budgets and recurrent budget data for nursing As the major employer of health-related staff and colleges since independence in 1975.7 the regulator of the health system, the NDoH has pri- Since 1999, the structure of the postsecondary edu- mary responsibility for planning health human resource cation and training system and the health-related sub- needs and communicating these needs to both the OHE component of the education and training system have and to individual training institutions. It also has a role evolved in significant ways. The University of Papua New in advocating with the Department of Treasury (Treasury) Guinea (UPNG) is responsible for medical (doctor) training and the DNP&M for these institutions to be adequately and other allied health training. The two nongovernment resourced. The budget division of Treasury is responsible universities emerged about this time: the Divine Word Uni- for recurrent budget finances (and final advice to Cabinet versity (DWU) in Madang and the Pacific Adventist Uni- on proposals for the expansion of the health workforce proposed by provinces and NDoH), while the DNP&M is 6  At independence in 1976 PNG had two universities—the formally responsible for development expenditures includ- University of Papua New Guinea, with its main campus in Port ing development assistance. The NDoH also engages with Moresby, and the University of Technology in Lae—both pub- all other stakeholders—including Development Partners licly owned and funded. Another publicly financed but privately (DPs)—s appropriate and also has a key role in ensuring administered university is the Divine Word University which was that training is of adequate quality and relevance. These established in Madang. A private university—the Pacific Adven- tist University outside Port Moresby—was also established in the issues are taken up in more detail in Chapter 5. 1990s. The government established the Goroka campus of the Another important context for this discussion is University of Papua New Guinea with a primary focus on sec- that PNG, between 1993 and 2009, has seen real recur- ondary teacher education. It also established the University of rent expenditures by government decline by almost Natural Resources and Environment using the Vudal agricul- tural training college infrastructure as its base. Each university one-third while the population has grown by 65 per- is established under its own national legislation. cent over the same period. Since independence in 1975, 7  This discussion draws heavily on background work and papers real per capita government expenditure has declined over undertaken for the PNG Universities Review 2010. 28  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 28 11/27/12 11:38 AM Figure 3‑1: Real Per Capita Expenditure on Nursing whole of the health-related training system. This will Colleges as Percentage of 2009 Expenditure be discussed in more detail in Chapter 4 as it relates to nurse and CHW training institutions. It is, however, an 400% important context for a discussion of the current capacity 300% of the health-related training system. Too often, specific issues with either specific schools or cadres are discussion 200% without adequate reference to the institutional context 100% within which individual institutions operate. Aspects of 0% the current institutional arrangements are dysfunctional, 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 which makes it all but impossible for individual institu- Source: PNG Universities Review 2010. tions or actors within the system to undertake reform in the interests of rational nationally focused outcomes. A versity (PAU) outside Port Moresby—both of which also more rational approach will require a “whole-of-govern- began teaching health-related subjects. The government ment” approach. This will be taken up in more detail in (NDoH) owned College of Allied Health Sciences which Chapter 7. was responsible for HEO training merged with the DWU and consequently became responsible for HEO training. Under the reconfigured government university system 3.3. A Brief Description of the Health- the University of Goroka (UoG) began teaching nurse- Related Training System Institutions and related undergraduate and postbasic qualifications train- Programs ing programs. The “old” Schools of Nursing (SoN) have essentially maintained their existing single course/pro- The health-related training institutions (including uni- gram structure—although some have begun the process versities) as they are currently configured include the of affiliating with a university. Similarly, the CHW train- following: ing programs have continued as single program institu- tions—albeit all managed by mission authorities. 3.3.1 Universities One recent expert report suggests that: “In gen- eral the nongovernment universities appear to be more 3.3.1.1 Government Institutions effectively led and governed, and have a sense of strat- University of Papua New Guinea. egy and vision as well as vibrancy that is lacking in the Course offerings include: public universities. They have smaller councils and these • Postgraduate degree programs (Masters) in emergency appear to be more strategically focused. They are also more medicine, obstetrics and gynecology, surgery, medi- focused on both external benchmarking and quality assur- cal sciences, and public health. ance. The public universities appear to be laboring under • Postgraduate diplomas in anesthesiology, child health, enormous pressures related to funding reductions—the obstetrics and gynecology and ortorhinolaryaryngology. NHEP III (draft National Higher Education Plan III) con- • Undergraduate medical programs (Bachelors) in: med- cludes that there has been a 51 percent real funding reduc- icine and surgery, pharmacy, dental surgery and oral tion between 1983 and 2003), a loss of direction, and a health. lack of policy certainty that are causing severe difficul- • Undergraduate allied health programs (Bachelors) ties which must have an impact on their quality. There is in medical imaging science and medical laboratory a lack of capital and equipment maintenance and renewal science. and inadequate funding for new capital works and general • Undergraduate clinical nursing (Bachelors) in mid- capital infrastructure to support teaching and research” wifery and pediatrics, mental health and acute care. (ACER and Masora Consulting 2010). • Undergraduate nursing (Bachelors) in community This picture, while focused more specifically on health and nursing administration and nursing admin- the more narrow university sector, also applies to the istration and education. Health Workforce Training Capacity and Issues   //  29 1305747 PNG HR Report 11-27-12.indb 29 11/27/12 11:38 AM • Diploma programs in allied health including anes- 3.3.2.2 Mission-Owned Schools of Nursing thetic science, community health, and medical lab- Mission-owned and managed SoNs include: (i) St Barn- oratory technology. abas (Milne Bay Province); (ii) Lae (Morobe Province); (iii) St Mary’s Vunapope (East New Britain Province); University of Goroka. (iv) Lutheran (Madang Province); and (v) Nazarene Course offerings include: (Western Highlands Province). • Undergraduate nursing (Bachelors) in maternal and child health. 3.3.3 Community Health Worker Training • Diploma programs in health teaching and health edu- Schools (All Mission Owned) cation (post-SoN qualifications). There are 12 CHW training schools—all mission owned. 3.3.1.2 Nongovernment Institutions They run a common curriculum and produce gradu- Divine Word University. DWU is publicly financed but ates with a Certificate in Community Health Work. The privately managed. The health-related courses offered CHW training schools are: (i) Kapuna (Gulf Province); include: (ii) Raihu (West Sepik Province); (iii) Rumginae (Western • Undergraduate programs (Bachelors) in rural health Province); (iv) Lemakot (New Ireland Province); (v) Sal- extension, environmental health, physiotherapy and amo (Milne Bay Province); (vi) St Gerard’s (Central Prov- health management. ince); (vii) Tinsley (Western Highlands Province); (viii) St • Diploma programs in eye care and accident and emer- Margaret’s (Oro Province); (ix) Braun (Morobe Province); gency medicine. (x) Kumin (Southern Highlands Province); (xi) Onamuga • The Health Extension Officer training program. (Eastern Highlands Province); and (xii) Gaubin (Madang Province). Pacific Adventist University. • Postgraduate program in public health planned. • Undergraduate program (Bachelors) in general nurs- ing with a midwifery program planned. 3.4. Preservice Enrollments and Outputs of Health Training Institutions 3.3.2 Schools of Nursing Systematic data on enrollments by institution and There are seven SoNs which run a diploma of general program are not centrally held and, as a consequence, nursing, a three-year course with a common curric- information is fragmented. The data in this section ulum. They are single-course institutions, the majority comes from a variety of sources, including a special sur- (five) operated by missions. The Madang School of Nurs- vey of health-training institutions carried out for this ing also runs a bachelors degree in midwifery. The diploma report and which is discussed in more detail in Chapter qualification is the basic course for nursing in Papua New 4. The universities, by and large, did not respond to this Guinea and is also a general qualification for entry, with survey and special efforts were made to visit and phone some experience, to postgraduate nursing courses run by institutions to generate the information contained in this the universities. In general, schools of nursing have, or are report. Data has also been cross checked for consistency in the process of, affiliating with a university. with other data held by the Human Resources Division of NDoH wherever possible. 3.3.2.1 Government-Owned Schools of Nursing The current capacity—enrollments and graduates— Government-owned institutions include: (i) the High- of the health-related education and training institutions lands Regional College of Nursing in the Eastern and programs are summarized in Table 3-1. Table 3A.1 Highlands Province; and (ii) Mendi in the Southern in Annex 3.2 presents detailed information on pre- and Highlands Province which re-opened in 2010 after 10 postgraduate training program enrollments and gradu- years closure. ates related to the health sector for 2009. 30  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 30 11/27/12 11:38 AM Table 3‑1: Summary of Key Preservice Training Enrollments and Graduates (2009) Institution Programs Male Female Total Length (Yrs) Graduates 2009 University Degrees Medical1 – – 192 6 49 Dental (Surgery) 1 – – – 6 16 Dental (Oral Health)1 – – – 4 2 Pharmacy 1 – – 74 5 25 Med. Lab Science 1 – – 51 4 20 Med. Imaging Science1 – – 36 4 18 Nursing General 25 66 91 4 202 Health Extension Officers – – – 3 46 Environmental Health 46 55 101 3 18 Physiotherapy 21 40 61 3 19 Health Management 62 40 102 3 17 Diplomas General Nursing – – 479 3 135 Community Medicine – – – 3 30 Certificates CHWs – – 449 2 149 Source: Health Training Institution Survey 2009 and data supplied by the various universities. Note: 1 One year science at UPNG main campus and other years of study at medical school campus. 2 Estimated. Key features of the supply situation summarized Diploma and Certificate Level Graduates in Table 3-1 include: • There were currently 165 diploma-level nursing grad- Graduates with Degrees uates. This represents just 5 percent of the current • There are less than 50 medical doctors currently grad- stock of 3,279 nurses in service delivery. uating annually—about 2.5 per province per year. As • There were 149 graduates with certificates as CHWs— one point of reference there are currently only 383 doc- equivalent to just 3.4 percent of those in service deliv- tors in service delivery and there are 62 in the head- ery (4,418). quarters of NDoH in largely administrative positions. • Half as many pharmacists are graduating (25) as there Annex 3.2 Table 3A.1 presents key information on are doctors (49). This compares to the existing stock the current training capacity of all the above referenced in service delivery of 92. health-related training institutions/programs. • There were 16 graduates in dental surgery and two in oral health. 3.4.1 Medical, Health Extension Officer, • There were 20 medical laboratory and 18 medical Nursing and Community Health Worker imaging graduates. This compares to an estimated Training Programs 240 currently in direct service delivery. • There were 46 rural health extension graduates (HEOs)—almost as many as there are doctors, com- 3.4.1.1 Medical Training pared to 460 currently employed in service delivery. Admission to the medical faculty at UPNG requires • There are fewer than 20 graduates per year in each students to satisfactorily complete the first year of the of health management, physiotherapy, and environ- general science program at the main campus. Admis- mental health. sion to university requires superior grades in the national Health Workforce Training Capacity and Issues   //  31 1305747 PNG HR Report 11-27-12.indb 31 11/27/12 11:38 AM Grade 12 examinations. A very significant proportion of the PNG health system of the future. Historically, students getting into university are believed to be sourced HEOs were typically in charge of health centers, and to from a very few high schools (essentially the five old some extent health subcenters, and were responsible for International Education Agency—now elite, private fee- supervising nurses and CHWs within their span of con- based—schools based in Port Moresby and a couple of trol. This was justified on the basis that PNG had a rural other urban areas and the four “old” National High Schools health service which was a “nondoctor” based model. which used to be the only government-funded schools HEOs were, to a significant extent, an alternative to doc- providing a Grade 12 education and entry to university tor supervision in the absence of an adequate supply programs). This means that provinces without an elite of doctors in rural areas. Over time, the entry require- school (the majority) will have very limited numbers of ments and qualifications of both nurses and CHWs have their students represented in the intakes. The extent to been increased. which trained doctors prefer to go back and practice in Many in the nursing profession and others argue their own province may be one big factor constraining that, with the advent of degree programs for nurses and regional equity in the supply of doctors. a range of postbasic nurse certificate and postgradu- There is considerable concern about the quality ate programs for nurses, there is no longer a justifica- of education currently provided, in large part driven tion for HEOs. This is an issue which will be taken up by the decline in real funding levels. This is reinforced in Chapter 5 when discussing demand issues. It is clear, by the ACER and Massaro review (2010). It noted that however, as is evident from the documentation of doc- there have been no internal or external reviews of medi- tors numbers, age structure and the proportion currently cal courses since 2000. The same report also cites a Coun- working in rural areas and the current supply situation try Profile of the PNG qualification system by Australian that, even with a decision to immediately ramp up doc- Education International (2010) which suggests bachelor tor training, the rural health system will operate without degrees in PNG are the equivalent of associate degrees in doctors as a primary service provider for at least another other universities. It is not clear that this general conclu- 15 to 20 years. sion should be applied to the PNG Medical School whose Enrollments in, and graduates of, the HEO pro- graduates are being accepted in Australia and/or as entry gram have generally declined over the past 15 years. to postgraduate medical programs. Nevertheless, even if Since 2004, graduate numbers have averaged 46 with a 3.5 the medical school is an island of excellence in an other- percent dropout rate over the course—fairly low by many wise fragile institution, there is a need to ensure student tertiary course standards in PNG and internationally. In graduate quality in a transparent manner. To this end, the 2009 there were, as discussed, 46 graduates. aforementioned report’s recommendation that: “an effec- tive quality assurance system must be implemented to HEO Curriculum Issues complement the existing accreditation system” and that Some critics within the health sector not only argue that “this should provide for regular internal quality assurance the role of HEOs, at least as originally envisaged, has and periodical quality assurance based on international changed significantly over the past decade but also argue peer review…” is appropriate. that the current HEO curriculum does not prepare grad- uates adequately with adequate competence, skills or con- 3.4.1.2 Health Extension Officer Training fidence in either health-facility management or as rural HEOs are now trained by DWU following the amalga- clinical officers. Most particularly, they are not prepared mation of the the College of Allied Health Sciences with to deal adequately with emergency obstetric care—a key DWU. Entry to the program at DWU is based on Grade rationale for their existence in the absence of doctors. Over 12 results. It is important to determine to what extent all the last part of the 1990s there have been at least two HEO provinces/regions are represented in the student intakes curriculum reviews conducted by DWU and one requested for the reasons discussed above for medical graduates. by the NDoH curriculum review committee. The reviews There has been considerable ambivalence in some done by the DWU have not, apparently, been made pub- quarters about the role and function of HEOs within lic or acted upon and the other by the NDoH has not been 32  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 32 11/27/12 11:38 AM acted upon. Perhaps there is a clear need to first generate were 479 students enrolled in the Diploma of Nursing in a national consensus on what the role of an HEO should 2009, with 386 (81 percent) enrolled in mission-owned be—if an HEO is required at all. Without such a consensus facilities and 93 (19 percent) enrolled in government on a core aspect of the rural health service-delivery strat- facilities. Over 70 percent of students are female. For the egy it is hard to see how agreement can be reached on an year 2010 there were 45 students enrolled in the newly appropriate HEO training curriculum. Clearly, this requires reopened Mendi School and approximately 30 expected a “whole-of-sector” approach to rebuild a consensus on graduates in 2012 from the first intake. the future of this—historically at least—very important In 2009 there were an estimated 135 graduates. cadre. This issue is taken up again in Chapter 5 (on human The average graduating class is less than 30. The average resource demand issues) and in Chapter 7. enrollment in all SoNs is only 80 students—with one hav- ing only 52 students. Thus the training institutions under 3.4.1.3 Preservice Nurse Training by Institution discussion are very small. and Program Entry requirements to general nursing programs The intake to nurse training has fluctuated over the are generally a Grade 12 completion with good grades period 2003 to 2008 from a low in 2003 of 114 to a high in mathematics and the sciences. Most institutions spec- of 407 in 2005 but, in overall terms, has declined steadily ify that students must be 18 years of age. since 1998. Since 2003, the average number of clinical nurse graduates has been 78.8 According to data from the 8  Submission on Education and Training to the Human Resource Training Institution Survey as shown in Table 3-2, there Forum from NDoH, June 2008. Table 3‑2: Nurse Preservice Training Enrollments and Graduates (2009) Institution Ownership Male Female Total Length (Yrs) Graduates 2009 1. University Degrees 1 University of PNG–NCD – – – – (–) Pacific Adventist–NCD 25 66 91 4 (–) University of Goroka–E. Highlands – – – – (–) Divine Word University–Madang – – – – (–) Subtotal Degree 25 66 91 (–) 2. Diplomas General Schools of Nursing 2a. Mission St Barnabas–Milne Bay 14 38 52 3 13 Lae–Morobe 8 50 58 3 30 St Mary’s–East New Britain – 90 90 3 26 Lutheran–Madang 50 58 108 3 42 Nazarene–W. Highlands 35 43 78 3 24 Subtotal Mission 107 279 386 135 2b. Government Mendi–S. Highlands – – – 3 Closed Highlands Regional–E. Highlands 27 66 93 3 0 Subtotal Government 27 66 93 0 TOTAL 159 411 570 135 Source: Health Training Institution Survey 2009 and data supplied by the various universities. Note: 1 Only PAU offers any preservice nurse training programs. UPNG, UOG, and DWU do not offer such programs. Health Workforce Training Capacity and Issues   //  33 1305747 PNG HR Report 11-27-12.indb 33 11/27/12 11:38 AM SoN In-service Training of October 2008 recommended that the NDoH establish From the responses of the Health Institution Training such a program. Survey of 2009, it would seem that there is little in the way of systematic in-service training programs con- 3.4.1.4 Community Health Worker Training by ducted. Vunapope indicated that they undertook infant Institution and young child feeding and counseling training, pre- Enrollments in, and graduations from, the CHW train- ceptor training, and accident and emergency training— ing programs in the twelve mission-run CHW train- all organized by the NDoH. Nazarene organized monthly ing schools in the recent past have been in the order in-service for clinicians and academic staff. Mendi indi- of 450–470 and 210–215 respectively. The data for 2009 cated they planned preservice preceptor workshops, acci- are reported in Table 3-3. In 2009 there were 209 gradu- dent and emergency and nutrition workshops. ates from the 12 institutions—meaning that the average graduating class is only 17 with actual graduating classes General Nurse Curriculum Issues varying between 0 and 36. More than half of all enroll- There is a core national curriculum for the general ments are female (55 percent). While complete datasets diploma in nursing which is used as the basis for all are not available for graduates by gender, it is not evident programs in all the SoNS. Basic training programs only that dropout rates among enrollees by gender are signif- contain training for care of normal pregnancy and do not icantly different—indicating that enrollments are a rea- prepare the graduates to handle abnormalities and obstet- sonable proxy for graduates. ric emergencies. The extent to which this is true needs to Generally, the minimum entry requirement is for be rectified. For the current and foreseeable future, nurses a Grade 10 pass with upper passes or credits in Eng- will remain a mainstay of the health service-delivery strat- lish, mathematics, science and social science. Students egy and in many rural areas are the primary or only health are generally expected to be in the age range 18–22 years cadre—other than much lower level trained CHWs—avail- and be of good character. Some specify a period of at least able in health centers and health subcenters which serve a two years in the community prior to entry to the pro- very significant share of the population—including preg- gram. Entry requirements for the CHW training pro- nant mothers. This suggests careful consideration should gram, as specified by principals of schools in the Health be given to skills that are included in the base curriculum Facility Training Institutions Survey 2009, are presented of nurse training programs. One option must be to con- in Appendix Table 3.1.3 in Volume 2. In more recent years sider a review of the curriculum to ensure that adequate an increasing proportion of students entering the various reproductive health content and skills are included. institutions have a Grade 12 education. There has been some discussion within the health While the increased number of students graduat- sector about the possibility of introducing a field-based ing from Grade 12 throughout the country may provide postbasic nurse diploma in (more) advanced midwifery an opportunity to reconsider the entry requirements skills than can be incorporated in the basic nurse pro- to CHW training, there are a number of factors which gram of six months. It is not clear how this relates to the need to be taken into account before such a decision. current 12 month graduate diploma in midwifery. Per- These include: (i) an increased level of general education haps it could be a first module—it requires more thought. on entry to CHW training programs would almost cer- However, the notion of a form of in-service training and/ tainly require adjustments to be made to the curriculum; or a postdiploma graduate program designed to enhance (ii) principals of the schools have indicated some dissat- reproductive and child health programs for rural areas is, isfaction with Grade 12 graduates as not settling down to we believe, appropriate if it is carefully thought through the curriculum and/or that Grade 12 students have expec- and developed appropriately—initially perhaps through tations of the programs that are not realistic; and (iii) if a a pilot program. It will not be costless, however the costs decision is made that there should be a wide geographic of a high IMR and MMR are also unacceptable if such an selection of students entering training programs on the approach could be demonstrated as effective. It is noted grounds this is an important method to encourage grad- that the national Health Human Resources Conference uates to serve in remote areas, it is necessary to ensure 34  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 34 11/27/12 11:38 AM there is an adequate supply of qualified graduates from number of institutions through the establishment of new all areas of the country. Grade 12 graduates are currently government training institutions. distributed very inequitably across the country and many areas are extremely unlikely to have an adequate supply CHW In-service Training of Grade 12 graduates. These issues are taken up in more It would seem that there are few systematic in-service detail in Chapter 7. training programs being conducted. Vunapope indicated At present the mission training institutions do not that they undertook infant and young child feeding and operate in all provinces—seven provinces and the NCD counseling training, preceptor training, and accident and do not have a CHW training institution. Nevertheless emergency training—all organized by the NDoH. Naz- the training institutions are widely disbursed across the arene organized monthly in-service for clinicians and country in a very decentralized manner, certainly com- academic staff, while Mendi indicated that they planned pared to all other training systems in the country. Thus, preservice preceptor workshops, accident and emergency they are well placed to meet the staff needs of a decen- and nutrition workshops. tralized delivery system at least within the provinces in which they operate. The church CHW training institu- CHW Curriculum Issues tions were, at least initially, primarily established to meet The CHW training curricula is a national competency- the needs of their own church health service. Over time, based program, however, the program only contains government institutions were closed and their role has training for care of normal pregnancy and delivery evolved somewhat. Clearly, the institutions cater for more care and does not prepare the graduates to handle preg- than their own needs, including other provinces but it is nancy abnormalities and any obstetric emergencies. The not clear that all needs are currently being met. Chapter program is also limited in the depth of training allocated 7 will discuss the options open to expand training in the to family planning and STI (sexually transmitted infec- future given the anticipated demand—including options tion) prevention. While international evidence supports to expand existing facilities and options to increase the the notion that there should be strict limits on the scope Table 3‑3: Community Health Worker Training Enrollments and Graduates (2009) Institution (All Mission Owned) Male Female Total Graduates 2009 St Margaret’s (Oro) 20 29 49 16 Salamo (Milne Bay)1 18 26 44 20 Tinsley (Western Highlands) 11 8 19 19 Braun (Morobe) , 1 2 11 14 25 12 Raihu (West Sepik) 22 34 56 23 St Gerard’s - Veifa’ (Central)1 21 23 44 22 Onamuga (Eastern Highlands) , 13 9 11 20 10 Kapuna (Gulf) 1 18 16 34 12 Gaubin (Madang) 29 25 54 26 Rumginae (Western)3 16 14 30 0 Kumin (Southern Highlands) 5 15 20 13 Lemakot (New Ireland) 24 30 54 36 TOTAL 204 245 449 209 Percentage 45.5 54.5 100 – Source: Health Training Institution Survey 2009.  Estimated graduates. Notes: 1 2 Enrollments by sex unknown: assumes proportions the same as for all other institutions. 3 No intake in 2008. No graduates in 2009. Health Workforce Training Capacity and Issues   //  35 1305747 PNG HR Report 11-27-12.indb 35 11/27/12 11:38 AM and function of this level of staff, it nevertheless remains cent health. This is often also used as another reason why clear that currently and for the foreseeable future CHWs government needs to re-establish its own CHW training will be a mainstay of the health service-delivery strategy schools. It is important that there is a national curricula and in many rural areas are the only health staff available and that training schools are subject to quality-assurance for a significant share of the population—including preg- programs. The state of quality-assurance programs—or the nant mothers. This suggests careful consideration should almost complete absence of systematic quality-assurance be given to skills that are included in the base curricula of programs—is discussed in more detail in Chapter 4 and CHWs. One option to consider is a review of the CHW options and the case for new government CHW training curriculum to ensure that adequate reproductive health schools is discussed in more detail in Chapter 5. content and skills are included. This review should be carefully nuanced with the possible increased supply of Grade 12 students entering CHW training programs— 3.5. Postgraduate Training Programs albeit balanced with the issues discussed above relating to Grade 12 entrants and the need to adequately supply The University of Papua New Guinea Medical School remote areas with core health staff. and other universities also provide a wide array of post- There has been considerable discussion within the graduate programs in a wide range of specialties. These health sector about the possibility of introducing a post- courses and programs make an important contribution basic CHW diploma in (more) advanced midwifery skills to the human resource needs of the health system—par- (as discussed above for nurses). This notion is appropriate ticularly for specialty doctors and nurses, including mid- if it is carefully thought through and developed appropri- wives. Table 3-4 summarizes the data collected for this ately—initially perhaps through a pilot program. Options report. While not complete, it does provide some indi- discussed have typically envisaged a six month course. It cation of the scale and diversity of programs offered. The might be appropriate to have a shorter program for more data for this table is based on Appendix 3.1 in Volume 2. CHWs—there has been virtually no in-service training/ refresher programs for CHWs. These are probably highly 3.5.1 Nurse Midwifery Programs desirable—particularly as a priority for those CHWs with at least five years before retirement and who are working in By and large this report, by design, is focused on the core the remoter areas of the country. It will not be cost-free but “basic or entry level” service-delivery staff—doctors, the costs of a high IMR and MMR are also unacceptable if HEOS, nurses, and CHWs because of a crisis in supply. such an approach could be demonstrated as effective. It is, These staff categories, particularly nurses and midwives, after all, a fact that most midwifery care—at least in rural form the backbone of rural health services. The emerging areas—is currently provided by either CHWs or a general supply constraints of these staff are in danger of ensur- nurse. The National Health Human Resources Conference ing the service-delivery structure is weakened even fur- of October 2008 recommended that the NDoH establish ther compared to its existing parlous state—as discussed such a postgraduate diploma program. throughout the NHP. The only postgraduate program that All CHW training is undertaken by church agen- is a particular focus in this report is that for midwives cies that make a critical contribution to health service because of their crucial potential role in enabling a clear, delivery—albeit with the vast majority of funding com- more immediate and decisive response to the appallingly ing from the public budget. There is considerable concern high maternal mortality rates and unacceptably high neo- in many quarters of the health sector that not all aspects of natal and infant mortality rates currently prevailing in the the national curricula for CHWs (or for that matter a range country—particularly throughout the highlands region. of basic package protocols in health facilities) are taught At present doctors is the only category of staff which adequately. It is often felt that some church agencies do has significant training in abnormal and emergency not teach family planning at all and many propagate very obstetric care as part of its core curriculum. Neither moralistic ideas and attitudes towards reproductive health the general nurse training program nor the CHW train- issues—for example HIV and STI prevention and adoles- ing program covers abnormal and/or emergency obstetric 36  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 36 11/27/12 11:38 AM Table 3‑4: Postgraduate Training Program Graduates (2009) Institution / Program Male Female Total Length (Yrs) Graduates 2009 University Degrees Master of Medicine (Emerg. Med.) 2 0 2 – 2 Master of Medicine 0 2 2 – 2 (Obs & Gynecology) Master of Medicine (Surgery) 4 0 4 – 4 Master of Medical Sciences 2 0 2 – 2 Masters in Public Health 1 0 1 1 1 Bachelor of Clinical Nursing – – – 1 49 Bachelor of Nursing – – – 1 30 Maternal Child Health – – – 1 – Diplomas Anesthetics – – – 1 1 Obstetrics & Gynecology – – – – 3 Eye Care – – – 1 1 Accident & Emergency – – – 1 1 Health Teaching – – (–  1 – Health Education – – – 1 – Certificates Any  – – – – – TOTAL 9 2 11 n.a. 96 Source: Mission data collection and Health Training Institution Survey 2009. care. HEO training on other than normal deliveries is also The NDoH responded in 2006 by initiating a pro- perfunctory. As currently configured, midwifery training gram of curriculum development with external assis- takes place at university level as a postgraduate program tance. Initially an external review of the midwifery training with students entering the program after completion of a and deployment was undertaken by the end of 2007. This general diploma in nursing from one of the SoNs and gen- report confirmed that none of the current midwifery pro- erally following some work experience. Current capacity grams were adequate and meeting the professional reg- for the training of midwives is about 90 from the current istration requirements for skilled clinical midwives. A four university postgraduate programs. recommendation was made that a new national curricu- No new midwives have, however, been formally reg- lum be developed and implemented. NDoH subsequently istered since 2000 because the curricula that have been attempted to collaboratively develop a structure and basis introduced by the various universities that are respon- for a national midwifery program by supporting staff of sible for training midwives since 2001 have never been the four midwifery training programs, facilitated by the approved or certified by the Nursing Council of PNG international expert who had conducted the initial review. (NC of PNG). The key issues are apparently that none of The universities have subsequently been developing the curricula fulfill the statutory requirements in terms a national curriculum, however, this process seems to of clinical skills training. Further, the curriculum is also have bogged down. In large part this is because of a per- very theoretical and contains insufficient training time ceived conflict between the “rights” of each university to in the clinical area, and often includes insufficient super- determine curriculum and the “right” of the NC of PNG vised clinical placements during training. and of the NDoH to drive the curricula—notwithstanding Health Workforce Training Capacity and Issues   //  37 1305747 PNG HR Report 11-27-12.indb 37 11/27/12 11:38 AM they are the major employer of graduates. One school of 3.6. A Note on Subsequent Chapters of thought in the universities seems to be that NDoH does not the Report have a right to make any input to curriculum development. Another thread of discussion, perhaps even inertia, seems Chapter 4 looks in some detail at the unit costs of stu- to be that there is resistance to reforming the theoretical dent training, school revenues, operational processes unregisterable curricula. It is hoped by the time this report and issues, including quality issues of the teaching is distributed that this impasse will have been overcome. process, organizational issues, and the ability of stu- It is the view of this report that NDoH should not dents to find jobs at graduation. It also reviews aspects hesitate to decisively push the universities to produce of current training facility assets and infrastructure qual- appropriate midwives—ones who meet the needs of the ity and condition. This chapter forms a background to national service delivery strategy. A system that does not Chapters 5 and 6 which discuss options and costs to allow midwives to be registered by the NC of PNG is not enhance the supply of health human resources to meet fair to individual graduates who, in good faith, enter pro- needs. It also reports the full details of the Health Train- grams expecting to be registered. More importantly, it is ing Facility Survey of 2009. Those wishing to just review a waste of national resources and the extent to which this the demand for health human resources and then look constrains service delivery and reduced IMRs and MMRs at the estimated imbalances in supply and demand for is unethical. The NDoH should put a strong case to the the core service-delivery cadres can proceed directly to OHE and government more generally, that student spon- Chapters 5 and 6. sorships for new intakes to universities not cooperating on the revised national curriculum for midwives should be withdrawn forthwith. 38  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 38 11/27/12 11:38 AM Annex 3.1 Figure 3A.1: Real Per Capita Recurrent and Figure 3A.2: Real Recurrent Expenditure on Nursing Development Expenditure (1975–2009) Colleges (1975–2009) 800 40 600 30 PGK, millions Kina 400 20 200 10 0 0 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Recurrent expenditure Development expenditure Source: PNG Universities Review 2010. Source: PNG Universities Review 2010. 39 1305747 PNG HR Report 11-27-12.indb 39 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 40 11/27/12 11:38 AM Annex 3.2 Table 3A.1: Health Training Institutions, Programs, Enrollments and Graduates (2009) Graduates 2009 Enrollments 2009 Program Name Males Females Total Males Females Total A. UNIVERSITIES 1. University of Papua New Guinea (UPNG) Postgraduate Programs Master of Medicine (Emergency Medicine) 2 0 2 – – – Master of Medicine (Obstetrics & Gynecology) 0 2 2 – – – Master of Medicine (Surgery) 4 0 4 – – – Master of Medical Sciences 2 0 2 – – – Masters in Public Health 1 0 1 – – – Postgrad. Dip. Anesthesiology 1 0 1 – – – Postgrad. Dip. Child Health 5 0 5 – – – Postgrad. Dip. Obstetrics & Gynecology 1 2 3 – – – Postgrad. Dip. Ortorhinolaryngology 1 0 1 – – – Undergraduate Medical Programs (Allied Health) Bachelor of Medicine & Surgery 39 10 49 – – 192 Bachelor of Pharmacy 8 17 25 – – 74 Bachelor of Dental Surgery * 6 10 16 – – – Bachelor of Oral Health * 2 0 2 – – – Undergraduate Programs (Allied Health) Bachelor of Medical Imaging Science 10 8 18 – – 36 Bachelor of Medical Laboratory Science 10 10 20 – – 51 Bachelor of Clinical Nursing (Postgrad Nursing Diploma. Program) ** Bachelor of Clinical Nursing (Midwifery) *** – – – – – – Bachelor of Clinical Nursing (Pediatric) *** – – – – – – Bachelor of Clinical Nursing (Midwifery & Pediatrics) 16 33 15 – – – Bachelor of Clinical Nursing (Mental Health) 9 – – – Bachelor of Clinical Nursing (Acute Care) 25 – – – Bachelor of Nursing (Postgrad. Program) ** Bachelor of Nursing (Community Health & Nursing Admin) 9 21 11 – – – Bachelor of Nursing (Nursing Administration) *** – – – – Bachelor of Nursing (Nursing Education) *** – – – – Bachelor of Nursing (Nursing Admin & Education) 19 – – – (continued on next page) 41 1305747 PNG HR Report 11-27-12.indb 41 11/27/12 11:38 AM Table 3A.1: Health Training Institutions, Programs, Enrollments and Graduates (2009)  (continued) Graduates 2009 Enrollments 2009 Program Name Males Females Total Males Females Total Diploma Programs (Allied Health) Diploma in Anaesthetic Science 11 10 21 – – – Diploma in Community Health 13 17 30 – – – Diploma In Medical Laboratory Technology 6 0 6 – – – 2. Pacific Adventist University (PAU) Master Public Health (Postgrad. Program) Program not commenced Bachelor of Midwifery (Postgrad. Program) – – – – – – Bachelor of General Nursing – – – – – 91 3. Divine Word University (DWU) Diploma in Eye Care 1 6 7 1 5 6 Diploma in Accident & Emergency Medicine 1 0 1 0 0 0 Bachelor of Rural Health Extension 22 24 46 104 141 245 Bachelor of Environmental Health 8 10 18 46 55 101 Bachelor of Physiotheraphy 4 15 19 21 40 61 Bachelor of Health Management 12 5 17 62 40 102 4. University of Goroka (UOG) Bachelor of Maternal Child Health (Postgrad. Program) – – – – – – Diploma in Health Teaching – – – – – – Diploma in Health Education – – – – – – B. SCHOOLS OF NURSING 1. St Barnabas (MBP) Diploma in General Nursing – – 13 – – 52 2. Lae (Morobe Province) Diploma in General Nursing – – 30 – – 58 3. St Mary’s Vunapope (ENBP) Diploma in General Nursing – 26 26 – 90 90 4. Lutheran ( Madang Province) Bachelor of Midwifery – – – – – – Diploma in General Nursing – – 42 50 58 108 5. Highlands Regional (EHP) Diploma in General Nursing – – – 27 66 93 6. Mendi (SHP) Diploma in General Nursing Closed 7. Nazarene (WHP) Diploma in General Nursing 8 16 24 – – 78 C. CHW TRAINING SCHOOLS 1. Kapuna (Gulf Province) Certificate in Community Health Work – – – – – 34 (continued on next page) 42  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 42 11/27/12 11:38 AM Table 3A.1: Health Training Institutions, Programs, Enrollments and Graduates (2009)  (continued) Graduates 2009 Enrollments 2009 Program Name Males Females Total Males Females Total 2. Raihu (Sandaun Province) Certificate in Community Health Work – – 23 – – 56 3. Rumginae (Western Province) Certificate in Community Health Work – – 16 – – 30 4. Lemakot (NIP) Certificate in Community Health Work – – 36 24 30 54 5 Salamo (MBP) Certificate in Community Health Work – – – 18 26 44 6. St Gerards (Central Province) Certificate in Community Health Work – – – 21 23 44 7. Tinsley ( WHP) Certificate in Community Health Work – – 19 11 8 19 8. St Margarets (Oro Province) Certificate in Community Health Work – – 16 20 29 49 9. Braun (Morobe Province) Certificate in Community Health Work – – – – – 25 10. Kumin (SHP) Certificate in Community Health Work 5 8 13 5 15 20 11. Onamuga (EHP) Certificate in Community Health Work – – – – – 20 12. Gaubin (Madang Province) Certificate in Community Health Work – – 26 29 25 54 Source: National Health Training Institution Survey 2009.  