Tracking Universal Health Coverage: 2017 Global Monitoring Report Tracking Universal Health Coverage: 2017 Global Monitoring Report Tracking universal health coverage: 2017 global monitoring report ISBN 978-92-4-151355-5 © World Health Organization and the International Bank for Reconstruction and Development / The World Bank 2017 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https:// creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO or The World Bank endorse any speci c organization, products or services. The use of the WHO logo or The World Bank logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO) or The World Bank. WHO and The World Bank are not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. Tracking universal health coverage: 2017 global monitoring report. World Health Organization and International Bank for Reconstruction and Development / The World Bank; 2017. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, gures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO or The World Bank, concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The maps in this work have been produced by the WHO. The boundaries, colors or other designations or denominations used in these maps and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. The mention of speci c companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO or The World Bank, in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO and The World Bank, to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO or The World Bank, be liable for damages arising from its use. The ndings, interpretations and conclusions expressed in this publication do not necessarily re ect the views of WHO or The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this publication Printed in Switzerland. CONTENTS Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Executive summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Monitoring coverage of essential health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Financial protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Monitoring UHC in the SDG era . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi What UHC does and does not mean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii The 2017 global monitoring report on progress towards UHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv Chapter 1. Coverage of essential health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Key measurement concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Effective service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Tracer indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Proxy indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Index of essential health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Inequalities in service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Operationalizing SDG indicator 3.8.1: an index of essential health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Guiding principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Criteria for tracer indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Selected tracer indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Calculating the index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 First findings on SDG indicator 3.8.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Gaps in health service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Time trends in service coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Inequalities in maternal and child health services in low- and lower-middle-income countries . . . . . . . . . . . . 16 Trends in maternal and child health service coverage inequalities over time . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Next steps for an index of essential health services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 iii Chapter 2. Financial protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Measures of financial protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Catastrophic spending on health (SDG and non-SDG indicators) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Impoverishing spending on health (non-SDG indicators) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Operationalizing measures of financial protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Defining and measuring out-of-pocket spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Defining and measuring income and consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Defining and measuring ability to pay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Poverty lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Global Data – and dealing with ‘missing data’. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Household surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Missing data, global and regional estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Levels and trends in catastrophic spending: the SDG 3.8.2 indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Cross-country variation in catastrophic spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Global and regional estimates of catastrophic spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Trends in catastrophic spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Inequalities in catastrophic spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Levels and trends in catastrophic spending: non-SDG indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Nonfood spending as a measure of ability to pay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Levels and trends in impoverishment due to out-of-pocket spending: non-SDG indicators . . . . . . . . . . . . . . . 39 Cross-country variation in impoverishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Global and regional estimates of impoverishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Trends in impoverishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Depth of impoverishing health spending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Annexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Annex 1. UHC indicators (service coverage and financial protection) by country. . . . . . . . . . . . . . . . . . . . . . . 48 Annex 2. Current values of the UHC index of coverage of essential health services and values of each of the tracer indicators used to calculate the index, by country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Annex 3. List of countries by United Nations regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Annex 4. UHC service coverage index by WHO and World Bank regions, 2015 . . . . . . . . . . . . . . . . . . . . . . . . 65 Annex 5. Financial protection indicators by WHO and World Bank regions . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Table 1. Incidence of catastrophic health spending SDG indicator 3.8.2: 10% threshold . . . . . . . . . . . . . . . . 66 Table 2. Incidence of catastrophic health spending SDG indicator 3.8.2: 25% threshold . . . . . . . . . . . . . . . 67 Table 3. Incidence of impoverishing health spending: 2011 PPP $1.90-a-day poverty line . . . . . . . . . . . . . . . 68 Table 4. Incidence of impoverishing health spending: 2011 PPP $3.10-a-day poverty line . . . . . . . . . . . . . . . 69 iv TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT PREFACE T his year’s joint Universal Health Coverage Monitoring Report is being published at a crucial moment. Never before has there been as much political momentum for universal health coverage as there is right now. And never before has there been greater need for commitment to health as a human right to be enjoyed by all, rather than a privilege for the wealthy few. Ensuring that all people can access the health services they need – without facing financial hardship – is key to improving the well-being of a country’s population. But universal health coverage is more than that: it is an investment in human capital and a foundational driver of inclusive and sustainable economic growth and development. It is a way to support people so they can reach their full potential and fulfil their aspirations. This is why we, as the leaders of the World Bank Group and the World Health Organization, have made the achievement of universal health coverage a priority for both our institutions. Part of that commitment is this joint 2017 UHC Global Monitoring Report. The report reveals that at least half the world’s population still lacks access to essential health services. Furthermore, some 800 million people spend more than 10 per cent of their household budget on health care, and almost 100 million people are pushed into extreme poverty each year because of out-of-pocket health expenses. This is unacceptable. But what gives us hope is that countries across the income spectrum are leading and driving progress towards UHC, recognizing that it is both the right and the smart thing to do. We are also encouraged that – although data availability and analysis are still a challenge – most countries are already generating credible and comparable data on health coverage. We would like to acknowledge the role of the Organisation for Economic Co-operation and Development (OECD) and the United Nations Children’s Fund (UNICEF) in making this happen. Our data have revealed major gaps. The more we know about those gaps – and how different countries are bridging them – the closer we come to identifying what we must do to improve health coverage. But if the world is serious about meeting its goal of achieving Universal Health Coverage by 2030, we all need to be far more ambitious. To this end, the World Bank Group and the World Health Organization are committed to working with countries to increase access to essential health services, ensure that people don’t fall into poverty because of health expenses, and move closer to our goal of Universal Health Coverage by 2030. That won’t be easy, but it’s possible. We are ready to make it happen. Jim Yong Kim Tedros Adhanom Ghebreyesus President Director General The World Bank Group World Health Organization v CONTRIBUTORS The principal contributors to this report were: Inputs were provided by the following WHO consultants: Jonathan Cylus (European Observatory on Health Systems World Health Organization (WHO): Gabriela Flores, and Policies), Sayem Ahmed and Kateryna Chepynoga Daniel Hogan, Gretchen Stevens, Justine Hsu, Tessa Tan- for financial protection and Nicole Bergen, Maria Clara Torres Edejer, Sarah Thomson, Tamás Evetovits, Agnès Restrepo, Arne Rückert and Bin Zhou for service coverage. Soucat, John Grove The World Bank consultant for financial protection was Marc Smitz. Hajer Aounallah-Skhiri, Mohamed Hsairi, and and the World Bank: Patrick Hoang-Vu Eozenou, Adam Olfa Saidi of the Tunisia Health Examination Survey 2016 Wagstaff. team made available the results, and this was facilitated by the WHO Tunisia country office and the Eastern Specific sections, data collection and analyses or reviews Mediterranean Regional Office. were contributed by staff from: The report was undertaken under the overall guidance World Health Organization (Somnath Chatterji, Richard of Naoko Yamamoto (World Health Organization) and Cibulskis, Alison Commar, Christopher Fitzpatrick, Marta Timothy G. Evans (the World Bank). Gacic-Dobo, Philippe Glaziou, Laurence Grummer-Strawn, Ahmad Hosseinpoor, Rick Johnston, Teena Kunjumen, Writing and editing assistance was supplied by David Annet Retno Mahanani, Ann-Beth Moller, Vladimir Bramley, Jane Parry and Amanda Milligan. Design and Poznyak, Florence Rusciano, Anne Schlotheuber and the layout were done by L’IV Com Sàrl, Villars-sous-Yens, WHO regional focal points for the 2017 global monitoring Switzerland. Overall project management was done by report) Gaël Kernen. The World Bank (João Pedro Wagner De Azevedo, Caryn Financial support for the preparation and production of Bredenkamp, Jishnu Das, Tania Dmytraczenko, Dean this report was provided by the Government of Japan, Mitchell Jolliffe, Rose Mungai, Minh Cong Nguyen, Espen the Rockefeller Foundation and the International Health Beer Prydz, Marco Ranzani, Umar Serajuddin and Owen Partnership for UHC 2030 (UHC2030). WHO also K. Smith) acknowledges financial support from the UK Department for International Development (DFID). We are grateful for Organisation for Economic Co-operation and Development the data on access and quality provided by the OECD, and (Chris James and Nick Tomlinson) for their advice and guidance during the development of this report. United Nations Children’s Fund (David Hipgrave) and the Joint United Nations Programme on HIV/AIDS (Juliana Daher and Kim Marsh). Chapters 1 and 2 and the Executive summary adapt and expand on the material from the following articles published in Lancet Global Health: Hogan DR, Stevens GA, Hosseinpoor AR, Boerma T. An index of the coverage of essential health services for monitoring Universal Health Coverage within the Sustainable Development Goals. Lancet Global Health. 