Bachelor of Dentistry is divided into Oral Health and Dental Surgery. Notes: * ** Nursing programs have been categorized into two (Clinical Nursing and Nursing) in the graduates register at UPNG. *** These programs were merged in 2005. Health Workforce Training Capacity and Issues   //  43 1305747 PNG HR Report 11-27-12.indb 43 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 44 11/27/12 11:38 AM CHAPTER 4 Characteristics of Schools of Nursing and CHW Training Institutions 4.1. Introduction draw a wide range of data together. As previously noted, it was also collected as background information for the Gar- This chapter reviews in considerable detail the char- naut-Namaliu Review of Papua New Guinea Universities.10 acteristics of two categories of health training institu- tions—the Schools of Nursing and Community Health Worker Training Schools. The information reported in this 4.2. Nurse and CHW Training Institution chapter is derived from the Health Training Facilities Sur- Expenditures 1995–2009 vey of 2009 which, as discussed, was the primary source of data on enrollments and graduates from the health training This section documents and discusses the total recur- system reported in Chapter 2.9 Nevertheless, these institu- rent costs, average recurrent costs and the break-up of tions are responsible for supplying the core of the health recurrent operating costs of Schools for Nursing and delivery system in PNG as discussed in Chapter 2 and CHW Schools in turn. subsequently in Chapters 5 and 6. To this end Chapter 4 documents and discusses: (i) expenditures of the training 4.2.1 Schools of Nursing schools; (ii) staff-student ratios and the unit costs of stu- dents in each of the nurse and CHW training institutions; The Schools of Nursing are relatively small institutions. (iii) characteristics of nurse and CHW trainers—qualifi- The recurrent budgets for each of the seven SoNs are pre- cations and teacher training qualification, age and sex and sented in Table 4-1 for the period 2005–2009. The total years of experience of the teaching force in each of the two recurrent cost of operating all SoNs has increased from types of training schools. Section 4.5 reports the details of K2.8 million in 2005 to K3.7 million in 2009 in nominal a qualitative and quantitative assessment of the conditions under which these training schools operate (as reported by 9  The Health Training Institutions Review of 2009 was con- principals)–including quality of teaching support and teach- ducted jointly with the Secretariat of the Papua New Guinea ing processes and the quality of buildings and equipment. Universities Review 2010 (forthcoming). By and large the Uni- The data in this chapter has never been documented versities did not respond to the survey—as a consequence only systematically before—including cost and expenditure data on the SoNs and CHW Training Schools can be reported. 10  There are considerable additional detailed data for individual data and, in particular, the assessment of the teaching training institutions derived from the survey available within the process and the quality of the buildings. In this regard the Human Resources and Policy Division of the National Depart- Training Facilities Survey of 2009 is the first real attempt to ment of Health. 45 1305747 PNG HR Report 11-27-12.indb 45 11/27/12 11:38 AM Table 4‑1: Recurrent Expenditures of All Schools of Nursing 2005–2009 (Kina) School 2005 2006 2007 2008 2009 St Barnabas 479,239 506,098 592,598 615,690 646,658 St Mary’s 225,000 234,000 449,000 484,000 633,000 Lutheran 1,040,467 1,066,883 1,115,258 909,885 840,155 Nazarene 482,131 495,492 529,604 558,572 580,653 Lae 214,473 247,020 236,187 237,333 271,423 Highlands Regional 379,081 477,125 445,982 458,178 511,897 Mendi n.a. n.a. n.a. n.a. 236,951 Total Expenditures 2,820,391 3,026,618 3,368,629 3,263,658 3,720,737 Average Expenditures/SoN 470,050 504,369 561,438 543,943 531,534 Source: Health Training Institution Survey 2009. terms. In real terms budgets will have fallen significantly. are not accorded a priority for quality-enhancing school The average size of recurrent expenditures of SoNs in 2009 expenditures. The details of individual school budgets are was less than K532,000. presented in Appendix Tables 4.1.1 to 4.1.7. The composition of nursing school salary expen- The break-up of school operational costs is also ditures is instructive (Tables 4-2 and 4-3). Salary costs instructive with administrative operational costs remain- of nursing schools are estimated to have remained at ing at a fairly modest level of 5 percent of the total. around K1.2 million over the period 2005 to 2009 and Maintenance has received an increasing share of the oper- have declined as a proportion of total recurrent costs from ating budget—rising from 6 percent in 2005 to 9 percent 44 percent in 2005 to 32 percent in 2009.11 Teaching costs in 2009. At face value this could be perceived as a good have been estimated at about the same level as salaries— and sound trend. However, much of the qualitative data about K1.3 million in 2008 and 2009 while operational reported in Section 4.5 indicates that both buildings and costs have increased somewhat from K0.6 million in 2005 plant and equipment are, by-and-large, in a parlous state. to K1.2 million in 2009. Thus teaching costs, including stu- The rising maintenance budgets no doubt reflect the cri- dent boarding costs, have represented just over one-third sis many schools are facing keeping their operations con- of school operating budgets—student boarding costs being tinuing. Vehicle and plant costs represent about 3 percent the largest cost in this category and representing about of school operating budgets—while power, water, and tele- one-quarter of a typical school’s operating cost budget. phone accounted for an additional 5 percent of expendi- Student travel (to pay students to get from home to the tures in 2009. school and return for holidays) and allowances for books account for a further 5 percent of total operating costs— 4.2.2 CHW Training Schools expenses usually paid together with boarding costs by the scholarship system from the public budget. The CHW training schools are very small institu- Teaching aids and related material only consti- tions—typically situated in mission compounds or other tute 2 percent of total operating costs with the costs of mission-owned and managed health facilities. The reported computing and the internet constituting about 1 per- recurrent operating budgets for each of the 12 schools are cent of total costs although this reached 2 percent in presented in Table 4-4 for the period 2005–2009. The total 2009. Significantly, no resources have been expended on school libraries. All in all teaching quality-enhancing 11  Thequality of data for 2009 is estimated to be more robust expenditures constitute only 5 percent—perhaps 6 per- than other years. As a consequence this trend in the propor- cent if allowances for student books are included in this tion of wages in total expenditure is probably quite accurate estimate. Libraries, as also noted below in Section 4.5, (see Appendix 4.4 for how these estimates were constructed). 46  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 46 11/27/12 11:38 AM Table 4‑2: Total Expenditures on All SoNs by Expenditure Categories 2005–2009 (Kina) Expenditures 2005 2006 2007 2008 2009 Salaries 1,252,990 1,221,181 1,259,170 998,547 1,190,958 Teaching Costs Teaching Aids and Material 82,347 93,593 95,904 96,000 78,667 Student Boarding/Lodging 674,176 785,869 919,607 970,531 855,965 Student Travel and Books 141,089 181,386 174,067 178,955 204,138 IT (Computing and Internet) 20,712 18,412 19,202 33,552 79,346 Library 12,762 13,333 13,333 12,093 14,629 Other 20,000 9,333 29,333 78.017 66,680 Subtotal Teaching Costs 951,086 1,101,926 1,251,446 1,369,148 1,299,425 Operational Costs Administration Costs 126,000 131,917 182,288 184,984 188,718 Maintenance 181,143 227,457 219,600 269,935 353,184 Vehicle and Plant Costs 77,192 85,016 102,725 97,911 106,186 Power 98,867 120,360 105,533 105,533 98,133 Water 26,421 29,100 24,667 34,667 28,667 Telephone 30,911 31,144 25,333 28,000 23,333 Other 75,780 78,517 197,867 174,933 432,133 Subtotal Operational Costs 616,314 703,511 858,013 895,963 1,230,354 Total Expenditures 2,820,390 3,026,618 3,368,629 3,263,658 3,720,737 Source: Health Training Institution Survey 2009. Table 4‑3: Total Expenditures on All SoNs by Expenditure Categories 2005–2009 (%) Expenditures 2005 2006 2007 2008 2009 Salaries 44 40 37 31 32 Teaching Costs Teaching Aids and Material 3 3 3 3 2 Student Boarding/Lodging 24 26 27 30 23 Student Travel and Books 5 6 5 5 5 IT (Computing and Internet) 1 1 1 1 2 Library 0 0 0 0 0 Other 1 0 1 2 2 Subtotal Teaching Costs 34 36 37 41 34 Operational Costs Administration Costs 4 4 5 6 5 Maintenance 6 8 7 8 9 Vehicle and Plant Costs 3 3 3 3 3 Power 4 4 3 3 3 Water 1 1 1 1 1 Telephone 1 1 1 1 1 Other 3 3 6 6 12 Subtotal Operational Costs 22 24 26 28 34 Total Expenditures 100 100 100 100 100 Source: Calculated from Table 4-2. Characteristics of Schools of Nursing and CHW Training Institutions   //  47 1305747 PNG HR Report 11-27-12.indb 47 11/27/12 11:38 AM recurrent costs of all institutions indicate there was a signif- dent travel (for home leave) and books account for an icant increase in expenditures between 2005 and 2006 from additional 4–5 percent of total expenditures. K1.6 million to K2.6 million but that since 2006 expendi- Teaching aids and associated materials are reported tures have been fairly constant at around K2.2 to 2.5 mil- at about 9 percent in 2009—substantially higher than lion per annum in nominal terms. In real terms they will for nurse schools—and have been generally increasing have declined significantly. The average cost of running a as a proportion of total operating costs. Internet con- CHW school has increased in nominal terms from K135,000 nections account for 1 percent or less over the period—as in 2005 to K214,000 in 2009 with a few schools spending reflected in Section 4.5 below—and many CHW schools around one-half of the average expenditure per school. do not have any access to the internet. As with nursing The composition of CHW school budgets pro- schools, no expenditures are reported for libraries—clearly vides an important insight into the current cost struc- libraries are either nonexistent or in a very parlous state ture of teaching community health workers (Tables which is not a good situation for a good teaching environ- 4-5 and 4-6). Salary costs in nominal terms have fluctu- ment. Overall direct teaching quality-enhancing expen- ated between K351,000 in 2005 to K475,000 in 2007 and ditures constitute about 10 percent of total operating down to K395,000 in 2009—a substantial decline since expenditures—an amount that could and should, at face 2006 in real terms. value, be higher. As a proportion of total recurrent costs, salaries Operational costs for CHW schools are reported at have declined from 22 percent in 2005 to 15 percent in about one-half of total recurrent expenditures. In 2005– 2009—having been around 20 percent in 2007 and 2009. 2007 maintenance accounted for a very high proportion Teaching costs over the period 2005–2009 have declined of overall expenditures and in 2007–2009 the unspecified from 47 percent in 2005 to 34 percent in 2009. The main “other” has increased significantly—particularly in 2009. cost under this heading is student boarding which showed As discussed under the section on nurse schools, the high a declining proportion over this period from 30 percent levels of maintenance expenditure probably constitute in 2005 to 16 percent in 2009. Nevertheless, it continues an effort to deal with a buildup of past neglect of main- to constitute a substantial proportion of total costs. Stu- tenance—see also Section 4.5 below. Vehicle and plant Table 4‑4: Total Recurrent Costs CHW Schools 2005–2009 (Kina) CHW School 2005 2006 2007 2008 2009 St Margaret’s 176,669 102,170 153,660 144,279 184,756 Salamo 57,433 91,000 129,184 195,110 194,993 Tinsley 156,344 297,891 193,562 189,655 107,153 Braun 159,994 301,541 197,212 210,822 244,470 Raihu 142,253 153,410 171,965 162,765 539,502 St Gerard’s 174,240 315,787 211,458 225,068 258,716 Onamuga 141,900 283,447 179,118 192,728 226,376 Kapuna 111,004 176,710 154,967 206,352 137,953 Gaubin 110,907 124,174 128,012 144,881 188,693 Rumginae 127,011 130,276 159,997 171,757 238,845 Kumin 103,428 403,472 396,640 109,488 129,418 Lemakot 165,503 233,564 135,207 338,065 119,255 Total Expenditure All Schools 1,626,686 2,613,443 2,210,983 2,290,969 2,570,131 Average Expend/School 135,557 217,786 184,248 190,914 214,177 Source: Health Training Institution Survey 2009. 48  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 48 11/27/12 11:38 AM Table 4‑5: Total Expenditures of All CHW Schools by Expenditure Categories 2005–2009 (Kina) Expenditures 2005 2006 2007 2008 2009 Salaries 350,647 516,803 475,293 450,315 394,694 Teaching Costs Teaching Aids and Material 113,156 108,469 126,765 149,435 237,799 Student Boarding/Lodging 490,868 370,030 383,548 437,135 403,954 Student Travel and Books 76,961 123,068 81,804 116,733 95,105 IT (Computing and Internet) 8,400 13,200 6,000 11,750 13,800 Library 5,000 5,000 2,250 4,000 5,000 Other 68,860 80,504 91,971 119,668 125,222 Subtotal Teaching Costs 763,245 700,271 692,338 838,721 880,880 Operational Costs Administration Costs 59,377 99,367 115,344 119,110 139,600 Maintenance 190,328 808,491 432,582 123,850 123,978 Vehicle and Plant Costs 116,244 124,588 103,701 243,465 141,276 Power 83,304 80,162 101,815 170,843 99,252 Water 7,838 14,430 28,145 32,761 50,521 Telephone 34,300 39,562 33,734 38,570 29,651 Other 21,403 229,769 228,031 273,334 710,279 Subtotal Operational Costs 512,794 1,396,369 1,043,352 1,001,933 1,294,557 Total Expenditures 1,626,686 2,613,443 2,210,983 2,290,969 2,570,131 Source: Health Training Institution Survey 2009. Table 4‑6: Total Expenditures of All CHW Schools by Expenditure Categories 2005–2009 (Percent) Expenditures 2005 2006 2007 2008 2009 Salaries 22 20 21 20 15 Teaching Costs Teaching Aids and Material 7 4 6 7 9 Student Boarding/Lodging 30 14 17 19 16 Student Travel and Books 5 5 4 5 4 IT (Computing and Internet) 1 0 0 1 1 Library 0 0 0 0 0 Other 4 3 4 5 5 Subtotal Teaching Costs 47 26 31 37 35 Operational Costs Administration Costs 4 4 5 5 5 Maintenance 12 31 20 5 5 Vehicle and Plant Costs 7 5 5 11 5 Power 5 3 5 7 4 Water 0 0 1 1 2 Telephone 2 1 2 2 1 Other 1 9 10 12 28 Subtotal Operational Costs 31 53 48 43 50 Total Expenditures 100 100 100 100 100 Source: Calculated from Table 4-5. Characteristics of Schools of Nursing and CHW Training Institutions   //  49 1305747 PNG HR Report 11-27-12.indb 49 11/27/12 11:38 AM costs and power costs constitute an important share of tion for nursing schools (Table 4-7) reports some interest- total operational costs—generally between 9 and 12 per- ing observations. Average unit costs for all nurse training cent of total expenditures over the period 2005–2009. schools was K7,479 in 2007. In nominal terms unit costs have increased by about K1,000 over the years 2007–2009, however, there is a significant differences between the unit 4.3. Unit Costs of Nurse Schools and costs of mission and the government-managed school. The CHW Training Institutions average unit costs for mission schools in 2009 was K7,874 while that for government schools was K5,504 in 2009. Unit costs are measured by the schools’ total recurrent The average unit costs of students in mission schools have expenditure divided by student enrollments (the latter is been about K2,400 per student more expensive (about drawn from those reported in Chapter 3). The informa- 43 percent more expensive) than the government school. Table 4‑7: Recurrent Unit Costs per Student for Schools of Nursing 2006–2009 (Kina) School/Costs 2006 2007 2008 2009 St Barnabas Total Recurrent Costs 506,098 592,598 615,690 646,658 Enrollments (EFTS)* 51 51 57 52 Unit Cost 9,923 11,619 10,802 12,436 St Mary’s Total Recurrent Costs 234,000 449,000 484,000 633,000 Enrollments (EFTS)* 59 84 89 90 Unit Cost 3,966 5,345 5,438 7,033 Lutheran Total Recurrent Costs 1,066,883 1,115,258 909,885 840,155 Enrollments (EFTS)* 168 133 120 108 Unit Cost 6,350 8,385 7,582 7,779 Nazarene Total Recurrent Costs 495,492 529,604 558,572 580,653 Enrollments (EFTS)* 83 78 77 78 Unit Cost 5,970 6,790 7,254 7,444 Lae Total Recurrent Costs 247,020 236,187 237,333 271,423 Enrollments (EFTS)* (–) 48 60 58 Unit Cost (–) 4,920 3,956 4,680 Highlands Regional Total Recurrent Costs 477,125 445,982 458,178 511,897 Enrollments (EFTS)* (–) 88 91 93 Unit Cost (–) 5,068 5,035 5,504 Average Unit Costs – All (–) 7,021 6,677 7,479 Average Unit Costs – Church 6,552 7,411 7,006 7,874 Average Unit Costs – Government (–) 5,068 5,035 5,504 Source: Health Training Institution Survey 2009. Notes: Effective Full-Time Equivalent Student Enrollments. Excludes Mendi because it had only partial enrollments in 2009 and was not operational in previous years. 50  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 50 11/27/12 11:38 AM These are significant differences—assuming all a mission school would be about K7,000 more expensive enrolled students graduate and that the nurse course at K23,600 per graduate. While one needs to be careful is three years. The average cost of a graduate from a gov- how to interpret these results—there could, for example, ernment school would be about K16,500 while that from be variations in quality of student outcomes and/or drop- Table 4‑8: Average Recurrent Unit Costs CHW Schools 2006–2009 (Kina) School 2006 2007 2008 2009 St Margaret’s Total Recurrent Costs 102,170 153,660 144,279 184,756 Enrollments (EFTS)* 19 20 20 49 Unit Cost 5,377 7,683 7,214 3,771 Salamo Total Recurrent Costs 91,000 129,184 195,110 194,993 Enrollments (EFTS)* 34 37 41 44 Unit Cost 2,676 3,491 4,759 4,432 Tinsley Total Recurrent Costs 297,891 193,562 189,655 107,153 Enrollments (EFTS)* (–) (–) 49 19 Unit Cost (–) (–) 3,871 5,640 Braun Total Recurrent Costs 301,541 197,212 210,822 244,470 Enrollments (EFTS)* 22 (–) 25 25 Unit Cost 13,706 (–) 8,433 9,779 Raihu Total Recurrent Costs 153,410 171,965 162,765 539,502 Enrollments (EFTS)* 47 42 19 56 Unit Cost 3,264 4,094 8,567 9,634 St Gerard’s Total Recurrent Costs 315,787 211,458 225,068 258,716 Enrollments (EFTS)* 30 30 30 44 Unit Cost 10,526 7,049 7,502 5,880 Onamuga Total Recurrent Costs 283,447 179,118 192,728 226,376 Enrollments (EFTS)* 25 25 25 20 Unit Cost 11,338 7,165 7,709 11,319 Kapuna Total Recurrent Costs 176,710 154,967 206,352 137,953 Enrollments (EFTS)* (-) 15 n.a. 34 Unit Cost (-) 10,331 n.a. 4,057 Gaubin Total Recurrent Costs 124,174 128,012 144,881 188,693 Enrollments (EFTS)* 51 49 54 54 Unit Cost 2,435 2,612 2,683 3,494 Rumginae Total Recurrent Costs 130,276 159,997 171,757 238,845 Enrollments (EFTS)* 30 19 15 30 Unit Cost 4,343 8,421 11,450 7,962 Kumin Total Recurrent Costs 403,472 396,640 109,488 129,418 Enrollments (EFTS)* 38 40 13 20 Unit Cost 10,618 9,916 8,422 6,471 Lemakot Total Recurrent Costs 233,564 135,207 338,065 119,255 Enrollments (EFTS)* 59 69 62 54 Unit Cost 3,959 1,960 5,453 2,208 Average Unit Costs – All** 6,824 6,272 6,915 6,221 Source: Health Training Institution Survey 2009. Note: *Effective Full-Time Equivalent Student Enrollments. ** Based on schools for which we have data. Characteristics of Schools of Nursing and CHW Training Institutions   //  51 1305747 PNG HR Report 11-27-12.indb 51 11/27/12 11:38 AM out and graduation rates may vary—the numbers do sug- tion qualifications; teaching qualifications; and years of gest government authorities produce nurse graduates at experience in a professional capacity since beginning lower unit costs. work. Nurse schools are discussed, followed in turn by The unit costs of CHW schools—all of which are CHW training managed by mission authorities—are presented in Table 4-8. The average unit costs are estimated at around 4.4.1 Nursing Schools K6,500 in nominal price terms over the period 2006–2009 with unit costs in 2009 being the lowest recorded over The network of seven nursing schools in PNG is a rel- this period at K6,221. Given that a CHW course is two atively small system with a total teaching staff of only years and assuming dropouts are not significant (and 58 (Table 4-9). The average school has only eight teach- available evidence suggests they are not significant) the ing staff. Still, within the context of PNG, the teaching average cost of producing a CHW graduate is about staff of nurse schools are well trained and have significant K12,450. It is noteworthy that there is a fair degree of experience which is not easy to replicate—and will not variation in the unit costs between the various CHW be easy to expand without decisive policy interventions schools. In 2009 the range varied between K2,208 for should a decision be taken to expand the nurse training Lemakot to K11,319 for Onamuga—quite an extreme system (see Chapter 7). At present there are 36 females variation. Nevertheless, the majority of schools are within (about two-thirds of the total) teaching in nursing schools a range of K3,800–K8,000. and 22 males (one-third). One school—St Mary’s—has an all-female staff while the newly established government- managed school in Mendi has a 70 percent complement 4.4. Staffing Numbers and Other of male staff. This is probably because of the difficult liv- Characteristics of Nurse and CHW ing conditions in Mendi and the Southern Highlands. Training Schools The age structure of nursing school teaching staff is important because there is a need to balance experi- Important factors which affect the cost structures of ence and to ensure that enough replacement teachers schools but also—and perhaps more importantly—the can be found when individual teachers reach retirement quality of the teaching experience for students and the age. The age structure of the overall health workforce is quality of outcomes from the schools are the num- aging very quickly. About 16 percent (nine teachers) will bers of teachers and their educational, training and be well past retirement age in the next decade (that is work experience. This section reports on teacher num- they are 55 years of age or older) and that another 26 per- bers by school; their sex and age; their formal educa- cent (15 teachers) are aged 45–54 years and will thus Table 4‑9: Total Full-Time Nursing School Teaching Staff by Sex (2009) Nursing School Female Male Total Percent Female Percent Male St Barnabas (Milne Bay) 4 4 8 50 50 St Mary’s (East New Britain) 9 0 9 100 0 Lutheran (Madang) 5 4 9 56 44 Nazarene(Western Highlands) 7 3 10 70 30 Lae (Morobe) 3 3 6 50 50 Highlands Regional (Eastern Highlands) 6 3 9 67 33 Mendi (Southern Highlands) 2 5 7 29 71 All Schools Total 36 22 58 62 38 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. 52  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 52 11/27/12 11:38 AM Table 4‑10: Total Full-Time Nursing School Teaching Staff by Age (2009) by Age Group 200 Nursing School <35 Years 35–44 Years 45–54 Years 55–64 Years 65 + Years St Barnabas (Milne Bay) 1 3 4 0 0 St Mary’s (East New Britain) 2 3 3 1 0 Lutheran (Madang) 1 6 2 0 0 Nazarene(Western Highlands) 2 8 0 0 0 Lae (Morobe) 1 3 1 3 0 Highlands Regional (Eastern Highlands) 0 1 2 3 1 Mendi (Southern Highlands) 0 3 3 1 0 All Schools Total 7 27 15 8 1 All Schools (%) 12 46 26 14 2 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. become eligible for retirement in 10–15 years. Almost The quality of teaching in schools, including nurs- 60 percent (34 teachers) are less than 45 years—these ing schools, is very likely to be enhanced by teaching teachers will become the core experienced staff over the staff having a teaching qualification as well as a profes- next decade and as efforts are made to expand the nurse sional qualification. The data in Table 4-12 presents this training system. information for nurse training colleges. This shows that The profile of the formal education qualifications about one-third (20 staff) do not have a teaching qualifi- of the six reporting nurse-training institutions is pre- cation while 60 percent (35 staff) have a Diploma in Edu- sented in Table 4-11. It is noteworthy that over 50 per- cation. A further 5 percent (3 staff) have a Bachelors of cent of nurse trainers (27) have a Bachelors degree and Education degree. that almost a further 40 percent (20) have a Masters Another important dimension of teaching staff degree. Only 8 percent have a diploma (4). This indicates as a part proxy for quality of teaching is the expe- that nurse trainers are quite well qualified. In fact, three rience of staff teaching. The data in Table 4-13 indi- of the six colleges do not have any trainers with less than cates that qualified staff (with a teaching qualification) a Bachelors degree and no school has more than 25 per- have twice the work experience of unqualified teaching cent of its staff with only a diploma. staff—16 years compared with eight years. This holds Table 4‑11: Education Qualifications of Full-Time SoN Teaching Staff (2009) (Percent) School of Nursing * Certificate Diploma Bachelors Masters PhD Total Percent Total No. St Barnabas 0 25 25 50 0 100 8 St Mary’s 0 0 89 11 0 100 9 Lutheran 0 0 44 56 0 100 9 Nazarene 0 0 30 70 0 100 10 Lae 0 17 50 33 0 100 6 Highlands Regional 0 11 78 11 0 100 9 All Schools Total 0 4 27 20 0 100 51 All Schools % 0 8 53 39 0 100 51 Source: Health Training Institution Survey 2009. Note: * Data for Mendi not available. **Part-time staff not included. Characteristics of Schools of Nursing and CHW Training Institutions   //  53 1305747 PNG HR Report 11-27-12.indb 53 11/27/12 11:38 AM Table 4‑12: Teaching Qualifications of Full-Time Nursing School Teaching Staff (2009) (Percent) School of Nursing Nil Certificate Diploma Bachelors Masters Total % Total No. St Barnabas 37.5 – 62.5 – – 100 8 St Mary’s 44.4 – 44.4 11.1 – 100 9 Lutheran 22.2 – 77.8 – – 100 9 Nazarene 80 – 20 – – 100 10 Lae 33.3 – 66.7 – – 100 6 Highlands Regional 11.1 – 77.8 11.1 – 100 9 Mendi – – 85.7 14.3 – 100 7 All Schools % 34.5 0.0 60.3 5.2 0.0 100 58 All Schools No. 20 0 35 3 0 100 58 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. more or less true for all schools. The apparent reason per school. One school has only three teaching staff and for this is that staff teaching in nursing schools for lon- five schools have six staff. ger periods have been rewarded in the past with access In total there are 60 teaching staff—about the same to scholarships to gain a teaching qualification—an issue number of teachers as there are in all seven nursing taken up in Chapter 7. schools. Some 60 percent of the teaching staff is female and 40 percent male. There is often a significant differ- 4.4.2 CHW Training Schools ence in the gender distribution of staff between individ- ual schools—with Lemakot in New Ireland Province being The total teaching staff of all CHW training schools is 100 percent female and Gaubin in Madang Province being presented in Table 4-14. Each of these schools is quite only 25 percent female. small—with there being an average of five teaching staff The age structure of the teaching force is impor- tant—both in terms of ensuring there is adequate experience within the institution and to form the basis of future planning of teaching staff as they retire or Table 4‑13: Average Years of Experience of Full-Time reach retirement age. The age structure of the overall Nursing School Teaching Staff (2009) health workforce—particularly of CHWs is aging very Qualified Unqualified quickly. The age structure of the teaching workforce is School of Teaching Teaching younger than for the nurse teaching workforce or for Nursing Staff1 Staff1 Total Staff CHWs in the workforce (Table 4-15). Some 70 percent St Barnabas 16 13 8 of the workforce is less than 44 years—so if there is no St Mary’s 14 5 9 rapid turnover of CHW school teaching staff the core Lutheran 12 8 9 of the teaching staff will not reach retirement for 15 or Nazarene 17 8 10 more years. Only 12 percent of the teaching workforce Lae 17 9 6 is more than 45 years and thus due for retirement by the Highlands Regional 19 9 9 end of the next decade. The age of some six staff (10 per- Mendi 17 0 7 cent) are not known. All Schools 16 8 58 The formal education qualifications of the staff Average teaching in the CHW training institutions are presented Source: Health Training Institution Survey 2009. in Table 4-16. Some 14 of the teaching staff (23 percent) Note: 1 Staff with and without formal teaching qualifications. Part-time staff not included. have a Certificate level education qualification while 36 54  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 54 11/27/12 11:38 AM Table 4‑14: Total Full-Time CHW School Teaching Staff by Gender (2009) CHW Schools Female Male Total Percent Female Percent Male St Margaret’s (Oro) 4 1 5 80 20 Salamo (Milne Bay) 2 3 5 40 60 Tinsley (Western Highlands) 2 2 4 50 50 Braun (Morobe) 2 2 4 50 50 Raihu (Sandaun) 3 3 6 50 50 St Gerard’s (Central) 5 1 6 83 17 Onamuga (Eastern Highlands) 2 1 3 67 33 Kapuna (Gulf) 3 2 5 60 40 Gaubin (Madang) 1 3 4 25 75 Rumginae (Western) 3 3 6 50 50 Kumin (Southern Highlands) 3 3 6 50 50 Lemakot (New Ireland) 6 0 6 100 0 All Schools Total 36 24 60 60 40 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. Table 4‑15: Total Full-Time CHW School Teaching Staff by Age (2009) CHW School <35 Years 35–44 Years 45–54 Years 55–64 Years 65 + Years Unknown St Margaret’s (Oro) 3 1 1 0 0 0 Salamo (Milne Bay) 0 4 0 1 0 0 Tinsley (Western Highlands) 1 3 0 0 0 0 Braun (Morobe) 0 1 0 0 0 3 Raihu (Sandaun) 1 1 1 0 0 3 St Gerard’s (Central) 1 2 2 1 0 0 Onamuga (Eastern Highlands) 1 2 0 0 0 0 Kapuna (Gulf) 5 0 0 0 0 0 Gaubin (Madang) 0 2 2 0 0 0 Rumginae (Western) 2 3 1 0 0 0 Kumin (Southern Highlands) 3 3 0 0 0 0 Lemakot (New Ireland) 1 2 3 0 0 0 All Schools Total 18 24 10 2 0 6 All Schools % 30 40 17 3 0 10 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. staff (60 percent of the total) have a Diploma level quali- only a Certificate and that three schools are fully staffed fication. In addition, eight teaching staff (13 percent) have by Diploma qualified staff. a Bachelors Degree level education qualification. Two As is the case with nursing schools, teaching staff teaching staff member have a Masters Degree. It is inter- with a formal teaching qualification, all other things esting that six CHW schools have no teaching staff with being equal, are likely to be better teachers—at least Characteristics of Schools of Nursing and CHW Training Institutions   //  55 1305747 PNG HR Report 11-27-12.indb 55 11/27/12 11:38 AM Table 4‑16: Education Qualifications of Full-Time CHW Schools Teaching Staff (2009) (Percent) CHW Schools Certificate Diploma Bachelors Masters PhD Total % Total No. St Margaret’s (Oro) 20 80 0 0 0 100 5 Salamo (Milne Bay) 60 40 0 0 0 100 5 Tinsley (W. Highlands) 0 75 25 0 0 100 4 Braun (Morobe) 0 100 0 0 0 100 4 Raihu (Sandaun) 0 50 50 0 0 100 6 St Gerard’s (Central) 67 – 16.5 16.5 0 100 6 Onamuga (E. Highlands) 0 100 0 0 0 100 3 Kapuna (Gulf) 60 20 20 0 0 100 5 Gaubin (Madang) 0 100 0 0 0 100 4 Rumginae (Western) 16.5 67 16.5 0 0 100 6 Kumin (S. Highlands) 0 83 0 17 0 100 6 Lemakot (New Ireland) 33 50 17 0 0 100 6 All Schools Total 14 36 8 2 0 100 60 All Schools % 23 60 13 3 0 100 60 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. early in their careers. Data compiled from the survey is noteworthy that a number of schools have a high pro- and presented in Table 4-17 indicate that 62 percent of portion of teaching staff without an education teaching teaching staff at CHW schools have an education teach- qualification—Kapuna (Gulf Province) has 60 percent ing qualification—with 5 percent having a certificate to unqualified as teachers; Salamo (Milne Bay Province) teach and 57 percent having a Diploma in Education. It has 80 percent unqualified as teachers and St Gerard’s Table 4‑17: Teaching Qualification of Full-Time CHW Schools Teaching Staff (2009) (Percent) CHW Schools Nil Certificate Diploma Bachelors Masters Total % Total No. St Margaret’s (Oro) 20 0 80 0 0 100 5 Salamo (Milne Bay) 80 20 0 0 0 100 5 Tinsley (Western Highlands) 25 0 75 0 0 100 4 Braun (Morobe) 50 0 50 0 0 100 4 Raihu (Sandaun) 16.5 16.5 67 0 0 100 6 St Gerard’s (Central) 100 0 0 0 0 100 6 Onamuga (Eastern Highlands) 0 0 100 0 0 100 3 Kapuna (Gulf) 60 0 40 0 0 100 5 Gaubin (Madang) 25 0 75 0 0 100 4 Rumginae (Western) 17 0 83 0 0 100 6 Kumin (Southern Highlands) 16.5 16.5 67 0 0 100 6 Lemakot (New Ireland) 33 0 67 0 0 100 6 All Schools Total 23 3 34 0 0 100 60 All Schools % 38 5 57 0 0 100 60 Source: Health Training Institution Survey 2009. Note: Part-time staff not included. 56  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 56 11/27/12 11:38 AM (Central Province) has 100 percent without a teaching K4.5 million. The student-to-staff ratios vary from 12:1 qualification. to 7:1. Interestingly, the school with the lowest student- The experience of staff teaching in CHW schools to-staff ratio (St Barnabas, Milne Bay Province) does not is presented in Table 4-18. The average experience of have the highest unit costs while the school with the high- qualified teaching staff is 10 years while that for unqual- est student-to-staff ratio school—Lutheran (Madang Prov- ified teaching staff is eight years. It is interesting that the ince)—does not have the lowest average unit cost. average length of service of qualified teaching staff in two In Community Health Worker Schools 449 students schools—Gaubin (Madang Province) and Lemakot (New are taught by about 60 staff at an average student-to- Ireland Province)—is significantly above the overall aver- staff ratio of 8:1 (Table 4-20). The average unit cost per age as is Raihu (Sandaun Province) for unqualified teach- enrolled student in 2009 was K5,724. The total recurrent ing staff. This may indicate that that Gaubin and Lemakot budget expended for all 12 schools in 2009 was K2.6 mil- face a renewal problem for qualified personnel due to the lion. The school with the highest student-to-staff ratio aging of their staffing cohort. (Gaubin, Madang Province with 14:1) does have one of the lowest unit costs at about K3,500 while the school with the lowest unit cost at K2,208 (Lemakot, New Ireland Prov- 4.5. Key Indicators for Nurse and CHW ince) also has a higher than average student-to-staff ratio. Schools: A Summary This section summarizes some of the key indicators for 4.6. A Qualitative and Quantitative each of the nursing schools and CHW training schools Assessment of Nurse and CHW Training for 2009. Schools: A Summary Report of the In the Schools of Nursing 479 students were taught Survey of Principal’s Views by 54 staff at an average student-to-staff ratio of 9 to 1 (Table 4-19). The average unit cost per student per year The survey also sought key information on the views in 2009 was K9,481 with total recurrent expenditures of of the principals of nurse training schools on a range Table 4‑18: Average Years of Experience of Full-Time CHW School Teaching Staff (2009) Qualified Teaching Unqualified Teaching CHW School Staff1 Staff1 Total Staff St Margaret’s (Oro) 10.5 1 5 Salamo (Milne Bay) 7.5 9 5 Tinsley (Western Highlands) 5 10 4 Braun (Morobe) – – 4 Raihu (Sandaun) 7 21 6 St Gerard’s (Central) 0 12 6 Onamuga (Eastern Highlands) 7 0 3 Kapuna (Gulf) 2 1 5 Gaubin (Madang) 27 7 4 Rumginae (Western) 8 4 6 Kumin (Southern Highlands) 6 1 6 Lemakot (New Ireland) 15 6 6 All Schools Average 8.6 6.5 60 Source: Health Training Institution Survey 2009. Note: 1 Staff with and without formal teaching qualifications. Part-time staff not included. Characteristics of Schools of Nursing and CHW Training Institutions   //  57 1305747 PNG HR Report 11-27-12.indb 57 11/27/12 11:38 AM Table 4‑19: Key Indicators for SoNs (2009) Student: Recurrent Unit Costs School of Nursing No. of Students No. of Staff Staff Ratio Expenditure (Kina) (Kina) St Barnabas (Milne Bay) 52 8 7:1 646,658 12,436 St Mary’s (East New Britain) 90 10.6 9:1 633,000 7,033 Lutheran (Madang) 108 9 12:1 840,155 7,779 Nazarene (W. Highlands) 78 10.5 8:1 580,653 7,444 Lae (Morobe) 58 7 9:1 1,092,323 18,833 Highlands Regional (E. Highlands) 93 9 11:1 511,897 5,504 All Schools 479 54.1 9:1 4,541,637 9,481 Source: Health Training Institution Survey 2009. Note: St Mary’s (one-third), Nazarene (one-half), and Lae (one-half) of their total part-time staff had been added to total permanent staff. Table 4‑20: Key Indicators for CHW Schools (2009) Student: Staff Recurrent Unit Costs CHW School No. of Students No. of Staff Ratio Expenditure (Kina) (Kina) St Margaret’s (Oro) 49 5 10:1 184,756 3,771 Salamo (Milne Bay) 44 5 9:1 194,993 4,432 Tinsley (W. Highlands) 19 4 5:1 107,153 5,640 Braun (Morobe) 25 4 6:1 244,470 9,779 Raihu (Sandaun) 56 6 9:1 539,502 9,634 St Gerard’s (Central) 44 6.5 7:1 258,716 5,880 Onamuga (E. Highlands) 20 3 7:1 226,376 11,319 Kapuna (Gulf) 34 5 7:1 137,953 4,057 Gaubin (Madang) 54 4 14:1 188,693 3,494 Rumginae (Western) 30 6 5:1 238,845 7,962 Kumin (S. Highlands) 20 6 3:1 129,418 6,471 Lemakot (New Ireland) 54 6 9:1 119,255 2,208 All CHW Schools 449 60.5 8:1 2,570,130 5,724 Source: Health Training Institution Survey 2009. Note: One-half of total part-time staff had been added to St Gerard’s total permanent staff. of aspects of the schools and the educational processes vey respondents vary somewhat between institutions within schools including: (i) student entry, quality of cur- but can be summarized as: (i) Grade 12 education riculum and of students and graduates; and (ii) school with sound grades in core subjects including Eng- assets infrastructure and its quality.12 lish, Mathematics, and Science; (ii) aged 18 years. 4.6.1 Nursing School Student Entry 12  Detailed information on each Nursing School and Commu- Requirements, Quality of Curriculum and of nity Health Worker Training Center is available from the Human Graduates Resources Division or the Planning Division of the National Department of Health. This section of the report summarizes the results of the survey but the detailed data will be available (i) Formal Entry Requirements: General Schools of for detailed planning for the supply response needed to respond Nursing. Entry requirements according to the sur- to the conclusions of this report. 58  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 58 11/27/12 11:38 AM Entry requirements for nonschool leavers are sim- ularly for lower grade point averages at entry. One ilar to those for school leavers but applications are observed that there were severe time and capacity considered on individual merit. limitations for remedial work while others observed (ii) Application of Entry Requirements at Institution that needs in English, Mathematics and Science Level. Institutions were also how they apply the cri- were fairly common. Two principals commented teria for entry at the institution level. This produced that the current Grade 12 students are not as good some interesting results that indicate each institu- as the previous Grade 10 students in both prepara- tion is also responding to perceived local needs. One tion and attitude. institution (St Barnabas) indicated that 50 percent (v) Principals Perceived Strengths and Weaknesses of students were selected from the local province— of Students at Entry. On the positive side there Milne Bay—and 50 percent from other provinces. was a general response that many students have Lae indicated that (“qualified”) nonschool leavers an improved ability to analyze problems, in part from remote areas of Morobe are given priority over because they are more mature (including those who others. Mendi (about to open) prioritizes the South- have done previous courses such as CHWs with ern Highlands as many facilities have closed due to Grade 12 and mature-entry students) and in part lack of staff. One institution (Lutheran) indicated because of an emphasis on grades and Mathemat- that they were beginning to recruit CHW gradu- ics/Science requirements. The weaknesses perhaps ates with Grade 12 and good recommendations offer more insights into the problems schools may from employers to their program. have teaching students. Negative issues enumerated (iii) Official Student Selection Guidance and Support include a problem with students enrolling and then Received from NDoH and or OHE. Generally, the dropping out after a few weeks, lack of motivation OHE provides direction for selection, including grade and bad attitudes/discipline of students (including point averages and lists of students giving the specific uncontrolled use of mobile phones), poor English school their first choice. The NDoH has only a lim- and writing skills, false recommendations and late ited role and does not advise on selection but does payment of student fees. pay some of the costs for school principals attend- (vi) Ability to Attract Female Students. All of the schools ing the selection conference (accommodation and responding to this question (six of seven) indicated airfares) while the OHE pays meal costs and some that there was no problem attracting female stu- boarding. One commented that there is no oppor- dents to their programs. Respondents were also tunity to discuss the numbers of students for which asked if safety issues for female students were a con- there will be scholarships even if the school feels it cern. Nearly all respondents indicated this was not a has the capacity. major concern. A number indicated that they gave (iv) Principals Perceived Quality of Students at Entry. students advice on precautions and or explained Quality of entry of students entering SoNs and their school rules and regulations associated with CHW training institutions in 2010 was rated on a this issue. One college indicated they required con- five point scale as follows: excellent 4; good 3; fair sent forms from parents/guardians before “out-of- 2; poor 1; and unsatisfactory 0. The overall results province” students could spend weekends locally seem to indicate principals have a reasonably good off campus. One facility (Nazarene in the Western view of student quality at entry with four principals Highlands Province) indicated that high commu- (of six) indicating quality of entry rated excellent or nity respect for their hospital and their students good. Principals were also asked to rate how student enhanced campus safety. quality was trending over time. Here a less flattering (vii) Qualitative Assessment of Nonstaff Inputs to Teach- picture emerged. Three principals believed quality ing and Curriculum Delivery. A wide range of fac- was improving while the remaining four believed tors can affect the quality of the teaching process. quality was declining. Some principals did acknowl- The survey covered principals’ views of the quality edge a considerable need for remedial work, partic- of the library, quality of teaching aids, number and Characteristics of Schools of Nursing and CHW Training Institutions   //  59 1305747 PNG HR Report 11-27-12.indb 59 11/27/12 11:38 AM quality of computers and quality of internet access placements were inadequate while four indicated (if any). With reference to the library one school they were adequate and another did not respond. indicated it was totally unsatisfactory (the new When the survey questionnaire was being devel- school at Mendi) and four indicated it was poor, oped, discussions with principals often centered on another rated it as fair and only one (Vunapope) the fragility of their capacity to adequately support indicated it was good. None indicated they had an the curriculum and that clinical/rural placements excellent library. and their supervision taxed their abilities to meet The quality of teaching aids was rated as curriculum requirements. being less than wholly satisfactory. One school (ix) Approval Authority for Curriculum and Princi- (Mendi) indicated they were unsatisfactory, three pals’ Assessment of the Quality of the Curriculum. schools indicated they were poor, two fair and one Each institution expressed a view of who is involved (Vunapope) good. With the exception of Vunapope in the approval of the standard curriculum—with with 36 computers available for teaching and 16 for a general consensus that the final authority is the the Lutheran school all the others had extremely low Nursing Council of PNG for the centrally approved (0–3) numbers of computers for teaching. With the curriculum. Nevertheless, a wide range of stake- exception of Vunapope—which rated the quality holders, including school governing councils, aca- of their computers as excellent (together with their demic boards of universities (where colleges are in access to the internet)—most other schools indi- the process of affiliating), and the NDoH (Curricu- cated their computers were of fair quality. Three lum Committee Training Division) were seen as key schools had no access to the internet, one had poor stakeholders. The overarching view of principals is access and two other schools rated the internet as that the curriculum is either good (5) or excellent fair. Overall this indicates that schools have a great (2). Suggestions to improve the curriculum include need for investments in basic efforts to support improving the connection between practice and the- teaching and the curriculum. oretical aspects of the curriculum, a need to review (viii) Assessment of Overall Ability to Teach Curriculum the curriculum to ensure what is implemented is and Adequately Support Clinical/Rural Placements. standard across the country, and a need for audit- Notwithstanding the apparent quality constraints ing bodies (Nursing Council, NDoH and universi- in support of teaching identified above, only one ties) to do annual school visits. school indicated their overall ability to teach the (x) Ability to Qualify for Professional Registration and curriculum was poor, another two indicated it was Date of Last School Evaluation Post 2000. All prin- fair, three indicated it was good and another (Vun- cipals who responded (six of seven) indicated that apope) indicated it was excellent. The tapestry of their student graduates could automatically qualify comments is worth noting. Three schools indi- for professional registration. Some schools noted cated getting good quality and qualified teaching administrative problems affecting registration while staff was an issue—a matter taken up in Chapter 4. one indicated there were processes to enable reme- Inadequate equipment to undertake clinical proce- dial actions for those who do not initially pass all dures was mentioned by two schools, including Lae subjects. Four schools had been evaluated in 2007 which reported an overall ability to teach rating of or 2008, Mendi in 2009 and another was evalu- poor. Vunapope, clearly a better resourced school, ated in 2004. One school, Vunapope, had not been indicated that adequate support comes from insti- audited or visited/evaluated by the Nursing Council tution and hospital administration for training and since the late 1980s. The Nazarene school has been that they had good relationships with specialist staff evaluated by the Nazarene Churches International in rural health facilities. Board of Education. Another, Lutheran, indicated Each school indicated they had access to hospi- it was reviewed in 2008, 2009, with a further review tals and rural facilities for clinical and rural place- in 2010 as part of the institutional affiliation pro- ments. Two schools indicated that clinical and rural cess with a university (Divine Word). One princi- 60  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 60 11/27/12 11:38 AM pal (Lae) calls for a Professional Registration Team 4.6.2 Nursing School Assets and to monitor schools annually and to provide feed- Infrastructure Quality back to make improvements. (xi) Professional Standards Reviews: Issues and Solu- (i) Land Ownership Levels and How Configured with tions Identified. Two schools indicated that they Other Institutions (if any). There was little clear did not have problems in reaching registration information available on how much land each insti- requirements for their graduates (Vunapope and tution owned—in part because most are on either Lutheran, although Vunapope has not been assessed mission land or hospital land. All schools except the for more than 20 years) while the remainder indi- Lutheran School are part of larger institutions—five cated they did face problems. A range of suggestions are part of hospital compounds. to improve student quality were made, including the (ii) Value and Condition of Land and Assets. Little cost of achieving the improvements. Suggestions reliable information was available on the value of included employment of additional staff (or filling land on which schools are located. Comments on of all vacancies—particularly for clinical nurses the state of electricity, water, sewerage and vehicles and/or clinical preceptors for field supervision, staff indicate that these assets are either very run down development programs), improvement of condi- or are nonexistent. For example, many facilities have tions of service (particularly since increased grad- inadequate water and electricity connections and ing of service-delivery staff) and adequate libraries many houses do not have connections. and teaching aids including computers and access (iii) Value and Adequacy of Offices and Buildings. Three to the internet. Two institutions (St Barnabas and schools valued their buildings in the range K0.5– Lutheran) identified infrastructure development K1 million while others were unable to adequately as key (costing K5-6 million. Excluding the infra- answer the question. Four schools have indicated structure, the total annual costs represented an addi- their buildings are inadequate while three indicated tional annual cost of about K500,000 per institution. they are adequate. Those indicating their build- It is noteworthy that principals are very focused on ings were inadequate blamed staff housing, student efforts which they feel will directly support teach- dormitories or major refurbishments while one of ing and education. These items have been starved those with adequate buildings noted that the need for funds over the past decade or more due to the for maintenance funding. squeeze on recurrent costs—particularly nonsal- (iv) Number, Value and Quality of Staff Housing. A ary recurrent costs. majority of schools indicated that they had less staff (xii) Information on Availability of Jobs at Graduation. housing than teaching staff although two facilities Principals’ knowledge of what happens to students at indicated they had more housing than teaching graduation is mixed. Overall, they estimated that a staff. By and large, principals were unable to value range of 50 percent to 100 percent had jobs at grad- the teaching staff housing for the survey. The qual- uation in 2009. Different schools had different pro- ity of teaching staff housing was rated as poor by portions going to government and church services three schools; satisfactory by three and excellent but one facility (Lae) indicated that most went to by one. The additional comments indicate much of private facilities and another (Vunapope) indicated the housing probably needs refurbishment and/or that almost 100 percent went to church health ser- replacement. vices. All facilities indicated that in the past there (v) Boarding Places and Their Quality. All institutions have been no problems with graduates finding jobs. are 100 percent boarding institutions. Clearly some It is, however, noted that delays in professional reg- schools have significant shortages of dormitories istration with the Nursing Council constrain the and/or they are in relatively poor condition. Two ability for graduates to find jobs immediately. This schools rate their dormitories as being in poor con- is probably the reason why many did not get jobs dition, one as unsafe, three as satisfactory and one immediately after graduation in 2009. as excellent (Vunapope). Although most indicate Characteristics of Schools of Nursing and CHW Training Institutions   //  61 1305747 PNG HR Report 11-27-12.indb 61 11/27/12 11:38 AM that their dormitories are safe for female students, 4.6.3 CHW School Student Entry a number still require adequate fencing. Those on Requirements, Quality of Curriculum and of institutions (such as hospitals) are often better off Graduates from a safety perspective. (vi) Amount Spent on Maintenance at Nursing Schools (i) Formal Entry Requirements: CHW Schools. Entry in 2009. For the schools reporting maintenance requirements according to the survey respondents budgets (four) in 2009 the average spend was vary somewhat between institutions but can be sum- about K40,000 with Mendi—being prepared for marized as: (i) Grade 10 or, increasingly, a Grade opening—spending K250,000. This was consid- 12 education with sound grades (upper passes) in ered significantly underfunded given the mainte- core subjects, including English, Mathematics, and nance backlog. Science; (ii) aged 18–25 years (some schools say (vii) Nurse School Teaching Staff Accommodation 20 years); (iii) a few schools emphasize good attitudes Arrangements. The number of staff living off school and/or one to three years living in the community. premises is a fraction of total staff numbers (less (ii) Application of Entry Requirements at Institution than 20 percent) except for the Highlands Regional Level. Principals were also asked how they apply the College which has a significant majority living off criteria for entry at the institution level. This pro- campus. A few colleges have staff in rented accom- duced some interesting results that indicate each modation. institution is also responding to perceived local needs (viii) Nursing School Administrative Staff Accommoda- and/or priorities within the region and/or with their tion Arrangements. A significant share of nursing overall mission service. Many institutions indicated school administrative staff lives off the school cam- they reserved some proportion of student places for pus—most live in rented accommodation. students from their province or neighboring prov- (ix) Nursing School Housing Rented Out for Rental inces. One institution (Kapuna) indicated that they Income Purposes. Housing rentals to nonstaff by were forced to accept local Grade 10s to boost the schools are not a significant source of revenue for local health workforce. Others indicated that being most schools although one facility (Lutheran) earns Christian and not married were given weightings K62,000 per year and acknowledges this helps with in the selection criteria. budget shortfalls. Some institutions, at least the gov- (iii) Official Student Selection Guidance and Support ernment-owned ones, collect rental incomes from Received from NDoH and/or OHE. The official salaries. The level of funds available from this source guidelines for entry are set out in the “Curricu- to schools remains uncertain. lum Information Handbook” (final draft October (x) Land Constraints to Nursing School Expansion. 2001) that was issued by the NDoH Curriculum School principals were asked if, assuming resources Development Unity. By and large, institutions seem were made available, land constraints would affect to report that they conduct the selection process the ability of schools to expand. Two schools (St with School Boards approving the final list, inde- Barnabas and Highlands Regional College) indi- pendent of NDoH (except where they are sponsor- cated they did not have enough land for the exist- ing students)—in this context NDoH approves the ing program. All others indicated they did. When list of students selected. The OHE does not have any asked if they could cope with a 100 percent expan- role in the selection process. sion Lae indicated there was a land dispute with (iv) Principals Perceived Quality of Students at Entry. their hospital, Mendi indicated they would need to Of the 11 schools admitting students in 2010,13 one build upwards and/or buy land elsewhere within principal assessed students entering as poor, four the township (indicating this new site may not be as good, four as excellent and two did not report. optimal) and Nazarene indicated they had enough A number of principals, including those rating stu- land but that the hospital would need to expand to cope with extra students. 13  Kapuna did not admit students in 2010. 62  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 62 11/27/12 11:38 AM dents as good, noted that remedial work with many three fair, four as good and one as excellent. Most students was needed. One school (Kumin), which schools only had one or two computers available for rated entering students as good, also indicated that teaching (one school had none, one had four and the selection process had been fraught with cheat- another six). The quality of these computers was also ing. When asked if the quality of students at entry seen by principals as variable. Seven of the schools was improving or declining, three indicated it was (over one-half) indicated they had no access to the the same, six indicated it was improving and one internet and those having access reported problems that indicated quality was declining. Interestingly, most of the time. one principal (Lemakot) claimed that Grade 12 stu- (viii) Assessment of Overall Ability to Teach Curricu- dents have more behavioral problems than Grade lum and Adequately Support Clinical/Rural place- 10 entrants. ments. Principals had a relatively high assessment (v) Principals Perceived Strengths and Weakness of Stu- of their schools ability to teach the curriculum. Two dents at Entry. On the positive side, there is general schools rated their capacity to teach the curriculum agreement that many students have an improved as excellent, six as good and two as fair (two did not ability to analyze problems, in part because they report on this question). Clinic placements are an are more mature (including those who have done important part of the curriculum. A vast major- previous courses such as CHWs with Grade 12 ity (nine of ten) reporting schools indicated they and mature-entry students) and in part because of were able to mount adequate clinical placements. an emphasis on grades and Mathematics/Science It is apparent from general comments that schools requirements. The weaknesses perhaps offer more have a good relationship with nearby service pro- insights into the problems schools may have teach- viders for student placements. On the other hand, ing students. Negative issues enumerated include a transport and staff for field-based supervision is a problem with students enrolling and then dropping constraining factor. out after a few weeks, lack of motivation and bad (ix) Approval Authority for Curriculum and Princi- attitudes/discipline of students (including uncon- pals’ Assessment of the Quality of the Curriculum. trolled use of mobile phones), poor English and By and large, schools responded that they teach the writing skills, false recommendations and late pay- standard curriculum and a number of schools have ment of student fees. mechanisms to oversee its implementation. The (vi) Ability to Attract Female Students. A significant overall rating of curriculum quality by principals majority of responding principals (eight of ten) was positive—with four indicating it was excellent, indicated that they had no major problems attract- six as good and one as fair. Observations on how to ing female students. Of the two schools indicating improve the curriculum included removal of repet- it was a problem, one (Tinsley) indicated that secu- itive content, reduction of theory relative to prac- rity and safety was an issue. Others acknowledged tice; more external auditing of the curriculum, and security as an issue but it is clear schools have strat- more staff. egies to alleviate safety concerns—including work- (x) Ability to Qualify for Professional Registration ing with local communities. and Date of Last School Evaluation Post2000. All (vii) Qualitative Assessment of Nonstaff Inputs to Teach- responding principals indicated that graduates could ing and Curriculum Delivery. Principals were asked qualify immediately for professional registration. to rate key quality inputs to the teaching program. Responses to the question about the number of eval- Of the ten who reported on the quality of the library, uations for professional registration since 2000 and one did not have a library, another indicated it was the time of their last evaluation revealed that three totally unsatisfactory, three indicated it was poor, schools had not been evaluated since 2000 and that a one rated it as fair, two as good and a further two felt further four had only been evaluated once, one had it was excellent. With respect to views on the qual- four evaluations and another nine. The last school to ity of teaching aids, one indicated they were poor, be evaluated was in 2008 (after more than 15 years) Characteristics of Schools of Nursing and CHW Training Institutions   //  63 1305747 PNG HR Report 11-27-12.indb 63 11/27/12 11:38 AM and only one other had been evaluated since 2006. health centers or on a mission compound. Thus they Given that the revised curriculum started in 2006 typically do not own the land themselves. this means there has been little review of the cur- (ii) Value and Condition of Land and Assets. The rent curriculum. information collected on land values is problem- (xi) Professional Standards Reviews: Issues and atic—largely because schools are part of larger Solutions Identified. Three principals indicated compounds owned by other institutions. The value that they did not have trouble meeting registra- of plant and equipment reported averages about tion requirements, while the other five respond- K275,000 per school. Quite a few schools indicated ing principals indicated they did have problems that basic infrastructure services (sewerage, water meeting requirements. The lack of workplace stan- and electricity) were a significant issue. While a few dards, equipment and facilities was cited by one schools report having vehicles for carrying students, school, shortage of staff by another and one men- most are eight to ten years old. tioned community social unrest as an issue. There (iii) Value and Adequacy of Offices and Buildings. The is a rich array of suggestions to improve student average estimated cost by principals who responded graduate quality, including an additional period of to the question of the value of offices and buildings practical training for students, staff development, was K688,000—with a majority indicating a value improved teaching materials (including comput- of K500,000 or below. Five of nine responses indi- ers and the internet), infrastructure (including cated that offices and buildings were adequate and houses, dormitories and facilities), transport, and four indicated they were not. Quite a few principals additional technical staff. were concerned at the level of maintenance required (xii) Information on Availability of Jobs at Gradua- to maintain very old buildings. tion. Principals report that the vast majority of stu- (iv) Number, Value and Quality of Staff Housing. On the dents got a job at graduation (75–100 percent) in basis of responses, about one-half of teaching staff 2009. It is noteworthy that 50 percent of graduates are housed by schools. Some have all staff housed of Kapuna went on for further training. If this is an and others have very limited stock. The value of emerging trend it could have implications for plan- staff housing reported is problematic with the aver- ning of student numbers. Government and church age staff housing being reported at K87,000 (from health services are the major employer, but at Kapuna a response covering 24 houses). The overall picture the principal noted that 20 percent of the gradu- on the quality of staff housing is that it is mixed— ates went to the private sector. The response to the with half of it being reported as either satisfactory question: “Have past graduates had trouble getting or excellent. a job” did not suggest there were major unemploy- (v) Boarding Places and Their Quality. All students are ment problems for graduates, but did suggest some boarders and live in school-owned dormitories. Five graduates could not go to rural services because of institutions rate the quality of dormitories as poor, a lack of vacancies and instead ended up working including two that also indicate they are not safe. in hospitals. Another school emphasized that there Five have responded that the dormitories are satis- were clear vacancies in remoter rural services and factory. A few schools are concerned about safety graduates went there. of female students. (vi) Amount Spent on Maintenance in 2009. The reported 4.6.4 CHW School Assets and Infrastructure average spent on building maintenance was about Quality K30,000 per school. The quality of administration and teaching buildings was rated as unsafe by one (i) Land Ownership Levels and How Configured with (St Gerards), partly poor or satisfactory/unsafe by Other Institutions (if any). All responding schools two others and the remaining five as satisfactory. indicated that they were part of another institution (vii) Teaching Staff Accommodation Arrangements. or compound—some are close to mission hospitals/ The vast majority of teaching staff are housed at the 64  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 64 11/27/12 11:38 AM school/host compound, with a handful of staff rent- 4.7. A Brief Conclusion to Chapter 4 ing accommodation off campus. (viii) Administrative Staff Accommodation Arrangements. This chapter has presented detailed information of The vast majority of administrative staff are housed costs and operations of nursing and CHW training at the school/host compound with a handful of staff schools. Some implications of the information presented renting accommodation off campus. in this chapter are drawn out in Chapter 7 which discusses (ix) Housing Rented Out for Rental Income Purposes. strategy and options to develop the health training sys- There are only nominal amounts earned from house tem in PNG to meet human resource requirements. The rental income at a few schools. next chapter will look at anticipated demand for health (x) Land Constraints to School Expansion (E2). All workers over the next 10 and 20 year periods based on responding schools indicated that there was enough assumptions in the NHP and the subsequent chapter land for existing programs and, with the excep- will look at the balance between existing supply (as dis- tion of one school, that there was enough land to cussed in Chapter 3) and the various demand scenarios at least double the program (assuming money was discussed in Chapter 5. made available. Characteristics of Schools of Nursing and CHW Training Institutions   //  65 1305747 PNG HR Report 11-27-12.indb 65 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 66 11/27/12 11:38 AM CHAPTER 5 Future Staffing Demand Scenarios for PNG’s National Health System 5.1. Introduction areas—will be reversed through higher per capita use of services—a not unreasonable assumption. It is also This chapter explores factors which may affect the noted that neither the NHP nor other reports on PNG demand for health service-delivery staff over the com- health service delivery have suggested that there should ing two decades. First, it notes that there is an expecta- be a major revision of the structure of health cadres tion that resources available to government and the health within the health system—although the NHP does sug- sector will be significantly better than in the immediate gest there should be an expansion of “rural doctors” by past. On the other hand, it is still too early to be sure what 40 per annum. The chapter also notes that there has been the full revenue implications will be to government of the considerable discussion over the years of whether HEO planned LNG projects and other potential major develop- cadres should be maintained or replaced by “Nurse Prac- ments. The NHP forecasts the population to grow at about titioners” or some similar cadre. 2.8 percent per annum over the next decade to 2020. The Five demand and supply scenarios for direct ser- PNGDSP expects population growth to reduce to about vice-delivery health staff are summarized for the period 2.5 percent over the period 2020 to 2025 and to 2.1 per- 2009 to 2030. Scenario 1 demonstrates that there is a cent between 2025 and 2030. health human resource crisis given expected popula- Population projections form the basis for all the sce- tion growth rates, expected workforce retirements and narios for demand and supply of health staff discussed the existing supply (training capacity) of the four main in this report. The PNGDSP forecast for declining pop- direct service-delivery cadres. Scenarios 2–4 present ulation growth is, however, probably ambitious unless three different demand for health staff scenarios, includ- decisive action is taken on the family planning front and ing: (i) the “aspirational targets” for service-delivery staff on education, particularly of girls. The NHP proposes to proposed by the PNGDSP; (ii) maintaining current popu- turn the recent significant decline in outpatient services lation to service-delivery staff numbers; and (iii) the sug- per capita in rural areas around. This will be achieved by gested WHO “Threshold” service-delivery staff numbers a variety of actions but, strategically, the proportion of required to achieve the MDGs. The recommended sce- the budget available for service-delivery staff will decline nario—Scenario 5—presents a suggested expansion of and that for nonsalary budget quality-enhancing expen- service-delivery staff which is affordable given the likely ditures will increase—albeit of an expanding budget in resources available for health. This scenario also suggests real terms per capita. there be a modest change in the structure of direct ser- The NHP envisages that the current low work- vice-delivery cadres. These are then discussed in more loads of direct service-delivery staff—at least in rural detail in Chapter 6. 67 1305747 PNG HR Report 11-27-12.indb 67 11/27/12 11:38 AM 5.2. Key Aspects of Demand for Direct cent of the rural population has access to safe water and Health Service Providers 42 percent to sanitation. While the situation in urban areas is better–88 percent have access to safe water and 67 per- 5.2.1 Historic Government Recurrent cent to sanitation facilities—anecdotal evidence indicates Resource Constraints that the situation in urban squatter settlements is deterio- rating and that urban settlements are growing faster than It was observed in Chapter 3 that real recurrent gov- official records indicate at face value (see below). ernment expenditure since independence has declined The most recent Demographic Household Sur- 45 percent per capita (from 1993 to 2009 it declined vey (2006) highlighted a key feature of the crisis in the by one-third). The health sector was not protected from health system—with results indicating that the MMR had these very significant reversals of fortune. Between 1997 climbed to 773/100,000 live births (from 370/100,000 and 2004 real recurrent expenditures on health declined live births a decade ago).16 UNICEF estimates that the 20 percent while the population continued to increase at average MMR for developing countries is 450. As the NHP around 2.8 percent per annum. Goods and services expen- notes: “This ranks PNG as second highest in the world in ditures and recurrent expenditures on capital—critical for maternal mortality, after Afghanistan and outside Sub- the quality of services—fell 27.4 percent and 77.4 percent Saharan Africa. The main causes of deaths related to preg- respectively over this period. Chapter 3 also notes that nancy are prolonged labor and excessive bleeding. A safe government real recurrent expenditures on nurse train- and accessible delivery environment could save many lives. ing facilities are around 25 percent of what they were in The risks for maternal deaths have increased due to the the 1980s. In short, the health system has been, and still still high fertility levels (many children) and shortened remains, in crisis. This, as discussed in all major reports on birth intervals over the past decade. The neonatal death the PNG health system in recent years, has been—together rate (within one month of birth), frequently the result of with dysfunctional national-provincial relations—the major poor maternal health and the delivery environment, has cause of the well-documented decline in health outcomes shown little change over the past decade.” in PNG over the past two decades.14 This is a significant The national population growth rate, as noted, part of the background setting for the NHP and the dis- remains high at 2.7 percent per annum (2.8 percent for cussion of future demand scenarios for health workers. the citizen population)17 due to the total fertility rate which remains high at 4.4 births per woman. The con- 5.2.2 Health Outcomes and Revealed traceptive prevalence rate is only 32 percent of the repro- Demand for Health Services The NHP notes that life expectancy at birth remains low 14  See, for example: World Bank, Asian Development Bank and AusAID. 2007. “Strategic Directions for Human Development at 57 years and recognizes that efforts to increase life in Papua New Guinea”. expectancy significantly over the coming years may well 15  The national population-based HIV Bio-behavioral Survey be severely challenged by the specter of the further sus- led by the World Bank in partnership with the NDoH, National tained spread of HIV. Consensus estimates indicate that AIDS Council Secretariat and other development partners cur- rently being implemented after years of earlier resistance by a HIV-positive cases may well represent 1.6 percent of the number of development partners, WHO and the UN System population aged 15–59 years.15 The IMR remains high at will provide the first national population-based estimate of the 57/1,000 live births as does the Child Mortality Rate (CMR) disease and its drivers. 16  There has been some discussion of technical aspects of these of 75/1,000 live births—although each has improved 20 per- estimates which suggest the 2006 estimate may be somewhat cent over the past decade. The NHP notes: “Preventable and overstated and that the 1996 estimate was somewhat understated treatable diseases, including malaria, pneumonia, diarrhea, but no one has suggested that these numbers are not indicative tuberculosis, HIV, and neonatal sepsis remain the most fre- of the trend or are inconsistent with other information avail- able—the MMR is unacceptably high. quent causes of childhood deaths. Adequate space between 17  Citizen population means nationals of PNG. Total popula- births provides for greater survival rates of children”. Risk tion includes the noncitizen population residing in country at factors to infectious disease are also significant—only 32 per- the time of the census. 68  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 68 11/27/12 11:38 AM ductive age group. Some 60 percent of expectant mothers person per annum declining from 2.39 in 1988 to 1.49 have at least one antenatal visit but a significant majority in 2003. According to the NHP, outpatient visits per cap- have two or antenatal visits or fewer and very few have ita per annum in rural areas on average for the years 2007 the recommended four antenatal visits. Postnatal visits are to 2008 are only 0.88.18 While the urban population may also low and only 37 percent of mothers have a supervised have been growing much faster than overall population delivery (it was estimated at 52 percent in 1991). Under growth (see below) and many rural dwellers are likely to the design of the service delivery system, CHWs are the be bypassing rural facilities to attend urban facilities, it is first point of contact but this cadre (as discussed in Chap- unlikely that total outpatient visits per capita per annum ter 3) does not have any competency in dealing with birth for the population in rural areas has not continued to complications and emergency obstetric care and basic level decline. In addition to ambulatory care, the MTEF back- nurses also have very little training related to birth com- ground tables indicate that there were an additional 0.44 plications and emergency obstetric care. This is a glaring visits per capita to health facilities for deliveries (84,400 health staffing issue as this means only a few doctors and visits); antenatal care (198,900 visits), child health visits a handful of nurses with postgraduate training in rural (1,474,000); and immunizations (2,203,000). Thus, overall areas (less than 250 doctors and nurses in total) are avail- visits to rural health facilities, for all purposes, is less than able and qualified in rural areas to deal with birth com- 1.4 visits per capita per annum. A well-functioning health plications and emergency obstetric care to deal with the system, with the disease profile of PNG could expect 2.5 more than 5 million people living in rural areas. to 3.5 visits per capita per annum just for ambulatory care. Mother and child nutritional issues remain a cause According to the NHP, for the years 2007–2008 the of concern. Ten percent of babies have a birthweight of most common single reason for an ambulatory care less than 2,500 grams. Some 26 percent of children under visit was malaria (29 percent of the total); skin disease five are underweight. Meanwhile, immunization rates do (10.8 percent); simple cough (9.7 percent); pneumonia not reach levels likely to give populations adequate herd (7.7 percent); diarrhea (4.4 percent); and other respira- immunity—only 61 percent of children get the third dose tory conditions (4.1 percent). Overall, the top six condi- of the Diptheria, Pertusis, and Tetanus vaccine and 62 per- tions leading to ambulatory care visits account for about cent get a measles vaccination—well below the desirable 66 percent of all visits. The next largest category—acci- target of 80 to 90 percent. The NHP also highlights the dents—accounts for an additional 3.4 percent of visits. It significant differences that exist between provinces for is interesting to note that yaws, pertussis and measles—the almost all indicators. Chapter 2 demonstrates the ineq- first easily treated and eradicated by antibiotics and the latter uitable distribution of staff across provinces. two preventable through vaccination—are among the top 15 One very important indicator of the effective causes of ambulatory care visits. This fact further underpins demand for health services is outpatient visits per capita the notion of fragility in the health status of the population. per annum. This is an indicator of the overall use of the To round off this discussion of the demand for health health system given the state of the health system (fund- services, the NHP records that hospital admissions ing, staffing, pharmaceuticals and other medical supplies) resulted in 1.5 million hospital bed days in 2008. The and the disease burden of the population. As discussed, largest cause of admissions was normal delivery (14.8 per- recurrent resourcing of the health system has been falling cent of the total); tuberculosis (12.5 percent); accidents considerably. The disease burden has not changed signif- and injury (10.5 percent); pneumonia/Acute Respiratory icantly—with the significant exception of the emergence Infections (ARI) (9.1 percent); malaria and other vector- of HIV and a growing importance of some lifestyle dis- borne diseases (6.8 percent); obstetric and maternal con- eases. There is little doubt that ambulatory care visits per ditions (5.2 percent); perinatal conditions (5.1 percent); capita has been decreasing and that this, without a rever- and diarrhea and enteric conditions (4.2 percent). In other sal, will be having a negative impact on health outcomes. 18  The data collected for urban areas (mainly hospitals) is unre- The existing disease burden does not suggest that liably reported and as such is not reported. Unfortunately we demands on the health system should be falling on a do not have access to rural data for the years 1998 and or 2003 per capita basis with the number of outpatient visits per to make a direct comparison. Future Staffing Demand Scenarios for PNG’s National Health System   //  69 1305747 PNG HR Report 11-27-12.indb 69 11/27/12 11:38 AM Table 5‑1: Official Population Estimates and Projections 2000–2030 (Selected Years ‘000) Source 2000 2009 2010 2015 2020 2025 2030 PNG DSP 2010–2030 5,191 6,637 – 7,846 8,090 8,950 9,860 MTDP 2010–2015 5,191 – – 7,280 – – 9,880 NHP 2010–2020 5,191 – 6,829 7,846 9,025 – – Growth Rates 2.8 – 2.8 2.8 2.8 – – Mission Projections – – – – 9,025 10,211 11,329 Growth Rates – – – – – 2.5 2.1 Source: Papua New Guinea NHP 2011–2020 and the Papua New Guinea Development Strategic Plan 2010–2030. words over two-thirds of hospital days are accounted for The NHP presents population projections for the by the top eight conditions. period 2010 to 2020. These essentially assume the citizen In summary, health outcomes in PNG have seen population will grow at about 2.8 percent over the period very little improvement over the past 35 years and evi- to 2020. These may well be a little pessimistic—that is dence from the past decade indicates extreme fragility population growth rates may well start to decline before of health outcomes, a rising MMR and sustained pop- 2020.19 This report has, nevertheless, used these popula- ulation growth. New diseases, including HIV and life- tion projections as the basis for the report. For the period style-related diseases are also emerging. This has happened 2020–2030 this report has assumed that in these outer years while government recurrent expenditures on health have the rate of population increase will start to decline along declined rapidly in real terms per capita—including key the lines assumed in the PNGDSP as shown in Table 5-1. expenditures on quality-enhancing nonsalary budgets. The population in 1980 was estimated by census at There has also been a dysfunctional national-provincial 3.1 million and 5.2 million in 2000—an average popu- relations arrangement bedeviling management of health lation growth of 2.8 percent over the 20 year period to staff and of finances for health more generally. Demand 2000. Thus a key driver or component of the decline in for health services—as measured by ambulatory care has real per capita recurrent expenditure on health discussed fallen steadily—at least in rural areas. above was the rate of population growth itself—not just the decline in recurrent resources available to government 5.2.3 Population Growth Notes: Projections, (in large part caused by the unplanned closure of the Bou- Rural-Urban Distribution and Mobility gainville copper mine and in the rapid decline in budget support from Australia to support recurrent expenditures). The total population of PNG has been growing, as dis- Another issue relevant to the discussion of future cussed, at 2.7 percent per annum—the citizen popu- demand for health services is the distribution of pop- lation has been growing at 2.8 percent from 1980 to ulation between rural and urban areas and between 2000. Table 5-1 presents actual census data for 2000 and provinces. Official census results and most official plan- estimates of the current population in 2009 and 2010 and ning documents suggest that PNG has not rapidly urban- projection for the years through 2030 from either official ized—that is that the urban population has not been rising planning documents or projections undertaken for this substantially faster than the total population. Census esti- report. One set of projections is presented in the PNGDSP mates suggest the urban population has grown by 1 per- 2010–2030—this shows the population growth rate being reduced quite significantly over the next few years (as a 19  Population estimates in PNG have been mired in problems result of interventions in the health system and due to a with the census estimation arising from the sheer magnitude of the effort required to implement a census in PNG—particularly decisive national population policy. These assumptions the 1990 census. There has been very little effort to systemati- have also been used to underpin the population estimates cally project population growth and trends in PNG by govern- used in the Medium-Term Development Plan. ment, academics or development partners. 70  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 70 11/27/12 11:38 AM cent per annum over the period 1980–2000. The 2000 at demand and population by province; rather it makes census reported that only 13.2 percent of the population the projections at a national level and makes suggestions resided in urban areas. It is probable that the 2000 census in Chapter 7 on how decisions should be made in allocat- has underestimated the urban population. The key reason ing staff between provinces. It also argues that the core of is likely to be because urban census boundaries were not this exercise should be updated annually to take account adjusted for the expansion in urban boundaries. As a con- of new trends as they emerge. sequence, recorded urban growth does not really include The information available on population mobility all those who really live in urban areas.20 in PNG is poorly documented and studied. When the Casual empiricism of population trends in the larger PNG health system architecture was established in the late PNG towns of Port Moresby, Lae, Mount Hagen and 1960s and 1970s the primary focus was primary health the emergence of new enclave developments through- care (before the Health for All declarations) delivered by out the country strongly suggest the urban population health centers, subcenters and aid posts—the latter staffed in 2000 was underestimated. Growth trends since 2000, by aid post orderlies who are now CHWs. At that time notwithstanding sluggish economic growth over most of cash cropping and access to cash incomes was quite con- the past decade, have also probably been very significant. strained—with less than 20 percent of the population being Port Moresby has certainly grown significantly over the significantly dependent on cash for their incomes. In this last few years. If the urban population has been growing context the population was not mobile and access to health at 3 percent per annum—only marginally faster than the services depended on services being located very close to total population—over the past decade then the urban pop- where they lived. For significant sections of the rural pop- ulation would be about 925,000 or about 13.2 percent of ulation this remains true. Nevertheless, cash cropping and the total population (notwithstanding the 2000 estimate informal employment opportunities have increased sig- is probably an underestimate). If, however, it has been nificantly—in short, market penetration into the lives of growing at say 5 percent (a not unreasonable conserva- rural communities has increased significantly. To partic- tive assumption) the urban population could have reached ipate in this commerce it is clear that growing sections of over 1.1 million—or about 16.2 percent of the total pop- many communities gather regularly at markets and other ulation. If this trend continued over the next decade the meeting sites throughout the country—often by the road- urban population could reach 1.8 million or 22 percent side and most visibly in the major towns. of the total population. If the 2000 estimate of the urban This trend of greater population mobility has been population is an underestimate then it is entirely feasible happening notwithstanding the near collapse of a sig- that PNG could have one-quarter of its population living nificant proportion of the rural road and coastal trad- in urban areas by 2020—by 2030 if urban areas grow at ing (shipping) infrastructure that used to ply the copra 5 percent per annum. Under this scenario the urban pop- and related trade in the maritime provinces. As we look ulation could reach about 3 million or one-third of the forward, government planning documents have a set of population. Trends of this nature, if they turn out to be policies designed to expand access to incomes and this realized, will bring new challenges and opportunities for will increase the daily mobility of the population. This health service-delivery strategies. raises the possibility for the health system to rethink and The forthcoming census will be critical for gaining a better understanding of the movement of the popu- 20  It is hoped that the census currently in advanced stages of lation as between provinces. The census results for 1980 planning looks very closely at the definition of urban. It would and 2000 indicate that over this period 11 provinces grew also be helpful if the census also carefully looked at the option of rural enclaves (areas which are not defined as urban but are set- at a rate slower than the average and that another two are tlements around commercial agriculture or industry supporting the same as average and seven grew faster than average. rural commercial activities and are not rural in the commonly It is not clear how much these trends can be extrapolated thought of way. The 1980 and earlier census’s had such a defi- nition. Often these areas become significant markets on some into the next two decades with any accuracy. Neverthe- day(s) of the week and many rural dwellers come to them on a less, the NHP presents population estimates based on very regular basis. This could as discussed below lead to a differ- these historic trends to 2015. This report does not look ent differentiated strategy for delivery health services. Future Staffing Demand Scenarios for PNG’s National Health System   //  71 1305747 PNG HR Report 11-27-12.indb 71 11/27/12 11:38 AM diversify its supply response to service delivery and supply scenario which assumes that GDP will grow at an average key services more centrally—at least in some areas. While of 5 percent in real terms over the period of the projec- serving populations in this way offers some important tions to 2030. This is significantly faster than the first sce- opportunities of how best to serve population it would, nario and would see per capita GDP almost double over of course, not be a reason to abandon the need to con- the period—assuming the likely trends in the population tinue to provide first line health services. A clear focus on growth rate discussed above. The third scenario is based referral systems may, however, achieve significant gains on the PNGDSP 2010–2030 which posits an “aspirational” in service-delivery quality delivered through larger facil- real GDP growth rate of 8.4 percent per annum over this ities with multiple cadres (allocated according to work- period. This would radically transform the economy and load). Larger facilities typically make it easier to create a see GDP per capita grow at almost 5 percent per annum. professional work environment for service-delivery staff. The three scenarios (projections) of real GDP growth are presented in Table 5-2 and discussed more fully below. 