2017. DOI: http://dx.doi.org/10.1016/S2214-109X(17)30472-2. Wagsta A, Flores G, Hsu J, Smitz M-F, Chepynoga K, Buisman LR, van Wilgenburg K and Eozenou P. Progress on catastrophic health spending: results for 133 countries. A retrospective observational study. Lancet Global Health. 2017. DOI: http://dx.doi.org/10.1016/S2214-109X(17)30429-1 Wagsta A, Flores G, Smitz M-F, Hsu J, Chepynoga K and Eozenou P. Progress on impoverishing health spending: results for 122 countries. A retrospective observational study. Lancet Global Health. 2017. DOI: http://dx.doi.org/10.1016/S2214-109X(17)30486-2 Creative commons licence: CC BY 3.0 IGO vi TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT EXECUTIVE SUMMARY Introduction tracer indicators of coverage of essential services – was developed to monitor SDG indicator 3.8.1. For the first A number of the 17 Sustainable Development Goals time, this report presents methods and baseline results (SDGs) adopted by the United Nations General Assembly for 183 countries for the index. The UHC service coverage in September 2015 have targets that relate to health. index is straightforward to calculate, and can be computed However, one goal – SDG 3 – focuses specifically on with available country data, which allows for country-led ensuring healthy lives and promoting well-being for all at monitoring of UHC progress. all ages. Target 3.8 of SDG 3 – achieving universal health coverage (UHC), including financial risk protection, access The levels of service coverage vary widely between to quality essential health-care services and access to countries (Fig. 1). As measured by the UHC service safe, effective, quality and affordable essential medicines coverage index, it is highest in East Asia (77 on the and vaccines for all – is the key to attaining the entire goal index) and Northern America and Europe (also 77). as well as the health-related targets of other SDGs. Sub-Saharan Africa has the lowest index value (42), followed by Southern Asia (53). The index is correlated Target 3.8 has two indicators – 3.8.1 on coverage of with under-five mortality rates, life expectancy and the essential health services and 3.8.2 on the proportion Human Development Index. Moving from the minimum of a country’s population with catastrophic spending index value (22) to the maximum value (86) observed on health, defined as large household expenditure on across countries is associated with 21 additional years health as a share of household total consumption or of life expectancy, after controlling for per capita gross income. Both must be measured together to obtain a clear national income and mean years of education among picture of those who are unable to access health care and adults. those who face financial hardship due to spending on health care. Since the SDGs aim to “leave no one behind”, Coverage of essential services has increased since 2000. indicators should be disaggregated by income, sex, age, Time trends for the UHC service coverage index are not race, ethnicity, disability, location and migratory status, yet available, but average coverage for a subset of nine wherever data allow. This report presents the results of the tracer indicators used in the index with available time latest efforts to monitor the world’s path towards UHC. series increased by 1.3% per annum, which is roughly a 20% increase from 2000 to 2015. Among these nine tracer indicators, the most rapid rates of increase were Service coverage seen in coverage of antiretroviral treatment for HIV (2% in 2000 to 53% in 2016) and use of insecticide-treated Monitoring coverage of essential health nets for malaria prevention (1% in 2000 to 54% in 2016). services Nevertheless, there is still a long way to go to achieve UHC. Although data limitations preclude precise measurement of the number of people with adequate Progress towards UHC is a continuous process service coverage, it is clear that at least half of the world’s that changes in response to shifting demographic, population do not have full coverage of essential services. epidemiological and technological trends, as well as Considering selected health services, over 1 billion people people’s expectations. The goal of the service coverage have uncontrolled hypertension, more than 200 million dimension of UHC is that people in need of promotive, women have inadequate coverage for family planning, preventive, curative, rehabilitative or palliative health and nearly 20 million infants fail to start or complete the services receive them, and that the services received are of primary series of diphtheria, tetanus, pertussis (DTP)- sufficient quality to achieve potential health gains. A UHC containing vaccine, with substantially more missing other service coverage index – a single indicator computed from recommended vaccines. vii Fig. 1. UHC service coverage index by country, 2015: SDG indicator 3.8.1 Index value by quintile ≥77 70–76 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and 62–69 the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 46–61 Not applicable any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0 850 1,700 3,400 Kilometers ≤45 Data not available SDG: Sustainable Development Goal; UHC: universal health coverage. Equity Financial protection Because of the lack of data, it is not yet possible to compare Many families worldwide suffer undue financial hardship the UHC service coverage index across key dimensions of as a result of receiving the health care that they need. inequality. Until these data gaps are overcome, inequalities UHC efforts in this area focus on two issues: “catastrophic in service coverage can be assessed by looking at a spending on health”, which is out-of-pocket spending narrower range of service coverage indicators, in particular (without reimbursement by a third party) exceeding a for maternal and child health interventions. For a set of household’s ability to pay; and “impoverishing spending seven basic services for maternal and child health, only on health”, which occurs when a household is forced 17% of mothers and infants in households in the poorest by an adverse health event to divert spending away wealth quintile in low-income and lower-middle-income from nonmedical budget items such as food, shelter countries in 2005–2015 received at least six of the seven and clothing, to such an extent that its spending on interventions, compared with 74% in the richest quintile. these items is reduced below the level indicated by the poverty line. Considering changes in large gaps in coverage over time, the median percentage of mother-child pairs that received The incidence of catastrophic spending on health is less than half of seven basic health services declined reported on the basis of out-of-pocket expenditures between 1993–1999 and 2008–2015 across all wealth exceeding 10% and 25% of household total income quintiles for 23 low- and lower-middle-income countries or consumption. This is the approach adopted for the with available data. Absolute reductions were larger in SDG monitoring framework. Across countries, the mean poorer wealth quintiles, and therefore absolute inequalities incidence of catastrophic out-of-pocket payments at the were reduced between these two time periods. 10% threshold is 9.2%. Incidence rates are inevitably lower at the 25% threshold with a mean of 1.8%. At Unless health interventions are designed to promote the global level (Fig. 2),it is estimated that in 2010, 808 equity, efforts to attain UHC may lead to improvements in million people incurred out-of-pocket health payments the national average of service coverage while inequalities exceeding 10% of household total consumption or income, worsen at the same time. Gaps in service coverage (some 11.7% of the world’s population), and 179 million remain largest in the poorest quintile, which reinforces incurred such payments at the 25% threshold (2.6% of the importance of structuring health services so that no the population). one is left behind. viii TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Fig. 2. Global and regional trends in catastrophic payments: SDG indicator 3.8.2 Africa Asia Europe Latin America and the Caribbean North America Oceania 900 808 800 741 700 589 600 500 400 300 179 200 155 113 100 0 2000 2005 2010 2000 2005 2010 10% threshold 25% threshold In 2010, Latin America and the Caribbean was the region Indicators of impoverishing spending on health are not part with the highest rate at the 10% threshold (14.8%). of the official SDG indicator of universal health coverage per Asia had the second-highest rate (12.8%) and was the se, but they link UHC directly to the first SDG goal, namely region where most people facing catastrophic payments to end poverty in all its forms everywhere. These indicators are concentrated. Both the percentage and the size of are based on international poverty lines – specifically the global population facing catastrophic payments 1.90 a day international dollars using 2011 purchasing have increased at all thresholds since 2000. At the power parity (PPP) for extreme poverty and 2011 PPP 10% threshold, the region with the fastest increase in 3.10 a day international dollars for moderate poverty. population facing catastrophic payments is Africa (+5.9% This report measures the incidence of impoverishment as per year on average) followed by Asia (+3.6% per year). the difference between the number of people in poverty North America is the only region where both the incidence with out-of-pocket spending included in household total and the population exposed have decreased (–0.9% consumption or income, and the number without. per year). An estimated 97 million people were impoverished on While monitoring SDG indicators of catastrophic health care at the 2011 PPP $ 1.90-a-day poverty line in expenditures is important, it is not the only way in which 2010, equivalent to 1.4% of the world’s population. At progress can be monitored, nor is it sufficient on its the 2011 PPP $ 3.10-a-day poverty line, the figure is 122 own to fully understand the picture as countries strive million (1.8%). At these two international poverty lines to provide financial protection. Catastrophic payments impoverishment rates in upper-middle-income countries can be measured in different ways. In addition, financial and high-income countries are close to or equal to zero. protection can also be measured using metrics other than At the 2011 PPP $ 1.90-a-day poverty line, the number and catastrophic spending. So, this report also provides global percentage of people globally impoverished fell between and regional results using complementary measures of 2000 and 2010 from 130 million (2.1%) to 97 million financial protection. (1.4%). By contrast, at 2011 PPP $ 3.10-a-day, both the percentage and number of people impoverished increased from 106 million (1.7%) to 122 million (1.8%), (Fig. 3). ix Fig. 3. Global and regional trends in impoverishment due to out-of-pocket payments: $1.90-a-day and $3.10-a-day poverty lines Africa Asia Europe Latin America and the Caribbean 140 130.4 122.3 120 115.6 115.8 106.1 97 100 80 60 40 20 0 2000 2005 2010 2000 2005 2010 $1.90-a-day poverty line $3.10-a-day poverty line In 2010, Asia and Africa had the highest rates of Monitoring UHC in the SDG era impoverishment at the 2011 PPP $ 1.90-a-day poverty line (1.9% and 1.4% respectively). Between 2000 and 2010, The monitoring efforts in this report relate directly to one Africa saw reductions in the incidence of impoverishing of the defining characteristics of the SDGs: promoting spending on health at both the 2011 PPP $ 1.90 and 2011 accountability by