5.2.4 Resources Available to Health: the The first scenario is fairly conservative but still Future shows GDP doubling in real terms over the period to 2030. It also shows that if the total government bud- Resources available for health (and the government get maintained its current share of GDP (about 30 per- more generally) are expected to increase significantly cent) it would also double over the period to 2030. If the over the next five years and beyond to 2030. In order to recurrent budget for health maintained its current (2009) estimate a reasonable resource envelope for the health sec- share of total government expenditure of about 14.7 per- tor over the period of the health human resource projec- cent then the recurrent budget for health would increase tions, three different scenarios have been developed. The by 100 percent. This would see only marginal changes in first is based on “historical trends” and assumes that real per capita GDP or recurrent health expenditure—thus GDP will grow at about the historical average of 3 per- making it extremely difficult to dramatically improve cent—marginally above the rate of growth of the popu- health services and population to health service-delivery lation. The second is a “central” or “medium” projection staff ratios. As discussed below, however, there is now an Table 5‑2: Potential Resource Scenarios for Public Expenditure on Health (Real 2009 Prices in ‘000 Kina) Growth Scenarios 2009 2015 2020 2025 2030 Gross Domestic Product Historic Trends 21,800 32,420 34,748 38,553 42,775 Central Trends 21,800 34,305 40,738 50,283 62,142 PNGDSP “Aspirational” 21,800 34,635 50,185 76,863 117,722 Total Budget (30 Percent of GDP) Historic Trends 6,540 9,726 10,424 11,566 12,833 Central Trends 6,540 10,292 12,221 15,085 18,643 PNGDSP “Aspirational” 6,540 10,391 15,056 23,059 35,317 Health Recurrent Budget (14.7% of Total Budget) Historic Trends 958.7 1,426 1,528 1,700 1,881 Central Trends 958.7 1,509 1,791 2,211 2,733 PNGDSP “Aspirational” 958.7 1,523 2,207 3,380 5,177 Source: Mission estimates. 72  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 72 11/27/12 11:38 AM expectation that PNG is at an important crossroads and it the MTDP. On the other hand, it recognizes that the plan- is reasonable to assume that government can expect rea- ning period for this exercise will more than likely be one sonable growth in the economy and thus an expanding which is a decisive shift from the past—historic trends budget and health budget. when per capita growth of both GDP and government The question is how fast the economy is likely to budgets were almost zero.21 grow given current and expected developments in the The third scenario—the PNGDSP 2010–2030 sce- economy. There has to be considerable uncertainty about nario—is very optimistic and should perhaps be better the future—in GDP growth rates, growth of the govern- considered an “aspirational” scenario. This scenario posits ment budget and in the share of the budget devoted to that GDP will grow at an average of 8.4 percent per annum health and, in particular, the recurrent health budget. over the period to 2030 and that this will result in more This in turn strongly influences—in fact it determines to than a five-fold increase in real GDP over the next 20 years. a very large extent—the number of health staff who can be This would mean that per capita GDP would increase sig- employed and thus the number of health staff who need nificantly given that population is only expected to increase to be trained. It is critical that the numbers being trained about 45 percent over this period. This scenario also sug- and are able to be employed is fiscally and economically gests that government revenues should maintain the equiv- sustainable. To this end the discussion of the likely growth alent of 30 percent of GDP—the current average—and that, scenario is critical to the recommendations in this report. as a consequence, government expenditure (recurrent and The second scenario—the “central” or “middle” sce- capital) can be expected to grow at about the same rate as nario—is one that is between the historic trends presented GDP. If health expenditures (not discussed except for that as Scenario 1 and the PNGDSP “aspirational” targets implied by the significant increases in staff posited) were presented in Scenario 3. This scenario (as for Scenarios to be maintained as a similar share of total government 1 and 3) recognizes that the economy will grow signifi- expenditure then health expenditure could be expected to cantly through 2015 as a result of major projects currently also increase at the same rate as GDP.22 under way. The MTDP and the NHP documents both dis- 21  It is a central recommendation of this report (see Chapter 7) cuss the medium-term fiscal future (compared with the that exercises such as this should be conducted annually. This long-term future which must be subject to greater levels of will enable trends in the economy and of governments fiscal and uncertainty) with implications for the resources likely to be development policies (including priorities for health) to be fac- available for health. The MTDP assumes economic growth tored into health workforce planning. 22  It is not clear that government revenues (and thus long-term (GDP) will average 8.5 percent per annum over the period expenditures) will be able to rise as fast as GDP under either 2010 to 2015 fueled in significant part by investment in this or the second scenario discussed below. A detailed analy- the LNG project. This compares to GDP growth of about sis of this is well beyond the scope of this report. However, there 7 percent over the previous four year period. The LNG proj- is a considerable downside risk that government revenues will not rise quite as fast as real GDP. This would mean that govern- ect is expected to come on line in 2014 enabling reason- ment might not be able to maintain government expenditure at able sustained growth through subsequent plan periods. 30 percent of GDP—it will depend on government revenue and The impact on government revenues will largely taxation policies and on the nature of tax expenditures (breaks) depend on the elasticity of government revenues with granted to the major companies developing oil, gas and mining projects. On the other hand, the overall priority given to health respect to GDP growth and be significantly determined will determine the share of the government budget available to by the fiscal impact of the LNG project—which is as health. The scenarios assume that the share of the health recur- yet not fully understood. The challenges to government rent budget will remain the same. It could vary either way and of sustaining a stable fiscal environment are likely to be will also depend on the share of the health budget which is allo- cated to recurrent spending and which to development expendi- significant given volatility in global commodity prices tures. For human resource planning purposes the health recurrent and exchange rate movements. To this end, Scenario 2 budget is critical. The NHP documents a huge backlog of capi- assumes that growth will fall back to about 4 to 4.3 per- tal expenditures over the next 10 years. To this end there will be pressures to constrain recurrent expenditures. Thus the assump- cent per annum over the subsequent period to 2030—in tions used to maintain the budget share of GDP and for health recognition that the growth rates posited under Scenario recurrent budget to maintain its share is a reasonably optimis- 3 must be much less certain than for the earlier period of tic set of assumptions. Future Staffing Demand Scenarios for PNG’s National Health System   //  73 1305747 PNG HR Report 11-27-12.indb 73 11/27/12 11:38 AM Other important and positive trends are noted average of K456 million over the period 2016–2020—an within the health budget. The NHP notes that total health increase of 23 percent or 4.6 percent per annum. This recurrent expenditure increased by 65 percent in nomi- implies maintaining the current average population to ser- nal terms between 2007 and 2010. With inflation running vice-delivery staff ratios that exist at present (see Chap- at about 6.5 percent per annum (about 26 percent over ter 2)–significantly below those implied by the PNGDSP this period) real health recurrent expenditure increases and the MTDP issued by the DNP&M and discussed in of about 3 percent have been achieved. At the same time, more detail below. expenditures on personnel, including on church health services, and provincial general hospitals has increased almost 60 percent while overall expenditures on opera- 5.3. Projecting the Demand for Health tional costs for rural health services has doubled between Cadres: Five Scenarios 2007 and 2010. This has been enabled by the changes to the intergovernmental financing which has seen a three-fold The current stock of health-related human resources increase in health functions grants to provinces over this is fixed, consisting primarily of doctors, HEOs, nurses period. This is a substantial reversal of trends identified (general and specialty nurses including midwives) and in the late 1990s and first half of the first decade begin- CHWs. The aggregate staff that can be employed in the ning in 2000 and which were responsible for a significant future will be constrained by the resource envelope avail- part of the collapse of rural health service delivery. Nev- able to health (the personnel budget) and the mix of cad- ertheless, the NHP clearly recognizes that there is a sus- res employed given their price. The NHP has focused on tained need to ensure more effective usage and allocation only modest increases in staff for the period 2010–2015 of financial resources over the next decade. and slight expansion in numbers through 2020 because of: The NHP estimates that personnel emoluments need (i) a view that real resources available for health will be rel- to increase from K371.3 million in 2010 to K417.4 mil- atively constrained over this NHP plan period; (ii) a recog- lion in 2016 in real (2010) terms—an increase of 12.4 per- nition that the health system needs a sustained real increase cent over five years or about 2.4 percent per annum—not in quality-enhancing nonsalary budgets to enable staff to quite keeping pace with estimated population growth. deliver services. This has started but it remains clear that The NHP provides estimates for the costing of the plan insufficient resources are yet allocated; (iii) infrastructure through 2020 (Table 5-3). This indicates that person- requirements—particularly refurbishments—are needed nel costs will increase from K371.3 million in 2010 to an because of an historic neglect of maintenance of existing Table 5‑3: Costing of the NHP 2010–2020 (Real Average per Year)(Millions of Kina 2010 Prices) Input Items 2010 2011–15 2016–20 2011–20 Personnel 371 397 456 426 Medical Supplies 149 177 218 197 Operating Costs 382 410 502 456 Total Recurrent 902 984 1,176 1,079 Capital 23 356 319 337 Total Expenditure Required 925 1,340 1,495 1,416 Funding Available 925 925 925 925 Government 631 631 631 631 Development Partners 294 294 294 294 Funding Shortfall 0 415 570 491 Source: Papua New Guinea NHP 2011–2020. 74  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 74 11/27/12 11:38 AM buildings and only limited new infrastructure over the age, while removing a retained placenta requires an emer- past decade; and (iv) a recognition (particularly following gency obstetric services package. the recent “Monash Study” on the operations of the rural Broadly speaking the existing operational struc- health system)23 that the existing complement of staff in ture and health cadres can deliver these technical the rural health facilities was significantly underutilized options—assuming they are appropriately financed (a and could handle (based on staff time available and used key objective of the NHP)–and the referral system works with existing demand) on average an additional 40 per- as envisaged. It would be a sound idea to carefully review cent throughput.24 In significant part this has resulted from this table and to make sure staff have the skills required the sustained decline in ambulatory care visits per cap- and that the health system delivers the right inputs to ita and a failure of the other services (such as MCH and enable this to happen. In this context, future planning immunization) to also be delivered at the desirable level. has to recognize that CHWs and nurses will provide the Broadly speaking, the NHP is seeking to improve first level health services for at least the next 15 years, irre- the productivity of the health system’s human resources. spective of whatever option for future staff development This means higher visits per capita per annum can be is adopted (see below). It will take at least this period–15 achieved without major expansion of staff.25 The NHP to 20 years—to produce enough doctors to place a doctor notes: “Overall service improvements, primarily in rural in each health center under even the best of conditions. health services, have been projected to reach 2004 levels The NHP has been silent on the question of the mix per population by 2015, or an increase of general service of staff it would like to see deployed in the future—this provisions by up to 40 percent. The overall rural health should be driven by the technical options that it believes services improvement is expected to reduce the load on the health system should be delivering and how. It is Primary Health Care (PHC) services at provincial gen- beyond the focus of this study to determine how best to eral hospitals, and thereby release capacity to enhance deploy cadres or even to consider abolishing a cadre. The and increase referral and specialized services” (NHP information in Table 5-4 would suggest most core ser- Volume 2A, pp125). The health system was historically vices can be delivered by the existing structure of cadres. configured on the principle that first contact with health On the other hand, this is not to say the skills currently services would (should) be delivered at the lowest-level imparted are optimal. There are a range of curriculum health facility and by the lowest level of trained staff able issues and gaps identified, but the most important seem to deal with the condition being dealt with. If the condi- to be the need for improved midwifery skills—neither the tion could not be dealt with patients would be referred up basic CHW or General Nursing Curriculum deals with the line to a higher category staff/facility as required. In more than normal deliveries and does not prepare grad- PNG this meant the CHWs staffing aid posts were seen uates adequately for abnormal deliveries and/or emer- as the first point of contact. gency obstetric care. Many professional concerns have A basic package of services to deliver or respond also been expressed about the level of practical training of to the core MDG needs within the existing structure of facilities and health cadres in rural areas is outlined in 23  Monash Study 2010. 24  here have been no similar studies of the hospital sector— Table 5-4.26 The NHP commits NDoH to delivery of the although some very useful work has been done as background MDGs. Table 5-4 lists, by MDG, where the health sector for the MTEF. can intervene, the technical options that are known to be 25  As discussed below and in Chapter 6 this is fortunate. The effective and the lowest level of the system that can deal current stock of the health workforce is aging and a substantial adequately with the observed condition. For example, proportion will retire over the next decade (Chapter 2) and the existing capacity of the health-related training schools (Chap- most maternal deaths are caused by hemorrhage, sepsis, ter 3) will be hard pressed over the next decade replacing exist- eclampsia, and labor complications. Each of these prob- ing staff. This means there is a breathing space of perhaps five lems has a series of technical options mapped into a pack- years which will enable services being delivered to substantially increase without major increases in staff numbers. age to prevent or treat them. Thus, the technical option 26  This table is drawn from Annex 2C of Chapter 2 World Bank, for preventing hemorrhage at delivery with oxytocic drugs Asian Development Bank and AusAID “Strategic Directions for fits into a reproductive health primary health care pack- Human Development in Papua New Guinea,” 2007. Future Staffing Demand Scenarios for PNG’s National Health System   //  75 1305747 PNG HR Report 11-27-12.indb 75 11/27/12 11:38 AM Table 5‑4: Technical Options for Interventions to Deliver the MDGs by Level of Service in PNG Lowest Level to MDG Problem How to Intervene Technical Options Implement Package Maternal Hemorrhage Prevent at delivery Drugs at delivery RH-PHC * Mortality Prevent severe antenatal Iron and folic supplements RH-PHC* anemia Detection and treatment RH-PHC* Drugs to prevent RH-PHC* Impregnated nets DC-Soc market** Reduce risk with high parity Tubal ligation RH-Hospital*** Access to temporary methods RH-PHC* RH-Soc market.**** Treat postpartum hemorrhage Drugs and resuscitation RH-PHC * RH-EOC ***** Manage retained placenta General anesthetic & remove RH-EOC ***** Sepsis Ensure clean delivery Kits for attendants RH-PHC * Supervised births RH-PHC * Treat Antibiotics RH-PHC * Eclampsia Treat Magnesium sulphate RH-EOC ***** Treat pre-eclampsia Antenatal care RH-PHC * Obstructed Manage promptly Vacuum extraction RH-EOC ***** Labor Cesarean section RH-EOC ***** Anticipate Antenatal care RH-PHC * Reduce risk Contraception for teenagers RH-PHC * Labor Treat Anesthetic, blood, theater RH-EOC **** complications Neonatal Sepsis Prevent (clean delivery) Supervised delivery RH-PHC * Mortality Supervised birth attendants RH-PHC * Asphyxia Resuscitate at birth Supervised delivery RH-EOC ***** HIV Prevent congenital infection ARVs to prevent transmission RH-EOC ***** 1 month – 5 year Pneumonia Treat early Antibiotics CH-PHC * Mortality Prevent measles Vaccination CH-PHC * Reduce smoke exposure House construction; stoves DC-public health Diarrhea Improved water + sanitation Wells, water supply, latrines DC-public health Improved hygiene at home Health education DC-public health Oral rehydration solution Dispense sachets CH-PHC * Malaria Reduce transmission Impregnated mosquito nets DC-Soc market ** Treat Treat promptly with ACTs RH-PHC * Meningitis Treat Treat promptly with antibiotics RH-PHC * CH-IPD Infectious Malaria Vector control Impregnated nets DC-Social marketing Diseases Treatment Effective drugs CC-PHC Treat severe disease CC-IPD AH-IPD (continued on next page) 76  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 76 11/27/12 11:38 AM Table 5‑4: Technical Options for Interventions to Deliver the MDGs by Level of Service in PNG  (continued) Lowest Level to MDG Problem How to Intervene Technical Options Implement Package Tuberculosis Prevent BCG vaccine CH-PHC Contact tracing CH-PHC AH-PHC Treat Short course CH-PHC AH-PHC HIV and AIDS Change behavior in key populations Safe sex with condoms +peer HIV-prevention education Supply of condoms at high-risk HIV-prevention sites (bars, hotels, clubs) Reduce risk of infection Prompt treatment of STDs + HIV-prevention condom promotion Counseling with VT + condom HIV-clinical promotion Awareness raising Mass media condom promotion & HIV-prevention partner reduction Reduce morbidity and delay death Treat common infections AH-PHC * TB outpatient treatment AH-PHC * ARV drugs for life HIV-clinical Source: Table 2C.1 MDGs-Technical Options for Interventions by Level of Service in “Strategic Directions for Human Development in Papua New Guinea”. Notes: RH-PHC = Reproductive Health – Primary Health Care; ** DC-Social Marketing = Direct to Community Social Marketing *** RH-Hospital = Reproductive Health at Hospital Level; **** RH – Social marketing = Reproductive Health Social Marketing; ***** RH-EOC = Reproductive Health – Emergency Obstetric Care midwives. These issues need to be addressed and are dis- considerable value and intrinsic merit, are not system- cussed further in Chapter 6 and in more detail in Chap- atic and are not well documented and recoded on indi- ter 7. They do not suggest that there needs to be a major vidual personnel records—at least in a form that can be change to the configuration of the different health ser- analyzed as part of the Human Resource Planning system. vice-delivery cadres. There is a plan within the Health Vision 2050: Another study undertaken in 200027 and focused on Directions for the Next Forty Years and the NHP to hospitals examined the work of auxiliary-level nurses progressively upgrade/introduce Community Health (CHWs) in hospitals. It revealed that: (i) many staff were Posts (CHPs) and district hospitals. CHPs will “become trained long ago and had had no in-service training; the new outer periphery of the health system staffed by (ii) many hospital CHWs are performing nursing proce- three health workers skilled in maternal and child health, dures for which they have not been trained; and (iii) the midwifery, health promotion, and community awareness new CHW curriculum targets rural health services and programs. These facilities will be slightly larger than the does not include a number of nursing skills considered current Aid Posts, and differ by the inclusion of a labor very important for hospital practice. It is quite obvious room. It is anticipated that supervised deliveries and rou- that there has been very little systematic in-service train- tine immunization will be conducted from these points” ing of health staff over at least the past decade—nurse and CHWs training schools have certainly reported very little 27  CHW Pre-Service Curriculum Review and Revision Work- involvement in in-service training. A lot of ad hoc in-ser- ing Group: Community Health Worker Roles and Responsibili- ties. Report on findings (results) from stakeholder consultation vice training has almost certainly been financed by devel- in Lae, Finschafen and Tewae/Siassi Districts in Morobe Prov- opment partners related to a desire to implement different ince; 18–22 September 2000 as reported in Papua New Guinea programs (such as DOTS and MCH) but these, while of Human Resources for Health: Policy Issues Review Paper, 2002. Future Staffing Demand Scenarios for PNG’s National Health System   //  77 1305747 PNG HR Report 11-27-12.indb 77 11/27/12 11:38 AM (NHP). District hospitals will be gradually introduced • The first scenario is the “No-Change-in-Supply to most districts over the next 40 years to be staffed by a Scenario”. This scenario shows the implications doctor. In order to implement this it is planned to estab- of no change in human resource supply capaci- lish a new rural doctors program with this service-deliv- ty over the period 2010–2030. It demonstrates the ery strategy in mind. Much more work needs to be done “crisis” to expect from declining total health ser- before both of these concepts are implemented on other vice-delivery staff numbers which will result from than a pilot basis—there will be five pilot CHPs estab- a “Do Nothing” strategy on the supply side (that is lished under the NHP and district hospitals will be slightly no change in the current throughputs of the health upgraded health centers with the defining difference being training system). the presence of a doctor equipped to undertake more com- • The second scenario is the “PNGDSP-Posited Aspi- plex obstetric care at the local level. rational Scenario”. This scenario is a set of projections One important aspect of this discussion is how based on the number of health staff and/or popula- services can best be cost-effectively delivered. Differ- tion-to-health staff ratios posited in the PNGDSP ent cadres have different skills and very different resource 2010–2030 to be achieved by 2030. In this report we implications given existing remuneration rates (and costs use the absolute staff numbers for each of Medical of production as discussed in Chapter 3). Doctors are obvi- Officers, Nurses and CHWs as the population pro- ously capable of undertaking a far wider range of medi- jections used in this report are different from those cal and general health interventions (assuming they are used in the PNGDSP (and would thus produce dif- appropriately equipped and have functioning facilities) ferent population-to-staff ratios to those derived for than a CHW. On the other hand, the salary of one doctor the PNGDSP. within the health system could employ 4.5 CHWs, three • The third scenario is the “Maintaining Current midwives and 3.3 general nurses (Table 5-5). It is, there- Population-to-Staff Ratios Scenario”. This scenario fore, important on cost-effective grounds to get the mix assumes that core direct service-delivery health cad- of health cadres right. res maintain both their current share of the workforce and the current (2009) population-to-staff ratios over the period 2009–2030. The fundamental driver of the 5.4. Future Health Human Resource demand for health staff in this example is growth of Demand Scenarios 2010–2030 the population. • The fourth scenario is the “WHO ‘threshold’ Ser- This section presents five scenarios of the demand and vice-Delivery Staff Scenario”. This scenario is based possible supply options for health personnel (core direct on the WHO “threshold” density of 2.28 per 1,000 service-delivery cadres) directly delivering services and population (or population-to-staff ratio of 439 to costs them in real 2009 prices (using unit annual salary 1) of doctors, nurses (registered and enrolled) and costs as documented in Table 5-5): midwives. According to WHO, coverage of essential Table 5‑5: Relative Costs of Core Health Service-Delivery Cadres (2009) Health Cadre Annual remuneration (Unit) costs (Kina) Number who can be Employed per Doctor Doctors 75,000 1.0 HEOs 25,600 2.9 General Nurses 22,673 3.3 Midwives (nurse 4)* 25,000 3.0 CHWs 16,800 4.5 Source: Health Medium-Term Public Expenditure Plan background documents and payroll information. Note: * Nurse 4 on the civil service salary scale. 78  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 78 11/27/12 11:38 AM Table 5‑6: Scenario 1: No Change in Supply Capacity 2010–2030 Indicator 2009 2015 2020 2025 2030 Population (‘000) 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 595 636 656 Population per Doctor 17,512 15,243 15,166 16,046 17,277 Nurse Numbers 3,252 3,051 2,971 2,912 2,869 Population per Nurse 2,041 2,572 3,038 3,506 3,949 CHW Numbers 4,398 3,899 3,621 3,556 3,537 Population per CHW 1,509 2,012 2,492 2,871 3,203 HEO Numbers 411 503 553 585 607 Population per HEO 16,148 15,609 16,322 17,452 18,663 Total Service-Delivery Staff 8,440 7,968 7,740 7,690 7,669 Population per Staff 786 985 1,166 1,328 1,477 Source: Summarized from Chapter 6: Scenario 1 – Tables 6-1 to 6-5. interventions below this level, including those neces- to 7,740 in 2020 and 7,669 in 2030 (Table 5-6). The pop- sary to reach the health-related MDGs, is not likely.28 ulation-to-staff ratio would increase significantly from • The fifth scenario is the recommended scenario 786 to one staff to 985 to one staff in 2015, 1,166 to each which envisages: (i) a gradual reduction in the staffer in 2020 and 1,477 to each staff by 2030—almost population-to-doctor, nurse and CHW ratios; doubling the population per staff member ratio over the and (ii) maintaining nurses and CHWs as the next 20 years. Total service-delivery staff would decline backbone of the service delivery system—partic- by 1 percent per annum over the period 2009–2015 and ularly rural service delivery. It is constrained by: decline a further 0.6 percent per annum for the period to (i) the growth in the resource envelope likely to be 2020 (Table 5-7). available for health and service-delivery staff; and The most significant decline under this scenario (ii) the feasibility and speed with which preservice would be with CHWs and nurses who form the back- training can be ramped up to meet the demands bone of rural service delivery. Over the period 2009 to of attrition from the workforce and the needs of a 2015 CHWs can be expected to decline by 2.3 percent growing population. per annum and a further 1.4 percent per annum over the period 2015 to 2020. Thus the total CHWs which can be 5.4.1 Scenario 1: The No-Change-in-Supply expected to be available to the health system is 3,621 in Scenario 2020 or some 777 less than the 4,398 in service in 2009— an overall decline of 18 percent at a time when the popu- This scenario shows the implications to the overall sup- lation is expected to increase 36 percent. This will result ply of direct health service delivery over the period to in the expected population per CHW to increase from 2030 if: (i) there is no change in the current preservice 1,509 in 2009 to 2,492 per CHW in 2020. A similar picture training capacity (described in Chapter 3) for doctors, also holds for nurses. The number of nurses is expected to nurses, HEOs and CHWs; and (ii) the expected retire- decline from 3,252 in 2009 to 3,051 in 2015 and 2,971 in ments from the workforce described in Chapters 2 and 2020. The population per nurse can be expected to increase 6 are realized. Overall, this scenario shows that the health system would face a major staff supply crisis and that there 28  WHO. 2006. This report notes specifically that countries with would have to be a major decline in staff ratios relative to densities of doctors, nurses, and midwives below 2.28 per 1,000 the population. The total number of direct service-deliv- population fail, on average, to achieve 80 percent coverage for ery staff could be expected to decline from 8,440 in 2009 live deliveries by skilled birth attendants. Future Staffing Demand Scenarios for PNG’s National Health System   //  79 1305747 PNG HR Report 11-27-12.indb 79 11/27/12 11:38 AM Table 5‑7: Scenario 1: No Change in Supply Capacity 2010–2030 Indicator 2009 2015 2020 2025 2030 Population 2.8 2.8 2.8 2.5 2.1 Doctor Numbers n.a. 5.9 3.1 1.4 0.6 Nurse Numbers n.a. –1.0 –0.5 –0.4 –0.3 CHW Numbers n.a. –1.9 –1.4 –0.4 –0.1 HEO Numbers n.a. 3.7 2.0 1.2 0.8 Total Service-Delivery Staff n.a. –1.0 –0.6 –0.1 –0.1 Source: Calculated from Table 5-6. from 2,041 in 2009 to 2,572 in 2015 and 3,038 in 2020. The ing scenarios explore different demand options for the number of doctors and HEOs would expand slightly but growth of staff and their implications for growth of the the population per doctor and HEO would not improve workforce relative to expected population growth and substantially over the period because of fast population their costs. Chapter 6 explores in more detail these sum- growth (Table 5-6). mary projections and reviews their sustainability from a The costs of the direct service-delivery staff using fiscal perspective. the unit costs presented in Table 5-5 for each category of staff under this scenario are presented in Table 5-8. 5.4.2 Scenario 2: The PNGDSP-Posited This table shows that total costs for service-delivery staff in Aspirational Scenario real 2009 prices would not increase significantly over the period to 2030—it would only increase from K191.3 mil- The PNGDSP proposes an extremely ambitious plan for lion in 2009 to K195.2 million in 2020 and K198.3 million the expansion of human resources for health together in 2030. If, as discussed above, the budget for health can with very ambitious health outcome targets. Over the be expected to grow at around 5 percent in real terms per 20-year period to 2030 the target is to reduce the IMR annum over the 20 years projection period it is expected from 57/1,000 live births to 20; the under-five mortality that much more rapid growth of staff numbers would be rate from 75/1,000 to 20; the MMR from 733/100,000 to sustainable (see Scenario 1 in Chapter 6). The follow- 100 and to increase life expectancy from 57 to 70 years. Table 5‑8: Scenario 1: No Change in Supply Capacity 2010–2030 Indicator Unit Cost 2009 2015 2020 2025 2030 Population (‘000) n.a. 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 595 636 656 Total Doctor Costs 1,2 K90,000 34.1 46.3 53.6 57.2 59.0 Nurse Numbers 3,252 3,051 2,971 2,912 2,869 Total Nurse Costs1,2 K22,700 73.8 69.3 67.4 66.1 65.1 CHW Numbers 4,398 3,899 3,621 3,556 3,537 Total CHW Costs1,2 K16,600 73.0 64.7 60.1 59.0 58.7 HEO Numbers 411 503 553 585 607 Total HEO Costs K26,000 10.7 13.1 14.4 15.2 15.8 Total Service-Delivery Staff 8,440 7,968 7,740 7,690 7,669 Total Service-Delivery Staff Costs1,2 191.6 193.4 195.5 197.5 198.6 Source: Summarized from Chapter 6: Scenario 1 – Tables 6-1 to 6-5. Note: 1 In millions of Kina. 2 Errors due to rounding. 80  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 80 11/27/12 11:38 AM Table 5‑9: Scenario 2: Public Sector Health Workforce Envisioned by PNGDSP 2010–2030 Indicator 2009 2015 2020 2025 2030 Population (‘000) 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 646 1,641 4,184 Population per Doctor 17,512 15,243 13,969 6,224 2,707 Nurse Numbers 3,252 3,026 4,116 8,477 19,526 Population per Nurse 2,041 2,593 2,193 1,205 580 CHW Numbers 4,398 3,899 6,284 12,010 18,795 Population per CHW 1,509 2,012 1,436 850 603 HEO Numbers* 411 400 300 250 200 Population per HEO 16,148 19,615 30,083 40,840 56,645 Total Service-Delivery Staff 8,440 7,840 11,346 22,378 42,705 Population per Staff 786 1,001 795 456 265 Source: Summarized from Chapter 6: Scenario 2 – Tables 6-6 to 6-9. Note: Assumes no new HEOs and that the existing stock gradually retires. It also hopes to reduce the incidence of tuberculosis from vice-delivery workforce to 9.8 percent of the workforce. 246/100,000 to 100 and the percentage of children under To achieve this, it is planned to increase the number of five with moderate to high malnutrition from 50 percent doctors from the current 379 in 2009 to 4,184 by 2030 an to 5 percent. This is to be achieved by the rapid expan- increase of over 1,190 percent. As discussed in Chapter 6, sion of most key service-delivery health cadres (Table 5-9). this is almost impossible to achieve in a 20 year period. It Using the population projections for the five-year periods would imply, as documented in Table 5-10 a growth rate and the base service-delivery staff complement known for of over 30 percent per annum in the graduation of doc- 2009, the PNGDSP envisages the total service-delivery tors over the period 2020–2030. Given the delay between staff (comprising doctors, nurses, CHWs and, while not a need to invest in additional training facilities and then mentioned in the PNGDSP, HEOs) rising from a base of the training of doctors which takes six years an immedi- 8,440 in 2009 to more than 40,000 by 2030—an increase ate decision to ramp up doctor training would not alter of about 375 percent. the supply of doctors for the workforce until well into The population-to-service delivery staff ratio the next Health Plan period 2016–2020. It would be very would improve from 786 per staff to just 265. This is unlikely that PNG could reach a doctor population of about to be achieved by radically altering the composition of 4,200 by 2030 as posited in Table 5-9—simply because of the health workforce—in particular by radically increas- the time it would take to establish a new medical school ing the number of doctors from 4.5 percent of the ser- and/or radically expand the existing medical school and Table 5‑10: Scenario 2: Annual Public Sector Health Workforce Growth Rates Envisioned by PNGDSP 2010–2030 Indicator 2009 2015 2020 2025 2030 Population 2.8 2.8 2.8 2.5 2.1 Doctor Numbers n.a. 5.9 5.1 30.8 31.0 Nurse Numbers n.a. –1.2 7.2 21.2 26.1 CHW Numbers n.a. –1.9 12.2 18.2 11.3 HEO Numbers n.a. –0.5 –5.0 –3.3 –4.0 Total Service-Delivery Staff n.a. 0.6 6.0 19.5 18.2 Source: Calculated from Table 5-9. Future Staffing Demand Scenarios for PNG’s National Health System   //  81 1305747 PNG HR Report 11-27-12.indb 81 11/27/12 11:38 AM it is unclear there are enough well-qualified science-based need to increase at over 12 percent per annum and over students graduating from high schools. Under this sce- the period 2020–2030 it would need to grow at between nario the population-to-doctor ratio would improve from 18 and 21 percent per annum (Table 5-10). This would 17,512 to one in 2009 to 2,707 to one in 2030. entail the population-to-CHW ratio improving from the The PNGDSP scenario proposes that there be more 1,509 to one in 2009 to about 600 to one CHW in 2030. than 19,500 nurses available for service delivery by The PNGDSP scenario for the health sector is 2030—up from 3,252 in 2009. This would result in the silent on the future plans for HEOs. For the purposes population-to-nurse ratio improving from 2,041 to one of this scenario it is assumed that a decision would be nurse to 580 to one nurse in 2030 (Table 5-9). To achieve made to stop production of HEOs immediately and that this, the number of nurses would need to increase by their role would be taken over either by increased num- 7.2 percent per annum over the period 2015–2020 and at bers of doctors and/or specially trained nurses. The exist- well over 20 percent per annum over the period 2020–2030 ing numbers of HEOs in the workforce would, therefore, (Table 5-10). This would, of course, also require a massive gradually decline as they retired. It is estimated that there expansion in training capacity—starting immediately. It would still be about 200 HEOs in the workforce in 2030 should also be noted that nurse numbers are expected to but after this the numbers would fall to zero over the sub- decline over the period 2009–2015 by –1.2 percent per sequent 5 years. annum. This is because attrition rates (including retire- In aggregate terms total health service-delivery ments) from the existing workforce are likely to be higher staff would be expected to grow from 8,440 in 2009 to than the current flow of new graduates into the workforce. about 43,000 in 2030—an increase of over 400 percent It will take at least four years to significantly expand nurse in 21 years—and dramatically faster than the increase output—one year to build the required infrastructure and expected in the population growth rate. As a consequence, recruit the needed teaching staff and three years of school- the PNGDSP envisages the population-to-service deliv- ing to generate additional nurse workforce entrants. ery staff ratio falling from 786 to 265 in 2030 (Table 5-9). The PNGDSP envisages there being nearly 19,000 This would involve service-delivery staff numbers grow- CHWs employed by 2030—up from the current estimate ing marginally in 2009–2015 (at less than 1 percent per of 4,398 in 2009 or an increase of 327 percent (Table 5-9). annum) and ramping up to 6 percent per annum over the To reach this level of CHWs available for the workforce period 2015 to 2020 and to between 18 and 19 percent per the numbers employed between 2015 and 2020 would annum over the ten-year period 2020–2030 (Table 5-10). Table 5‑11: Scenario 2: Costs of Public Sector Health Workforce Envisioned by PNGDSP 2009–2030 Indicator Unit Cost 2009 2015 2020 2025 2030 Population (‘000) n.a. 6,637 7,846 9,025 10,210 11,329 Doctors Numbers 379 515 646 1,641 4,184 Total Doctor Costs1 K90,000 34.1 46.3 58.1 147.7 376.6 Nurse Numbers 3,252 3,026 4,116 8,477 19,526 Total Nurse Costs 1 K22,700 73.8 68.7 93.4 192.4 443.2 CHW Numbers 4,398 3,899 6,284 12,010 18,795 Total CHW Costs1 K16,600 73.0 64.7 104.3 199.4 312.0 HEO Numbers 411 400 300 250 200 Total HEO Costs1 K26,000 10.7 10.4 7.8 6.5 5.2 Total Service-Delivery Staff 8,440 7,840 11,346 22,378 42,705 Total Service-Delivery Staff Costs 1,2 191.6 190.1 263.6 546.0 1,137.0 Source: Summarized from Chapter 6: Scenario 2 – Tables 6-6 to 6-9. Note: 1 In millions of Kina. 2 Errors due to rounding. 82  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 82 11/27/12 11:38 AM The staff costs (remuneration) of this strat- The population of nurses would grow from 3,252 in 2009 egy—assuming all staff are employed by government— to 3,844 in 2015 and 5,551 in 2030. Similarly, the number would increase from K191.6  million for the 8,440 of CHWs would increase from 4,398 in 2009 to 7,507 in service-delivery staff employed in 2009 to K1,135 million 2030 and the number of HEOs would increase from 411 for 42,705 staff in 2030 (in 2009 prices). This is costed at in 2009 to 702 in 2030. Total service-delivery staff would the average 2009 unit costs of each cadre estimated by the increase from 8,440 in 2009 to 9,977 in 2015, 11,477 in Monash Study and by background working tables for the 2020 and 14,407 in 2030 (Table 5-12). Key issues associ- Health MTEF. This implies an increase in the budget of ated with meeting these workforce targets from a supply almost 500 percent in 21 years in real terms or an increase side are discussed in more detail in Chapter 6. of over 20 percent per year. Clearly, Scenario 2 is unlikely The real (2009) remuneration costs associated with to be achievable given both the resources likely to be avail- Scenario 3 are presented in Table 5-13. This estimates able over this period or feasible given extremely likely con- that at 2009 salary costs the personnel budget would grow straints and the ability to expand training this dramatically. from K191.3 million estimated in 2009 to K326.5 million This issue is discussed in more detail in Chapter 6. in 2030—an increase of just overr 3.3 percent per annum or the estimated rate of population increase. 5.4.3 Scenario 3: Maintaining Current Population-to-Staff Ratios Scenario 5.4.4 Scenario 4: The WHO-Recommended “Threshold” Service-Delivery Staff The third scenario assumes the core direct service-deliv- Scenario ery health cadres (doctors, nurses, CHWs and HEOs) maintain both their current share of the workforce and The fourth scenario is based on the WHO-recommended that the current population-to-staff ratios reported “threshold” density of 2.28 per 1,000 population (or a in 2009 are sustained over the period 2009–2030. The population-to-staff ratio of 439 to 1) of doctors, nurses fundamental driver of the demand for health staff in this (registered and enrolled), and midwives. Below this example is growth of the population. Under this scenario level, according to WHO, coverage of essential inter- the number of each cadre will, in effect, grow at the same ventions, including those necessary to reach the health- rate as the population. The number of doctors would related MDGs, is not likely. Overall, this would mean an increase from 379 in 2009 to 515 in 2020 and 647 in 2030. increase in total staff from 8,440 in 2009 to almost 25,000 Table 5‑12: Scenario 3: Health Sector Service-Delivery Workforce Maintaining Current (2009) Population-to-Staff Ratios 2009–2030 Indicator 2009 2015 2020 2025 2030 Population (‘000) 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 448 515 583 647 Population per Doctor 17,512 17,512 17,512 17,512 17,512 Nurse Numbers 3,252 3,844 4,422 5,003 5,551 Population per Nurse 2,041 2,041 2,041 2,041 2,041 CHW Numbers 4,398 5,199 5,980 6,766 7,507 Population per CHW 1,509 1,509 1,509 1,509 1,509 HEO Numbers 411 486 559 632 702 Population per HEO 16,148 16,148 16,148 16,148 16,148 Total Service-Delivery Staff 8,440 9,977 11,476 12,984 14,407 Population per Staff 786 786 786 786 786 Source: Summarized from Chapter 6: Scenario 3 (Tables 6-10 to 6-14). Future Staffing Demand Scenarios for PNG’s National Health System   //  83 1305747 PNG HR Report 11-27-12.indb 83 11/27/12 11:38 AM Table 5‑13: Scenario 3: Costs of Health Service-Delivery Staff Workforce When Maintaining Current (2009) Population Staff Ratios (2009–2030) Indicator Unit Cost 2009 2015 2020 2025 2030 Population (‘000) n.a. 6,637 7,846 9,025 10,210 11,329 Doctors Numbers 379 448 515 583 647 Total Doctor Costs 1,2 K90,000 34.1 40.3 46.4 52.5 58.2 Nurse Numbers 3,252 3,844 4,422 5,003 5,551 Total Nurse Costs1,2 K22,700 73.8 87.3 100.4 113.6 126.0 CHW Numbers 4,398 5,199 5,980 6,766 7,507 Total CHW Costs 1,2 K16,600 73.0 86.3 99.3 112.3 124.6 HEO Numbers 411 486 559 632 702 Total HEO Costs 1,2 K26,000 10.7 12.6 14.5 16.4 18.3 Total Service-Delivery Staff 8,440 9,977 11,476 12,984 14,407 Total Service-Delivery Staff Costs 1,2 191.6 226.5 261.0 295.2 327.5 Source: Summarized from Chapter 6: Scenario 3 – Tables 6-10 to 6-14. Note: 1 In millions of Kina. 2 Errors due to rounding. in 2030—an increase of over 190 percent over 21 years elements—an argument that this needs to consist of a (Table 5-14). Doctor numbers would need to rise 6 per- doctor density of 0.55 doctors per 1,00029 and a non- cent per annum over the period 2009–2015 and between doctor staff density of 1.73. To this end, Scenario 3 has 23 and 29 percent per annum over the subsequent 15 years the doctor workforce growing to achieve the 0.55/1,000 (Table 5-15). This scenario, while ambitious, results in target in 2030 and the combined nurse and CHW ratio about 40 percent less staff than envisaged by the PNGDSP growing to reach the 1.73/1,000 target by 2030 but with in 2030 (Scenario 1). the two cadres maintaining the same proportion as cur- Scenario 4 does not distinguish between staff rent (2009) staffing. For simplicity purposes, we have let employed by government and the private sector—so HEOs grow at the same rate as the population because strictly speaking the numbers recommended by WHO they represent a small proportion of the total. It is prob- do not mean that all will be employed and/or financed able that HEOs could substitute for a proportion of the by government. The PNGDSP is silent on the role of the doctors as a minor variation of this scenario. private sector and it is assumed from the documentation This scenario implies a very significant rise in the that it was staff numbers for the government to finance absolute number of doctors (a rise from 379 in 2009 to (government and mission/church). Until now this report 515 in 2015, 1,173 in 2020 and over 6,200 in 2030. It has been relatively silent on the role of the private sector. would also involve a more than proportionate increase in This is because there are no numbers available on staff the doctors in the workforce—from 4.5 percent in 2009 employed in the private sector. This does not mean that to over 24 percent in 2030. This would result in a rapid the private sector is either insignificant now or will be decline in the ratio of population per doctor from 17,512 insignificant in the future. It is probable that the private in 2009 to 1,818 per doctor in 2030. Given the discussion sector employs 10–15 percent of the total health service- about the staff and facilities required to deliver a core MDG delivery workforce and that this may well rise to 20 per- package, discussed above and summarized in Table 5-3, cent or more over the next decade if the private sector grows in the ways envisioned in the PNGDSP. This issue 29  The work of Scheffler et al (2009) found that in order to achieve the system performance desired (80 percent coverage of will be taken up in more detail in Chapter 6. live births by a skilled attendant) a doctor density of 0.55/1,000 The WHO “threshold” density of 2.28 service-deliv- population would be required ranging from 0.41 to 0.61 based ery staff per 1,000 population is made up of two key on a 95 percent confidence interval. 84  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 84 11/27/12 11:38 AM Table 5‑14: Scenario 4: Achieving WHO “Threshold” Health Service-Delivery Staff Density (2009–2030) Indicator 2009 2015 2020 2025 2030 Population (‘000) 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 1,173 2,553 6,231 Population per Doctor 17,512 15,234 7,692 4,000 1,818 Nurse Numbers 3,252 3,295 4,964 6,229 8,497 Population per Nurse 2,041 2,381 1,818 1,639 1,333 CHW Numbers 4,398 4,315 6,137 8,271 10,310 Population per CHW 1,509 1,818 1,471 1,235 1,099 HEO Numbers 411 486 559 632 702 Population per HEO 16,148 16,148 16,148 16,148 16,148 Total Service-Delivery Staff 8,440 8,611 12,833 17,685 25,739 Population per Staff 786 911 703 577 440 Source: Summarized from Chapter 6: Scenario 4 – Tables 6-15 to 6-19. it is not clear that this number of doctors would be cost per annum between 2015 and 2020 and around 6.8 per- effective—particularly given the relative costs of doctors cent per annum over the period 2020 to 2030 (Table 5-15). and other service-delivery staff described in Table 5-4. HEOs, as discussed, grow at the same rate as the popula- The scenario also involves a quite significant expan- tion in this scenario. sion of other cadre numbers. The number of nurses would Table 5‑15: Scenario 4: Service-Delivery Staff Growth increase from 3,252 in 2009 to almost 5,000 in 2020 and Rates Required to Achieve WHO “Threshold” Service- to about 8,500 in 2030—an increase of about 160 per- Delivery Staff Density 2009–2030 Source: Calculated cent (Table 5-10). This would mean a rapid expansion from Table 5-14. of nurses by 10.1 percent per annum between 2015 and The cost of employing doctors to implement Sce- 2020, 5.1 percent between 2020 and 2025 and 7.3 percent nario 3 would rise from K34.1  million in 2009 to between 2025 and 2030 (Table 5-14). The population-to- K560.8 million in 2030—an increase of nearly 17 times nurse ratio would improve from 2,041 per nurse in 2009 the cost of doctors in 2009 (Table 5-16). The share of to 1,333 per nurse in 2030. For CHWs the numbers would doctor costs in total remuneration costs would go from increase from 4,398 in 2009 to 6,137 in 2020 and 10,310 about 18 percent of the total remuneration budget to about in 2030. This would imply CHWs growing at 8.4 percent 60 percent in 2030. It is very unlikely, in the PNG context, Table 5‑15: Scenario 4: Service Delivery Growth Rates Required to Achieve WHO “Threshold” Service-Delivery Density 2009–2030 Indicator 2009 2015 2020 2025 2030 Population 2.8 2.8 2.8 2.5 2.1 Doctor Numbers n.a. 6.0 25.6 23.8 28.8 Nurse Numbers n.a. –1.3 10.1 5.1 7.3 CHW Numbers n.a. –1.9 8.4 7.0 4.9 HEO Numbers n.a. 3.0 3.0 2.6 2.2 Total Service-Delivery Staff n.a. 3.4 9.8 7.5 8.0 Source: Calculated from Table 5-14. Future Staffing Demand Scenarios for PNG’s National Health System   //  85 1305747 PNG HR Report 11-27-12.indb 85 11/27/12 11:38 AM Table 5‑16: Scenario 4: Costs of Achieving WHO “Threshold” Health Service-Delivery Staff Density 2009–2030 Indicator Unit Cost 2009 2015 2020 2025 2030 Population (‘000) n.a. 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 1,173 2,553 6,231 Total Doctor Costs 1,2 K90,000 34.1 46.3 105.6 229.7 560.8 Nurse Numbers 3,252 3,295 4,964 6,229 8,497 Total Nurse Costs1,2 K22,700 73.8 74.8 112.7 141.4 192.9 CHW Numbers 4,398 4,315 6,137 8,271 10,310 Total CHW Costs1,2 K16,600 73.0 71.6 101.9 137.3 171.1 HEO Numbers 411 486 559 632 702 Total HEO Costs 1,2 K26,000 10.7 12.6 14.5 16.4 18.3 Total Service-Delivery Staff 8,440 8,611 12,833 17,685 25,740 Total Service-Delivery Staff Costs1,2 191.6 205.3 334.7 524.8 943.1 Source: Summarized from Chapter 6: Scenario 4—Tables 6-15 to 6–19. Note: 1 In millions of Kina. 2 Errors due to rounding. (even with considerable additional work to demonstrate doctor ratio would go from 17,512 per doctor in 2009 to the optimum number of doctors required in PNG) that 13,004 in 2020 and to 7,380 per doctor in 2030. this would be a cost-effective investment. The number of nurses would increase from 3,252 in 2009 to 4,277 in 2020 (an increase of about 32 per- 5.4.5 Scenario 5: The Recommended cent) and to just over 8,000 by 2030 an increase of Scenario about 146 percent over current numbers by 2030. The growth rate of nurse numbers will be negative through This scenario envisages: (i) a gradual reduction in the 2015 (due to the attrition rate from the workforce and the population to doctor, nurse and CHW ratios; (ii) main- time needed to ramp up training) but would then ramp taining nurses and CHWs as the backbone of the ser- up to be much faster than the rate of population increase vice delivery system—particularly rural service delivery. over the years 2015 to 2030. Over this period the nurs- Rural service delivery is constrained by: (i) the growth in ing numbers would increase at an average of over 10 per- the resource envelope likely to be available for health and cent per annum. Thus the population-to-staff ratio would service-delivery staff; and (ii) the feasibility and speed with improve from around 2,041 to one nurse in 2009 to 1,414 which preservice training can be ramped up to meet the to one by 2030. demands of attrition from the workforce and the needs Under this scenario the number of CHWs would of a growing population. It is also recommended that increase from 4,398 to 5,133 in 2020 and about 8,250 there is careful monitoring of effective demand for staff in 2030. The population to CHW numbers would decline at the provincial, district and hospital levels based on staff from around 1,500 to a CHW in 2009 to about 1,370 per workloads. The key results of this scenario are presented CHW over the next two decades—however, the rapidly in Table 5-17. rising population, high attrition rates and the time needed Under Scenario 5 doctor numbers are planned to to ramp up training capacity means that the population rise from 379 to 515 in 2015, 694 in 2020, 1,069 in 2025 to CHW ratio will decline through 2015, increase mod- and a target of 1,535 in 2030. This would imply the doc- estly over the next 15 years re-establishing current levels tor workforce growing at 6.0 percent per annum between by 2025. Overall, CHW numbers would grow from 2015 2009 and 2015; 6.9 percent per annum between 2015 and at over twice the rate of growth of the population—about 2020; 10.8 percent between 2020 and 2025 and fall to 6.7 percent per annum (Table 5-18). This scenario assumes 8.7 percent between 2025 and 2030. The population-to- the existing capacity for producing HEOs is sustained over 86  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 86 11/27/12 11:38 AM Table 5‑17: Scenario 5: Recommended Scenario for Direct Service-Delivery Health Staff 2010–2030 Indicator 2009 2015 2020 2025 2030 Population (‘000) 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 694 1,069 1,535 Population per Doctor 17,512 15,234 13,004 9,551 7,380 Nurse Numbers 3,252 3,136 4,277 6,082 8,012 Population per Nurse 2,041 2,502 2,110 1,679 1,414 CHW Numbers 4,398 4,111 5,133 6,671 8,256 Population per CHW 1,509 1,909 1,758 1,531 1,372 HEO Numbers 411 498 546 580 604 Population per HEO 16,148 15,755 16,529 17,603 18,756 Total Service-Delivery Staff 8,440 8,260 10,650 14,402 18,407 Population per Staff 786 950 847 709 616 Source: Summarized from Chapter 6: Scenario 5 – Tables 6-20 to 6-24. the period—on the grounds that they provide an impor- 2015 and then begin to decline through the next 15 years tant management and supervisory cadre—particularly to 616 to one staff by 2030 (Table 5-17). for rural health services (see also discussion in Chap- It is expected that the remuneration budget for all ters 3, 6 and 7). staff would increase from K191.6 million in 2009 to Total service-delivery staff numbers under Scenario K472.8 million in 2030 (Table 5-19). The share of the 5 would rise from 8,440 in 2009 to 18,407 in 2030—an budget allocated to doctors would increase from 18 per- increase of 118 percent. The total workforce can be cent in 2009 to 29 percent of the budget in 2030. This sce- expected to decline over the period 2009 to 2015 due to nario is more fully explained in Chapter 6. the expected attrition from the workforce and the time it Under Scenario 5—the recommended sce- will take to ramp up training, starting immediately. Over- nario—more careful attention has been paid to how all, the growth of the workforce would be around 8.2 per- fast training can be ramped up for the different cadres cent per annum from 2015 (Table 5-17). The population to and to likely budget constraints. Each cohort of cadres direct service-delivery staff ratio is still expected, despite has a different gestation period from input to the train- the growth of the workforce proposed, to worsen from ing school to the workforce and some allowance in time the current 786 to one staff in 2009 to 950 to one staff in has to be made to allow for construction of additional Table 5‑18: Scenario 5: Direct Service-Delivery Health Workforce Growth Rates For Recommended Scenario 2010–2030 Indicator 2009 2015 2020 2025 2030 Population 2.8 2.8 2.8 2.5 2.1 Doctor Numbers n.a. 6.0 6.9 10.8 8.7 Nurse Numbers n.a. –0.7 7.3 8.4 6.3 CHW Numbers n.a. –1.1 5.0 6.0 4.8 HEO Numbers n.a. 3.5 1.9 1.2 0.8 Total Service-Delivery Staff n.a. –0.4 5.8 7.0 4.0 Source: Calculated from Table 5-17. Future Staffing Demand Scenarios for PNG’s National Health System   //  87 1305747 PNG HR Report 11-27-12.indb 87 11/27/12 11:38 AM Table 5‑19: Scenario 5: Costs of Recommended Scenario for Direct Service-Delivery Health Staff 2010–2030 Indicator Unit Cost 2009 2015 2020 2025 2030 Population (‘000) n.a. 6,637 7,846 9,025 10,210 11,329 Doctor Numbers 379 515 694 1,069 1,535 Total Doctor Costs 1 K90,000 34.1 46.3 62.5 96.2 138.2 Nurse Numbers 3,252 3,136 4,277 6,082 8,012 Total Nurse Costs 1 K22,700 73.8 71.2 97.1 138.1 181.9 CHW Numbers 4,398 4,111 5,133 6,671 8,256 Total CHW Costs1 K16,600 73.0 68.2 85.2 110.7 137.0 HEO Numbers 411 498 546 580 604 Total HEO Costs 1 K26,000 10.7 12.9 14.2 15.1 15.7 Total Service-Delivery Staff 8,440 8,260 10,650 14,402 18,407 Total Service-Delivery 191.6 198.6 259.0 360.1 472.8 Staff Costs1 Source: Summarized from Chapter 6: Scenario 5 – Tables 6-20 to 6-24. Note: 1 In millions of Kina. facilities for the increased capacity. There is a need to be vice-delivery staff. This will help ensure that demand for very realistic about what can be achieved on the supply health services as expressed by outpatient visits per cap- side even with important decisions being taken over the ita per annum and other services, including antenatal care next 18 months. and immunizations also increase. Without the health sys- The NHP indicates that there are plans to expand tem being responsive to the real needs of the population as the number of doctors trained—particularly “rural” expressed by increased demand for services, the demand doctors to be trained under a new training program, for health staff will not be as great as that assumed in any probably undertaken by Divine Word University. This of these scenarios. is discussed further in Chapter 7. Certainly data pre- Given the demand for services documented in the sented in Chapter 2 indicates that there are very few Monash Report for 2009, existing staff numbers are some doctors working in rural areas—in 2009 there were only 40 percent over the required numbers—at least in rural 51 doctors in rural areas. The justification for expansion areas. There is considerable scope to increase rural ser- in doctor numbers proposed here would be in order to vices with existing staff. On the other hand, existing ser- expand numbers working in rural areas. Nevertheless, vices demanded are well below what should be demanded much would need to be done to encourage doctors to if the health system was responding to the burden of dis- practice in rural areas. ease within the population. If the demand for services increases as projected by the NHP, existing staff will not be underemployed. It is unlikely, however, that demand 5.5. Conclusions for services will change significantly overnight—it will be a more gradual process. This is just as well since the current This chapter has explored some of the issues and fac- training system is not producing—particularly for nurses tors which affect efforts to project the demand for ser- and CHWs—enough graduates to enable a replacement vice-delivery staff. The final effective demand for health of the expected retirements from the workforce and to workers will depend in no insignificant part on the efforts meet even a gradual expansion of the workforce. In fact, of the health system to increase the quality-enhancing it is expected that the workforce for nurses and CHWs nonsalary budgets—which need to increase faster than will decline over the NHP period to 2015 because of the total expenditure on health and of expenditure on ser- age of the current workforce. 88  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 88 11/27/12 11:38 AM Chapter 6 looks at the five scenarios described in for health is uncertain. The outer years are thus subject to this chapter in both more detail (year to year over the significant variations (up and down). Such a review pro- period 2009 to 2030). In particular, it looks at the impli- cess would enable changes in the fiscal situation, consid- cations for the ramping up of training capacity that would erations of the role of cadres and progress with increasing be required to try and achieve each of the four scenarios training capacity to be taken into account and appropri- (2–5) involving increased training. Scenario 1 holds exist- ate adjustments made in outer years. In fact, as discussed ing training capacity constant and shows what no action above and in more detail in Chapter 6, the most imme- on the training front involves. To this end it discusses in diate decisions need not constrain options in the future. more detail the supply constraints that effectively rule out Without action to expand training, particularly of Scenarios 2 and 4 as realistic possibilities—in the near nurses and CHWs, the existing outputs of training insti- term. Scenario 3—growth at the same rate as the pop- tutions will not be enough to replace expected attrition ulation—is certainly feasible but the resource envelope from the workforce. Provided adequate initial planning likely to be available for health will allow much more to goes into planning the expansion of the training system happen. Scenario 5—the recommended strategy—con- (including strategic planning for possible stage 2 and strains overall expenditure on health staff to that which 3 expansion of the system) the issues facing the supply can be afforded. It assumes about a 7 percent real increase side will constrain options. One major issue is the expan- in the health budget each year through 2030 as discussed sion of secondary education and particularly of Grade 12 above. Space is left to expand the quality-enhancing recur- entrants to the training system. As discussed in Chapter rent expenditures on health—particularly the financing 3 it is important that entrants to training programs come of pharmaceuticals, medical supplies and vaccines and from the provinces where staff will be required—in order other quality-enhancing expenditures on transport, train- to attract staff to the needed areas of the country. In fact, ing and referral systems. Without this it is unlikely that with some of the more ambitious scenarios, it is unlikely effective demand for services per capita will reverse and that enough qualified high school graduates—at either the later increase as desired. micro (province or district level) or at the macro level—will As further discussed in both Chapters 6 and 7 it is be available to enter the training schools over the next few argued that this building of scenarios should be done years—particularly if the demand for skilled (high school annually as part of the health MTEF processes. The educated) workers required for the LNG and other devel- likely future resource envelope available to government opments come to fruition. Future Staffing Demand Scenarios for PNG’s National Health System   //  89 1305747 PNG HR Report 11-27-12.indb 89 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 90 11/27/12 11:38 AM CHAPTER 6 Supply and Demand: Key Health Cadre Supply Gaps 6.1. Introduction Chapter 5); and (v) share of the total service-delivery staff costs of the expected total recurrent budget. These tables This chapter presents the detailed demand and supply also show the combined costs of nurse and CHW training projections for the five scenarios described in Chapter 5 which are also expressed as a percentage of the expected and, for each scenario, the implications for each cadre. recurrent budget for health—for 2009 unit costs and for the In addition to showing the detailed (year by year) growth quality-enhanced unit costs of training. Finally, the impli- of the direct service-delivery staff (by cadre) implied for cations for policy on the expansion of direct service-deliv- each scenario the detailed tables also show: (i) the expected ery staff training programs are discussed for each scenario. trend in population to staff and staff per 1,000 population ratios; (ii) the expected attrition from the workforce; (iii) the expected outputs from the training schools in the initial 6.2. The Range and Scale of Supply and years of the NHP (prior to implementing any expansion Demand Gaps by Key Health Worker plans and allowing for new trainees to graduate) and how Cadre fast graduations from training schools will need to ramp upwards to the employment targets implied by each of the The range and scale of the supply and demand gaps for scenarios; (iv) the first year intakes to respective train- each of the direct service-delivery cadres (and for all ing schools required to ensure, given expected dropout service-delivery staff) are set out in the five scenarios rates for each school, the needed graduates to reach the described in Chapter 5: employment targets for the specific scenario; and (v) the • Scenario 1: No Change in Human Resource Sup- costs in 2009 prices of employing all he graduates given ply Capacity 2010–2030. This scenario is designed average remuneration costs of each cadre. The tables also to show the implications of a “Do Nothing” Strat- show the recurrent costs of training nurses and CHWs at: egy on the supply side (that is no change in the cur- (i) 2009 unit costs; and (ii) quality-enhanced unit costs rent throughputs of the health training system) given for the training of nurses and CHWs. the expected retirements of the workforce due to its Each set of scenario tables is collated into a summary aging and the sustained growth in the population and table looking at the “Total Service-Delivery Staff” picture. in the expected budget for health. It clearly demon- This includes the: (i) staff to be employed; (ii) population strates the current crisis facing the human resource to service-delivery staff ratios and total service-delivery requirements of the health sector in PNG. staff per 1,000 population ratios; and, critically; (iii) sal- • Scenario 2: Aspirational Targets Envisioned by ary costs for direct service-delivery staff under each sce- PNGDSP 2010–2030. The second set of projections is nario; (iv) expected total recurrent budget (as discussed in based on the number of staff and/or population-to-staff 91 1305747 PNG HR Report 11-27-12.indb 91 11/27/12 11:38 AM ratios posited in the PNGDSP 2010–2030. These are to 2020. This essentially assumes the citizen population in the form of absolute staff numbers for each of med- will grow at about 2.8 percent per annum over this period. ical officers, nurses and CHWs. The report argues that These numbers have been projected forward for the period this scenario: (i) is not affordable on current likely to 2025 at 2.5 percent per annum and from 2025 to 2030 resource envelope projections; and (ii) does not ade- at 2.1 percent per annum.31 quately reflect the likely needed composition of cad- Another common element in each of the five sce- res in the future health workforce. narios is the resource envelope used for health recurrent • Scenario 3: Maintaining Existing Population to Ser- budget growth over the period 2010–2030. A range of vice-Delivery Staff Ratios. The third scenario assumes scenarios are fully discussed in Chapter 5–which cover a the core direct service-delivery health cadres maintain combination of expected GDP growth, growth of the gov- both their current share of the workforce and the cur- ernment recurrent budget and the share of the recurrent rent (2009) population-to-staff ratios over the period budget allocated to health—all significant variables with 2009–2030. The fundamental driver of the demand for significant uncertainty. Nevertheless, the future expecta- health staff in this example is population growth. This tions for growth of the economy and of the government scenario is probably affordable but it is also probably not budget and of health is significantly more positive than the right mix of cadres required for the health workforce. over the past 10–15 years when growth of both the econ- • Scenario 4: WHO Recommended “Threshold” Ser- omy and the government budget in real terms has been less vice-Delivery Staff Density Targets. The fourth sce- than the growth of the population. For the purpose of the nario is based on the WHO “threshold” density of discussion in Chapter 6 we have assumed the health bud- 2.28 per 1,000 population (or population to staff get will grow at about 5 percent in real terms—the mid- ratio of 439 to 1) of doctors, nurses (registered and dle-growth scenario—over the period of the projections. enrolled) and midwives below which, according to This growth rate varies over time, as discussed in Chap- WHO, coverage of essential interventions, including ter 5, but broadly follows the estimates of the IMF and the those necessary to reach the health-related MDGs, is Ministry of Finance (MoF) over the period 2010–2015 and not likely.30 This scenario is not affordable—particu- an average growth of 4–5 percent over subsequent periods. larly in the outer years—and recommends a doctor/ population ratio which is probably not feasible from a supply constraint perspective and which is lower than 6.3. Detailed Scenario Analysis is need to meet the health needs of the population. • Scenario 5: A Suggested Preservice Training Scenario 6.3.1 Scenario 1: No Change in Human for Direct Service-Delivery Staff. The fifth scenario— Resource Supply Capacity 2010–2030 which is the broadly recommended scenario—envisages: This scenario looks at what will happen to the size of (i) a gradual reduction in the population to doctor, nurse the health workforce with no change in the current out- and CHW ratios; (ii) maintaining nurses and CHWs as the backbone of the service-delivery system—par- 30  WHO, 2006. This report notes specifically that countries with ticularly rural service delivery constrained by: (a) the densities of doctors, nurses, and midwives below 2.28 per 1,000 growth in the resource envelope likely to be available population fail, on average, to achieve 80 percent coverage for for health and service-delivery staff; and (b) the fea- live deliveries by skilled birth attendants. sibility and speed with which preservice training can 31  As discussed in Chapter 5 the population projections in the be ramped up to meet the demands of attrition from NHP may overestimate population growth slightly for the period 2010–2020 but have been used for consistency purposes for the the workforce and the needs of a growing population. rest of the NHP. The population projections for the period 2020– 2030 are based on the assumptions for population growth for this One common element of each scenario is the pop- period used by the PNGDSP. These growth rate reductions are prob- ably ambitious and without formal population projections—badly ulation projection used. This is discussed more fully in needed for PNG but outside the scope of this report—it is hard to Chapter 5 but the population projections are based on have too much certainty about the projections. Nevertheless, the the ones used in the NHP (Volume 2) for the years 2010 average trend over the period 2020–2030 is probably reasonable. 92  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 92 11/27/12 11:38 AM puts of the health-related training system—for medi- working longer over the next decade. Working against cal officers, nurses, CHWs and HEOs. The picture that this may well be increased attrition to the private sec- emerges is one of an impending workforce crisis that will tor if this expands as many believe it will—particularly result in a fast-shrinking service-delivery workforce. Each with an expanding urban population and rising GDP of the health service-delivery cadres are discussed in turn from the LNG project and associated developments. and then the total service-delivery staff outcome over the (v) The line for first year student intake required repre- period 2009–2030 is summarized, together with remuner- sents the number of students needing to enter the ation costs to the health budget and remuneration costs program six years earlier given the current estimated as a share of the expected recurrent budget. attrition rate (10 percent) from medical programs over The projections for medical officers are presented in the six years of the program. Thus, to achieve grad- Table 6-1. The key characteristic of this scenario is that uations of 49 the medical school needs to be enroll- it is assumed that new medical graduates remain at the ing about 54 students each year. current 49 per annum over the 20 year projection period. (vi) Costs of employing this number of doctors would rise The key points about Scenario 1 for medical officers in real terms from K34 million in 2009 to K54 mil- as set out in Table 6-1 can be summarized as follows: lion in 2020 and K59 million in 2030. (i) The total number of medical officers employed would rise from 379 in 2009 to 515 in 2015 and to nearly The projections for nursing officers for Scenario 1 595 by 2020. This would increase more modestly to are presented in Table 6-2. The key characteristic of this 636 in 2025 and 656 in 2030. scenario is that it is assumed that new nursing officer (ii) This scenario of a modestly increasing doctors work- graduates remain at the number expecting to graduate force would result in the population to medical offi- from 2012 (following the re-establishment of the Mendi cer ratio falling from 17,512 to 1 medical officer in School of Nursing in 2010. From 2012 it is expected that 2009 to 15,078 to 1 medical officer in 2018 but over there will be 165 graduates per annum over the projec- the next decade or more the population to medical tion period to 2030. officer ratio would increase to 17,277 to 1 in 2030– The key points about Scenario 1 for nurses can be approximately the same as in 2009. summarized as follows: (iii) One reason for this trend ((ii) above) is that popu- (i) The total number of nurses employed would decline lation growth, as discussed above, remains relatively from 3,252 in 2009 to 3,051 in 2015 and to 2,971 by high. The growth in doctor numbers available for the 2020 and would continue to fall to 2,869 in 2030–a workforce is not growing as fast as population growth. decrease of 383 by 2030 or about 12 percent. (iv) A second very significant reason for this is the rate of (ii) This scenario of declining nurse numbers in the work- attrition from the workforce. It is assumed that attrition force would result in the population-to-nurse ratio from the existing doctor workforce will be 6 percent rising from 2,041 to 1 nursing officer in 2009 to 3,038 per annum rising to 7 percent per annum from 2021 to to 1 nursing officer in 2020 and to almost 4,000 per 2030. Thus the number of retirees from the workforce nursing officer in 2030. is estimated to rise from 23 per year in 2009 to about (iii) One reason for this trend ((ii) above) is that population 36 in 2020 and to 46 per year in 2030. At the same time growth, as discussed above, remains relatively high. the new entrants is only 49 per year. We do not have Together with declining nurse numbers, this means studies of the attrition rate of the health workforce. the population-to-nurse ratio is falling dramatically. The last estimate was done in about 2000 but this was (iv) A second very significant reason for this trend is the also a guesstimate. We do know, however, that about rate of attrition from the workforce. For the period of 47 percent of medical officers will reach retirement age the projection, attrition from the workforce is expected by 2020 so this alone could account for attrition from to be greater than the number of new graduates enter- the workforce postulated. To this end the assump- ing the workforce. It is assumed that attrition from tion underlying 6 percent attrition in the next decade the existing workforce will be 6 percent per annum assumes that policies are considered to keep doctors over the period of the projection scenario to 2030. Supply and Demand: Key Health Cadre Supply Gaps   //  93 1305747 PNG HR Report 11-27-12.indb 93 11/27/12 11:38 AM Table 6‑1: Scenario 1: Medical Officers (MO): No Change in Supply Capacity (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 94 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Employed 379 405 430 453 475 495 515 533 550 566 581 Population per MO 17,512 16,852 16,330 15,929 15,625 15,401 15,243 15,142 15,089 15,078 15,105 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 Attrition From MO Workforce 23 24 26 27 28 30 31 32 33 34 35 New Graduates Required 49 49 49 49 49 49 49 49 49 49 49 1st Year Student Intake Required 54 54 54 54 54 54 54 54 54 54 (6 years earlier) MO Employment Cost (Kmn) 34.1 36.5 38.7 40.8 42.7 44.6 46.3 48.0 49.5 50.9 52.3 (Unit cost = K90,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 94  //  PNG Health Workforce Crisis: A Call to Action MOs Employed 595 608 621 626 632 636 641 645 649 652 656 Population per MO 15,166 15,206 15,272 15,515 15,774 16,046 16,269 16,504 16,751 17,008 17,277 MO per 1,000 Population 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From MO Workforce 36 37 43 44 44 45 45 45 45 46 46 New Graduates Required 49 49 49 49 49 49 49 49 49 49 49 1st Year Student Intake Required 54 54 54 54 54 54 54 54 54 54 54 (6 years earlier) MO Employment Cost (Kmn) 53.6 54.8 55.9 56.4 56.8 57.3 57.7 58.0 58.4 58.7 59.0 11/27/12 11:38 AM Table 6‑2: Scenario 1: Nursing Officers: No Change in Supply Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 Nurses Employed 3,252 3,192 3,135 3,112 3,091 3,070 3,051 3,033 3,016 3,000 2,985 1305747 PNG HR Report 11-27-12.indb 95 Population per Nurse 2,041 2,140 2,239 2,319 2,401 2,486 2,572 2,660 2,751 2,844 2,940 Nurse per 1,000 Population 0.49 0.47 0.45 0.43 0.42 0.40 0.39 0.38 0.36 0.35 0.34 Attrition From Nurse 195 192 188 187 185 184 183 182 181 180 179 Workforce New Graduates Required 135 135 135 165 165 165 165 165 165 165 165 1st Year Student Intake 139 139 139 170 170 170 170 170 170 170 170 Required (3 years earlier) Nurse Employment Cost (Kmn) 73.8 72.5 71.2 70.6 70.2 69.7 69.3 68.8 68.5 68.1 67.8 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.0 3.0 3.0 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 (b) Quality-Enhanced Cost 5.0 5.0 5.0 5.9 5.9 5.9 5.9 5.9 5.9 5.9 5.9 (K12,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 Nurses Employed 2,971 2,958 2,945 2,933 2,922 2,912 2,902 2,893 2,885 2,877 2,869 Population per Nurse 3,038 3,128 3,220 3,313 3,409 3,506 3,592 3,679 3,768 3,857 3,949 Nurse per 1,000 Population 0.33 0.32 0.31 0.30 0.29 0.29 0.28 0.27 0.27 0.26 0.25 Attrition From Nurse 178 177 177 176 175 175 174 174 173 173 172 Workforce New Graduates Required 165 165 165 165 165 165 165 165 165 165 165 1st Year Student Intake 174 174 174 174 174 174 174 174 174 174 174 Required (3 years earlier) Nurse Employment Cost (Kmn) 67.4 67.1 66.9 66.6 66.3 66.1 65.9 65.7 65.5 65.3 65.1 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 Supply and Demand: Key Health Cadre Supply Gaps   //  95 (b) Quality-Enhanced Cost 6.3 6.3 6.3 6.3 6.3 6.3 6.3 6.3 6.3 6.3 6.3 (K12,000) 11/27/12 11:38 AM (v) As discussed above, we do not have studies of the From 2012 it expected that there will be 165 graduates attrition rate of the health workforce. We do know, per annum over projection period to 2030. however, that about 50 percent of nurses will reach The key points about Scenario 1 for CHWs can be retirement age by 2020–suggesting that retirements summarized as follows: alone could account for the workforce attrition rates (i) The total number of CHWs employed would decline postulated. The assumption underlying 6 percent from 4,398 in 2009 to about 3,900 in 2015 and to 3,621 attrition in the next decade assumes that policies by 2020 and would continue to fall to about 3,537 in are considered to keep nurses working longer over 2030–a decrease of 861 by 2030 or about 20 percent the next decade. Working against this may well in the total workforce. be increased attrition to the private sector if this (ii) This scenario of declining CHW numbers in the expands as many believe it will—particularly with workforce would result in the population to CHW an expanding urban population and rising GDP ratio rising from 1,509 to 1 CHW in 2009 to 2,492 to from the LNG project and associated developments. 1 CHW in 2020 and to over 3,200 per CHW in 2030. (vi) The line for first year student intake required repre- (iii) One reason for this trend ((ii) above) is that popu- sents the number of students needing to enter the lation growth, as discussed above, remains relatively program three years earlier given the current esti- high. Together with significantly declining CHW mated attrition rate (5 percent) from nurse train- workforce numbers, this means the population-to- ing programs over the three years of the program. CHW ratio is falling dramatically. Thus to achieve nurse graduations of 165 the nurse (iv) Second a very significant reason for this trend is the training schools need to enroll about 174 students significant rate of attrition from the workforce. For into the first year program. the period of the projection attrition from the work- (vii) The cost of employing this number of nurses (in force is expected to be greater than the number of new real terms) would decline significantly from about graduates entering the workforce. It is assumed that K74 million in 2009 to K67 million in 2020 and attrition from the existing workforce will be 7 percent K65 million in 2030. per annum over the period 2009–2020 and 6 percent (viii) Finally, the total recurrent costs of training of over the period 2020–2030. As discussed above, we nurses are presented. First, the total recurrent costs do not have studies of the attrition rate of the health of training based on the 2009 estimate of recur- workforce. We do know, however, that about 56 per- rent unit costs of training (K7,273 in Table 4-7) cent of CHWs will reach retirement age by 2020– are presented. In the case of Scenario 1 these costs suggesting that retirements alone could account for remain constant as numbers under training remain the workforce attrition rates postulated. The assump- constant as do the costs at about K3.7 million per tion underlying 6 percent attrition in the next decade annum from 2012. Second, the costs of training assumes that policies are considered to keep CHWs with a significant improvement in the quality of working longer over the next decade. Working against training (represented by increased unit costs to this may well be increased attrition to the private sec- K12,000 to allow for improved teaching inputs, tor if this expands as many believe it will—particularly library, internet connection, boarding costs and with an expanding urban population and rising GDP maintenance) is presented. This represents an from the LNG project and associated developments. increase in total costs for a constant capacity to (v) The line for first year student intake required repre- K6.1 million from the K3.7 million estimate of sents the number of students needing to enter the current quality teaching. program two years earlier given the current estimated attrition rate (3–5 percent) from nurse-training pro- The projections for CHWs for Scenario 1 are pre- grams over the three years of the program. Thus to sented in Table 6-3 below. The key characteristic of this achieve CHW graduations of 209 the CHW train- scenario is that it is assumed that new CHW graduates ing schools need to enroll about 215 students into remain at the number currently graduating in 2009. the first year program. 96  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 96 11/27/12 11:38 AM Table 6‑3: Scenario 1: Community Health Workers (CHWs): No Change in Supply Capacity Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 CHWs Employed 4,398 4,299 4,207 4,122 4,042 3,968 3,899 3,835 3,776 3,721 3,669 1305747 PNG HR Report 11-27-12.indb 97 Population per CHW 1,509 1,589 1,669 1,751 1,836 1,923 2,012 2,104 2,197 2,293 2,392 CHW per 1,000 Population 0.66 0.63 0.60 0.57 0.54 0.52 0.50 0.48 0.46 0.44 0.42 Attrition From CHW Workforce 308 301 295 289 283 278 273 268 264 260 257 New Graduates Required 209 209 209 209 209 209 209 209 209 209 209 1st Year Student Intake 215 215 215 215 215 215 215 215 215 215 215 Required (2 years earlier) CHW Employment Cost (Kmn) (Unit 73.0 71.4 69.8 68.4 67.1 65.9 64.7 63.7 62.7 61.8 60.9 cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 (b) Quality-Enhanced Costs 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 (K8,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 CHWs Employed 3,621 3,577 3,571 3,566 3,561 3,556 3,552 3,548 3,544 3,540 3,537 Population per CHW 2,492 2,586 2,655 2,725 2,797 2,871 2,935 3,000 3,067 3,134 3,203 CHW per 1,000 Population 0.40 0.39 0.38 0.37 0.36 0.35 0.34 0.33 0.33 0.32 0.31 Attrition From CHW Workforce 253 215 214 214 214 213 213 213 213 212 212 New Graduates Required 209 209 209 209 209 209 209 209 209 209 209 1st Year Student Intake 215 215 215 215 215 215 215 215 215 215 215 Required (2 years earlier) CHW Employment Cost (Kmn) (Unit 60.1 59.4 59.3 59.2 59.1 59.0 59.0 58.9 58.8 58.8 58.7 cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 (b) Quality-Enhanced Costs 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 (K8,000) Supply and Demand: Key Health Cadre Supply Gaps   //  97 11/27/12 11:38 AM (vi) Costs of employing this number of CHWs (in real new entrants to the workforce is only 49 per year. We terms) would decline significantly from about K73 mil- do not have studies of the attrition rate of the health lion in 2009 to K60 million in 2020 and K59 mil- workforce. The last estimate was done in about 2000 lion by 2030. but this was also a guesstimate. We do know, how- (vii) The total recurrent costs of training based on the 2009 ever, that over 50 percent of HEOs will reach retire- estimate of recurrent units costs of training (K6,221 ment age by 2020–suggesting that retirements alone in Table 4-8) are presented. In the case of Scenario 1 could account for the workforce attrition rates pos- these costs remain constant as numbers under train- tulated. To this end the assumption underlying 7 per- ing remain constant as do the costs at about K2.7 mil- cent attrition in the next decade assumes that policies lion per annum. Second, the costs of training from are considered to keep HEOs working longer over a significant improvement in the quality of training the next decade. Working against this may well be (represented by increased unit costs to K8,000 to increased attrition to the private sector if this expands allow for improved teaching inputs, library, internet as many believe it will—particularly with an expand- connection, boarding costs and maintenance) is pre- ing urban population and rising GDP from the LNG sented. This represents an increase in total costs for a project and associated developments. constant capacity to K3.4 million from the K2.7 mil- (v) The line for first year student intake required repre- lion estimate of current quality teaching. sents the number of students needing to enter the pro- gram four years earlier given the current estimated The projections for HEOs are presented in Table 6-4 attrition rate (7 percent) from the HEO program over below. The key characteristic of this scenario is that it the four years of the program. Thus to achieve grad- is assumed that new HEO graduates remain at the cur- uations of 46 the medical school needs to enroll 49 rent 49 per annum over the 20 year projection period. at the start of the program. The key points about Scenario 1 for HEOs can be summarized as follows: Finally, Table 6-5 for Scenario 1 shows the total (i) The total number of HEOs employed would rise from number of service-delivery staff which are estimated 411 in 2009 to 503 in 2015 and to 553 by 2020. This to be available for the health workforce over the pro- would increase to 585 in 2025 and about 607 in 2030. jection period to 2030 assuming no change in existing (ii) This scenario of a modestly increasing HEO work- training capacity. force would result in the population-to-HEO ratio Specifically, the following can be observed from improving from 16,148 to 1 HEO in 2009 to 15,609 Table 6-5: to 1 HEO in 2015 but over the next period of the pro- (i) If existing training capacity is maintained over the jection to 2030 the population-to-HEO ratio would period and the assumptions outlined for each of the rise to 16,322 to 1 HEO in 2020 and to 18,663 to 1 cadres are as discussed above, the total service-deliv- HEO in 2030—a significantly higher population-to- ery workforce for health would decline from 8,440 in HEO ratio than in 2009. 