encouraging countries to commit to PPP $ 3.10 lines, while Asia saw a marked reduction at reporting of their progress. Most of the data provided the 2011 PPP $ 1.90 line and an increase at the 2011 PPP in the following pages have been subject to an official $ 3.10 line. consultation with World Health Organization (WHO) Members States carried out in 2017. Countries are the The report also focuses on the depth of poverty, taking into main actors in monitoring and evaluation, and national account the monetary impact of out-of-pocket payments ownership is key to the success of achieving the SDGs. on those pushed, and further pushed, into poverty due to Each country’s process of monitoring and evaluation spending on health. will take account of national and potentially subnational priorities. Countries can also contribute to regional SDG Note that a low incidence of catastrophic or impoverishing monitoring frameworks. It is hoped that by developing spending on health could result from people being metrics and reporting internationally comparable data, protected from financial hardship, but it could also result this report may encourage countries and regions to refine from people not getting the care they need because they and tailor them to their local circumstances. cannot access it or because they cannot afford it. Financial protection always needs to be jointly monitored with As the data show in this report, the process is fraught with service coverage. challenges, not just in reaching the targets themselves, but also in terms of measuring progress towards them. The road to UHC is long, but the global commitment to achieving and measuring it is underway. x TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT INTRODUCTION The goal of universal health coverage (UHC) is to ensure Health Report, Health systems financing: the path to universal that every individual and community, irrespective of coverage shows, countries across the world have for some their circumstances, should receive the health services time been heeding the call and implementing reforms they need without risking financial hardship. In the last geared to accelerating progress towards UHC (3). 10 years or so, calls for increased efforts to achieve UHC have grown noticeably. In a September 2017 Lancet Global The momentum behind UHC was reflected in the Health editorial, Tedros Adhanom Ghebreyesus, Director- United Nations General Assembly (UNGA) decision General of the World Health Organization (WHO), called of September 2015 to adopt health as one of the 17 UHC an ethical question, asking: “Do we want our fellow sustainable development goals (SDGs) (4) and UHC as citizens to die because they are poor?” (1). an SDG health target (SDG 3.8: “achieve universal health coverage, including financial risk protection …”). The Jim Yong Kim, President of the World Bank Group, UHC target lies at the core of the other 12 health targets, addressing the May 2013 World Health Assembly said: and the health goal itself is closely interlinked with the “We can bend the arc of history to ensure that everyone other 16 SDGs, in some cases making inputs into them in the world has access to affordable, quality health and in others being dependent on their progress for its services in a generation” (2). And as WHO’s 2010 World attainment (Fig. 1). Fig. 1. Health is central to the SDG agenda xi Box 1. Definitions of UHC, SDG target 3.8, and SDG indicators 3.8.1 and 3.8.2 Universal health coverage means that all people receive the health services they need, including public health services designed to promote better health (such as anti-tobacco information campaigns and taxes), prevent illness (such as vaccinations), and to provide treatment, rehabilitation and palliative care (such as end-of-life care) of su cient quality to be e ective, while at the same time ensuring that the use of these services does not expose the user to nancial hardship (12). SDG target 3.8: Achieve universal health coverage, including nancial risk protection, access to quality essential health-care services and access to safe, e ective, quality and a ordable essential medicines and vaccines for all. SDG indicator 3.8.1: Coverage of essential health services (de ned as the average coverage of essential services based on tracer interventions that include reproductive, maternal, newborn and child health; infectious diseases; noncommunicable diseases; and service capacity and access; among the general and the most disadvantaged population). SDG indicator 3.8.2: Proportion of population with large household expenditures on health as a share of total household expenditure or income. It was not until July 2017 that the UNGA adopted specific Both indicators must be measured together to capture indicators for measuring the SDGs, including UHC (SDG the complete picture, and in particular not to miss those target 3.8). These were based on the recommendations of who are unable to access health care at all (and therefore the United Nations (UN) Inter-agency and Expert Group do not pay for it at the point of use), and those who (IAEG) on Sustainable Development Goal Indicators, receive low-quality care (12). WHO is the designated composed of national statisticians from 27 countries (5). In custodian agency for both SDG 3.8 indicators, with the case of SDG target 3.8, the IAEG found a high degree of the United Nations Children’s Fund (UNICEF), United consensus among technical experts, civil society, national Nations Population Fund (UNFPA) and United Nations governments and UN agencies, thanks to a consultative Department of Economic and Social Affairs (UN DESA) process spanning 2013 and 2014 involving all relevant Population Division as partner agencies for 3.8.1 and the experts and stakeholders (6–8). World Bank for 3.8.2. This process on UHC monitoring built on a collaborative Equity is key to the SDGs in general and to UHC specifically, effort by WHO and the World Bank, announced at the requiring as it does that everyone – irrespective of their February 2013 WHO-World Bank ministerial level meeting circumstances – gets the services they need without on universal health coverage (9), to develop a monitoring experiencing financial hardship (12). To measure UHC, it framework to support countries in tracking their progress is therefore necessary to assess not only access to use of towards the goal of UHC. This work led to the publication health services and the direct cost of care for a country’s of a discussion paper in December 2013 (10), and the population overall, but also that different segments of the launch in 2014 of the WHO-World Bank global monitoring population, particularly the most disadvantaged, are not framework for UHC (7, 8). left behind – in line with the SDG spirit. This has led to an increased emphasis on monitoring distributions across In their 2017 declaration, the G20 ministers of health invited dimensions of inequality, as well as averages. Accordingly, “the WHO to identify appropriate indicator frameworks and SDG goal indicators are to be disaggregated by income, to monitor progress on HSS [health systems strengthening] sex, age, race, ethnicity, disability, geographical location and UHC worldwide, working jointly with the World Bank, and migratory status, as applicable (5). the OECD and other relevant stakeholders” (11). The framework used in this report builds on two SDG UHC indicators: What UHC does and does not mean UHC means that everyone – irrespective of their living 3.8.1 which captures the population coverage dimension standards – receives the health services they need, and that of UHC (that everyone – irrespective of their living using health services does not cause financial hardship. standards – should receive the health services they need); and Progress towards UHC means that more people – especially the poor, who are currently at greatest risk of 3.8.2 which captures the financial protection dimension not receiving needed services – get the services they need. of UHC (use of health services should not lead to Implicit in the definition of UHC is that the services are financial hardship) (Box 1). high quality, meaning that people are diagnosed correctly xii TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Fig. 2. Investing in health systems to reach UHC and the SDGs SDG 1: No poverty SDG 3: Equitable health SGD 8: Inclusive economic SDG 4: Quality education outcomes and well-being; growth and decent jobs Impact on SDGs SDG 5: Gender equality global public health security SDG 16: Inclusive societies and resilient societies Universal health coverage SDG Target 3.8 All people and communities receive the quality health services they need, Determinants of health without nancial hardship Actions Health systems strengthening Source: adapted from Kieny et al., 2017 WHO Bulletin (13). and receive the interventions currently agreed to be on nearly three years’ worth of collaborative work between necessary. Progress towards UHC means a lowering of WHO and the World Bank, dating back to the February barriers to seeking and receiving needed care: for example, 2013 WHO-World Bank ministerial-level meeting on out-of-pocket payments, distance, poorly equipped universal health coverage, and leading to the joint WHO- facilities and poorly trained health workers. World Bank global monitoring framework for UHC which underpinned the first global monitoring report (7–9). But UHC also means that getting needed health services is associated less and less with financial hardship; that This report comes shortly after the UNGA’s adoption of people receiving health services are still able to afford food the two specific UHC indicators 3.8.1 and 3.8.2 earlier in and other necessities, and do not place their families at 2017, and therefore places a strong emphasis on their risk of poverty by getting the care they need. measurement. Initial analyses on, or in support of, these indicators were reported in the first global monitoring UHC does not mean that health care is always free of report, but are given greater prominence here (12). charge, merely that out-of-pocket payments are not so high as to deter people from using services and causing In the 2015 report (12), a set of individual tracer indicators financial hardship. Nor is UHC solely concerned with were used to paint a picture of the coverage of essential financing health care. In many poorer countries, lack services, while in the current report an index is computed of physical access to even basic services remains an from tracer indicators to summarize the coverage of enormous problem. Health systems have a role to play essential services using one number, consistent with the in achieving progress towards UHC. Health systems definition of SDG indicator 3.8.1. On financial protection, strengthening – enhancing financing but also strengthening this report expands the geographical scope considerably. governance, the organization of the health-care workforce, Whereas the 2015 report was based on financial service delivery, health information systems and the protection data from 37 countries covering less than provision of medicines and other health products – is 20% of the world’s population, the current report draws central to progressing towards UHC (Fig. 2). on data from 132 countries representing over 90% of the world’s population. The 2017 global monitoring report While the two UHC SDG indicators are important, they are a subset of the indicators used to monitor progress on progress towards UHC towards UHC and part of a broader UHC monitoring agenda, which draws on a wider range of established This joint report by the World Bank and World Health indicators, often tailored to specific regions and countries. Organization on progress towards UHC is the second This report, goes beyond the official SDG UHC indicators. in the series. The first, launched in December 2015 (12), Thus, in addition to reporting on ‘catastrophic’ out-of- shortly after the adoption of UHC as an SDG target, built pocket expenditures (SDG 3.8.2), the report also reviews xiii progress towards reducing impoverishment due to out- 4. Resolution 70/1. Transforming our world: the 2030 Agenda for of-pocket expenditures. This second aspect of financial Sustainable Development. In: Seventieth session of the United Nations General Assembly, New York, 21 October 2015. New protection is not an official SDG indicator for UHC, but York: United Nations; 2015 (https://undocs.org/A/RES/70/1, it links directly to the very first SDG goal, namely to end accessed 2 December 2017). poverty in all its forms everywhere. 