2009 to 7,968 in 2015 and to 7,740 by 2020–a decline (iii) One reason for this trend ((ii) above) is that popu- of 700 or 8 percent. This declining trend would con- lation growth, as discussed above, remains relatively tinue with the service-delivery health workforce esti- high. The growth in HEO numbers available for the mated to be 7,689 in 2025 and to 7,669 in 2030. workforce is not growing as fast as population growth. (ii) As a consequence of the sustained increase in pop- (iv) A second very significant reason for this is the rate of ulation and the decline in the total service-delivery attrition from the workforce. It is assumed that attri- workforce, the population to service-delivery staff tion from the existing HEO workforce will be 7 per- ratio under this scenario can be expected to rise sub- cent per annum through 2030. Thus the number of stantially from 786 to 1 service-delivery staff in 2009 retirees from the workforce is estimated to rise from to 985 to 1 service-delivery staff in 2015 to 1,166 to 1 29 per year in 2009 to about 38 in 2020 and to 42 per staff in 2020. Over the subsequent decade the popula- year in 2030. At the same time the current capacity for tion to service-delivery staff ratio would rise to 1,477 98  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 98 11/27/12 11:38 AM Table 6‑4: Scenario 1: Health Extension Officers (HEOs): No Change in Supply Capacity Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 HEOs Employed 411 428 445 461 476 490 503 514 525 535 544 1305747 PNG HR Report 11-27-12.indb 99 Population per HEO 16,148 15,948 15,761 15,641 15,581 15,573 15,609 15,686 15,799 15,944 16,119 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 29 29 30 31 32 33 34 35 36 37 37 New Graduates Required 46 46 46 46 46 46 46 46 46 46 46 1st Year Student Intake Required 49 49 49 49 49 49 49 49 49 49 49 (4 years earlier) Cost of HEO Employment (Kmn) 10.7 11.1 11.6 12.0 12.4 12.7 13.1 13.4 13.7 13.9 14.2 (Unit cost = K26,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 HEOs Employed 553 561 568 574 580 585 590 595 599 603 607 Population per HEO 16,322 16,494 16,706 16,937 17,185 17,452 17,666 17,895 18,138 18,394 18,663 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.05 Attrition From HEO Workforce 38 39 40 40 41 41 41 42 42 42 42 New Graduates Required 46 46 46 46 46 46 46 46 46 46 46 1st Year Student Intake Required 49 49 49 49 49 49 49 49 49 49 49 (4 years earlier) Cost of HEO Employment (Kmn) 14.4 14.6 14.8 14.9 15.1 15.2 15.3 15.5 15.6 15.7 15.8 (Unit cost = K26,000) (2009) Supply and Demand: Key Health Cadre Supply Gaps   //  99 11/27/12 11:38 AM Table 6‑5: Scenario 1: Total Service-Delivery Staff: No Change in Supply Capacity Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 100 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Employed 379 405 430 453 475 495 515 533 550 566 581 Nurses Employed 3,252 3,192 3,135 3,112 3,091 3,070 3,051 3,033 3,016 3,000 2,985 CHWs Employed 4,398 4,299 4,207 4,122 4,042 3,968 3,899 3,835 3,776 3,721 3,669 HEOs Employed 411 428 445 461 476 490 503 514 525 535 544 Total Service-Delivery Staff 8,440 8,324 8,217 8,148 8,084 8,023 7,968 7,915 7,867 7,822 7,779 Population per Service Staff 786 820 854 886 918 951 985 1,019 1,055 1,091 1,128 Service Staff Per 1,000 Pop. 1.27 1.22 1.17 1.13 1.09 1.05 1.02 0.98 0.95 0.92 0.89 Total Staff Cost (Kmn) 191.6 191.1 191.0 191.5 192.1 192.6 193.4 193.5 194.0 194.4 194.8 Total Recurrent Budget 513.0 551.5 601.1 634.2 646.9 672.7 807.3 823.4 852.2 886.3 921.8 S.D Staff as % Recurrent Budget 37.3 34.7 31.8 30.2 29.7 28.6 23.9 23.5 22.8 21.9 21.1 100  //  PNG Health Workforce Crisis: A Call to Action Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 5.7 5.7 5.7 6.3 6.3 6.3 6.3 6.3 6.3 6.3 6.3 Quality-Enhanced Costs (Kmn) 8.5 8.5 8.5 9.3 9.3 9.3 9.3 9.3 9.3 9.3 9.3 2009 Costs % Recurrent Budget 1.1 1.0 1.0 1.0 1.0 1.0 0.8 0.8 0.8 0.7 0.7 Quality-Enhanced % Recurrent 1.6 1.5 1.4 1.5 1.5 1.4 1.2 1.2 1.1 1.1 1.0 Budget (continued on next page) 11/27/12 11:38 AM Table 6‑5: Scenario 1: Total Service-Delivery Staff: No Change in Supply Capacity Scenario (2010–2030)  (continued) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Employed 595 608 621 626 632 636 641 645 649 652 656 Nurses Employed 2,971 2,958 2,945 2,933 2,922 2,912 2,902 2,893 2,885 2,877 2,869 1305747 PNG HR Report 11-27-12.indb 101 CHWs Employed 3,621 3,577 3,571 3,566 3,561 3,556 3,552 3,548 3,544 3,540 3,537 HEOs Employed 553 561 568 574 580 585 590 595 599 603 607 Total Service-Delivery Staff 7,740 7,704 7,705 7,699 7,695 7,689 7,685 7,681 7,677 7,672 7,669 Population per Service Staff 1,166 1,201 1,231 1,262 1,295 1,328 1,357 1,386 1,416 1,446 1,477 Service Staff Per 1,000 Pop. 0.86 0.83 0.81 0.79 0.77 0.75 0.74 0.72 0.71 0.69 0.68 Total Staff Cost (Kmn) 195.5 195.6 196.5 196.8 197.1 197.5 197.6 197.8 198.0 198.2 198.6 Total Recurrent Budget 958.7 999.9 1,042.9 1,087.7 1,134.5 1,183.3 1,234.1 1,287.2 1,342.6 1,400.3 1,460.5 S.D Staff as % Recurrent Budget 20.4 19.6 18.8 18.1 17.4 16.7 16.0 15.4 14.7 14.2 13.6 Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 Quality Enhanced Costs (Kmn) 9.7 9.7 9.7 9.7 9.7 9.7 9.7 9.7 9.7 9.7 9.7 2009 Costs % Recurrent Budget 0.7 0.6 0.6 0.6 0.6 0.5 0.5 0.5 0.5 0.5 0.4 Quality-Enhanced % Recurrent 1.0 1.0 0.9 0.9 0.9 0.8 0.8 0.8 0.7 0.7 0.7 Budget Supply and Demand: Key Health Cadre Supply Gaps   //  101 11/27/12 11:38 AM to 1 in 2030. This would mean that in 2030 the popu- for 2030 assumed by the PNGDSP is lower by 1.5 million lation to service-delivery staff ratio would be almost (Table 5-1). Under Scenario 2 the PNGDSP target of 4,900 twice that in 2009, in other words that each staffer doctors would be achieved by 2031 if the growth in doc- would, on average, be responsible for almost twice tor numbers was achieved as outlined. the number as at present. Clearly this is an alarm- The key points about Scenario 2 for medical offi- ing scenario. cers can be summarized as follows: (iii) The share of the remuneration costs in the expected (i) The total number of medical officers are planned to health recurrent budget would decline from the cur- grow from 379 in 2009 to 4,184 in 2030 and about rent 37.3 percent to 13.6 percent in 2030. 4,900, as discussed, in 2031. This would result in the (iv) Finally, the costs of the training of both nurses and population per doctor improving from 17,512 to one CHWs are presented both at 2009 unit costs and at to about 2,700 to one in 2030. quality-enhanced unit costs (see above). The total (ii) It is assumed that attrition from the existing work- recurrent costs of training for both nurses and CHWs force will be 6 percent per annum rising to 7 percent are estimated at about K6.4 million (from 2012 when per annum from 2021 to 2030. We do not have stud- the Mendi School of Nursing will be running at ies of the attrition rate of the health workforce. The planned capacity). The costs of a quality-enhanced last estimate was done in about 2000 but this was teaching program for nurses and CHWs are also pre- also a guesstimate. We do know, however, that about sented. This would cost about K9.6 million per annum 47 percent of medical officers will reach retirement (in 2009 prices). age by 2020-suggesting that retirements alone could (v) As a cost of the expected recurrent budget the total account for the workforce attrition rates postulated. costs of nurse and CHW training in 2009 was about The assumption underlying 6 percent attrition in the 1.1  percent. With no change in capacity and no next decade assumes that policies are considered to improvement in unit costs, the share of the expected keep doctors working longer over the next decade. health budget allocated to training of these two cadres Working against this may well be increased attrition would fall to 0.7 percent by 2020 and to 0.4 percent to the private sector if this expands as many believe by 2030. With no change in capacity and at enhanced it will—particularly with an expanding urban pop- unit costs, the training costs of these two cadres would ulation and rising GDP from the LNG project and represent about 1.6 percent of the health budget in associated developments. 2009. This could be expected to fall to 1.0 percent by (iii) New graduates in this scenario are simply the 2020 and to 0.7 percent by 2030. numbers of graduates required, after allowing for: (a) some time to immediately ramp up training; and 6.3.2 Scenario 2: Aspirational Health (b) the length of the course—six years. New medi- Workforce Targets Envisioned in the cal graduates would need to rise from 49 per year in PNGDSP 2010–2030 2009 to 100 in 2019, 200 in 2021 and 700 per year from 2025. Simply put, it is extremely unlikely that This scenario sets targets for medical officers, nurses the infrastructure or staffing required for such an and CHWs for 2030 in either absolute numbers and/ expansion of training would be feasible. In addi- or rates per 100,000 population. It does not consider tion, the state of the education system, and partic- either the timeline or the feasibility of expanding train- ularly senior secondary education, would make it ing of doctors and other staff to achieve these targets. This extremely unlikely that enough students with an scenario is, however, included in this report because it is adequate science background would be available specified in an important national document as at least to enter medical school. an important aspiration. (iv) The line for first year student intake required repre- The formal PNGDSP target for medical officers is sents the number of students needing to enter the 4,900 by 2030 or 50 per 100,000 population (PNGDSP) program six years earlier, given the current estimated (Table 6-6). As discussed in Chapter 5, the population base attrition rate (10 percent) from medical programs over 102  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 102 11/27/12 11:38 AM Table 6‑6: Scenario 2: Medical Officers (MOs): Aspirational Targets Envisioned by PNGDSP (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Plan by PNGDSP 379 405 430 453 475 495 515 533 550 566 581 Population per MO 17,512 16,862 16,327 15,934 15,625 15,415 15,234 15,137 15,085 15,075 15,103 1305747 PNG HR Report 11-27-12.indb 103 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 Attrition from MO Workforce 23 24 26 27 28 30 31 32 33 34 35 New Graduates Required 49 49 49 49 49 49 49 49 49 49 100 1st Year Student Intake 54 54 54 54 54 54 54 54 54 111 Required (6 years earlier) MO Employment Cost (Kmn) (Unit 34.1 36.5 38.7 40.8 42.7 44.6 46.3 48.0 49.5 50.9 52.3 cost = K90,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Plan by PNGDSP 646 707 865 1,004 1,334 1,641 2,226 2,770 3,276 3,747 4,184 Population per MO 13,969 13,084 10,961 9,680 7,468 6,222 4,684 3,843 3,317 2,962 2,707 MO per 1,000 Population 0.07 0.08 0.09 0.10 0.13 0.16 0.21 0.26 0.30 0.34 0.37 Attrition From MO Workforce 39 42 61 70 93 115 156 194 229 262 293 New Graduates Required 100 200 200 400 400 700 700 700 700 700 700 1st Year Student Intake 111 222 222 444 444 778 778 778 778 778 778 Required (6 years earlier) MO Employment Cost (Kmn) 58.1 63.7 77.8 90.4 120.1 147.7 200.3 249.3 294.8 337.2 376.6 Supply and Demand: Key Health Cadre Supply Gaps   //  103 11/27/12 11:38 AM the six years of the program. Thus, to achieve grad- length of the course—three years. New nurse gradu- uations of 200 in 2021, medical schools would need ates would need to rise from 135 per year in 2009 (or to be enrolling at least 221 students into first year by 165 per year from 2012 when the new capacity from 2015 and 778 by 2018. In other words, investment the Mendi School of Nursing comes on line) to 300 decisions would need to be made during the current in 2015, 1,000 in 2021, 1,500 in 2022 and 3,000 per NHP to make this target at all feasible. year from 2025. Simply put, it is extremely unlikely that the infrastructure or staffing required for such The cost of employing this number of doctors— an expansion of training would be feasible. In addi- aside from issues of real need to achieve the health tion, the state of the education system, and partic- outcomes desired—also raises feasibility questions. In ularly senior secondary education, would make it 2009 expenditure on doctors is estimated at K34.1 mil- extremely unlikely that enough students with an lion. This would increase to over K376 million by 2030 or adequate science background and aptitude would be about 26 percent of the expected health recurrent budget available to enter nursing schools. This issue would in 2030 (see Table 6-9 below) compared to K34.1 million be particularly acute for the remoter areas—where a in 2009 (7 percent of the total health recurrent budget). policy of recruiting locally is a recommended strat- The projections for nurses for Scenario 2 are pre- egy to get nurses going to areas currently lacking in sented in Table 6-7 below. The PNGDSP sets the target staff—unless a decision was made to reduce school for nurses as 19,500 nurses by 2030 or 200 nurses per qualifications required for entry to nursing. 100,000 population. As discussed above, the population (iv) The line for first year student intake required repre- estimates used in this report are 1.5 million higher than sents the number of students needing to enter the the PNGDSP report. Thus we use the target of 19,500 program three years earlier—given the current esti- nurses as the specific one for this scenario for nurses. mated attrition rate (5 percent) from nurse training The key points about Scenario 2 for nurses can be programs over the three years of the program. Thus, summarized as follows: to achieve nurse graduations of 300 in 2015 nurse (i) The total numbers of nurses is planned to grow from training schools would need to be enrolling at least 3,252 in 2009 to 19,526 in 2030. This would result in 309 students into first year by 2012, increasing to 515 the population per nurse ratio improving from 2,041 by 2014, 1,500 by 2022 and 3,000 by 2025. This is pos- to one to about 580 to one. sible with an emergency program to ramp up train- (ii) It is assumed that attrition from the existing workforce ing. In other words, investment decisions would need will be 6 percent per annum over the period of the to be made immediately and of considerable scale in projection scenario to 2030. As discussed above, we the current NHP to make this target at all feasible. do not have studies of the attrition rate of the health (v) The cost of employing this number of nurses—aside workforce. We do know, however, that about 50 per- from issues of real need to achieve the health outcomes cent of nurses will reach retirement age by 2020–sug- desired—also raises feasibility questions. In 2009, gesting that retirements alone could account for the expenditure on nurses is estimated at K73.8 million. workforce attrition rates postulated. The assump- This would increase to over K443.2 million by 2030 tion underlying 6 percent attrition in the next decade or about 30 percent of the expected health recurrent assumes that policies are considered to keep nurses budget in 2030 (Table 6-9) compared to K34.1 mil- working longer over the next decade. Working against lion in 2009 (6.6 percent of the total health recur- this may well be increased attrition to the private sec- rent budget). tor if this expands as many believe it will—particularly (vi) The total recurrent costs of training based on the 2009 with an expanding urban population and rising GDP estimate of recurrent units costs of training (K7,273 from the LNG project and associated developments. in Table 4-7) are presented (Table 6-7). In the case of (iii) New graduates in this scenario are simply the num- this scenario these costs would increase from about bers of graduates required, after allowing for: (i) some K3 million in 2009 to about K10.9 million in 2020. time to immediately ramp up training; and (ii) the These costs would increase to about K65.5 million by 104  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 104 11/27/12 11:38 AM Table 6‑7: Scenario 2: Nursing Officers (NOs): Aspirational Targets Envisioned by PNGDSP (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 Nurses Plan by PNGDSP 3,252 3,192 3,135 3,082 3,062 3,044 3,026 3,144 3,256 3,560 3,847 Population per Nurse 2,041 2,140 2,239 2,342 2,423 2,507 2,593 2,566 2,548 2,396 2,281 1305747 PNG HR Report 11-27-12.indb 105 Nurse per 1,000 Population 0.49 0.47 0.45 0.43 0.41 0.40 0.39 0.39 0.39 0.42 0.44 Attrition From Nurse Workforce 195 192 188 185 184 183 182 189 195 214 231 New Graduates Required 135 135 135 165 165 165 300 300 500 500 500 1st Year Student Intake Required 139 139 139 170 170 170 309 309 515 515 515 (3 years earlier) Nurse Employment Cost (Kmn) 73.8 72.5 71.2 70.0 69.5 69.1 68.7 71.4 73.9 80.8 87.3 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.0 3.0 3.0 3.6 3.6 3.6 6.5 6.5 10.9 11.2 11.2 (b) Quality-Enhanced Costs 4.9 4.9 4.9 5.9 5.9 5.9 10.8 10.8 18.5 18.5 18.5 (K12,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 Nurses Plan by PNGDSP 4,116 4,369 5,107 6,300 7,422 8,477 10,968 13,310 15,512 17,581 19,526 Population per Nurse 2,193 2,117 1,857 1,543 1,342 1,205 950 800 701 631 580 Nurse per 1,000 Population 0.46 0.47 0.54 0.65 0.75 0.83 1.05 1.25 1.43 1.58 1.72 Attrition From Nurse Workforce 247 262 306 378 445 509 658 799 931 1,055 1,172 New Graduates Required 500 1,000 1,500 1,500 1,500 3,000 3,000 3,000 3,000 3,000 3,000 1st Year Student Intake 526 1,053 1,579 1,579 1,579 3,158 3,158 3,158 3,158 3,158 3,158 Required (3 years earlier) Nurse Employment Cost (Kmn) 93.4 99.2 115.9 143.0 168.5 192.4 249.0 302.1 352.1 399.1 443.2 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 10.9 21.8 32.7 32.7 32.7 65.5 65.5 65.5 65.5 65.5 65.5 (b) Quality-Enhanced Cost 18.5 36.0 54.0 54.0 54.0 108.0 108.0 108.0 108.0 108.0 108.0 (K12,000) Supply and Demand: Key Health Cadre Supply Gaps   //  105 11/27/12 11:38 AM 2025 and remain constant as nurse training intakes ondary education, would make it extremely unlikely stabilized at about 3,000 students per year. Second, that enough students with an adequate science back- the costs of training from a significant improvement ground and aptitude would be available to enter CHW in the quality of training (represented by increased schools unless there is effort to increase the propor- unit costs to K12,000 to allow for improved teach- tion of Grade 12 students entering the program and a ing inputs, library, internet connection, boarding decision is made to revert to all students coming from costs and maintenance) is presented. The costs of an a Grade 10 background (see discussion in Chapter 4). “enhanced” quality of nursing graduate would increase (iv) The line for first year student intake required repre- from about K4.9 million in 2009 (with an enhanced sents the number of students needing to enter the unit cost of K12,000 per annum) to almost K11 mil- program two years earlier given the current esti- lion by 2015 and K18 million by 2020. It would reach mated attrition rate (3–5 percent) from CHW train- almost K108 million by 2025 and remain at this level ing programs over the two years of the program. through 2030. Thus, to achieve CHW graduations of 500 in 2015 CHW training schools would need to be enrolling The projections for CHWs for Scenario 2 are pre- at least 515 students into first year by 2013 (possible sented in Table 6-8. The PNGDSP sets the target for at a stretch) and 1,031 by 2015, 1,684 by 2019 and CHWs as 20,000 by 2030 (it does not specify a CHW 2,211by 2022. In other words, investment decisions per 100,000 population ratio). Thus we use the target of would need to be made immediately and of consid- 20,000 CHWs as the basis for this scenario for CHWs. erable scale in the current NHP to make this target The key points about Scenario 1 for CHWs can be at all feasible. At present all CHW training is con- summarized as follows: ducted by church agencies. It is highly unlikely that (i) The total number of CHWs is planned to grow from these would be prepared to expand so rapidly for gov- 4,398 in 2009 to 18,795 in 2030 and almost 20,000 by ernment facilities—thus much of the training would 2031. This would result in the population per CHW need to be done by new government training insti- ratio improving from 1,509 to one to about 603 to one. tutions established from scratch. (ii) It is assumed that attrition from the existing workforce (v) The cost of employing this number of CHWs—aside will be 6 percent per annum over the period to 2020 from issues of real need to achieve the health out- and 5 percent through 2030. As discussed above, we comes desired—also raises feasibility questions. In do not have studies of the attrition rate of the health 2009 expenditure on CHWs is estimated at K73.0 mil- workforce. We do know, however, that about 56 per- lion (about the same as nurses in 2009). This would cent of CHWs will reach retirement age by 2020–sug- increase to over K312 million by 2030 or about 21 per- gesting that retirements alone could account for the cent of the expected health recurrent budget in 2030 workforce attrition rates postulated. The assumption (Table 6-9) compared to K73 million in 2009 (14.2 per- underlying 6 percent attrition in the next two decades cent of the total health recurrent budget). assumes that policies are considered to keep nurses (vi) Finally, the total recurrent costs of training of CHWs working longer over the next decade. based on the 2009 estimate of recurrent unit costs of (iii) New graduates in this scenario are simply the numbers training (K6,221 in Table 4-8) are presented. In the of graduates required, after allowing for: (i) some time case of Scenario 2 these costs would increase from to immediately ramp up training; and (ii) the length about K2.7 million in 2009 to K6.4 million in 2015 and of the course—two years. New CHW graduates would almost K13 million by 2017 when first year intakes need to rise from 209 per year in 2009 to 500 in 2015, would be 1,031–up from about 215 in 2009. With 1,000 in 2017, 1,600 in 2021 and 2,100 per year from the training program ramping up to meet the desired 2024. Simply put, it is extremely unlikely that the infra- targets (intake of almost 1,700 by 2021) the costs of structure or staffing required for such an expansion training would increase to K21 million in 2021 and of training would be feasible. In addition, the state to K27.5 million by 2024 and stabilize with student of the education system, and particularly senior sec- graduates at about 2,100 per year. Second, the costs of 106  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 106 11/27/12 11:38 AM Table 6‑8: Scenario 2: Community Health Workers (CHWs): Aspirational Targets Envisioned by PNGDSP (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 CHWs Plan by PNGDSP 4,398 4,299 4,207 4,122 4,042 3,968 3,899 4,126 4,338 5,034 5,682 Population per CHW 1,509 1,589 1,669 1,751 1,836 1,923 2,012 1,955 1,913 1,695 1,544 1305747 PNG HR Report 11-27-12.indb 107 CHWs per 1,000 Population 0.66 0.63 0.60 0.57 0.54 0.52 0.50 0.51 0.52 0.59 0.65 Attrition From CHW Workforce 308 301 295 289 283 278 273 289 304 352 398 New Graduates Required 209 209 209 209 209 209 500 500 1,000 1,000 1,000 1st Year Student Intake Required 215 215 215 215 215 215 515 515 1,031 1,031 1,031 (2 years earlier) CHW Employment Cost (Kmn) 73.0 71.4 69.8 68.4 67.1 65.9 64.7 68.5 72.0 83.6 94.3 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 2.7 2.7 2.7 2.7 2.7 6.4 6.4 12.8 12.8 12.8 (b) Quality-Enhanced Costs 3.4 3.4 3.4 3.4 3.4 3.4 8.2 8.2 16.5 16.5 16.5 (K8,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 CHWs Plan by PNGDSP 6,284 6,844 8,102 9,297 10,432 12,010 13,510 14,934 16,288 17,573 18,795 Population per CHW 1,436 1,352 1,170 1,045 955 850 772 713 667 631 603 CHWs per 1,000 Population 0.70 0.74 0.85 0.96 1.05 1.18 1.30 1.40 1.50 1.58 1.66 Attrition From CHW Workforce 440 342 405 465 522 601 675 747 814 879 940 New Graduates Required 1,000 1,600 1,600 1,600 2,100 2,100 2,100 2,100 2,100 2,100 2,100 1st Year Student Intake 1,031 1,684 1,684 1,684 2,211 2,211 2,211 2,211 2,211 2,211 2,211 Required (2 years earlier) CHW Employment Cost (Kmn) (Unit 104.3 113.6 134.5 154.3 173.2 199.4 224.3 247.9 270.4 291.7 312.0 cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 12.8 21.0 21.0 21.0 27.5 27.5 27.5 27.5 27.5 27.5 27.5 (b) Quality-Enhanced Costs 16.5 26.9 26.9 26.9 35.4 35.4 35.4 35.4 35.4 35.4 35.4 (K8,000) Supply and Demand: Key Health Cadre Supply Gaps   //  107 11/27/12 11:38 AM training with a significant improvement in the qual- available) rise from K5.7 million in 2009 to K12.9 mil- ity of training (represented by increased unit costs to lion in 2015 and K24 million by 2017 (Table 6-9). It K8,000 to allow for improved teaching inputs, library, would further increase to K43 million by 2021 (as a con- internet connection, boarding costs and maintenance) sequence of rapid student expansion) and to level off at is presented. The costs of this “quality” enhanced about K93 million by 2025 through 2030. As a share of teaching program would increase under this scenario the expected recurrent health budget the costs of train- from K3.4 million in 2009 to K8.2 million by 2015, ing nurses and CHWs at 2009 unit costs would increase K16.5 million by 2017, K27 million by 2021 and to from 1.1 percent of the budget to 1.6 percent by 2015 and K35 million by 2024 when student intakes would be 2.8 percent by 2017. It would eventually rise to about 8 per- 2,100 compared to 215 in 2009. cent of the expected budget by 2025 and then decline to about 6.6 percent of the expected budget in 2030. This scenario does not have a set of proposals for If we use a “reasonable quality-enabling” unit HEOs—for the purposes of Scenario 2 we assume the cost (see discussion above) the total costs of training program for HEOs is shut down and the capacity for would increase from K8.3 million in 2009 to K19 mil- this training is used to ramp up medical officer train- lion by 2015 and K34.5 million by 2020. It would then ing. If doctors were expanded at the rate proposed there increase dramatically to about K63 million by 2021 and would be little rationale for sustaining the HEO cadre. Thus to K143 million by 2025–this level of expenditure would they would phase out as they retired. This is the assump- be sustained through 2030. The costs of nurse and HEO tion underlying the data on HEOs included in Table 6-9 training would increase as a share of the expected health which looks at trends in the total service-delivery cadres budget from 1.6 percent in 2009 to 4 percent by 2017 and proposed under the PNGDSP. 12.1 percent by 2025 when it would then begin to decline The data presented in Table 6-9 shows clearly that to about 10 percent of the expected health budget by 2030. this plan is extremely ambitious. The total number of service-delivery staff by 2030 would be about 42,700–up 6.3.3 Scenario 3: Maintaining Existing from 8,440 in 2009. It would reach 8,183 by 2016, 11,346 Population-to-Service Delivery Staff Cadre by 2020 and then rapidly ramp up to almost 22,400 by Ratios 2010 to 2030 2025 and 42,700 by 2030. This would enable the popu- lation to total service-delivery staff ratio to shift from The discussion of Scenarios 3 to 5 does not repeat the 786 to one in 2009 to about 795 to one in 2020 and then assumptions included in the relevant scenario except ramp up rapidly, as training capacity is expanded, to about where they vary from the more detailed description 456 to one in 2025 and 265 to one in 2030. These ratios of Scenarios 1 and 2. Thus, this and subsequent scenar- imply far lower staff-to-population ratios than those pro- ios assumes: (i) the same population projections; (ii) the posed by WHO (see Scenario 3 below) for a country at same attrition rates from the workforce for each cadre as the stage of the epidemiological transition that PNG cur- described above for Scenarios 1 and 2; and (iii) the same rently finds itself. dropout rates from the respective training programs. A critical part of the data presented in Table 6-9 The fundamental drive for this scenario is popula- is the cost implied by trying to employ all the planned tion growth which assumes the core direct service-deliv- staff. In 2009 the total expenditure on direct service- ery health cadres maintain both their current share of delivery staff was K191.6 million or about 37 percent of the workforce and the current (2009) population-to- the total health recurrent budget. If we assume the budget staff ratios over the period 2009–2030. It suggests, there- grows at an average of 5 percent per annum (the exam- fore, that additional demand for services can be achieved ple shown)32 the total cost of service-delivery staff would by using the existing staff more efficiently (the “Monash” rise to 78 percent of the expected total recurrent budget. Study suggests existing rural staff could increase services Using the recurrent unit costs of training of both 40 percent) and letting the workforce grow at the rate of nurses and CHWs, this program would see training costs (excluding doctors and HEOs for which data is not 32  See discussion above and in Chapter 5. 108  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 108 11/27/12 11:38 AM Table 6‑9: Scenario 2: Total Service-Delivery Staff: Aspirational Targets Envisioned by PNGDSP (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Plan by PNGDSP 379 405 430 453 475 495 515 533 550 566 581 Nurses Plan by PNGDSP 3,252 3,192 3,135 3,082 3,062 3,044 3,026 3,144 3,256 3,560 3,847 1305747 PNG HR Report 11-27-12.indb 109 CHWs Plan by PNGDSP 4,398 4,299 4,207 4,122 4,042 3,968 3,899 4,126 4,338 5,034 5,682 HEOs Existing Stock Only 411 440 440 420 415 411 400 380 360 340 320 Total Service-Delivery Staff 8,440 8,336 8,212 8,077 7,994 7,918 7,840 8,183 8,504 9,500 10,430 Population per Service Staff 786 819 855 894 928 964 1,001 986 976 898 841 Service Staff per 1,000 Pop. 1.27 1.22 1.17 1.12 1.08 1.04 1.00 1.01 1.02 1.11 1.19 Total Staff Cost (Kmn) 191.6 191.4 190.8 189.8 189.8 189.9 190.1 197.4 204.4 223.8 241.9 Total Recurrent Budget 513.0 551.5 601.1 634.2 646.9 672.7 807.3 823.4 852.2 886.3 921.8 S.D Staff as % Recurrent Budget 37.3 34.7 31.7 29.9 29.3 28.2 23.5 24.0 24.0 25.3 26.2 Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 5.7 5.7 5.7 6.3 6.3 6.3 12.9 12.9 24.0 24.0 24.0 Quality-Enhanced Costs (Kmn) 8.3 8.3 8.3 9.3 9.3 9.3 19.0 19.0 34.5 34.5 34.5 2009 Costs % Recurrent Budget 1.1 1.0 1.0 1.0 1.0 1.0 1.6 1.6 2.8 2.7 2.6 Quality-Enhanced % Recurrent 1.6 1.5 1.4 1.5 1.5 1.4 2.4 2.3 4.0 3.9 3.7 Budget (continued on next page) Supply and Demand: Key Health Cadre Supply Gaps   //  109 11/27/12 11:38 AM Table 6‑9: Scenario 2: Total Service-Delivery Staff: Aspirational Targets Envisioned by PNGDSP (2010–2030)  (continued) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 1305747 PNG HR Report 11-27-12.indb 110 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Plan by PNGDSP 646 707 865 1,004 1,334 1,641 2,226 2,770 3,276 3,747 4,184 Nurses Plan by PNGDSP 4,116 4,369 5,107 6,300 7,422 8,477 10,968 13,310 15,512 17,581 19,526 CHWs Plan by PNGDSP 6,284 6,844 8,102 9,297 10,432 12,010 13,510 14,934 16,288 17,573 18,795 HEOs Existing Stock Only 300 290 280 270 260 250 240 230 220 210 200 Total Service-Delivery Staff 11,346 12,210 14,354 16,871 19,448 22,378 26,944 31,244 35,296 39,111 42,705 Population per Service Staff 795 758 661 576 512 456 387 341 308 284 265 Service Staff Per 1,000 Pop. 1.26 1.32 1.51 1.74 1.95 2.19 2.58 2.94 3.25 3.52 3.77 Total Staff Cost (Kmn) 263.6 283.5 335.0 394.1 467.7 546.0 678.7 804.0 921.5 1,137 1,135.1 Total Recurrent Budget 958.7 999.9 1,042.9 1,087.7 1,134.5 1,183.3 1,234.1 1,287.2 1,342.6 1,400.3 1,460.5 S.D Staff as % Recurrent Budget 27.5 28.4 32.1 36.2 41.2 46.1 55.0 62.5 68.6 73.7 77.7 110  //  PNG Health Workforce Crisis: A Call to Action Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 24.0 42.8 53.7 55.4 60.2 93.0 96.4 96.4 96.4 96.4 96.4 Quality-Enhanced Costs (Kmn) 34.5 62.9 80.9 83.8 89.4 143.9 149.1 149.1 149.1 149.1 149.1 2009 costs % Recurrent Budget 2.5 4.3 5.1 4.9 5.3 7.9 7.5 7.2 6.9 6.6 6.6 Quality-Enhanced % Recurrent 3.6 6.3 7.8 7.4 7.9 12.1 11.6 11.1 10.7 10.2 10.2 Budget 11/27/12 11:38 AM Table 6‑10: Scenario 3: Medical Officers (MOs): Maintaining Existing Population to Staff Ratio (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Required 379 390 401 412 424 436 448 461 474 487 501 Population per MO 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 1305747 PNG HR Report 11-27-12.indb 111 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From MO Workforce 23 23 24 25 25 26 27 28 28 29 30 New Graduates Required 49 34 35 36 37 38 39 40 41 43 44 1st Year Student Intake Required 38 39 40 41 42 44 45 46 47 49 (6 years earlier) MO Employment Cost (Kmn) 34.1 35.1 36.1 37.1 38.2 39.2 40.3 41.5 42.7 43.8 45.1 (Unit cost = K90,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Required 515 528 541 555 569 583 595 608 621 634 647 Population per MO 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 17,512 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From MO Workforce 31 32 38 39 40 41 42 43 43 44 45 New Graduates Required 45 45 51 52 54 55 54 55 56 57 59 1st Year Student Intake Required 50 50 57 58 60 61 60 61 62 64 65 (6 years earlier) MO Employment Cost (Kmn) 46.4 47.5 48.7 49.9 51.2 52.5 53.6 54.7 55.9 57.1 58.2 (Unit cost = K90,000) (2009) Supply and Demand: Key Health Cadre Supply Gaps   //  111 11/27/12 11:38 AM Table 6‑11: Scenario 3: Nursing Officers (NOs): Maintaining Existing Population to Staff Ratios (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 112 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 Nurses Required 3,252 3,346 3,440 3,537 3,636 3,739 3,844 3,953 4,065 4,181 4,300 Population per Nurse 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 Nurse per 1,000 Population 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 Attrition From Nurse Workforce 195 201 206 212 218 224 231 237 244 251 258 New Graduates Required 135 165 165 309 318 327 336 346 356 366 377 1st Year Student Intake 139 304 310 318 328 337 347 357 367 378 389 Required (3 years earlier) Nurse Employment Cost (Kmn) 73.8 76.0 78.1 80.3 82.5 84.9 87.3 89.7 92.3 94.9 97.6 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.0 6.4 6.5 6.7 6.9 7.1 7.3 7.5 8.0 8.0 8.2 (b) Quality-Enhanced Cost 4.9 10.6 10.8 11.1 11.4 11.8 12.1 12.5 12.8 13.2 13.6 112  //  PNG Health Workforce Crisis: A Call to Action (K12,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 Nurses Required 4,422 4,533 4,646 4,762 4,881 5,003 5,108 5,216 5,325 5,437 5,551 Population per Nurse 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 2,041 Nurse per 1,000 Population 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 Attrition From Nurse Workforce 265 272 279 286 293 300 306 313 320 326 333 New Graduates Required 388 383 392 402 412 422 412 420 429 438 447 1st Year Student Intake 408 403 413 423 434 444 433 442 452 461 471 Required (3 years earlier) Nurse Employment Cost (Kmn) 100.4 102.9 105.5 108.1 110.8 113.6 116.0 118.4 120.9 123.4 126.0 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 8.5 8.3 8.6 8.8 9.0 9.2 9.0 9.2 9.4 9.6 9.8 (b) Quality-Enhanced Cost 14.7 14.5 14.9 15.2 15.6 16.0 15.6 15.9 16.3 16.6 16.9 (K12,000) 11/27/12 11:38 AM Table 6‑12: Scenario 3: Community Health Workers (CHW): Maintaining Existing Population to Staff Ratios (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 CHWs Required 4,398 4,525 4,652 4,783 4,918 5,057 5,199 5,346 5,498 5,654 5,815 Population per CHW 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1305747 PNG HR Report 11-27-12.indb 113 CHWs per 1,000 Population 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 Attrition From CHW Workforce 308 317 326 335 344 354 364 374 385 396 407 New Graduates Required 209 444 453 466 479 493 507 522 536 552 568 1st Year Student Intake Required 215 458 467 480 494 508 522 538 553 569 585 (2 years earlier) CHW Employment Cost (Kmn) 73.0 75.1 77.2 79.4 81.6 83.9 86.3 88.7 91.3 93.9 96.5 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 5.7 5.8 6.0 6.1 6.3 6.5 6.7 6.9 7.1 7.3 (b) Quality-Enhanced Costs 3.4 7.3 7.5 7.7 7.9 8.1 8.4 8.6 8.8 9.1 9.4 (K8,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 CHWs Required 5,980 6,130 6,283 6,440 6,601 6,766 6,908 7,053 7,202 7,353 7,507 Population per CHW 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 1,509 CHWs per 1,000 Population 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 0.66 Attrition From CHW Workforce 419 368 377 386 396 406 415 423 432 441 450 New Graduates Required 584 517 530 543 557 571 557 568 580 592 605 1st Year Student Intake 602 533 547 560 574 589 574 586 598 611 624 Required (2 years earlier) CHW Employment Cost (Kmn) 99.3 101.8 104.3 106.9 109.6 112.3 114.7 117.1 119.5 122.1 124.6 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (kmn) (a) 2009 Costs (K6,221) 7.5 6.6 6.8 7.0 7.1 7.3 7.1 7.3 7.4 7.6 7.8 (b) Quality-Enhanced Costs 9.6 8.5 8.7 9.0 9.2 9.4 9.2 9.4 9.6 9.8 10.0 (K8,000) Supply and Demand: Key Health Cadre Supply Gaps   //  113 11/27/12 11:38 AM population growth. Thus it does not imply maintaining the (v) The total recurrent costs of training based on the existing service levels in rural areas but envisages signif- 2009 estimate of recurrent unit costs of training icantly increased service levels and sustaining them over (K7,273 in Table 4-7) are presented. In Scenario 3 time. Further, it is consistent with many of the implications these costs would increase from about K3 million in the NHP that the existing cadres remain relevant to health in 2009 to about K7 million in 2013 to “catch up” service delivery. It is noted that no major case is made in with training capacity to the population growth rate. the NHP to adjust the structure of service-delivery cadres. These costs would increase to about K9 million by The key observations about this scenario for med- 2024 and to K10 million by 2030 as student entry to ical officers are: nurse training increases to ensure supply of gradu- (i) The number of doctors would increase from 379 in ates grows to enable the population-to-nurse ratio 2009 to 515 in 2020 and to 647 in 2030. This would sus- to be sustained. Second, the costs of training with a tain a population-to-doctor ratio of 17,512 per doctor. significant improvement in the quality of training (ii) The number of graduates required from medical (represented by increased unit costs to K12,000 to school(s) to sustain these ratios would decrease from allow for improved teaching inputs, library, internet 49 at present to between the mid-thirties to mid-for- connection, boarding costs and maintenance) is pre- ties over the next decade and gradually rise to about sented. The costs of an “enhanced” quality of nurs- 60 per year by 2030. In other words, sustaining cur- ing graduate would increase from about K5 million rent capacity would achieve the target implied. in 2009 to K12.5 million by 2015 and K14.4 million (iii) The costs of employing the staff would increase from by 2020. It would reach almost K16 million by 2025 K34.1 million in 2009 to 58.2 million in 2030. and K17 million by 2030. The key observations about the projections for The key observations about the projections for nurses under Scenario 3 are that: CHWs under Scenario 3 are that: (i) The number of nurses would increase from 3,252 in (i) The number of CHWs would increase from 4,398 in 2009 to 4,422 in 2020 and 5,551 in 2030. This would 2009 to 5,980 in 2020 and over 7,500 in 2030. This sustain the population-to-nurse ratio at 2,041 per would sustain the population-to-CHW ratio at 1,509 nurse over this period. per CHW over this period. (ii) The number of graduates from the nursing schools (ii) The number of graduates from the CHW training would need to increase from the expected numbers schools would need to increase from the expected graduating in 2009 to around 300 immediately and numbers graduating in 2009 of 209 to around 450 this would need to rise to 336 in 2015, 388 by 2020 immediately and this would need to rise to over 500 and almost 450 by 2030. in 2015, 584 by 2020 and over 600 by 2030. (iii) It should be noted that the projected attrition from (iii) It should be noted that the projected attrition from the nurse workforce is significantly greater than the the CHW workforce is significantly greater—about 2009 output. Without expansion of training above and 50 percent—than the current 2009 output. This sug- beyond the opening of the Mendi School of Nursing, gests that there is a real emerging crisis on the supply which should increase nursing diploma graduates to side for CHWs. Specifically, (and this applies to all about 165 per annum, the population to nurse ration scenarios), it can be expected that there may be a fall will not be maintained. Thus this scenario is suggest- in the total CHW workforce unless the attrition rate ing that nurse training capacity increase from 165 in from the workforce can be reduced (see Scenario 1). 2012 to 388 by 2020. New graduates would need to This might be achieved by encouraging CHWs to increase a further 16 percent over the 2020 to reach postpone retirement. Thus this scenario is suggest- 447 per year by 2030. ing that CHW training capacity, given expected work- (iv) The cost of the nurse expansion to the budget would force attrition, needs to grow about 190 percent from grow from K 73.8 million in 2009 to K100 million in current 2009 output of 209 graduates to over 580 by 2020 and to K126 million in 2030. 2020 and over 600 by 2030. It will be a significant 114  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 114 11/27/12 11:38 AM Table 6‑13: Scenario 3: Health Extension Officers (HEOs): Maintaining Existing Population-to-Staff Ratios (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 HEOs Required 411 423 435 447 460 473 486 500 514 528 543 Population per CHW 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 1305747 PNG HR Report 11-27-12.indb 115 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 29 30 30 31 32 33 34 35 36 37 38 New Graduates Required 46 42 42 44 45 46 47 49 50 52 53 1st Year Student Intake Required 49 45 45 47 48 50 51 52 54 55 57 (4 years earlier) Cost of HEO Employment (Kmn) 10.7 11.0 11.3 11.6 11.9 12.3 12.6 13.0 13.4 13.7 14.1 (Unit cost = K26,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 HEOs Required 559 573 587 602 617 632 646 659 673 687 702 Population per HEO 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 39 40 41 42 43 44 45 46 47 48 49 New Graduates Required 55 54 55 57 58 60 58 60 61 62 64 1st Year Student Intake Required 59 58 60 61 63 64 63 64 66 67 68 (4 years earlier) Cost of HEO Employment (Kmn) 14.5 14.9 15.3 15.6 16.0 16.4 18.8 17.1 17.5 17.8 18.3 (Unit cost = K26,000) (2009) Supply and Demand: Key Health Cadre Supply Gaps   //  115 11/27/12 11:38 AM challenge in the short run just expanding produc- to 13,000 by 2025 and reach 14,400 by 2030. This would tion to replace the existing workforce. enable the population to total service-delivery staff ratio (iv) The cost of the CHW expansion to the budget would to remain at 786 to one over the projection period to 2030. grow from K73 million in 2009 to almost K100 mil- The cost of employing all the staff under this lion in 2020 and to K125 million in 2030. scenario is also presented in Table 6-14. In 2009 the (v) The total recurrent costs of training of CHWs—based total expenditure on direct service-delivery staff was on the 2009 estimate of recurrent unit costs of train- K191.6 million or just over 37 percent of the total health ing (K6,221 in Table 4-8)–would increase from about recurrent budget and these costs would rise to nearly K2.7 million in 2009 to K6.5 million in 2015 and almost K327 million by 2030 in real terms. With the recurrent K7.5 million by 2020 when first year intakes would budget assumed to grow at about 5 percent in real terms be 602–up from about 215 in 2009. The cost of train- over the projection period, the share of total service-deliv- ing would be around K7 million by 2023 and would ery staff cost to the budget would fall from 37 percent to expand marginally to K7.8 million by 2030. Second, the about 22 percent of the estimated total health recurrent costs of training assuming a significant improvement budget. The primary driver of this decline is that health in the quality of training (represented by increased expenditures are expected to grow faster than the pop- unit costs to K8,000 to allow for improved teaching ulation growth rate. inputs, library, internet connection, boarding costs Using the recurrent unit costs of training of both and maintenance) is presented. The cost of this qual- nurses and CHWs, this program would see training costs ity-enhanced teaching program would increase under (excluding doctors and HEOs for which data is not avail- this scenario from K3.4 million in 2009 to K8.4 mil- able) rise from K5.7 million in 2009 to K13.8 million lion by 2015, K9.6 million by 2020 and this would be in 2015 and K16.0 million by 2020. It would gradually roughly sustained over the subsequent decade. rise to K17.6 million by 2030. As a share of the expected recurrent health budget the costs of training nurses and The key observations on the projections for HEOs CHWs at 2009 unit costs would increase from 1.1 per- under Scenario 3 are that: cent of the budget to 1.7 percent by 2015. This would be (i) The number of HEOs would increase from 411 in 2009 sustained through 2020 and would then decline as a per- to 559 in 2020 and about 700 in 2030. This would sus- cent of the expected health budget to 1.2 percent by 2030– tain the population-to-HEO ratio at 16,148 per HEO largely as a consequence of the expected growth of the over this period. recurrent health budget. (ii) The number of graduates from the DWU training If we use a “reasonable quality-enabling” unit cost program for HEOs would need to increase from the (see discussion above) the total costs of training would numbers graduating in 2009 of 46 to around 55 by increase from K8.5 million in 2009 to K20.8 million by 2020 and about 64 by 2030. 2015 and K24 million by 2020. It would then increase to (iii) It should be noted that the projected attrition K25.4 million by 2025 and to almost K27 million in 2030. from the HEO workforce would account for about The cost of nurse and HEO training would increase as a 70 percent of the required output of HEOs under share of the health budget from 1.6 percent in 2009 to this scenario. 2.6 percent by 2015 and this would be sustained through (iv) The cost of HEOs to the budget would grow from 2020 and it would then fall away to 1.8 percent of the K10.7 million in 2009 to K14.1 million in 2020 and expected recurrent budget by 2030. to K18.7 million in 2030. 6.3.4 Scenario 4: WHO “Threshold” Service Table 6-14 for Scenario 3 shows the total num- Delivery Staff Density Targets by 2030 ber of service-delivery staff which would be required to achieve the targets set for the scenario. Total ser- This fourth scenario is based on the WHO “threshold” vice-delivery staff would need to increase from 8,440 in density of 2.28 per 1,000 population (or a population- 2009 to almost 10,000 by 2015 to almost 11,500 by 2020, to-staff ratio of 439 to 1) of doctors, nurse (registered 116  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 116 11/27/12 11:38 AM Table 6‑14: Scenario 3: Total Service-Delivery Staff: Maintaining Existing Staff to Population Ratios for all Cadres (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Required 379 390 401 412 424 436 448 461 474 487 501 Nurses Required 3,252 3,346 3,440 3,537 3,636 3,739 3,844 3,953 4,065 4,181 4,300 1305747 PNG HR Report 11-27-12.indb 117 CHWs Required 4,398 4,525 4,652 4,783 4,918 5,057 5,199 5,346 5,498 5,654 5,815 HEOs Existing Stock Only 411 423 435 447 460 473 486 500 514 528 543 Total Service-Delivery Staff 8,440 8,684 8,928 9,179 9,438 9,705 9,977 10,260 10,551 10,850 11,159 Population per Service Staff 786 786 786 786 786 786 786 786 786 786 786 Service Staff per 1,000 Pop. 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 Total Staff Cost (Kmn) 191.6 196.8 202.4 208.1 213.9 219.9 226.1 232.6 239.1 245.9 252.9 Total Recurrent Budget 513.0 551.5 601.1 634.2 646.9 672.7 807.3 823.4 852.2 886.3 921.8 S.D Staff as % Recurrent Budget 37.3 35.7 33.7 32.8 33.1 32.7 28.0 28.2 28.1 27.7 27.4 Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 5.7 12.1 12.3 12.7 13.0 13.2 13.8 14.2 14.9 15.1 15.5 Quality-Enhanced Costs (Kmn) 8.3 17.9 18.6 18.8 19.3 19.9 20.5 21.1 21.6 22.3 23.0 2009 Costs % Recurrent Budget 1.1 2.2 2.1 2.0 2.0 2.0 1.7 1.7 1.7 1.7 1.7 Quality-Enhanced % Recurrent 1.6 3.3 3.1 3.0 3.0 3.0 2.5 2.6 2.5 2.5 2.5 Budget (continued on next page) Supply and Demand: Key Health Cadre Supply Gaps   //  117 11/27/12 11:38 AM Table 6‑14: Scenario 3: Total Service-Delivery Staff: Maintaining Existing Staff to Population Ratios for all Cadres (2010–2030)  (continued) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 1305747 PNG HR Report 11-27-12.indb 118 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Required 515 528 541 555 569 583 595 608 621 634 647 Nurses Required 4,422 4,533 4,646 4,762 4,881 5,003 5,108 5,216 5,325 5,437 5,551 CHWs Required 5,980 6,130 6,283 6,440 6,601 6,766 6,908 7,053 7,202 7,353 7,507 HEOs Existing Stock Only 559 573 587 602 617 632 646 659 673 687 702 Total Service-Delivery Staff 11,476 11,764 12,057 12,359 12,668 12,984 13,257 13,536 13,821 14,111 14,407 Population per Service Staff 786 786 786 786 786 786 786 786 786 786 786 Service Staff per 1,000 Pop. 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 1.27 Total Staff Cost (Kmn) 259.7 266.3 273.0 279.7 286.7 293.9 300.1 306.4 312.8 319.3 326.0 Total Recurrent Budget 958.7 999.9 1,042.9 1,087.7 1,134.5 1,183.3 1,234.1 1,287.2 1,342.6 1,400.3 1,460.5 S.D Staff as % Recurrent Budget 27.1 26.7 26.2 25.8 25.3 24.9 24.3 23.8 23.3 22.8 22.4 118  //  PNG Health Workforce Crisis: A Call to Action Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 16.0 14.9 15.4 15.8 16.1 16.5 16.1 16.5 16.8 17.2 17.6 Quality-Enhanced Costs (Kmn) 24.3 23.0 23.6 24.2 24.8 25.4 24.8 25.3 25.9 26.4 26.9 2009 costs % Recurrent Budget 1.7 1.5 1.5 1.4 1.4 1.4 1.3 1.3 1.3 1.2 1.2 Quality-Enhanced % Recurrent 2.5 2.3 2.3 2.2 2.2 2.1 2.0 2.0 1.9 1.9 1.8 Budget 11/27/12 11:38 AM Table 6‑15: Scenario 4: Medical Officers (MO): WHO “Threshold” Service Delivery Staff Density Target (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Required 379 405 430 453 475 495 515 533 550 617 878 Population per MO 17,512 16,862 16,327 15,934 15,625 15,415 15,234 15,137 15,085 13,828 9,994 1305747 PNG HR Report 11-27-12.indb 119 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.10 Attrition From MO Workforce 23 24 26 27 28 30 31 32 33 37 53 New Graduates Required 49 49 49 49 49 49 49 49 49 100 313 1st Year Student Intake Required 53 54 54 54 54 54 54 54 54 111 348 (6 years earlier) MO Employment Cost (Kmn) 34.1 36.5 38.7 40.8 42.7 44.6 46.3 48.0 49.5 55.5 79.0 (Unit cost = K90,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Required 1,173 1,388 1,659 1,944 2,241 2,553 3,128 3,726 4,347 5,548 6,231 Population per MO 7,692 6,664 5,715 5,000 4,445 4,000 3,332 2,856 2,500 2,000 1,818 MO per 1,000 Population 0.130 0.150 0.175 0.200 0.225 0.250 0.300 0.350 0.400 0.500 0.550 Attrition From MO Workforce 70 83 100 117 134 153 188 224 261 333 374 New Graduates Required 366 298 371 401 432 464 763 821 882 1,534 1,057 1st Year Student Intake Required 394 320 399 431 465 499 820 883 949 1,649 1,136 (6 years earlier) MO Employment Cost (Kmn) 105.6 124.9 149.3 174.9 201.7 229.7 281.5 335.3 391.2 499.3 560.8 (Unit cost = K90,000) (2009) Supply and Demand: Key Health Cadre Supply Gaps   //  119 11/27/12 11:38 AM and enrolled) and midwives. Below this, according to The projections for nurses for Scenario 4 are pre- WHO, coverage of essential interventions, including sented in Table 6-16. For the purposes of this scenario it those necessary to reach the health-related MDGs, is not is assumed that the 1.73 per 1,000 population for nurses likely. This scenario assumes the same population pro- and CHWs are proportionately equal and that HEOs jections, attrition rates from the workforce, and drop- grow at the same rate as the population (see Chapter 5). out rates from training programs as described above for The key observations about this scenario are that: Scenario 1 and 2. (i) The number of nurses would increase from 3,252 in The fundamental driver for this scenario is number 2009 to 3,295 in 2015 and 4,964 in 2020. This would of service-delivery staff that WHO believes are needed rise to 6,229 in 2025 and almost 8,500 in 2030. This per 1,000 population (see discussion in Chapter 5). In rate of expansion would see the population-to-nurse summary, this scenario proposes: (i) that 0.55 per 1,000 ratio improve from 2,041 to one nurse in 2009 to 1,818 population should be doctors; and (ii) 1.73 per 1,000 pop- in 2020, 1,639 in 2025 and 1,333 per nurse in 2030 ulation should be “other service-delivery staff ”. Scenario (or 0.75 nurses per 1,000 population). 3 aims to reach these ratios by 2030 and demonstrates the (ii) Clearly, the number of graduates from the nursing substantial additional staffing and costs this would require. schools would need to increase dramatically from The key observations about the projections for med- the expected numbers graduating in 2012 of 165 to ical officers under Scenario 4 are: around 423 by 2015 and to nearly 700 by 2020 and (i) The number of doctors would increase from 379 in over 1,200 by 2030. The feasibility of achieving this 2009 to 515 in 2015 and to 1,173 in 2020. It would is in extreme doubt. then need to more than double to 2,553 by 2025 and (iii) It should be noted, as discussed in the previous sce- to 6,231 by 2030. This would result in the popula- nario, that the projected attrition from the nurse tion-to-doctor ratio improving from 17,512 to one workforce is significantly greater than the current doctor in 2009 to 15,243 to one in 2015, and 7,692 2009 outputs and is expected to remain well above to one doctor by 2020. It would reach 1,818 per doc- the expected outputs likely from 2012 to 2014 (with tor—the aspirational target—by 2030. the recent opening of the Mendi School of Nursing, (ii) The number of graduates required from medical nursing diploma graduates will increase to about school(s) to achieve these massive increases in doc- 165 per annum from 2012. This scenario suggests tor numbers would need to expand dramatically. Spe- that nurse training capacity, given expected work- cifically, graduations would need to increase from force attrition, needs to grow about 325 percent from the current 49 per year in 2009 to 100 in 2018 (this the expected 2012 output of 165 graduates to about would entail first year enrollments for medical school 700 by 2020. New graduate numbers would need to rising to 111 immediately) and dramatically ramp increase a further 77 percent over the 2020 rate to up in each subsequent year. By 2025 new graduates reach 1,240 per year by 2030. required would be over 460 per year and by 2030 this (iv) The cost of the nurse expansion to the budget would would need to double-a clearly unachievable target. grow from K73.8 million in 2009 to nearly K75 mil- It would not be feasible to build the infrastructure, lion in 2015 (because the workforce declines for a few develop a reasonable quality faculty and/or recruit years due to attrition unless older nurses are encour- enough science graduates from the senior secondary aged not to retire) and to K113 million in 2020 and system with adequate training in science. to K193 million in 2030. (iii) The costs of employing this number of doctors would (v) The total recurrent cost of training of nurses—based increase from K34.1 million in 2009 to K46.3 million on the 2009 estimate of recurrent unit costs of train- in 2015 and to more than K105 million in 2020. It ing (K7,273) would increase from about K3 million in would then dramatically increase to almost K230 mil- 2009 to about K9.5 million in 2015 and K15.7 million lion in 2025 and K560 million in 2030. The impli- by 2020. These costs would increase to about K16 mil- cations of this are discussed below and are shown lion by 2024 and to K28 million by 2030. Second, in Table 6-19. the cost of training with a significant improvement 120  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 120 11/27/12 11:38 AM Table 6‑16: Scenario 4: Nursing Officers (NOs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 Nurses Required 3,252 3,192 3,135 3,112 3,091 3,070 3,295 3,631 3,900 4,266 4,563 Population per Nurse 2,041 2,140 2,239 2,319 2,401 2,486 2,381 2,222 2,128 2,000 1,923 1305747 PNG HR Report 11-27-12.indb 121 Nurse per 1,000 Population 0.49 0.47 0.45 0.43 0.42 0.40 0.42 0.45 0.47 0.50 0.52 Attrition From Nurse Workforce 195 192 188 187 185 184 198 218 234 256 274 New Graduates Required 135 135 135 165 165 165 423 553 503 623 571 1st Year Student Intake Required 139 139 139 170 170 170 436 570 518 642 588 (3 years earlier) Nurse Employment Cost (Kmn) 73.8 72.5 71.2 70.6 70.2 69.7 74.8 82.4 88.5 96.8 103.6 (Unit cost = K22,700)(2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.0 3.0 3.0 3.7 3.7 3.7 9.5 12.4 11.3 14.0 12.8 (b) Quality-Enhanced Cost 5.0 5.0 5.0 6.1 6.1 6.1 15.7 20.5 18.7 23.1 21.2 (K12,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 Nurses Required 4,964 5,180 5,405 5,637 5,977 6,229 6,568 6,919 7,281 7,767 8,497 Population per Nurse 1,818 1,786 1,754 1,724 1,667 1,639 1,587 1,538 1,493 1,429 1,333 Nurse per 1,000 Population 0.55 0.56 0.57 0.58 0.60 0.61 0.63 0.65 0.67 0.70 0.75 Attrition From Nurse Workforce 298 311 324 338 359 374 394 415 437 466 510 New Graduates Required 699 527 549 571 699 625 733 766 800 952 1,239 1st Year Student Intake Required 735 555 577 601 736 658 772 806 842 1,002 1,305 (3 years earlier) Nurse Employment Cost (Kmn) 112.7 117.6 122.7 128.0 135.7 141.4 149.1 157.1 165.3 176.3 192.9 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 16.0 12.1 12.6 13.1 16.0 14.4 16.8 17.6 18.4 21.9 28.5 (b) Quality-Enhanced Cost 26.5 20.0 20.8 21.6 26.5 23.7 27.8 29.0 30.3 36.1 47.0 (K12,000) Supply and Demand: Key Health Cadre Supply Gaps   //  121 11/27/12 11:38 AM in the quality of training (represented by increased (K6,221) would increase from about K2.7 million in unit costs to K12,000 to allow for improved teach- 2009 to K9.2 million in 2015, K11.1 million by 2020, ing inputs, library, internet connection, boarding and K12.1 million by 2030. Second, the cost of train- costs and maintenance) is presented. The cost of an ing assuming a significant improvement in the qual- “enhanced” quality of nursing graduate would increase ity of training (represented by increased unit costs to from about K5 million in 2009 to K15.7 million by K8,000 to allow for improved teaching inputs, library, 2015, almost K26 million by 2020 and would reach internet connection, boarding costs and maintenance) K47 million by 2030. is presented. The cost of this “quality” enhanced teach- ing program would increase under this scenario from The key observations for the CHW projections in K3.4 million in 2009 to almost K12 million by 2015 Scenario 4 are that: and to over K14 million by 2020. This would increase (i) The number of CHWs would increase from 4,398 further to about K15 million by 2025 and this would in 2009 to 6,137 in 2020 and over 10,300 in 2030. be roughly sustained over the five years to 2030. Expansion of CHWs to this schedule would see the population per CHW improve from 1,509 in 2009 The key observations about the projections for to 1,471 in 2020 (the slow improvement due to attri- HEOs under Scenario 4 are that: tion being higher than current CHW training out- (i) The number of HEOs would increase from 411 in puts) and to a population of 1,235 per CHW in 2025 2009 to 559 in 2020 and about 700 in 2030. This and to 1,099 in 2030. would sustain the population-to-HEO ratio at 16,148 (ii) The number of graduates from the CHW training per HEO over this period. This is the same as Sce- schools would need to increase from the expected nario 3 for HEOs. numbers graduating in 2009 of 209 to around 714 by (ii) The number of graduates from the DWU training 2015 (intakes to CHW training schools would need program for HEOs would need to increase from the to expand to about 190 by 2013) and this would need numbers graduating in 2009 of 46 to around 55 by to rise to over 774 by 2018, and to about 900 by 2025 2020 and about 64 by 2030. and almost 950 by 2030. (iii) It should be noted that the projected attrition from the (iii) It should be noted that projected attrition from the HEO workforce would account for about 70 percent CHW workforce is significantly greater—about 50 per- of the required output of HEOs under this scenario. cent—than the current 2009 output. This suggests a (iv) The cost of HEOs to the budget would grow from real emerging crisis on the supply side for CHWs. K10.7 million in 2009 to K14.9 million in 2020 and Specifically, (and this applies to all scenarios), a fall to K18.7 million in 2030. may be expected in the total CHW workforce unless the attrition rate from the workforce can be reduced. To achieve the targets set for Scenario 4, total ser- This might be achieved by encouraging CHWs to vice-delivery staff would need to increase from 8,440 postpone retirement. Thus, Scenario 4 suggests that in 2009 to 8,611 by 2015 to 12,833 by 2020, to 17,685 CHW training capacity, given expected workforce by 2025 and reach about 25,739 by 2030. This would attrition, needs to grow about 350 percent from cur- enable the population to total service-delivery staff ratio rent 2009 output of 209 graduates to almost 950 by to improve from 786 to one in 2009 to 703 to one in 2020 2030. It will be a significant challenge in the short and to 440 to one in 2030. term just expanding production to replace the exist- The cost of employing all the staff under this sce- ing workforce. nario is K191.6 million or just over 37 percent of the (iv) The cost of the CHW expansion to the budget would total health recurrent budget and these costs would rise grow from K73 million in 2009 to about K102 mil- to about K942 million by 2030 in real terms (where it lion in 2020 and to K171 million in 2030. would represent about 65 percent of the expected health (v) The total recurrent costs of training of CHWs based recurrent budget). Even assuming much higher health on the 2009 estimate of recurrent unit costs of training recurrent budget growth, this scenario would seem to 122  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 122 11/27/12 11:38 AM Table 6‑17: Scenario 4: Community Health Workers (CHWs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 CHWs Required 4,398 4,299 4,200 4,101 4,003 3,904 4,315 4,680 4,978 5,375 5,704 Population per CHW 1,509 1,589 1,672 1,760 1,854 1,955 1,818 1,724 1,667 1,587 1,538 1305747 PNG HR Report 11-27-12.indb 123 CHWs per 1,000 Population 0.66 0.63 0.60 0.57 0.54 0.51 0.55 0.58 0.60 0.63 0.65 Attrition From CHW Workforce 308 301 294 287 280 273 302 328 348 376 399 New Graduates Required 209 202 195 188 181 174 714 692 647 774 728 1st Year Student Intake Required 215 208 201 194 187 180 736 713 667 797 750 (2 years earlier) CHW Employment Cost (Kmn) 73.0 71.4 69.7 68.1 66.4 64.8 71.6 77.7 82.6 89.2 94.7 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 2.6 2.5 2.4 2.3 2.2 9.2 8.9 8.3 9.9 9.3 (b) Quality-Enhanced Costs 3.4 3.3 3.2 3.1 3.0 2.9 11.8 11.4 10.7 12.8 12.0 (K8,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 CHWs Required 6,137 6,475 6,922 7,386 7,870 8,271 8,757 9,261 9,672 9,986 10,310 Population per CHW 1,471 1,429 1,370 1,316 1,266 1,235 1,190 1,149 1,124 1,111 1,099 CHWs per 1,000 Population 0.68 0.70 0.73 0.76 0.79 0.81 0.84 0.87 0.89 0.90 0.91 Attrition From CHW Workforce 430 389 415 443 472 496 525 556 580 599 619 New Graduates Required 863 727 862 908 956 897 1,012 1,059 992 913 942 1st Year Student Intake Required 889 749 888 936 985 925 1,043 1,092 1,023 942 971 (2 years earlier) CHW Employment Cost (Kmn) 101.9 107.5 114.9 122.6 130.6 137.3 145.4 153.7 160.6 165.8 171.1 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 11.1 9.3 11.1 11.6 12.3 11.5 13.0 13.6 12.7 11.7 12.1 (b) Quality-Enhanced Costs 14.2 12.0 14.2 15.0 15.8 14.8 16.7 17.5 16.4 15.1 15.5 (K8,000) Supply and Demand: Key Health Cadre Supply Gaps   //  123 11/27/12 11:38 AM Table 6‑18: Scenario 4: Health Extension Officers (HEOs): WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 124 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 HEOs Required 411 423 435 447 460 473 486 500 514 528 543 Population per HEO 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 29 30 30 31 32 33 34 35 36 37 38 New Graduates Required 46 42 42 44 45 46 47 49 50 52 53 1st Year Student Intake Required 49 45 45 47 48 50 51 52 54 55 57 (4 years earlier) Cost of HEO Employment (Kmn) 10.7 11.0 11.3 11.6 12.0 12.3 12.6 13.0 13.4 13.7 14.1 (Unit cost = K26,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 124  //  PNG Health Workforce Crisis: A Call to Action HEOs Required 559 573 587 602 617 632 646 659 673 687 702 Population per HEO 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 16,148 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 39 40 41 42 43 44 45 46 47 48 49 New Graduates Required 55 54 55 57 58 60 58 60 61 62 64 1st Year Student Intake Required 59 58 60 61 63 64 63 64 66 67 68 (4 years earlier) Cost of HEO Employment (Kmn) 14.5 14.9 15.3 15.7 16.0 16.4 16.8 17.1 17.5 17.9 18.3 (Unit cost = K26,000) (2009) 11/27/12 11:38 AM be totally infeasible. This would remain the same even if increased allocations of both support staff and quality- we also assume a fairly rapid growth of the private sec- enhancing nonsalary budgets so necessary for improved tor—which in 2009 may have accounted for 15 percent of health outcomes. The scenario sees staff growth con- doctors and nurses and a much smaller share of CHWs. strained by: (i) the likely level of resources available for Using the recurrent unit costs of training of both health and service-delivery staff; and (ii) the feasibility and nurses and CHWs, this program would see training speed with which preservice training can be ramped up costs (excluding doctors and HEOs for which data is to meet the demands of attrition from the workforce and not available) rise from K5.7 million in 2009 to almost the needs of a growing population. This scenario assumes K19 million in 2015 and K27 million by 2020. It would the same population projections, attrition rates from the remain at about this level through 2025 and would then rise workforce, and dropout rates from training programs as to about K40 million by 2030. As a share of the expected for the other scenarios. recurrent health budget, the costs of training nurses and The key observations about the projections for med- CHWs at 2009 unit costs would increase from 1.1 percent ical officers for Scenario 5 are: of the budget to 2.3 percent by 2015 and 2.8 percent by (i) The number of doctors would increase from 379 2020 and this would be sustained through 2030. in 2009 to 515 in 2015 and to nearly 700 in 2020. It If we use a “reasonable quality-enabling” unit cost would then increase by about 370 to 1,069 by 2025 (see discussion above) the total costs of training would and by another 466 to 1,535 by 2030. This would increase from K8.5 million in 2009 to almost K28 mil- result in the population-to-doctor ratio improving lion by 2015 and K40.2 million by 2020. It would then from 17,512 to one doctor in 2009 to 15,243 to one increase to K45.9 million by 2025 and to K70.3 million in in 2015, to 13,004 to one doctor by 2020. It would 2030. The cost of nurse and HEO training would increase then improve more dramatically to be 9,556 to one as a share of the expected health budget from 1.6 percent in 2015 and 7,380 to one in 2030. in 2009 to 3.4 percent by 2015 and 4.2 percent by 2020 and (ii) The number of graduates required from medical this would be roughly sustained through 2030. school(s) to achieve this increase in doctor numbers would need to rise from the current 49 per year in 6.3.5 Scenario 5: A Recommended 2009 to 100 in 2019 (this would entail first year enroll- Preservice Training Scenario to Meet Key ments for medical school rising to 111 in 2013) and Health Human Resource Needs increasing graduates to 150 by 2023 and reaching 200 through the last years of the projection period. The fifth scenario—which is the broadly recommended This build up of medical training would be feasible, scenario—is one which envisages: (i) a gradual reduc- at a stretch, although the efforts and costs to build tion in the population-to-doctor ratio (from 17,512 to the infrastructure, develop a reasonable quality fac- 1 doctor in 2009 to 7,380 to 1 doctor in 2030); (ii) a ulty and/or recruit enough science graduates from reduction in the population-to-nurse ratio (from 2,041 the senior secondary system should not be under- to 1 nurse in 2009 to 1,414 to 1 nurse in 2030), (iii) a estimated. This strategy to increase doctor num- reduction in the current population-to-CHW ratio bers should be focused on getting doctors into rural (1,509 to 1 CHW in 2009) to 1,372 to 1 in 2030; and areas—a matter discussed in Chapter 7. It would (iv) no expansion in the number of HEOS produced need to be phased in—depending on the future per- annually which would result in the HEO-to-popu- formance of the economy. lation ratio falling from 16,148 to 1 HEO in 2009 to (iii) The cost of employing this number of doctors would 18,756 to 1 in 2020.. increase from K34.1 million in 2009 to K46.3 mil- This preservice strategy is affordable, responds to lion in 2015 and to more than K62 million in 2020. the demand requirements for staff from the health sys- It would then increase to about K96 million in 2025 tem and leaves space for recurrent health resources to and K138 million in 2030. This assumes doctors in be allocated to a significant expansion of training— rural areas would receive allowances equivalent to the both preservice and in-service. It also leaves space for average that doctors working in hospitals currently Supply and Demand: Key Health Cadre Supply Gaps   //  125 1305747 PNG HR Report 11-27-12.indb 125 11/27/12 11:38 AM Table 6-19: Scenario 4: Total Service-Delivery Staff: WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 126 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Required 379 405 430 453 475 495 515 533 550 617 878 Nurse Plan by PNGDSP 3,252 3,192 3,135 3,112 3,091 3,070 3,295 3,631 3,900 4,266 4,563 CHW Plan by PNGDSP 4,398 4,299 4,200 4,101 4,003 3,904 4,315 4,680 4,978 5,375 5,704 HEO Existing Stock Only 411 423 435 447 460 473 486 500 514 528 543 Total Service-Delivery Staff 8,440 8,319 8,200 8,113 8,029 7,942 8,611 9,344 9,942 10,786 11,688 Population per Service Staff 786 821 856 890 924 961 911 864 835 791 751 Service Staff per 1,000 Pop. 1.27 1.22 1.17 1.12 1.08 1.04 1.10 1.16 1.20 1.26 1.33 Total Staff Cost (Kmn) 191.6 191.0 190.6 190.8 191.0 191.1 205.3 220.7 233.6 254.9 290.9 Total Recurrent Budget 513.0 551.5 601.1 634.2 646.9 672.7 807.3 823.4 852.2 886.3 921.8 S.D Staff as % Recurrent Budget 37.3 34.6 31.7 30.1 29.5 28.4 25.4 26.8 27.4 28.8 31.6 126  //  PNG Health Workforce Crisis: A Call to Action Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 5.7 5.6 5.5 6.1 6.0 5.9 18.7 21.3 19.6 23.9 22.2 Quality-Enhanced Costs (Kmn) 8.4 8.3 8.2 9.2 9.1 9.0 27.5 31.9 29.4 35.9 33.2 2009 Costs % Recurrent Budget 1.1 1.0 0.9 1.0 0.9 0.9 2.3 2.6 2.3 2.7 2.4 Quality-Enhanced % Recurrent 1.6 1.5 1.4 1.5 1.4 1.3 3.4 3.9 3.4 4.0 3.6 Budget (continued on next page) 11/27/12 11:38 AM Table 6-19: Scenario 4: Total Service-Delivery Staff: WHO “Threshold” Service-Delivery Staff Density Targets (2010–2030)  (continued) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs WHO “Threshold” 1,173 1,388 1,659 1,944 2,241 2,553 3,128 3,726 4,347 5,548 6,231 1305747 PNG HR Report 11-27-12.indb 127 Nurses Plan WHO Threshold 4,964 5,180 5,405 5,637 5,977 6,229 6,568 6,919 7,281 7,767 8,497 CHWs Plan WHO Threshold 6,137 6,475 6,922 7,386 7,870 8,271 8,757 9,261 9,672 9,986 10,310 HEOs Existing Stock Only 559 573 587 602 617 632 646 659 673 687 702 Total Service-Delivery Staff 12,833 13,616 14,573 15,569 16,705 17,685 19,099 20,565 21,973 23,988 25,740 Population per Service Staff 703 679 651 624 596 577 546 518 495 463 440 Service Staff per 1,000 Pop. 1.42 1.47 1.54 1.60 1.68 1.73 1.83 1.93 2.02 2.16 2.27 Total Staff Cost (Kmn) 334.7 364.3 401.7 440.6 483.5 524.8 592.1 662.5 733.9 858.5 943.1 Total Recurrent Budget 958.7 999.9 1,042.9 1,087.7 1,134.5 1,183.3 1,234.1 1,287.2 1,342.6 1,400.3 1,460.5 S.D Staff as % Recurrent Budget 34.9 36.4 38.5 40.5 42.6 44.3 48.0 51.5 54.7 61.3 64.5 Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 27.1 21.4 23.7 24.7 28.3 25.9 29.8 31.2 31.1 33.6 40.6 Quality-Enhanced Costs (Kmn) 40.7 32.0 35.0 36.6 42.3 38.5 44.5 46.5 46.7 51.2 62.5 2009 Costs % Recurrent Budget 2.8 2.1 2.3 2.3 2.5 2.2 2.4 2.4 2.3 2.4 2.8 Quality-Enhanced % Recurrent Budget 4.2 3.2 3.5 3.4 3.7 3.3 3.6 3.6 3.5 3.7 4.3 Supply and Demand: Key Health Cadre Supply Gaps   //  127 11/27/12 11:38 AM Table 6‑20: Scenario 5: Medical Officers (MOs): A Suggested Preservice Training Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 128 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Employed 379 405 430 453 475 495 515 533 550 566 632 Population per MO 17,512 16,862 16,327 15,934 15,624 15,415 15,234 15,137 15,085 15,075 13,884 MO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 Attrition From MO Workforce 23 24 26 27 28 30 31 32 33 34 38 New Graduates Required 49 49 49 49 49 49 49 49 49 49 100 1st Year Student Intake Required 54 54 54 54 54 54 54 54 54 54 111 (6 years earlier) MO Employment Cost (Kmn) 34.1 36.5 38.7 40.8 42.7 44.6 46.3 48.0 49.5 50.9 56.9 (Unit cost = K90,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 128  //  PNG Health Workforce Crisis: A Call to Action MOs Employed 694 752 807 901 988 1,069 1,144 1,214 1,329 1,436 1,535 Population per MO 13,004 12,301 11,749 10,786 10,082 9,551 9,113 8,767 8,177 7,727 7,380 MO per 1,000 Population 0.08 0.08 0.09 0.09 0.10 0.10 0.11 0.11 0.12 0.13 0.14 Attrition From MO Workforce 42 45 57 63 69 75 80 85 93 100 107 New Graduates Required 100 100 100 150 150 150 150 150 200 200 200 1st Year Student Intake Required 111 111 111 167 167 167 167 167 222 222 222 (6 years earlier) MO Employment Cost (Kmn) 62.5 67.7 72.7 81.1 88.9 96.2 102.9 109.2 119.6 129.2 138.2 (Unit cost = K90,000) (2009) 11/27/12 11:38 AM Table 6‑21: Scenario 5: Nursing Officers (NOs): A Suggested Preservice Training Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 Nurses Employed 3,252 3,192 3,135 3,112 3,091 3,070 3,136 3,298 3,450 3,743 4,018 Population per Nurse 2,041 2,140 2,239 2,319 2,401 2,486 2,502 2,447 2,405 2,280 2,184 1305747 PNG HR Report 11-27-12.indb 129 Nurse per 1,000 Population 0.49 0.47 0.45 0.43 0.42 0.40 0.40 0.41 0.42 0.44 0.46 Attrition From Nurse Workforce 195 192 188 187 185 184 188 198 207 225 241 New Graduates Required 135 135 135 165 165 165 250 350 350 500 500 1st Year Student Intake Required 139 139 139 170 170 170 258 361 361 515 515 (3 years earlier) Nurse Employment Cost (Kmn) 73.8 72.5 71.2 70.6 70.2 69.7 71.2 74.9 78.3 85.0 91.2 (Unit cost = K22,700)(2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 3.0 3.0 3.0 3.7 3.7 3.7 5.6 7.9 7.9 11.2 11.2 (b) Quality-Enhanced Cost 5.0 5.0 5.0 6.1 6.1 6.1 9.3 13.0 13.0 18.6 18.6 (K12,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 Nurses Employed 4,277 4,621 4,943 5,347 5,726 6,082 6,517 6,926 7,311 7,672 8,012 Population per Nurse 2,110 2,002 1,918 1,818 1,740 1,679 1,600 1,537 1,487 1,446 1,414 Nurse per 1,000 Population 0.47 0.50 0.52 0.55 0.57 0.60 0.63 0.65 0.67 0.69 0.71 Attrition From Nurse Workforce 257 277 297 321 344 365 391 416 439 460 481 New Graduates Required 500 600 600 700 700 700 800 800 800 800 800 1st Year Student Intake Required 526 632 632 737 737 737 842 842 842 842 842 (3 years earlier) Nurse Employment Cost (Kmn) 97.1 104.9 112.2 121.4 130.0 138.1 147.9 157.2 166.0 174.2 181.9 (Unit cost = K22,700) (2009) Recurrent Unit Cost of Training (Kmn) (3 year course) (a) 2009 Costs (K7,273) 11.5 13.8 13.8 16.1 16.1 16.1 18.4 18.4 18.4 18.4 18.4 (b) Quality-Enhanced Cost 18.9 22.7 22.7 26.5 26.5 26.5 30.3 30.3 30.3 30.3 30.3 (K12,000) Supply and Demand: Key Health Cadre Supply Gaps   //  129 11/27/12 11:38 AM receive. The implications of this are discussed below in the quality of training (represented by increased and are shown in Table 6-24. unit costs to K12,000 to allow for improved teach- ing inputs, library, internet connection, boarding The key observations about the projections for costs and maintenance) is presented. The cost of this nurses under Scenario 5 are that: “enhanced” quality of nursing graduate would increase (i) The number of nurses would increase from 3,252 in from about K5 million in 2009 to K9.3 million by 2009 to 3,136 in 2015 and 4,277 in 2020. This would 2015 and almost K19 million by 2018. It would reach rise to 6,082 in 2025 and just over 8,000 in 2030. This nearly K27 million by 2023 and K30 million by 2026 rate of expansion would see the population-to-nurse which would be sustained through 2030. ratio decline from 2,041 to a nurse in 2009 to around 1,414 per nurse over the period of the scenario to 2030. The key observations about the projections for (ii) The number of graduates from the nursing schools CHWs for Scenario 5 are that: under this scenario would still need to increase sig- (i) The number of CHWs would decrease from 4,398 in nificantly to achieve the nurse numbers proposed. 2009 to 4,111 in 2015 (as a result of expected attrition Specifically, the numbers expected to graduate in from the workforce given CHW training school out- 2012 (165) would need to increase to 250 by 2015, puts) and would, thereafter, increase to 5,133 in 2020 to 500 by 2020 and to 800 by 2026. The feasibility of even with the significant training expansion proposed achieving this is possible but would be a challenge—at (see below). Expansion of CHWs to this schedule least in the early years. It implies entry to nurse train- would see the population-per-CHW improve from ing schools to be 170 in 2012–an increase of 20 per- 1,509 to one CHW in 2009 to 1,909 in 2015, 1,758 in cent over current (2011) entry numbers. In the next 2020 and reach 1,372 to each CHW by 2030. health plan period the intake would need to increase (ii) The number of graduates from the CHW training to about 258 by 2015 and 361 by 2016. schools would need to increase from the expected (iii) It should be noted, as discussed in previous scenar- numbers graduating in 2009 of 209 to around 300 by ios, that the projected attrition from the nurse work- 2014 (intakes to CHW training schools would need force over most of the current NHP plan period will be to expand to about 310 by 2013 given assumed drop- greater than current nurse school outputs whithout a out rates)–an increase in graduations in 2015 over decision to expand outputs—hence it can be expected current output capacity by 44 percent. The number that the total number of nurses employed will fall of graduates would need to expand to 500 by 2016 over this period unless nurses approaching retire- and to 600 by 2020. In the subsequent decade grad- ment can be encouraged not to retire (also see dis- uate numbers would need to further increase to 700 cussion in Scenario 1 above). by 2023 and 800 by 2026. (iv) The cost of nurse employment to the budget in real (iii) It should be noted that the projected attrition from terms would decline from K73.8 million in 2009 to the CHW workforce is significantly greater—almost about K71 million in 2015 (because the workforce 50 percent—than the current 2009 output. This sug- declines for a few years due to attrition unless older gests, as discussed, a real emerging crisis on the sup- nurses are encouraged not to retire). This will increase ply side for CHWs. Specifically, (and this applies to to K97 million in 2020 to K138 million in 2025 and all scenarios), it can be expected that there may be a further to K182 million in 2030. fall in the total CHW workforce unless the attrition (v) The total recurrent cost of training of nurses based rate from the workforce can be reduced (see discus- on the 2009 estimate of recurrent unit costs of train- sion under scenario 1). This might be achieved by ing (K7,273) would increase from about K3 million encouraging CHWs to postpone retirement. Sce- in 2009 to about K5.6 million in 2015 and K11.2 mil- nario 5, therefore, suggests that CHW training capac- lion by 2018. These costs would increase to about ity, given expected workforce attrition, needs to grow K16 million by 2024 and to K18 million by 2030. about 200 percent from current 2009 output of 209 The cost of training with a significant improvement graduates to about 600 in 2020. It will be a signifi- 130  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 130 11/27/12 11:38 AM Table 6‑22: Scenario 5: Community Health Workers (CHWs): A Suggested Preservice Training Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 CHWs Employed 4,398 4,299 4,207 4,122 4,083 4,097 4,111 4,323 4,520 4,704 4,875 Population per CHW 1,509 1,589 1,669 1,751 1,818 1,862 1,909 1,866 1,836 1,814 1,800 1305747 PNG HR Report 11-27-12.indb 131 CHWs per 1,000 Population 0.66 0.63 0.60 0.57 0.55 0.54 0.52 0.54 0.54 0.55 0.56 Attrition From CHW Workforce 308 301 295 289 286 287 288 303 316 329 341 New Graduates Required 209 209 209 209 250 300 300 500 500 500 500 1st Year Student Intake Required 215 215 215 215 258 309 309 515 515 515 515 (2 years earlier) CHW Employment Cost (Kmn) 73.0 71.4 69.8 68.4 67.8 68.0 68.2 71.8 75.0 78.1 80.9 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 2.7 2.7 2.7 2.7 3.2 3.8 3.8 6.4 6.4 6.4 6.4 (b) Quality-Enhanced Costs 3.4 3.4 3.4 3.4 4.1 4.9 4.9 8.2 8.2 8.2 8.2 (K8,000) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 CHW s Employed 5,133 5,374 5,652 6,012 6,352 6,671 7,070 7,446 7,799 7,931 8,256 Population per CHW 1,758 1,721 1,678 1,616 1,568 1,531 1,475 1,430 1,393 1,399 1,372 CHWs per 1,000 Population 0.57 0.58 0.60 0.62 0.64 0.65 0.68 0.70 0.72 0.71 0.73 Attrition From CHW Workforce 359 322 339 361 381 400 424 447 468 476 495 New Graduates Required 600 600 600 700 700 700 800 800 800 800 800 1st Year Student Intake Required 619 619 619 722 722 722 825 825 825 825 825 (2 years earlier) CHW Employment Cost (Kmn) 85.2 89.2 93.8 99.8 105.4 110.7 117.4 123.6 129.5 131.7 137.0 (Unit cost = K16,600) (2009) Recurrent Unit Cost of Training (Kmn) (a) 2009 Costs (K6,221) 7.7 7.7 7.7 9.0 9.0 9.0 10.3 10.3 10.3 10.3 10.3 (b) Quality-Enhanced Costs 9.9 9.9 9.9 11.5 11.5 11.5 13.2 13.2 13.2 13.2 13.2 (K8,000) Supply and Demand: Key Health Cadre Supply Gaps   //  131 11/27/12 11:38 AM cant challenge in the short term just expanding pro- increase to 10,650 by 2020. Total staff numbers would sub- duction to replace the existing workforce. sequently increase to about 14,400 by 2025 and to 18,400 (iv) The total cost of the CHW workforce to the bud- by 2030. This means that the population to total service- get would decline from K73 million in 2009 due to delivery staff ratio would decline from 786 to one staff in a smaller workforce to about K68 million in 2015, 2009 to 950 to one in 2015. It would improve as supply before increasing to K85 million in 2020–only with capacity is ramped up to be around 847 to one in 2020, decisive action on the supply side. With the work- 709 to one by 2025 and 616 to one by 2030. force subsequently expanding, the remuneration cost The cost of employing all the staff under this sce- of CHWs would increase to K111 million in 2025 and nario was about K191 million in 2009 or just over K137 million in 2030. 37 percent of the total health recurrent budget and (v) The total recurrent cost of training of CHWs, based on these costs would increase to about K259 million by the 2009 estimate of the recurrent unit cost of train- 2020 in real terms (where it would represent 27 percent ing (K6,221) would increase from about K2.7 million of the expected health recurrent budget). In the subse- in 2009 to K3.8 million by 2014 and about K7.7 mil- quent decade the cost of staff would increase to K360 mil- lion by 2020. The cost of training would increase lion in 2025 (30 percent of the recurrent budget) and to to around K9 million by 2023 before plateauing at K472 million in 2030 or about 32 percent of the expected about K10 million from 2026. Assuming a signif- health recurrent budget. icant improvement in the quality of training (rep- Using the recurrent unit costs of training of both resented by increased unit costs to K8,000 to allow nurses and CHWs, this program would see training costs for improved teaching inputs, library, internet con- (excluding doctors and HEOs for which data is not avail- nection, boarding costs and maintenance), the cost able) rise from K5.7 million in 2009 to K9.4 million in of this “quality” enhanced teaching program would 2015 and almost K19 million by 2020. It would increase increase under this scenario from K3.4 million in 2009 to K25 million by 2023 and to more than K28 million from to almost K5 million by 2014 and to about K10 mil- 2026. As a share of the expected recurrent health budget, lion by 2020. This would increase further to about the cost of training nurses and CHWs at 2009 unit costs K11.5 million by 2023 and to K13.2 by 2026 and this would rise from 1.1 percent of the budget to 1.7 percent would be roughly sustained through 2030. by 2016 and 2 percent by 2018 which would be roughly sustained through the projection period to 2030. The key observations for the HEO projections for If we use a “reasonable quality-enabling” unit Scenario 5 are that: cost (see discussion above) the total cost of training (i) The number of HEOs would increase from 411 in 2009 would increase from K8.4 million in 2009 to K14.2 mil- to 546 in 2020 and about 600 in 2030. This would see lion by 2015 and K21.2 million by 2016. It would then the population-to-HEO ratio decline from the current increase to almost K27 million by 2018 and to K33 mil- 16,148 per HEO to 18,756 to one over this period. lion in 2021. It would reach about K38 million by 2023 (ii) The number of graduates from the DWU training and K44 million from 2026. The cost of nurse and CHW program for HEOs would need to remain at current training would increase as a share of the expected health capacity of about 50 per year. budget from 1.6 percent in 2009 to 1.8 percent by 2015 (iii) It should be noted that the projected attrition from and 3 percent by 2020 and this would be roughly sus- the HEO workforce would account for over 90 percent tained through 2030. of the required output of HEOs under this scenario. The fundamental driver of this scenario in the first (iv) The cost of HEOs to the budget would grow from decade of the scenario is an assessment of how fast train- K10.7 million in 2009 to K14.2 million in 2020 and ing capacity can be ramped up—not the desirable rate to K15.7 million in 2030. of training expansion. The historic neglect of managing the supply and demand for health staff over the last decade Under Scenario 5 total service-delivery staff num- means there is an emerging crisis in the supply of health bers would fall from 8,440 in 2009 to 8,260 by 2015 but personnel. In the subsequent decade the health sector 132  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 132 11/27/12 11:38 AM has a greater degree of freedom to maneuver. Neverthe- tial focus being on MCH and, in particular, training of less, underlying this scenario is a suggestion to alter the nurses and CHWs in birth complications; (iii) additional structure of the health service-delivery workforce some- staff for support services which follow from the decision what—particularly in the second decade of the scenario. to increase the number of doctors; and (iv) more qual- Specifically, provision is made for a reasonable expansion ity-enhancing nonsalary budget expenditures. To this of the number of doctors (to be targeted for rural facili- end we recommend this scenario as the starting point for ties) and an expansion of general nursing graduates rel- moving forward. This exercise needs to be repeated with ative to CHWs. vastly improved data on the workforce and training capac- Underlying this scenario is a firm suggestion that ity and as the expected economic boom emerges. At pres- there needs to be a very significant expansion of recur- ent it is not clear how fast revenues will become available rent (and capital) resources to support: (i) the expansion to government to expand recurrent expenditures. Future of pretraining; (ii) a very significant and decisive expan- repeats of this exercise can adjust for improved knowl- sion of in-service training (for all rural staff every five edge about the assumptions made in this report. This is years and some training for hospital staff) with the ini- further discussed in Chapter 7. Supply and Demand: Key Health Cadre Supply Gaps   //  133 1305747 PNG HR Report 11-27-12.indb 133 11/27/12 11:38 AM Table 6‑23: Scenario 5: Health Extension Officers (HEOs): A Suggested Preservice Training Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1305747 PNG HR Report 11-27-12.indb 134 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 HEOs Employed 411 428 444 459 473 486 498 509 519 529 538 Population per HEO 16,148 15,956 15,812 15,726 15,690 15,701 15,755 15,850 15,986 16,129 16,310 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 Attrition From HEO Workforce 29 30 31 32 33 34 35 36 36 37 38 New Graduates Required 46 46 46 46 46 46 46 46 46 46 46 1st Year Student Intake Required 49 49 49 49 49 49 49 49 49 49 49 (4 years earlier) Cost of HEO Employment (Kmn) 10.7 11.1 11.6 11.9 12.3 12.6 12.9 13.2 13.5 13.8 14.0 (Unit cost = K26,000) (2009) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 134  //  PNG Health Workforce Crisis: A Call to Action HEOs Employed 546 554 561 568 574 580 585 590 595 599 604 Population per HEO 16,529 16,697 16,901 17,110 17,355 17,603 17,821 18,041 18,265 18,524 18,756 HEO per 1,000 Population 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.05 0.05 Attrition From HEO Workforce 38 39 39 40 40 41 41 41 42 42 42 New Graduates Required 46 46 46 46 46 46 46 46 46 46 46 1st Year Student Intake Required 49 49 49 49 49 49 49 49 49 49 49 (4 years earlier) Cost of HEO Employment (Kmn) 14.2 14.4 14.6 14.8 14.9 15.1 15.6 15.2 15.5 15.6 15.7 (Unit cost = K26,000) (2009) 11/27/12 11:38 AM Table 6‑24: Scenario 5: Total Service-Delivery Staff: A Suggested Preservice Training Scenario (2010–2030) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Population 6,637,000 6,829,266 7,020,903 7,218,240 7,421,458 7,630,741 7,846,000 8,068,278 8,296,935 8,532,464 8,775,085 MOs Employed 379 405 430 453 475 495 515 533 550 566 632 Nurses Plan Employed 3,252 3,192 3,135 3,112 3,091 3,070 3,136 3,298 3,450 3,743 4,018 1305747 PNG HR Report 11-27-12.indb 135 CHWs Employed 4,398 4,299 4,207 4,122 4,083 4,097 4,111 4,323 4,520 4,704 4,875 HEOs Employed 411 428 444 459 473 486 498 509 519 529 538 Total Service-Delivery Staff 8,440 8,324 8,216 8,146 8,122 8,148 8,260 8,663 9,039 9,542 10,063 Population per Service Staff 786 820 854 886 914 936 950 931 918 894 872 Service Staff per 1,000 Pop. 1.27 1.22 1.17 1.13 1.09 1.07 1.05 1.07 1.09 1.12 1.15 Total Staff Cost (Kmn) 191.6 191.1 190.9 191.5 192.7 194.6 198.6 207.5 216.0 227.4 242.6 Total Recurrent Budget 513.0 551.5 601.1 634.2 646.9 672.7 807.3 823.4 852.2 886.3 921.8 S.D Staff as % Recurrent Budget 37.3 34.7 31.8 30.2 29.8 28.9 24.6 25.2 25.3 25.7 26.3 Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 5.7 5.7 5.7 6.4 6.9 7.5 9.4 14.3 14.3 17.6 17.6 Quality-Enhanced Costs (Kmn) 8.4 8.4 8.4 9.5 10.2 11.0 14.2 21.2 21.2 26.8 26.8 2009 Costs % Recurrent Budget 1.1 1.0 1.0 1.0 1.1 1.1 1.2 1.7 1.7 2.0 1.9 Quality-Enhanced % Recurrent 1.6 1.5 1.4 1.5 1.6 1.6 1.8 2.6 2.5 3.0 2.9 Budget (continued on next page) Supply and Demand: Key Health Cadre Supply Gaps   //  135 11/27/12 11:38 AM Table 6‑24: Scenario 5: Total Service-Delivery Staff: A Suggested Preservice Training Scenario (2010–2030)  (continued) Year 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 1305747 PNG HR Report 11-27-12.indb 136 Population 9,025,023 9,250,649 9,481,915 9,718,963 9,961,937 10,210,985 10,425,416 10,644,350 10,867,881 11,096,106 11,329,125 MOs Employed 694 752 807 901 988 1,069 1,144 1,214 1,329 1,436 1,535 Nurses Plan Employed 4,277 4,621 4,943 5,347 5,726 6,082 6,517 6,926 7,311 7,672 8,012 CHWs Employed 5,133 5,374 5,652 6,012 6,352 6,671 7,070 7,446 7,799 7,931 8,256 HEOs Employed 546 554 561 568 574 580 585 590 595 599 604 Total Service-Delivery Staff 10,650 11,301 11,963 12,828 13,640 14,402 15,316 16,176 17,034 17,638 18,407 Population per Service Staff 847 819 793 758 730 709 681 658 638 629 616 Service Staff per 1,000 Pop. 1.18 1.22 1.26 1.32 1.37 1.41 1.47 1.52 1.57 1.59 1.62 Total Staff Cost (Kmn) 259.0 275.8 292.8 316.5 338.7 360.1 382.8 404.7 429.8 449.9 472.8 Total Recurrent Budget 958.7 999.9 1,042.9 1,087.7 1,134.5 1,183.3 1,234.1 1,287.2 1,342.6 1,400.3 1,460.5 S.D Staff as % Recurrent Budget 27.0 27.6 28.1 29.1 29.9 30.4 31.0 31.4 32.0 32.1 32.3 136  //  PNG Health Workforce Crisis: A Call to Action Nurse & CHW Training Costs (Kmn) (Recurrent Costs) Using 2009 Unit Costs (Kmn) 19.2 21.5 21.5 25.1 25.1 25.1 28.7 28.7 28.7 28.7 28.7 Quality-Enhanced Costs (Kmn) 28.8 32.6 32.6 38.0 38.0 38.0 43.5 43.5 43.5 43.5 43.5 2009 Costs % Recurrent Budget 2.0 2.1 2.1 2.3 2.2 2.1 2.3 2.2 2.1 2.0 2.0 Quality-Enhanced % Recurrent 3.0 3.3 3.1 3.5 3.4 3.2 3.5 3.4 3.2 3.1 3.0 Budget 11/27/12 11:38 AM CHAPTER 7 Meeting Human Resource Needs: Options and Recommendations to Enhance Health Human Resource Supply Responses to meet Needs 7.1. Introduction 7.2. Critical Need for Improved Data on Health Human Resources This chapter presents the core recommendations of 7.2.1 The Stock of Health Human the report: Resources: Public and Private (i) future priorities for data system development for health human resources planning and management; The data presented in Chapter 2 on the total existing stock (ii) institutionalization of the documentation of the of human resources working for the publicly financed existing supply capacity for health human resources; component of the health system has been derived from (iii) curriculum issues for core health service-delivery a special survey of the health workforce undertaken by staff (including for emergency obstetric care); the Human Resources Division of the NDoH for the (iv) the core training system expansion recommenda- NHP. This head count survey collected data on: (i) the tions arising from Scenario 5 for the future devel- broad occupations employed in the health system (based opment of health service-delivery staffing and the on a list of occupations agreed for the definition of mini- implications of the health human resource crisis mum standards); (ii) their sex; and (iii) their location by which will arise from taking no action on increasing province and district for health staff working in all health preservice training capacity; facilities—government and mission run—except for staff (v) the establishment of a “Whole-of-Government” Task working for open aid posts and for the NDoH in Port Force to manage implementation of the key agreed Moresby. This survey also has a number of potential flaws recommendations from the report; and for which it is hard to crosscheck for accuracy—never- (vi) an extensive consultation process on the key results theless, it is the best available. of the report in order to establish a national consen- This special survey was undertaken for the NHP sus on the way forward. because the five existing systems which gather informa- 137 1305747 PNG HR Report 11-27-12.indb 137 11/27/12 11:38 AM tion on health human resources are in a state of utter employed workers and the mission payroll(s) should confusion and/or lack any reliability or credibility in be the basis for much of this type of reporting for the their current form because they have been allowed to church sector of the health system. decay and/or are not maintained to any semblance of (ii) The Annual Census of Health Workers undertaken professional standards for reliability and scope. A key by Human Resource Division of NDoH: While this problem is that efforts across both government and within is an annual exercise in theory, in practice it is not the NDoH to collect data on human resources are dupli- updated annually because the capacity to do this and cative—often the same data is collected with slightly dif- report it is not present in NDoH. It is not clear that ferent details (see Annex 2.1 to Chapter 2). In addition, the system for data collection is sufficiently formally each system requires specialized systems, including com- codified with checks on the quality of data being sup- puter systems and databases, which need to be main- plied by PHAs and hospitals. The reporting systems tained, notwithstanding that the key skills required for are in Microsoft Excel with different coding structures maintaining such a system are all but completely lack- for many of the provinces and hospitals (41 spread- ing. The result is that no system is complete and/or easy sheets in total). The data for 2009 had been put into to access. Standard tables are very rarely produced as a Microsoft Access and a set of complex programs writ- matter of course. ten (using Crystal Report Writer) to enable the data The five existing systems which could/should pro- to processed. A consultant had been hired to do this, vide reliable data are: however, the system files and data were not backed (i) The Payroll Systems: In 2004 this was the best up and all system and data files were lost except for source for numbers employed in the health sector, the original spreadsheets. This confirmed that the although as reported by the World Bank (2007) it systems are simply not robust enough or adequately was clear that in 2004 there were significant num- managed technically. bers of ghost workers on the payroll. At this time This report gained access to the original 2009 (and for earlier periods) the payroll was the sin- files and recreated some of the national crosstabs for gle best source of data on key characteristics of the the data used in this report. These have been used, health workforce. This is no longer the case. While as described in Annex 2.1 to Chapter 2, to generate some effort has been made to reform the payroll and (estimate) information on the characteristics of the introduce a new system (Concept) it would seem that workforce for the current stock of the workforce. If other data on staff which used to be part of the system the payroll system was working, this data set would have not been maintained (for example age, sex and not be necessary (at least for most purposes and cer- occupation) and there are large gaps in the database. tainly for medium-term human resource planning). The NDoH has also been in the process of Nevertheless, in the short term, it was critical to the decentralizing the payroll to both hospitals and information needs of this report and until the payroll to provinces—without updating the data on each is fixed another data source will be needed. individual prior to its devolution. This appears to have been a mistake because, as efforts are made to 33  The recommendation made was as follows: The NDoH should establish PHAs, these will need to be recentralized, stop the current efforts to decentralize the payroll system to hos- at least at the provincial level, and major efforts will pitals and provinces as has been initiated in late 2009, pending need to be made to update the files and information agreement on: (i) how the missing information on the payroll is collected on all employees. An interim report of this to be updated and maintained; and (ii) a detailed plan is reached to ensure its decentralization is appropriate and manageable. The study suggested in early 2010 that the decentraliza- payroll is fragile at best and the systems developed to decentral- tion not be proceeded with until these issues had been ize it need to be very robust to prevent over expenditures and resolved.33 This report had planned to use the payroll corruption. The legal issues surrounding the payroll (as with all decentralization issues in PNG) are both complex and it is not as the principal source of data for the study but the immediately clear that the role, responsibilities and duties of the decay in data made this unfeasible. It should be the Secretary of Health with respect to control of expenditures and major source of personnel data on the government- a range of other issues are well served by rapid decentralization. 138  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 138 11/27/12 11:38 AM (iii) The NDoH Health Information System: This gener- and information expert able to consider and advise how: ates information on the health system operations. (i) the agreed minimum set of human resource data can As noted in the health plan, it is not always able to be best collected, managed, analyzed and updated; (ii) the gather data from all facilities—including quite often above set of information systems can be rationalized; and major hospitals. The system is also cumbersome for (iii) data for health human resource planning as outlined in NDoH staff to interrogate. It gathers some data on this report can best be generated annually and as required. health staff. It would be appropriate to consider what Recommendation 2. An immediate first step should additional staff information could/should be gener- be to re-establish the Health Professionals Database(s) ated on/by this system and whether it is possible for and to make it operational. This is critical to get fur- it to replace the annual survey. This system needs to ther insight to the scale of the private sector and how it be overhauled so that available information can be is tracking over time. This will need immediate technical accessed and analysis carried out on the database eas- support (systems development and database development ily and on demand. and management) and can be undertaken while other (iv) The Health Professionals Registration Systems: actions related to Recommendation 1 are implemented. These are the only systems that, in theory, collect any information on health professionals working 7.2.2 Documentation of Current National in the private sector. The database for these systems Training Capacity of Health Workers (doctors, nurses, and pharmacy) is not maintained to an adequate technical standard. It is not clear that In December 2009 an extensive survey instrument of the database even separates the numbers on it by year health training institutions was developed with NDoH or registration renewal. This should be looked at as staff and the vast majority of principals/heads of each a matter of urgency and resources devoted to ensur- health-related training institution from across the coun- ing it is adequately maintained and the information try. This survey gathered all the information presented made available to policymakers. All health profession- in Chapter 3 on the capacity of the training system and als currently pay a registration fee to the NDoH. Part Chapter 4 on the operations and state of health training of this should be devoted to ensuring the database is facilities. The survey provides key information on student managed effectively. Perhaps it should be contracted numbers, unit costs, staff and staff qualifications, train- out. The professional registration fees could be used ing and experience, as well as qualitative information on to finance this database management. aspects of their programs, student intakes, and the state (v) The Establishment Register held by the Depart- of infrastructure. This had to be done as a one-off exer- ment of Personnel Management: The Establishment cise for this report. Register should contain a database for all public There was enthusiastic support for the consultative service positions and the qualifications and expe- approach adopted for the design of the questionnaire. It rience required, as well as location, occupation, is acknowledged that principals/heads of training institu- and pay level. This is not currently available but it is tions made a major contribution to the definition of prob- also not within the remit of the NDoH to manage— lems they wanted analyzed. The Bank team, NDoH staff being the responsibility of the Department of Per- and the secretariat staff from the Australia-PNG joint study sonnel Management. of University and Post-Secondary Education conducted by Sir Rabbie Namaliu and Professor Ross Garnaut worked Recommendation 1. Establish a committee of NDoH jointly to develop this questionnaire and all information management with appropriate support from qualified was mutually shared. The secretariat was located in OHE. technical NDoH staff. The committee will have the remit to Despite this effort and with considerable follow-up by sec- review human resource data requirements for management retariat staff in the OHE and by both NDoH staff and a and planning purposes and to decide how best to rational- World Bank consultant, little information was forthcom- ize current data system arrangements (within the control ing from the universities. This is a gap in this study and of NDoH). It would need to be supported by a database one which needs to be rectified. Meeting Human Resource Needs   //  139 1305747 PNG HR Report 11-27-12.indb 139 11/27/12 11:38 AM Recommendation 3. NDoH management together versity-trained graduates have inadequate practical with OHE management form an ad hoc committee to experience. It is not clear that the universities had an determine how best to generate the key information adequate internal or external quality-assurance pro- required on health training program enrollment poli- gram (a point also made by the Garnaut-Namaliu cies (including all universities), institution throughputs Review of Universities). by program, costs and other related key information. (iv) what in-service training programs should contain. This needs to be agreed at a high level, perhaps with an There are, for example, proposals for special modules “all of government approach” to reaching agreement on on emergency obstetric care for the in-service train- how best to collect this critical data. This exercise would ing CHWs and nurses—see also below. also need to involve the mission training facilities, partic- ularly the CHW training institutions as these are currently Recommendation 4. These curriculum issues should all outside of government and are under the oversight of be considered by the Executive of the NDoH and seri- NDoH—not the OHE. ous consideration given to how to proceed to deal with these “all-of-government” issues which arise. The evolv- ing institutional structures for health worker training have 7.3. Health Training Program Curriculum not been kept in sync. Clearly a solution to these prob- Issues lems needs to involve NDoH, OHE, universities and rep- resentatives of other health-related training institutions. A range of critical curriculum issues are identified A solution will also need to involve the central agencies of throughout the report—not the least of which is a wide government including the Treasury and its Budgets Divi- range of previous reports which have either failed to sion and the DNP&M. generate a consensus on what to do and/or they have not been followed up and implemented. The range of key curriculum issues has included, 7.4. Health Service-Delivery Cadre but is not limited to: Issues (i) the appropriate level of training on birth complica- tions and emergency obstetric care that should be It is beyond the scope of this report to make firm spe- included in general nurse and CHW training pro- cific suggestions about the need or otherwise to alter grams. At present this is very minimal—notwith- significantly the structure of health cadres in operation standing that the vast majority of pregnant mothers in PNG. The NHP makes no suggestions in this regard— only see a CHW and perhaps a nurse prior to giving although it implicitly suggests there should be a higher birth and this is not expected to change radically over proportion of doctors—and in particular that “rural doc- the coming 10 to 20 years in rural areas; tors” are needed in the workforce. A discussion on the need (ii) how and whether to monitor and ensure the full for particular cadres (such as HEOs) and on the relative agreed curriculum on family planning, STIs, and mix of health cadres should be led by a set of decisions HIV prevention are taught to nurses and CHWs on what the key objectives of the health system are. Dia- training in church-run (and government) train- logue with senior health managers within NDoH, includ- ing facilities. Chapter 4 notes that there is very little ing with the NDoH Executive, were only tentative on this systematic professional standards reviews undertaken issue. The recommendations for a bottom-up planning pro- of training facilities of NDoH despite their mandate cess province-by-province (see below) would assist with in this area with respect to professional standards; this. When considering this report it would be appropriate (iii) the fact that NDoH is the major employer and finan- for there to be an extensive discussion on this issue prior cier of health workers but there is little account- to finalizing the recommended human resource strategy ability of the universities to meet nationally set for direct service-delivery staff (see below). curriculum standards. No midwives, for example, This report has not directly dealt with: (i) non direct have been registered for almost a decade because uni- service-delivery staff; and (ii) postgraduate training (for 140  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 140 11/27/12 11:38 AM example the proportion of general nurses who should response to the current human resource crisis facing the do a postgraduate specialty such as midwifery) that health sector needs to deal with: (i) the immediate sup- are key issues which need to be addressed in the near ply-side crisis (quantity); (ii) the qualitative side, including future. The crisis in just sustaining the existing service- the development of a significant in-service training pro- delivery workforce needs primary attention—as Scenario gram (including emergency obstetric care for the existing 1 in Chapter 5 and 6 so clearly demonstrates. However, staff); and curriculum issues for the preservice training these issues need to be considered in the context of the programs so that emerging staff have appropriate skills; mix of cadres and how they are trained. For instance, and (iii) incentives to ensure staff are able to be deployed important decisions need to be considered on how best where needed, particularly in rural areas. to meet the needs of the crisis in maternal mortality in PNG. A key question is “what mix of the following do we implement”: (i) rapidly expand the number midwives; 7.5. Health Service-Delivery Staffing and/or (ii) increase the number of doctors in rural areas; (Demand) and Training (Supply) and/or (iii) focus on basic level training in this area for Scenarios: The Implications nurses and CHWs (discussed above) and make sure the referral system is working. This report also has not con- This report presents five scenarios (options) for the sidered the needs of hospitals in any depth. The specific future development of first line direct service-deliv- need of hospitals requires much further analysis. Apart ery staff. The demand scenarios are discussed in detail from how hospital needs might grow over the next two in Chapter 5 and the implications for expansion of the decades (and thus absorb staff) it is not clear that cur- training system are discussed in Chapter 6. In summary, rent cadres meet their real needs. For instance, CHWs the five scenarios are as follows: were primarily trained to provide services in rural areas. • Scenario 1: No Change in Human Resource Supply Is there a need for different training for positions in hos- Capacity 2010–2030. This scenario is designed to pitals that are currently filled by CHWs. show the implications of a “Do Nothing” Strategy Recommendation 5. The NDoH management needs on the supply side (that is no change in the current to consider and make decisions on the issues raised in throughputs of the health training system) given the this report following extensive consultation with key expected retirements of the workforce due to its aging stakeholders. Specifically, NDoH management needs to and the sustained growth in the population. It dem- decide: (i) the key immediate steps that need to be taken onstrates clearly the current crisis facing the human to respond to the impending crisis in direct service-deliv- resource requirements of the health sector in PNG. ery staff facing the health system in PNG (expand gen- • Scenario 2: Aspirational Targets Envisioned by eral nurse and CHW preservice training); and (ii) which PNGDSP 2010–2030. The second set of projections broad scenario discussed in this report (or another aris- is based on the number of staff and/or population- ing from further dialogue) should underpin the “emer- to-staff ratios posited in the PNGDSP 2010–2030. gency response” to human resources for the health sector. These are in the form of absolute staff numbers for “No change” on the training supply side is not a each of medical officers, nurses and CHWs. The report feasible option and it is critical that decisions are made argues that this scenario: (i) is not affordable on cur- immediately as the recommendations in this report rent likely resource envelope projections; and (ii) does require critical investment decisions to reinvigorate not adequately reflect the likely needed composition and, in most cases, expand the training of existing cad- of cadres in the future health workforce. res. A specific set of decisions need to be made on: (i) the • Scenario 3: Maintaining Existing Population to Ser- mix of cadres; (ii) the future of HEOs (and the use of spe- vice-Delivery Staff Ratios. The third scenario assumes cialty nurses or nurse practitioners); (iii) the relative bal- the core direct service-delivery health cadres main- ance of pre- and in-service training of nurses and CHWs; tain both their current share of the workforce and and (iv) how to deal with emergency obstetric care knowl- the current (2009) population-to-staff ratios over edge in the health workforce. This report suggests that the the period 2009–2030. The fundamental driver of Meeting Human Resource Needs   //  141 1305747 PNG HR Report 11-27-12.indb 141 11/27/12 11:38 AM the demand for health staff in this example is growth training; and (iii) expanded quality-enhancing nonsalary of the population. This scenario is probably afford- budget expenditures. able but it is also probably not the right mix of cad- If the key assumptions are accepted, the scenario res required for the health workforce. recommends that: • Scenario 4: WHO “Threshold” Service-Delivery Staff (i) Medical Officers: The number of graduates required Density Targets. The fourth scenario is based on the from medical school(s) would need to increase WHO “threshold” density of 2.28 per 1,000 popu- from the current 49 per year in 2009 to 100 in 2019 lation (or a population-to-staff ratio of 439 to one) (this would entail first year enrollments for med- of doctors, nurses (registered and enrolled) and ical school rising to 111 in 2013) and increasing midwives. According to WHO, at ratios less than graduate numbers to 150 by 2023 and 200 by 2028. this, coverage of essential interventions, including (ii) Nurses: The number of nurses graduating in 2012 those necessary to reach the health-related MDGs, is of 165 would need to increase to 250 by 2015, to not likely.34 This scenario is not affordable—particu- 350 by 2016 and to about 500 by 2018. In the fol- larly in the outer years—and recommends a doctor- lowing decade the scenario envisages supply rising to-population ratio which is probably not feasible to around 600 by 2021; to 700 by 2023 and stabilizing from a supply constraint perspective and which is at 800 per year from 2026. The feasibility of achiev- lower than is need to meet the health needs of the ing this is possible but only with difficulty—at least population. in the early years. It implies entry to nurse training • Scenario 5: A Suggested Preservice Training Sce- schools to be up to 260 as soon as 2012–an increase nario. The fifth scenario—which is the broadly of over 50 percent over current (2011) entry. recommended scenario—is one which envisages: (iii) Community Health Workers: The number of grad- (i) a gradual reduction in the population-to-doc- uates from the CHW training schools would need tor ratio (from 17,512 to one in 2009, to 7,380 to to increase from the expected numbers graduat- one in 2030); (ii) a reduction in the population-to- ing in 2009 of 209 to around 300 by 2014. Intakes nurse ratio (from 2,041 to one in 2009 to 1,414 to to CHW training schools would need to expand to one in 2030); (iii) allowing a modest decline in the about 310 by 2013 given assumed dropout rates— population-to-CHW ratio from 1,509 in 2009 to an increase in graduations in 2014 over current out- 1,372 in 2030; and (iv) no expansion in the num- put capacity of 44 percent. The number of graduates ber of HEOs produced annually which will see the would need to further expand to 500 by 2016 and to population to HEO ratio increase. This preservice 600 by 2020. In the subsequent decade capacity would strategy is affordable, responds to the demand require- need to expand to 800 by 2026. ments for staff from the health system and leaves (iv) Health Extension Officers: The number of graduates space for recurrent health resources to be allocated from the DWU training program for HEOs under to a significant expansion of training—both preser- this scenario is maintained at current capacity. There vice and in-service. would be room to increase HEO numbers (by say 100 percent) without major changes to other cadres. The NDoH management should consider each of A policy decision, as discussed, is needed on HEOs. the scenarios posited in this report and decide if the one proposed as Scenario 5–A Suggested Preservice Training One important implication of the analyses (in all Scenario—is broadly appropriate. It still needs further scenarios) is that there will be a serious drop in the exploration with NDoH officials but is suggested as the number of nurses and CHWs over the period of the broadly recommended scenario. It is feasible, at a stretch, from a supply capacity constraint perspective assuming 34  WHO. 2006. This report notes specifically that countries with decisive immediate action and it leaves space in the health densities of doctors nurses, and midwives below 2.28 per 1,000 recurrent budget to finance: (i) expanded training (partic- population fail, on average, to achieve 80 percent coverage for ularly of CHWs); (ii) a significant ramp-up of in-service live deliveries by skilled birth attendants. 142  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 142 11/27/12 11:38 AM current NHP and extending into the next NHP period. larly Treasury and DNP&M. The National NDoH Execu- This depends on how decisive is the action taken to adjust tive should consider and make decisions on such a report the supply side because attrition from the health work- annually at an agreed point in its policy consideration force, due to age and other reasons, will exceed the num- cycle. The next report needs to consider more explicitly bers being trained and entering the workforce.35 In the the implications of the private sector and its likely growth short run this may not be so much of a crisis because the over the coming 20 years. To this end the need to fix the “Monash Study” shows that there is excess staffing capac- health professional registration system is critical (see Rec- ity relative to existing demand for services in rural areas. ommendations 1 and 2). The NHP has an objective of constraining staff numbers Recommendation 7. Establishment of a Whole- and budgets for the hospital sector. However, as also dis- of-Government Taskforce. The implications of the rec- cussed in this report, the NHP and the Health MTEF, the ommendations for the expansion of the health training objective of the current policy settings in health is to dra- system are significant in that they imply a rapid and imme- matically turn the declining demand for services around. diate expansion of the medical officer, general nursing This may take time so there may be a little breathing space. and CHW training. This will require a “Whole-of-Gov- This holds true at the national level but there is no cur- ernment Approach”. rent mechanism to adjust staffing at facility level according A “Whole-of-Government Taskforce” should be to workloads within and between provinces (or between established immediately36 to review options to: mission and government services). (i) develop a costed plan to expand training capacity as There is also no operational process through the agreed. This should also explore with training insti- budget for the health sector (provinces) to request and tutions short-term options to expand supply capacity; have funded additional staff—an issue which needs to (ii) explore options and incentives to encourage exist- redressed at a “whole-of-government” level as this is ing staff retention through: (i) incentives to reduce outside the control on the NDoH. Further, it is not clear early retirement; and (ii) incentives to postpone how long hospital services can be constrained in subse- retirement; quent NHP periods after 2015 and the private health sector (iii) explore options to encourage redeployment of staff is now significant (probably at least 15 percent of doc- to rural areas and for new graduates to deploy to rural tors and 10 percent of nurses) and growing and is likely areas—particularly areas now with staff shortages; and to expand dramatically if some of the economic projec- (iv) as part of the short-term options plan to expand tions come to fruition. This will draw staff from the pub- training of nurses and CHWs there is a need to sig- lic sector and is an emerging major challenge. nificantly refurbish existing training facilities (see Recommendation 6. The exercise of matching supply information presented in Chapter 4 derived from the and demand or building scenarios for human resource survey of institutions). development needs to be done annually—from the top down and from the bottom up. This could be linked to This taskforce should also become responsible for the MTEF process (it gathers a lot of the information addressing the set of “Whole-of-Government” imple- required to do the scenarios already) and to the process of mentation issues outlined below. expanding and redeveloping the health system province- by-province. Each province should prepare a plan for the deployment of current and planned future staff and facili- 35  The assumptions used for workforce attrition are conserva- ties district-by-district showing how staff are to be deployed tive compared to those used by the 2000 report on health human resources. The workforce has subsequently aged considerably and how workloads can be expected to demand the need so this will lead to increasing retirements over time. The lack of for additional staff. This process should be driven by the data on this is a critical issue. 36  This should consist of at least NDoH, Church Health Agen- Policy Division of the NDoH but with the active partici- cies, OHE, relevant universities, DNP&M, Budgets Division of pation of the Human Resources Division. Both the MTEF Treasury, Economic Policy Division of Treasury, National Depart- and Human Resource planning exercises need to be done ment of Education, Prime Minister’s Department and Depart- in close consultation with the central agencies—particu- ment of Personnel Management. Meeting Human Resource Needs   //  143 1305747 PNG HR Report 11-27-12.indb 143 11/27/12 11:38 AM 7.6. Implementing the Human Resource (v) the role of church agencies: nurses and CHWs. The Plan: Need for “Whole-of-Government” partnership between NDoH and church health agen- Approach cies is critical to both service delivery and the training of core service-delivery staff. The training of health A central theme throughout this report is that the staff (scholarships) for nursing schools is largely sup- NDoH has responsibility for ensuring that the health plied/determined by the OHE and most are publicly system is functioning and adequately staffed but it does financed. The OHE has a policy of encouraging the not have the ability to directly control many of the key abolition of single-stream training colleges and their facets required to make things happen. This means the affiliation/amalgamation with a university. Consen- NDoH needs information with which to advocate and sus needs to be reached immediately, as part of the engage other agencies of government and its key part- development of a training program expansion plan ners. In the past this has been inadequate, but a lot of discussed above, on how to proceed with the devel- effort is currently going into being clear on what policy opment of training capacity. priorities are and their financial implications. This is a major step—and it is reflected in the new NHP and the health MTEF. 7.7. The Role of Development Partners NDoH management has to lead and advocate for a whole-of-government process which is required to Development partners have recognized the need for this implement key aspects of this human resource strat- study and have been concerned for some time about the egy including: impending crisis in human resources for health based on (i) how to distribute both existing staff across prov- the fragmentary evidence that was previously available. inces (and with mission partners) as demand for ser- This report clearly demonstrates that decisive actions on vices evolve and as new graduates become available the health human resource front are needed. It is recom- from training institutions. This involves provinces, mended that NDoH involve development partners in the new PHAs, Department of Personnel Management, review of this report. There is a need to fully review the and Budgets Division of Treasury; findings of this report across government and there will (ii) ensuring effective demand for new staff positions. be considerable scope for immediate development partner The process of increasing the budgeted health work- support to upgrade training and invest in the institutional force establishment for provinces and hospitals is not developments required to generate information required clear and transparent. Staff need to be allocated accord- to both improve this report and to ensure subsequent ing to health service needs—through a combination analysis of human resource needs are better informed. of bottom-up and top-down planning; In providing support it is critical that development (iii) engagement with the Commission/Office of Higher partners support government systems and do not under- Education. There is a need to ensure that scholarships mine the needed institutional reforms required, or the for student places in training institutions are adjusted long-term sustainability of training institutions (for to the emerging staffing needs of the health sector. example by temporary supply of scholarships to spe- The NDoH does not make these decisions except for cific training institutions outside of the human resource CHWs—which are all currently provided by church framework set by the OHE and DNP&M). One option health authorities; would be to establish a contestable fund that requires (iv) the relationship between the Commission of Higher training institutions to bid for resources from an appro- Education/OHE, universities and the NDoH. This priate group of government agencies. is currently fraught with institutional fragmenta- Recommendation 8. This draft report should be tion, unclear responsibilities and accountabilities widely discussed and disseminated—throughout NDoH, (for example for curriculum, eligibility for profes- government, and church agencies. It is suggested that sional registration and for setting the size of courses) presentations are made by the World Bank initially to: and no mechanism to deal with these problems; and (i) the National Executive of NDoH; (ii) core related agen- 144  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 144 11/27/12 11:38 AM cies including DNP&M, Treasury and OHE; (iii) a special these agencies might have on the report. This could be fol- meeting of church agencies; (iv) development partners lowed by a one to two day conference where the results are accompanied by senior NDoH staff who will have respon- discussed collectively. This could then lead to the devel- sibility for driving the implementation of the report’s rec- opment of a strategic plan to respond to the call for action ommendations in order to elicit the unique perspectives to redress the health workforce crisis. Meeting Human Resource Needs   //  145 1305747 PNG HR Report 11-27-12.indb 145 11/27/12 11:38 AM 1305747 PNG HR Report 11-27-12.indb 146 11/27/12 11:38 AM REFERENCES Aitken, I.W. 2002. Papua New Guinea: Training Institution Garnaut, R., and R. Namaliu. 2010. PNG Universities Review. Mimeo. Port Moresby, Papua New Guinea. Review: Report to Prime Ministers Somare and Rudd. Aitken, I.W.,Bhattarai, H. R., Kolehmainen-Aitken, R., Governments of Papua New Guinea and Australia. Port Newbrander, W., O’Neil, M., and Pollock, J. 2002a. Moresby, Papua New Guinea and Canberra, Australia. Papua New Guinea Human Resource Development Gannicott, K. 2010. Higher Education Finance in Papua New Study: Realigning and Enriching the Skills of the Work- Guinea. Mimeo. Produced for Independent Review of force. Mimeo. Port Moresby, Papua New Guinea. PNG Universities. Port Moresby, Papua New Guinea. Aitken, I.W.,Bhattarai, H. R., Kolehmainen-Aitken, R., GoPNG. 2003a. Public Expenditure Review and Ratio- Newbrander, W., O’Neil, M., and Pollock, J.2002b. nalization (PERR) – Overview of Discussion Papers. Papua New Guinea: Human Resources for Health Pol- Mimeo. Port Moresby, Papua New Guinea. icy Issues Review Paper. Mimeo. Port Moresby, Papua GoPNG. 2003b. Public Expenditure Review and Rational- New Guinea. ization (PERR) – Discussion Paper No 1 – Road Map to Aitken, I.W., Bhattarai, H. R., Kolehmainen-Aitken, R., Fiscal Sustainability. Port Moresby, Papua New Guinea. Newbrander, W., O’Neil, M., and Pollock, J. 2002c. GoPNG. 2003c. Public Expenditure Review and Rational- Papua New Guinea: Human Resource Development ization (PERR) – Discussion Paper No 2 – Civil Service Strategy Final Report. Mimeo. Size and Payroll. Port Moresby, Papua New Guinea. Appiah-Denkyira, E., Narasimham, M., Snow, R., Haddad, GoPNG. 2003d Public Expenditure Review and Rational- D., Phillips, D., and Scheffler, R. 2010. Needs Based ization (PERR) – Discussion Paper No 5 – Health Sector Implications of Observed HRH Outcomes. Mimeo. Trends and Options. Port Moresby, Papua New Guinea. Government of Ghana and World Bank. GoPNG. 2010a. Keeping the Doors Open: Budgeting Guide- Australian Council for Education Research and Mas- lines for Rural Health Services. 2nd Edition. National saro Consulting. 2010. Report on Needs and Anal- Department of Health. Mimeo. Port Moresby, Papua ysis of Courses and Curricula. A background paper New Guinea. to the Garnaut-Namaliu Review of PNG Universi- GoPNG. 2010b. National Health Plan 2011–2020: Volume ties. (unpublished). 1 Policies and Strategies. Government of Papua New Cibulskis, R., and G. Hiawalyer. 2002. Information Systems Guinea: Port Moresby. for Health Sector Monitoring in Papua New Guinea. GoPNG. 2010c. National Health Plan 2011–2020: Volume Bulletin of the World Health Organization, vol. 80 (9). 2 (Part A) Reference Data and National Health Profile. DNP&M. 2009. Population Projections for the Provinces and Government of Papua New Guinea: Port Moresby. Regions of Papua New Guinea 2000–2030, National GoPNG. 2010d. National Health Plan 2011–2020: Volume Report, Report No 1. Department of National Planning 2 (Part B) Reference Data and National Health Profile. and Monitoring, Port Moresby, Papua New Guinea. Government of Papua New Guinea: Port Moresby. 147 1305747 PNG HR Report 11-27-12.indb 147 11/27/12 11:38 AM GoPNG and UN. 2004. Papua New Guinea – Millennium Jointly produced by the National Department of Health Development Goals Progress Report 2004. Port Moresby, and the World Health Organization, Port Moresby, Papua New Guinea and the United Nations. Papua New Guinea. Hornby, P., D.K. Ray, P.J. Shipp, and T.L. Hall. 1980. Guide- NSO. 1995. Papua New Guinea: Demographic and Health lines for Health Manpower Planning. World Health Survey 1991. National Statistical Office, Port Moresby, Organization. Papua New Guinea. Inder, B., J. Spinks, P. Srivastrava, and R. Sweeney. 2009. NSO. 1997. Papua New Guinea: Demographic and Health Papua New Guinea: Modeling Costs and Efficiency of Survey 1996. National Statistical Office, Port Moresby, Primary Health Care Services. Mimeo. Port Moresby. Papua New Guinea. Massaro, V., L. Perry, and D. Edwards. 2010. Report on NSO. 2002. Papua New Guinea 2000 Census Report: Final Needs Analysis and Analysis of Course Costs and Cur- Figures. National Statistical Office, Port Moresby, ricula. Mimeo. Produced for the Independent Review Papua New Guinea. of PNG Universities, Port Moresby. NSO. 2003. Recent Fertility and Mortality Indices and Trends National Strategic Taskforce. 2010. Papua New Guinea in Papua New Guinea: A Report Based on the Anal- Vision 2050. Government of Papua New Guinea, ysis of 2000 Census Data. National Statistical Office, Port Moresby. Port Moresby, Papua New Guinea. NAC and NDoH. 2005. The report of the 2004 National Papua New Guinea Nurses’ Association and Department Consensus Workshop of Papua New Guinea. World of Personnel Management. 2006. Consolidated Agree- Health Organization and the National HIV/AIDS Sup- ment Governing Terms and Conditions of Nursing Per- port Project. National AIDS Council and the National sonnel in the Health Service. Mimeo. Port Moresby, Department of Health, Port Moresby. Papua New Guinea. NDoH. 2006a. Education and Training. National Depart- PNG DFP. 1986. National Manpower Assessment 1982– ment of Health Human Resource Forum Mimeo. Port 1992. Department of Finance and Planning, Port Moresby, Papua New Guinea. Moresby, Papua New Guinea. NDoH. 2006b Demand and Supply. National Department PNG Ministry of Health. 2001. Report on the Future of of Health Human Resource Forum. Mimeo. Port Nurse and Community Health Worker Education. Moresby, Papua New Guinea. Mimeo. Port Moresby, Papua New Guinea. NDoH. 2008a. The Workforce. National Department PNG NTC. 2005. National Human Resource Development of Health Human Resource Forum. Mimeo. Port Policy and Strategy. Papua New Guinea National Train- Moresby, Papua New Guinea. ing Council, Port Moresby. NDoH. 2008b. Ministerial Taskforce on Maternal Health PNG NEFC. 2003. Review of Intergovernmental Financing in Papua New Guinea. Mimeo. Port Moresby, Papua Arrangements – Interim Report and Proposals for a New New Guinea. Framework. Papua New Guinea National Economic and NDoH. 2008c. Ministerial Taskforce on Maternal Health Fiscal Commission, Port Moresby, Papua New Guinea. in Papua New Guinea: The Departments Response. World Bank. 2004a. Papua New Guinea: Poverty Assessment. Mimeo. Port Moresby, Papua New Guinea. World Bank. 2004b. Papua New Guinea: Public Expendi- NDoH. 2011. The Medium-Term Expenditure Framework ture and Service Delivery (PESD). for Health 2011–2013. Mimeo. Port Moresby, Papua World Bank, Asian Development Bank and Australian New Guinea. Agency for International Development. 2007. Stra- NDoH and WHO. 2008. A Report on the Work Value of tegic Directions for Human Development in Papua Nurses Employed in Public Health Facilities. Mimeo. New Guinea. 148  //  PNG Health Workforce Crisis: A Call to Action 1305747 PNG HR Report 11-27-12.indb 148 11/27/12 11:38 AM PNG Health Workforce Crisis: A Call to Action OCTOBER, 2011 1305747