5. Resolution 71/313. Work of the Statistical Commission pertaining to the 2030 Agenda for Sustainable Development. In: The monitoring efforts in this report relate directly to one Seventy-first session of the United Nations General Assembly, of the defining characteristics of the SDGs: promoting New York, 10 July 2017. New York: United Nations; 2017 (https:// undocs.org/A/RES/71/313, accessed 2 December 2017). accountability by encouraging countries to commit to reporting their progress. Most of the data provided in 6. Monitoring universal health coverage [website]. Geneva: World the following pages have been subject to an official Health Organization; 2017 (http://www.who.int/healthinfo/ universal_health_coverage/en/, accessed 2 December 2017). consultation with WHO Member States carried out in 2017. Countries are the main actors in monitoring and 7. World Bank Group, World Health Organization. Monitoring evaluation, and national ownership is key to the success of progress towards universal health coverage at country and global levels: framework, measures and targets. Washington achieving the SDGs. Each country’s process of monitoring DC, Geneva: The World Bank, World Health Organization; 2014. and evaluation will take account of national and potentially subnational priorities. Countries can also contribute to 8. Boerma T, Eozenou P, Evans D, Evans T, Kieny MP, Wagstaff A. Monitoring progress towards universal health coverage at regional SDG monitoring frameworks. It is hoped that country and global levels. PLoS Med. 2014;11(9):e1001731. by developing metrics and reporting internationally 9. World Bank Group, World Health Organization. Report on comparable data, this report may encourage countries the ministerial level roundtable on universal health coverage. and regions to refine and tailor them to their local In: WHO/World Bank ministerial-level meeting on universal circumstances. health coverage, 18–19 February 2013, Geneva, Switzerland (http://www.who.int/health_financing/ministerial_meeting_ report20130328.pdf, accessed 7 December 2017). As the subsequent pages show, the process is fraught with challenges, not just in reaching the targets themselves, 10. World Bank Group, World Health Organization. Monitoring progress towards universal health coverage at country and but also in terms of measuring progress towards them. global levels: a framework [joint WHO/World Bank Group The road to UHC is long, but the global commitment to Discussion Paper]. Washington DC, Geneva: The World Bank, achieving and measuring it is underway. World Health Organization; 2013. 11. Together today for a healthy tomorrow. Berlin Declaration of the G20 Health Ministers. Berlin: G20; 2017 (http://www. References bundesgesundheitsministerium.de/fileadmin/Dateien/3_ Downloads/G/G20-Gesundheitsministertreffen/G20_Health_ Ministers_Declaration_engl.pdf, accessed 2 December 2017). 1. Ghebreyesus TA. All roads lead to universal health coverage. 12. World Bank Group, World Health Organization. Tracking Lancet Glob Health. 2017 Sep;5(9): e839-e840. universal health coverage: first global monitoring report. Washington DC, Geneva: The World Bank Group; World Health 2. Poverty, health and the human future. Speech by Jim Yong Kim, Organization; 2015 (http://www.who.int/healthinfo/universal_ World Bank Group President, to the World Health Assembly, 21 health_coverage/report/2015/en/, accessed 2 December May 2013, Geneva, Switzerland. (http:/ /www.worldbank.org/ 2017). en/news/speech/2013/05/21/world-bank-group-president- jim-yong-kim-speech-at-world-health-assembly, accessed 2 13. Kieny MP, Bekedam H, Dovlo D, Fitzgerald J, Habicht J, Harrison December 2017). G et al. Strengthening health systems for universal health coverage and sustainable development. Bull World Health 3. The World Health Report. Health systems financing: the path to Organ. 2017;95(7):537-539. universal coverage. Geneva: World Health Organization; 2010. xiv TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT CHAPTER 1 COVERAGE OF ESSENTIAL HEALTH SERVICES Progress towards UHC is a continuous process that changes in response to shifting demographic, epidemiological and technological trends, as well as people’s expectations. The goal of the service coverage dimension of UHC is that people in need of promotive, preventive, curative, rehabilitative or palliative health services receive them, and that the services received are of sufficient quality to achieve potential health gains. Resource constraints mean that countries cannot provide all health services, but all countries should be able to ensure coverage of essential health services. This section presents methods and SDG baseline results for an index, which aims to summarize the coverage of essential health services with a single number, as well as estimates of gaps in service coverage and more detailed analyses of levels and trends in a subset of service coverage indicators by key dimensions of inequality. Health service coverage: Key findings Levels of service coverage vary widely across countries. The UHC service coverage index has a value of 64 (out of 100) globally, with values ranging from 22 to 86 across countries in 2015. As measured by the index, coverage of essential services is highest in the SDG regions of Eastern Asia (77) and Northern America and Europe (77), whereas sub-Saharan Africa has the lowest index value (42), followed by Southern Asia (53). High index values are associated with high life expectancy, even after controlling for national income and education. The index is correlated with under-5 mortality rates (ρ=-0.86), life expectancy (ρ=0.88), and the Human Development Index (ρ=0.91). Moving from the minimum index value (22) to the maximum index value (86) observed across countries is associated with 21 additional years of life expectancy after controlling for gross national income per capita and mean years of adult education. At least half of the world’s population does not have full coverage with essential health services… Precisely estimating this number is challenging, but based on a set of plausible sensitivity analyses, the number of people who are covered with most essential services ranged from 2.3 to 3.5 billion in 2015. This implies that at least half of the world’s 7.3 billion people do not receive the essential health services they need. …with substantial unmet need for a range of specific interventions. For example, more than 1 billion people live with uncontrolled hypertension; more than 200 million women have inadequate coverage for family planning; and almost 20 million infants fail to start or complete the primary series of DTP-containing vaccine, with substantially more missing other recommended vaccines. Coverage of essential services has increased since 2000. Time trends for the UHC service coverage index are not yet available, but average coverage for a subset of nine tracer indicators used in the index with available time series increased by 1.3% per annum, which is roughly a 20% relative increase from 2000 to 2015. Among these tracer indicators, the most rapid rates of increase were seen in coverage of antiretroviral treatment for HIV (2% in 2000 to 53% in 2016) and use of insecticide-treated nets for malaria prevention (1% in 2000 to 54% in 2016). Despite progress, large inequalities in basic maternal and child health services in low- and lower-middle-income countries persist. Absolute wealth inequalities in the coverage of seven basic maternal and child health services have declined; however, only 17% of those in households in the poorest wealth quintile in low- and lower-middle-income countries received at least six of seven basic interventions, as compared with 74% in the wealthiest quintile. 2 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Key measurement concepts Proxy indicators Effective service coverage For several important health areas, including NCDs, mental health, surgical and emergency care, as well as Effective service coverage is defined as the proportion routine health examinations, robust indicators of service of people in need of services who receive services of coverage are not always available. In these cases, proxy sufficient quality to obtain potential health gains (1). indicators must be used to reflect these important areas. Effective coverage indicators capture a country’s efforts Proxy indicators are correlated with the provision of to meet people’s needs for quality health services, and health services to those in need, and may be ‘upstream’ are the preferred indicators for monitoring the service or ‘downstream’ of (effective) service coverage. Indicators coverage dimension of UHC. As an example, an indicator of capacity, access or service utilization are upstream of effective coverage of treatment for HIV should measure – they represent either the availability of services for not just whether an individual is receiving antiretroviral those in need or the rate of use of such services, without therapy, but also whether her viral load is suppressed. providing information about the proportion of people in Unfortunately, for many important health areas, indicators need of a particular service that actually receive it. In of effective coverage are not widely available, either due to the other direction, ‘downstream’ indicators such as the lack of investment in data collection or difficulties around prevalence of a risk factor or mortality rate of a disease defining an operational indicator for a particular health or injury reflect the impact of service coverage, but are service. In these cases, other indicators associated with also a function of other factors that may be outside the effective coverage must be used. control of the health system, such as a country’s wealth or average education level. Service coverage Index of essential health services Indicators of service coverage, which is defined as the proportion of people in need of a service that receive it, The UHC service coverage index is a single indicator regardless of quality, are more commonly measured than that is computed based on tracer indicators (some effective coverage indicators. For example, the number of which are proxies of service coverage) to monitor of antenatal care visits can be ascertained by self-report coverage of essential health services. Essential health in a survey, but determining the quality of care received services are services that all countries, regardless of during those visits is more challenging. In the absence of their demographic, epidemiological or economic profile, information on effective coverage, these indicators are are expected to provide. This is what is intended by the often used for monitoring the coverage of health services, definition of SDG indicator 3.8.1, which is: at the expense of capturing information on the quality of the services received. There is not always a definitive line Coverage of essential health services (defined as the average coverage of essential services based on tracer separating effective service coverage and service coverage interventions that include reproductive, maternal, newborn for a given health service, and therefore in some cases and child health, infectious diseases, noncommunicable which label to use for an indicator may not be clear. This diseases and service capacity and access, among the general and the most disadvantaged population). report often uses ‘service coverage’ as short-hand for both. There are a number of methodological choices that Tracer indicators must be made when constructing an index, including the selection of tracer indicators and the calculations used to Countries will provide a wide range of services as they combine them into a final index value. There are a number progress towards UHC. It is not practical to monitor of examples of indexes meant to summarize population indicators for all of these services; therefore a manageable health (3–5), including for UHC (6–8), which often draw subset of indicators was chosen to represent overall inspiration from the Human Development Index. For the coverage (1, 2). Tracer indicators were selected based on first time, this report and accompanying journal article (9) several criteria, which are discussed in more detail below. operationalizes a measure of SDG indicator 3.8.1 on the It is important to note that these tracer indicators are not coverage of essential health services, presenting methods a recommended basket of services; rather they are chosen and baseline results for 183 countries. The UHC service to capture the breadth of health services within UHC in coverage index is straightforward to calculate, and can be a measurable way. computed with available country data, which allows for country-led monitoring of UHC progress. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 3 Inequalities in service coverage diseases and injuries. Following the definition of SDG 3.8.1, four categories of indicators were established: Inequalities in service coverage can be summarized by RMNCH, infectious diseases, noncommunicable diseases, calculating coverage levels in different subpopulations, service capacity and access. Lastly, the index should be for example by household wealth quintiles, educational disaggregated by key inequality dimensions. attainment, geographical region, age and sex. It is important to measure coverage across key dimensions of inequality since national averages can mask low coverage Criteria for tracer indicators levels in disadvantaged population groups. In each of the four categories described by the definition of SDG 3.8.1, tracer indicators were selected based on several Operationalizing SDG indicator 3.8.1: criteria (2) and ensuring that within each category the indicators reflect a range of programme service delivery an index of essential health services strategies. First, an indicator should be relevant, reflecting epidemiological burden and the presence of cost-effective Guiding principles interventions. Second, it must also be feasible, with current, comparable data available for most countries, The index was developed as part of a multi-year process which ideally can be disaggregated for equity analysis. that included global reviews, country case studies, Third, an indicator should be conceptually sound, with a consultations with ministry of health officials, and a measurable numerator and denominator, a clear target formal WHO country consultation with Member States and ideally, a definition that captures effective coverage in 2017 (1, 2, 10–15). The development of the index followed (16). Lastly, an indicator should be usable, in the sense it is four guiding principles, not all of which are fully achievable easy to communicate: indicators that are already reported given current data availability. The first guiding principle across countries, including those in the SDG monitoring concerned the preference for measures of effective framework, are appealing as they reduce reporting burden. service coverage. Second, in line with the definition of UHC, the index should include indicators for different Identifying indicators that fulfil these criteria is challenging types of services, namely: prevention, comprising health (Box 1.1 and Box 1.2), and few of the selected indicators fulfil promotion and illness prevention, as well as indicators all criteria. The greatest challenge is lack of available data for treatment, comprising curative services, rehabilitation for indicators of service coverage. These data limitations and palliation (2). Note, this includes public health services motivate the use of proxy indicators, in particular for NCD and interventions that are not implemented by the health treatment coverage, and by definition within the service sector but which have health improvement as a key capacity and access category. Use of proxy indicators motivation (1). Third, the index should cover all main ensures that the first two criteria, relevance and feasibility, health areas of reproductive, maternal, newborn and child are met for all indicators. health (RMNCH), infectious diseases, noncommunicable Box 1.1. Challenges of monitoring effective service coverage There are three key challenges associated with monitoring e ective service coverage, which is de ned as service coverage that results in the maximum possible health gains. The rst challenge is accurate measurement of the population in need of the service. Administrative records from service providers and self-reported prior diagnosis are often unreliable sources of information, as those who do not have access to health services remain undiagnosed. A full assessment of population need requires alternative sources of data, such as a set of survey questions or biomarkers collected in a household health examination survey. Because few conditions requiring treatment can be diagnosed in this way, this substantially limits the set of e ective coverage indicators that may be reliably monitored. Determining e ectiveness of service coverage – that is, the degree to which services result in health improvement – is a second challenge (a comprehensive discussion of measuring quality is discussed in Box 1.2). For some indicators, it is possible to directly measure quality of care. For example, monitoring of treatment for hypertension can include measurement of whether hypertension is e ectively controlled, and monitoring of cataract surgical coverage can include measurement of current visual acuity (17). However, generally speaking, measuring e ectiveness of care is more complicated than measuring service provision. The third key challenge is to monitor equity in access to quality health services. Making sure that no one is left behind as countries strive for UHC requires access to data disaggregated by inequality dimensions, such as wealth or geographical location. Disaggregated data are commonly available for RMNCH interventions, malaria prevention, and water and sanitation services in low- and middle-income countries, but may not be available for other health topics and indicators required for UHC monitoring. Therefore, investments are needed in data collection, especially for conducting regular household health examination surveys and developing electronic and harmonized facility reporting systems. In addition, it is crucial to build capacities for analysing and reporting health inequality data. Only then can countries tie this information to the policies they are implementing to improve health equity. 4 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Box 1.2. Measuring quality of care Measurement of health-care quality begins with understanding what is meant by quality, which is a multifaceted concept (18). The Health Care Quality Indicators project, initiated in 2001 by the OECD, which aims to develop and report common indicators for international comparisons (19), has distilled quality to three main dimensions: e ectiveness, patient safety, and responsiveness/people-centredness (19, 20). In countries with well-developed health information systems, data for monitoring are often derived from administrative reporting systems; in contrast, in low- and middle-income countries, such data are typically unavailable or unreliable, and instead specialized studies may be carried out. Effectiveness E ective service coverage is de ned as service coverage that results in the desired health gains. The WHO/World Bank monitoring framework has focused on integrating health service e ectiveness into monitoring tracer coverage indicators whenever possible (see Box 1.1 on monitoring challenges), but has also recognized that e ectiveness may be measured by using indicators other than coverage (2, 10). One approach takes the form of monitoring exposure to health risks, such as uncontrolled blood pressure, or health status as a proxy for e ective coverage. Many researchers have assessed health systems performance on the basis of mortality that should not have occurred if e ective care were provided (21–25). Such data re ect both health promotion and provision of e ective personal health care, but also factors outside the health system, such as environmental, social and economic in uences. In addition, high-quality data on mortality by cause of death are not available for many low- and middle-income countries. Another way to measure e ectiveness of care is to assess providers’ medical practice using medical vignettes (hypothetical cases that the provider ‘treats’) or standardized patients (actors recruited from the local community trained to present the same condition to multiple providers who are blinded from the study). For the conditions that have been studied, the standardized patient research consistently shows that less than half of patients receive what they needed for their condition, and typically less than 5% receive what they needed without additional and unnecessary medications, including antibiotics (26–27). Patient safety Patient safety is concerned with avoiding injuries to people who receive care. The OECD has identi ed two types of patient safety indicators: frequency of ‘never’ events that should never occur, such as failure to remove surgical foreign bodies at the end of a surgery; and frequency of ‘adverse’ events such as obstetric trauma, which can be reduced but not eliminated (28). Both types of indicators rely upon reporting mechanisms that are best-developed in some high-income countries. The OECD acknowledges that higher adverse event rates may simply signal more developed monitoring systems and a stronger patient safety culture, rather than worse care (28). In the absence of such reporting systems, the World Bank has recently tested a di erent approach, conducting a specialized study that observes medical practice; for instance, whether proven infection prevention and control actions are correctly carried out (29). Responsiveness/people-centredness This dimension of quality comprises patient experiences (providing care that responds to individual preferences, needs and values) and integratedness (seamless, continuous and holistic care, tailored to the patient’s needs) (19, 20). These are generally measured by interviewing patients about their health-care experiences, for example whether explanations provided by doctors were easy to understand. It is also important to note that what a patient perceives as good health care might not correspond to e ective health care (30). There is also concern that participation in patient satisfaction surveys can be biased by language and cultural barriers (31). The research from low-income countries typically shows very high levels of patient satisfaction, making the data hard to interpret (32). Selected tracer indicators of effective health promotion, screening and treatment programmes. Sixteen tracer indicators were selected, four for each of the four categories specified by the definition of SDG The service capacity and access category uses proxy indicator 3.8.1. Data availability was a major consideration indicators for the suite of coverage measures that cannot in the final list of indicators, with the expectation that currently be monitored due to data limitations (Box 1.1). substitutions will be made as new data become available. This includes important areas such as routine medical The list of tracer indicators, with information on their examinations, treatment for mental illnesses, emergency characteristics, data availability, rationale for inclusion, care and surgical procedures. The selected proxy limitations and possible refinements are provided in indicators in this category include hospital bed density, Table 1.1. the density of physicians, psychiatrists and surgeons, access to essential medicines, and compliance with the For indicators of cardiovascular disease prevention and International Health Regulations to reflect health security. diabetes management, no standardized data sets of effective coverage of cardiovascular disease and diabetes It should be noted that proxy measures like hospital bed treatment, nor treatment for elevated cardiovascular risk, density, physician density, as well as alternatives like are currently available. In the meantime, the prevalence service utilization rates, are difficult to interpret as the of normal blood pressure (including those whose blood optimal level for these indicators is unclear and they do pressure is controlled by medication) and mean fasting not relate to a specific need for services. Despite this, low plasma glucose (an indicator for diabetes) were selected levels for these indicators are indicative of poor access and as proxy measures (Table 1.1). These reflect the success use of essential health services. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 5 Table 1.1. Sixteen tracer indicators selected to monitor progress towards UHC on coverage of essential health services Countries Measurability with Primary of key primary Tracer data inequality data since Rationale, limitations and possible Tracer area indicator Type sources dimensionsa 2010 Data source re nements Reproductive, maternal, newborn and child health 1. Family Demand E ective Household W, E, R, A 112 UNPD Demand satis ed with a modern method is planning satis ed service survey estimates SDG indicator 3.7.1. It has a relatively complex with modern coverage (33) denominator derived from multiple survey method questions, and data collection often focuses on women in a union, as opposed to all sexually among active women. women 15–49 years who are married or in a union (%) 2. Pregnancy Antenatal Service Household W, E, R, A 98 WHO global Number of ANC visits captures contact with the and delivery care, four or coverage survey database health system but does not capture quality of care more visits (34) care received and may not lead to improved (ANC4) (%) mortality outcomes. Births attended by skilled health personnel (SDG indicator 3.1.2) is a preferred alternative; however, lack of standardized measurement of ‘skilled’ health personnel makes cross-country comparisons di cult. WHO/UNICEF e orts to improve comparability for reporting on SDG 3.1.2 should resolve these issues and allow 3.1.2 to replace ANC4 in the index. 3. Child One-year- Service Administrative W, E, R, S 183 WHO/UNICEF DTP3, which is identical to coverage with immunization old children coverage system, estimates pentavalent vaccine in most countries, is an who have household (35) indicator of a routine infant immunization received survey system. However, several other vaccines such as for measles (second dose), pneumococcal 3 doses of pneumonia and rotavirus diarrhoea, typically diphtheria- have lower coverage and the fraction of children tetanus- receiving all vaccines in a national schedule is pertussis typically much lower (although not possible to vaccine measure directly with existing data systems in (DTP3), (%) most countries). This indicator could be replaced with second dose of measles vaccine, following the recent recommendation of the Strategic Advisory Group of Experts on Immunization. 4. Child Care-seeking Service Household W, E, R, S 94 UNICEF Pneumonia is a leading cause of child illness treatment behaviour coverage survey global and death. Suspected pneumonia is determined for children database based on a series of survey questions about with (36) illnesses in the past two weeks, which may include mild respiratory illnesses; the indicator suspected does not currently capture the quality of care pneumonia received as a mother’s recall of treatment (%) speci cs tends to be poor. The main alternative indicator of child treatment that is widely measured is use of oral rehydration solution (ORS) therapy for child diarrhoea, which is also a leading cause of child death. The inclusion of the sanitation indicator in the index is relevant for diarrhoea prevention. 6 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Countries Measurability with Primary of key primary Tracer data inequality data since Rationale, limitations and possible Tracer area indicator Type sources dimensionsa 2010 Data source re nements Infectious diseases 1. TB e ective E ective Administrative (R) 179 WHO This indicator combines two more common Tuberculosis treatment service system, estimates ones – the case-detection rate and the treatment treatment coverage coverage household (37) success rate – to estimate the proportion of TB (%) survey cases that are detected and successfully treated. Calculation of the case-detection rate requires estimates of incident cases (including those not detected by the health system). Treatment- success rate is measured through administrative data, and includes all patients who successfully complete treatment without bacteriological evidence of failure. 2. HIV People living Service Administrative (R), (S), (A) 136 UNAIDS Provision of ART averts a substantial number of treatment with HIV coverage system, estimates deaths in high-burden HIV countries, and can be receiving household (38) a marker of how well a health system reaches ART (%) survey, marginalized populations with higher prevalence in lower-burden countries. Recent surveys have surveillance started measuring e ective coverage of ART system by collecting data on viral load suppression. The numerator – people on ART – is generally obtained from health facility data, while the denominator is often estimated from household surveys, sentinel surveillance sites and facility data. 3. Malaria Population Service Administrative W, E, R, S 29b WHO/ There are major ITN distribution programmes in prevention at risk coverage system, Malaria malaria-endemic countries. Coverage estimates sleeping household Atlas Project should account for geographical heterogeneity in under survey estimates malaria risk when analysing national household surveys. Due to net deterioration, e ective insecticide- (39)b coverage rates can decline without resupply. treated bednets (%) 4. Water and Households Service Household W, R 176 WHO/UNICEF While not always implemented by the health sanitation with access coverage survey estimates sector, access to clean water and safely to at least (40) managed sanitation are important public health basic interventions. The current indicator of at least basic sanitation is typically much lower than sanitation access to at least a basic water source, and (%) therefore is used as the tracer indicator for this area. This tracer indicator could be replaced with SDG 6.1.1 or 6.2.1, once they are more widely reported. Noncommunicable diseases 1. Prevention Prevalence Proxy Household (E), (R), S, A 85 NCD-RisC/ Hypertension is the leading risk factor for CVD. of of normal survey WHO The prevalence of normal blood pressure is the cardiovascular blood estimates sum of the percentage of individuals who do disease pressure, (41) not have hypertension, and the percentage of individuals whose hypertension is controlled regardless by medication. The absence of hypertension of treatment is a result of prevention e orts via promotion status (%)c of physical activity and healthy diets, as well as other factors. Hypertension controlled with medication is a result of e ective treatment. This indicator is thus a proxy for both e ective health promotion and e ective medical services. This indicator will be replaced with a measure of treatment coverage among people with hypertension, once the data become available. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 7 Countries Measurability with Primary of key primary Tracer data inequality data since Rationale, limitations and possible Tracer area indicator Type sources dimensionsa 2010 Data source re nements 2. Mean fasting Proxy Household (E), (R), S, A 6d WHO An individual’s FPG may be low because of Management plasma survey estimates e ective treatment with glucose-lowering of diabetes glucose (42) medication, or because the individual is not (FPG), diabetic as a result of health promotion activities or other factors such as genetics. Mean FPG (mmol/L)c is thus a proxy for both e ective promotion of healthy diets and behaviours and e ective treatment of diabetes. However, diabetes treatment guidelines do not recommend lowering blood glucose to non-diabetic levels for all patients, meaning that a population with a large prevalence of diabetes should not necessarily attain a low mean FPG. This indicator will be replaced with the proportion of people with diabetes receiving treatment once data become available. 3. Cancer Cervical Service Household — <30 Insu cient Data on this indicator are collected in some detection cancer coverage survey data household surveys, although not yet widely and screening currently enough to be used for global monitoring. The treatment among available indicator does not re ect whether e ective treatment is available. This indicator was chosen women aged over other potential cancer screening indicators, 30–49 years such as for breast or prostate cancer, because of (%) clearer guidelines for the former, and because cervical cancer screening is the only one included in the core indicator set of the NCD Global monitoring framework. 4. Tobacco Adults aged Proxy Household (W), (E), (R), 125 WHO Prevalence of smoking (SDG indicator 3.a.1) is a control ≥15 years survey S, (A) estimates proxy for adoption and enforcement of a suite of not smoking (43) e ective anti-tobacco measures. This indicator tobacco in could be replaced with a measure of e ective implementation of tobacco control policies. last 30 days (%)c Service capacity and access 1. Hospital Hospital Proxy Facility data (R) 158 WHO global This indicator is a proxy for coverage of the full access beds per database range of essential inpatient services. It has higher capita (w/ (44) data availability in low- and middle-income threshold) countries than inpatient admission rates, with which it is highly correlated (rho=0.84 in low- and middle-income countries). A threshold is used to capture low capacity levels; very high values are not necessarily desirable. Inpatient service utilization rates, subject to a threshold, could be used in place of hospital beds as more data become available. 2. Health Health Proxy Administrative (R) 180 WHO global Comparable data on outpatient utilization rates worker professionals system database are not currently available across low- and density per capita (45) middle-income countries. Due to this, physician (w/ density, part of SDG indicator 3.c.1, is included as a proxy for coverage of the full range of threshold): essential outpatient services not captured by physicians, tracer indicators included elsewhere in the index. psychiatrists Nurses and midwives are currently excluded due and to lack of comparable data across countries in surgeons existing global databases. Nurses and midwives could be included once comparable data become available. Psychiatrist and surgeon density are proxies for coverage of mental health and surgical and emergency care respectively. As with hospital beds per capita, a threshold is used to capture low densities for all three cadres. 8 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Countries Measurability with Primary of key primary Tracer data inequality data since Rationale, limitations and possible Tracer area indicator Type sources dimensionsa 2010 Data source re nements 3. Access to Proportion Proxy Facility (R) <30 Insu cient Medicines are the main intervention resulting essential of health survey data from clinical services, and their availability is medicines facilities currently a proxy for access to needed medications. This with WHO- available tracer will be included once data become widely available. recommended core list of essential medicines available 4. Health International Proxy Key — 181 WHO Since many health risks are rare, preparedness security Health informant database measures must be tracked to capture health Regulations (46) security as part of UHC. This indicator – SDG core capacity 3.d.1 – is based on key-informant reports to WHO, but could be informed by Joint External index Evaluations in the future. This indicator measures country capacity for early warning, risk reduction and management of national and global health risks, and serves as a proxy for the e ectiveness of those capacities. ART: antiretroviral therapy; CVD: cardiovascular disease; HIV: human immunode ciency virus; ITN: insecticide-treated nets; NCD: noncommunicable disease; SDG: sustainable development goal; TB: tuberculosis; UHC: universal health coverage; UNAIDS: Joint United Nations Programme on HIV/AIDS; UNPD: United Nations Population Division; UNICEF: United Nations Children’s Fund; WHO: World Health Organization. a W = household wealth quintile; E = educational attainment; R = place of residence (typically urban vs. rural); S = sex; and A = age. Letters in parentheses indicate that data sources exist to estimate coverage by the indicated dimension but that more analytical work is needed to prepare disaggregated estimates. b Only pertains to countries with highly endemic malaria. c Age-standardized. d Data availability for 178 countries is based on the 2011 analysis used to calculate the index (41). This analysis used predominantly older data, but included one data source collected in 2010. During the country consultation process, ve countries submitted recent data on mean FPG. Estimates of mean FPG have not been updated as the aim is to move toward a true coverage indicator as explained above. The NCD-RisC collaboration has estimated that recent (since 2010) national or subnational household survey data, including a measure of diabetes, are available for 87 countries or territories. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 9 Calculating the index antiretroviral treatment. However, there were several exceptions requiring further manipulation of the data, Indicators of cervical cancer screening coverage and which are explained in Fig. 1.1. The index is constructed access to essential medicines are currently excluded from geometric means of the tracer indicators; first, from the index calculations due to low data availability. within each of the four categories, and then across the Service coverage is typically measured on a scale of 0 to four category-specific means to obtain the final summary 100%, with 100% as the target, and therefore the UHC index (Fig. 1.1). Geometric means are used instead of service coverage index is presented on a scale of 0 to arithmetic means as they favour equal coverage levels 100. Most of the tracer indicators can be incorporated across services as opposed to higher coverage for some directly into the index on their natural scale, for example services at the expense of others. the percentage of people living with HIV who are receiving Fig. 1.1. Calculating the UHC service coverage index Reproductive, maternal, newborn and child health 1. Family planning (FP) 2. Antenatal care 4+ visits (ANC) RMNCH = (FP • ANC • DTP3 • Pneumonia) 1/4 3. Child immunization (DTP3) 4. Care seeking suspected pneumonia (Pneumonia) Infectious disease control 1. TB e ective treatment (TB) Infectious = (ART • TB • WASH • ITN) 1/4 if high risk malaria 2. HIV treatment (ART) 3. Insecticide-treated nets (ITN) Infectious = (ART • TB • WASH) 1/3 if low risk malaria 4. At least basic sanitation (WASH) Noncommunicable diseases 1. Normal blood pressure (BP)a 2. Mean fasting plasma glucose (FPG)b NCD = (BP • FPG • Tobacco) 1/3 3. Cervical cancer screeningc 4. Tobacco non-smoking (Tobacco)d Service capacity and access 1. Hospital bed density (Hospital)e 2. Health worker density (HWD)f Capacity = (Hospital • HWD • IHR) 1/3 3. Access to essential medicinesc 4. IHR core capacity index (IHR) UHC service coverage index = (RMNCH • Infectious • NCD • Capacity) 1/4 IHR: International Health Regulations; NCD: noncommunicable diseases; RMNCH: reproductive, maternal, newborn and child health; UHC: universal health coverage. a The percentage of the adult population with normal blood pressure is based on age-standardized estimates. These distributions are rescaled to provide ner resolution for the index, based on the observed minima across countries. The rescaled indicator = (X–50)/(100–50)*100, where X is the prevalence of normal blood pressure. b Mean fasting plasma glucose (FPG) is not measured on a scale bounded between 0 and 100%. While very high levels are unhealthy, very low levels are not expected to provide additional health bene ts or could even be harmful. To account for this range, while also providing a well-distributed range of indicator values across countries, from 0 to 100 after rescaling, estimates of national mean FPG were rescaled using a minimum of 5.1 mmol/L (the midpoint of minimum theoretical risk) and a maximum of 7.1 mmol/L (the maximum across national means). The rescaled indicator for mean FPG = (7.1–X)/(7.1–5.1), where X is mean FPG. c Cervical cancer screening and access to essential medicines are excluded due to low data availability. d As in (a), tobacco non-smoking is also based on age-standardized estimates, and is rescaled to provide ner resolution based on a minimum bound of 50%, so that the rescaled indicator = (X–50)/(100–50)*100, where X is prevalence of tobacco non-smoking. e Hospital bed density values were rescaled and capped based on a threshold of 18 per 10 000, based on minimum rates observed in high income OECD countries. Values below 18 per 10 000 are rescaled as X/18*100, where X is hospital beds per 10 000, and values above 18 per 10 000 are set to 100. f As in (e), health worker density (HWD) is rescaled and capped based on threshold values. Physician density has a threshold of 0.9 per 1000, psychiatrists have a threshold of 1 per 100 000, and surgeons have a threshold of 14 per 100 000. After rescaling these values (i.e., minimum (100, X/threshold*100), where X is the cadre-speci c density, they are combined into a HWD composite variable for entry into the above index calculations, computed as (physicians * psychiatrists * surgeons).1/3 10 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Data sources the coverage index. However, it should be noted that no country reports values for all tracer indicators in Common primary data sources used for indicators of every year. Simply excluding an indicator without data to service coverage include surveys, facility data and other compute the index creates expected bias as some services administrative data (Table 1.1). Nationally representative, tend to have higher coverage than others. The alternative population-based surveys are often the best source as is to use some form of imputation to fill these data gaps. they can enable the measurement of those who need an Most UN estimates of tracer indicators use statistical intervention, in addition to counting those who already or mathematical models to combine different data receive it, and allow for the disaggregation of service sources and fill data gaps to produce annual values for coverage by different subpopulations for equity analysis. each country. In cases where UN estimates were not The use of facility data or other administrative sources available, the most recent value from 2000 to 2015 for presents challenges as they may capture the number each country’s indicators was used to compute the index. of people receiving a service (the numerator) but fail to In cases where no country value was available from that count all those who need a service (the denominator), time period, a regional median from countries with data and typically do not collect variables relevant for equity was computed and used as the country value. More analyses other than geographical location. They may also details are available in Annex 2. be subject to reporting incentives. However, an advantage of administrative data sources is that they are often reported annually through routine systems, and therefore provide more timely data than household surveys, which First ndings on SDG indicator 3.8.1 are typically conducted every three to five years. Data availability on tracer indicators varied from country to country but was fairly similar across regions (Fig. 1.2) UN agencies lead substantial measurement and reporting and generally high, with countries having recent primary efforts for many of the selected tracer indicators, which data for 72% of tracer indicators on average. This figure feed into SDG reporting processes where relevant. reflects only primary data, not estimates computed to fill Therefore, priority was given to official UN estimates in data gaps. for the year 2015 to compute SDG baseline values for Fig. 1.2. Percentage of tracer indicators with primary data source available since 2010, by country Percentage (%) This map has been produced by WHO. The boundaries, colours 75–100 or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank 50–74 Data not available or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any 0–49 Not applicable endorsement or acceptance of such boundaries or frontiers. 0 850 1,700 3,400 Kilometers CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 11 Fig. 1.3. UHC service coverage index by country, 2015, for monitoring SDG indicator 3.8.1 Index value by quintile ≥77 70–76 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and 62–69 the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 46–61 Not applicable any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0 850 1,700 3,400 Kilometers ≤45 Data not available SDG: Sustainable Development Goal; UHC: universal health coverage. Current values for the UHC service coverage index increase in life expectancy. Over the range of observed ranged from 22 to 86 across 183 countries, with country values (22 to 86), this translates into a difference a median value of 65 (Fig. 1.3). The service coverage of 21 years in life expectancy. index is highly correlated with other measures of health and development, for example, under-5 mortality rates The service coverage index is constructed from subindices (ρ=-0.86), life expectancy (ρ=0.88) and the Human representing the four categories of RMNCH, infectious Development Index (ρ=0.91), and modestly correlated diseases, NCDs, and service capacity and access. Table with gross national income (GNI) per capita (ρ=0.65). 1.2 depicts these subindices, along with the full service High-income countries tend to have high values on the coverage index, across modified SDG regions weighted index, while the lowest values are seen among low- by population size. The UHC service coverage index is income countries and some countries affected by conflict highest in Europe and Northern America (77) and the (see Annex 2 for UHC service coverage index and tracer Eastern Asia region (77), while sub-Saharan Africa (42) indicator values by country). and Southern Asia (53) have the lowest average values. The strongest gradient across regions is for the service The UHC service coverage index is more predictive of life capacity and access subindex; the mean value for sub- expectancy than the GNI, and remains predictive of life Saharan Africa is only 27 compared with 99 in Eastern expectancy after controlling for GNI and mean years of Asia. The NCD subindex is fairly evenly distributed across adult education. For example, a regression of national life regions and less correlated with other categories. This is expectancy on the service coverage index, the log of GNI largely because tobacco use is low in some areas with per capita and mean years of adult education, indicates weaker health systems, such as sub-Saharan Africa and that going from 0 to 100 on the index is associated with Southern Asia, and high in Europe. a 32-year (95% confidence interval, CI: 25-39 years) 12 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Table 1.2. Regional (population-weighted) means for the UHC service coverage index and its four component subindices Area UHC service RMNCH Infectious NCDs Service capacity coverage index diseases and access Global 64 75 54 63 71 Africa 46 55 40 67 37 Northern Africa 64 73 50 62 77 Sub-Saharan Africa 42 51 37 69 27 Asia 64 75 51 63 71 Eastern Asia 77 86 64 64 99 Southern Asia 53 66 41 64 47 South-Eastern Asia 59 78 45 59 63 Central Asia 70 81 56 58 93 Western Asia 65 69 59 57 79 Europe and Northern America 77 88 73 58 96 Latin America and the Caribbean 75 81 65 68 88 Oceania 74 83 71 62 84 NCDs: noncommunicable diseases; RMNCH: reproductive, maternal, newborn and child health; UHC: universal health coverage. Small differences in country rankings are not meaningful, Gaps in health service coverage as many country values are close together and there is uncertainty in the measurement of tracer indicators, To communicate the magnitude of the task ahead to particularly for countries with low data availability increase health service coverage to improve health (Fig. 1.2), and in methods used to calculate the index outcomes and achieve the health-related SDGs, perhaps (9). Currently, the index does not adequately distinguish no single statistic is more in demand than the number of between countries with the highest level of service people with coverage of essential health services. These coverage provision. Therefore, country index values numbers are challenging to compute precisely (Box 1.3), of 80 and over are reported as ‘≥80’ for presentation and can be supplemented by considering the gaps in purposes, to avoid comparisons that are not meaningful selected essential health services. (see Annex 1 for country values). This should not be interpreted as a target. This is illustrated in Fig. 1.4, which shows the number of people with unmet need for each of nine UHC service coverage index tracer indicators that are measured on a percentage scale and have global estimates of unmet need available. Across selected individual indicators, unmet need ranges from 2.3 billion for at least basic household sanitation to 5 million for effective tuberculosis treatment. This set of nine indicators does not reflect the total unmet need for health services, but it provides evidence that very large gaps in coverage persist. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 13 Box 1.3. How many people are covered with essential health services? Fully answering this question is challenging because there is no dataset that contains information on all people’s needs for health services and whether they received those services. The rst UHC Global Monitoring Report stated that, in 2013, over 400 million people were not receiving at least one of seven essential health services that they needed. These represented Millennium Development Goal priority areas of family planning, antenatal care, births attended by skilled health personnel, DTP3 immunization, HIV treatment, TB treatment and ITN use among children. This calculation did not encompass the broad range of essential health services that individuals should receive. Here, we estimate the number of people who are covered with the broad and representative range of health services that they should receive in any country. A simple algorithm is used to estimate the number of people who have full coverage with essential health services. a First, a set of tracer coverage indicators are selected, based on those in the UHC service coverage index. These tracer indicators track, but do not de ne, a full package of essential health services, ranging across health areas (such as reproductive health and noncommunicable diseases) and service delivery platforms (such as community services, primary care and specialized services). Second, average coverage of these tracer indicators is calculated for each country. This number represents the average chance that an individual who needs an essential health service will receive it. It does not represent the percentage of people who are covered with all needed services, because any given individual may be covered with some services, but not others. The analysis of co-coverage of basic services in mother-child pairs (see the following section on inequalities in maternal and child health service coverage), however, o ers a way to estimate the relationship between average coverage of services and the proportion of people with full coverage with essential health services. A regression equation tted to these data is used to convert average coverage of essential services in each country to the proportion of people that are expected to have full coverage with essential services. To set a realistic goal for full coverage, this is operationalized as the percentage of mother-child pairs who receive at least six out of seven basic services. There is substantial uncertainty around this approach. Given a set of plausible sensitivity analyses,a the number of people with full coverage with essential services ranged from 2.3 to 3.5 billion in 2015. This implies that at least half of the world’s 7.3 billion people do not receive the essential health services they need. a For methodological details, please see technical note online: http://www.who.int/healthinfo/universal_health_coverage/report/2017/en/ Fig. 1.4. Number of people in need but not receiving a selected essential health servicea Sanitation (at least basic) Hypertension control Tobacco control Insecticide-treated nets (use) Family planning Antenatal care 4+ visits Immunization (DTP3) HIV treatment TB e ective treatment 1 10 100 1 000 10 000 Millions of people (log scale) DTP3: third dose of diphtheria-tetanus-pertussis containing vaccine; HIV: human immunode ciency virus; TB: tuberculosis. a 2016 estimates: TB e ective treatment, HIV treatment, Immunization (DTP3), Family planning, Insecticide-treated nets (use); 2015 estimates: Tobacco control, Hypertension control, Sanitation (at least basic); 2013 estimates: Antenatal care 4+ visits. 14 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Time trends in service coverage global public health in recent years. Progress in DTP3 Country and global trends are available for the same nine immunization coverage between 2000 and 2010 has tracer indicators in the UHC service coverage index that slowed in more recent years. Although normal blood are measured on a percentage scale and already discussed pressure indicators have improved slightly at the global (see ‘Gaps in health service coverage’ and Fig. 1.5). The level, this obscures diverse underlying trends at the average coverage of these indicators increased by 1.3% regional level. High-income countries have experienced per annum, which is roughly a 20% relative increase from substantial improvements, while the prevalence of normal 2000 to 2015. blood pressure has remained stable or even deteriorated in Eastern, Southern and South-Eastern Asia, Oceania The largest relative increases are seen in indicators (excluding Australia and New Zealand), and sub-Saharan for HIV, TB and malaria services, arguably reflecting Africa (41). resource allocation priorities that have dominated Fig. 1.5. Trends in global coverage of selected health service tracer indicators, 2000–2016 100 90 Immunization (DTP3) 80 Normal blood pressure Family planning Tobacco 70 non-smoking Sanitation (at least basic) 60 Coverage (%) 50 Antenatal care 4+ visits 40 TB e ective treatment 30 HIV treatment Insecticide-treated nets (use) 20 10 0 2000 2005 2010 2015 DTP3: diphtheria-tetanus-pertussis containing vaccine (third dose); HIV: human immunode ciency virus; TB: tuberculosis. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 15 Inequalities in maternal and child birth attendance, Bacillus Calmette–Guérin vaccination, health services in low- and lower- the third dose of a diphtheria-tetanus-pertussis containing middle-income countries vaccine, measles vaccination and access to improved drinking water in the household. All seven indicators With complete data, the UHC service coverage index could were calculated for children aged 12–59 months, using be computed and compared across different dimensions information available from their mothers’ most recent of inequality, for example across wealth and education pregnancy where relevant (for instance for ANC). This gradients, different geographical regions within a country, analysis shows what proportion and number of these and age and sex. This is currently not possible for all tracer basic services each mother-child pair received, and can be indicators due to data limitations; however, a subset of summarized across key dimensions of inequality. indicators can be used to illustrate variation in inequality across countries (9). The most readily available information In low- and lower-middle-income countries, large gaps on inequalities in health service coverage are for RMNCH in maternal and child health services persist and are indicators measured through household surveys. As these not evenly distributed across population groups (Fig. 1.6 indicators are measured at the individual level in a single and Fig. 1.7).1 While 39% of mother-child pairs in these survey, it is possible to assess the fraction of needed countries received at least six of seven basic interventions, services that each person receives. This measurement 4% of mother-child pairs received no interventions at approach is often referred to as ‘co-coverage’ (47). all. When the data are stratified by wealth quintile, significant inequalities emerge: overall, only 17% of those To assess levels and trends in inequalities in maternal in households in the poorest wealth quintile in their and child service coverage indicators, co-coverage of countries received at least six basic interventions, versus seven basic health services, collected in Demographic and Health Surveys (DHS) carried out in low- and lower- 1 In this paragraph and Fig. 1.6 and Fig. 1.7, all analyses were carried out using the most middle-income countries, were considered. The seven recent survey in each country during the time period 2005-2015. Data were available for 48 countries, covering 90% of 2010 live births in lower-middle and low-income countries; the services were: four or more antenatal care (ANC) visits, median survey year was 2012. To create estimates for all low- and lower-middle-income at least one tetanus vaccination during pregnancy, skilled countries, country data were weighted by the number of live births in 2010. Fig. 1.6. Mother-child pairs in low- and lower-middle-income countries, by number of basic interventions received out of seven, 2005–2015 50 Q5 (richest) 45 40 35 30 Mother-child pairs (%) 25 20 All 15 10 Q1 (poorest) 5 0 0 1 2 3 4 5 6 7 Number of interventions 16 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Fig. 1.7. Mean number of basic interventions that mother-child pairs receive out of seven, overall and by inequality dimensions, low- and lower-middle-income countries, 2005–2015 Q5 (richest) Q4 Wealth quintile Q3 Q2 Q1 (poorest) secondary Mother's education primary none rural Area urban Overall 0 1 2 3 4 5 6 7 Mean number of basic interventions 74% in the richest quintile. Those in the poorest wealth and 2008–2015. Data were available from 23 low- and quintile in each country were most likely to receive no lower-middle-income countries for all three periods, interventions, with 9% receiving none of them. The mean representing approximately 38% of live births in these number of interventions received ranged from three in regions; therefore, summary statistics should not be the poorest wealth quintile to six in the wealthiest, with considered representative. an overall average of five out of seven. Mother-child pairs living in rural areas had lower coverage than those living Considering large gaps in coverage, the median in urban areas. percentage of mother-child pairs that received three or fewer basic health services declined between 1993-1999 and 2008-2015 across all wealth quintiles among 23 low- Trends in maternal and child health and lower-middle-income countries with available data (Fig. 1.8). Absolute reductions were larger in poorer wealth service coverage inequalities over quintiles, and therefore absolute inequalities in missed time health services were reduced over this time period. The median percentage of mother-child pairs receiving three Unless health interventions are designed to promote or fewer of the basic services declined by 25 percentage equity, efforts to attain UHC may have the unintended points among the poorest wealth quintiles, falling from consequence of bringing early and accelerated gains for 56% to 24% across the time period. However, mother- the most-advantaged section of society, and at the same child pairs in the poorest wealth quintile were still far time leaving the most disadvantaged behind. As a result, more likely to experience a large gap in service coverage the national average of service coverage may improve, but than mother-child pairs in the wealthiest quintile: the inequalities may worsen at the same time (48). In order to median proportion of mother-child pairs receiving three assess time trends in inequalities in service coverage in or fewer basic interventions was only 3% in the wealthiest low- and lower-middle-income countries, survey data were quintile. This reinforces the importance of structuring subdivided into three periods: 1993–1999, 2000–2007 health services so that no one is left behind. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 17 Fig. 1.8. Percentage of mother-child pairs covered with three or fewer basic health services out of seven by within-country wealth quintilea 60 50 40 Mother-child pairs (%) 30 Q1 (poorest) 20 Q2 Q3 10 Q4 Q5 (wealthiest) 0 1993–1999 2000–2007 2008–2015 a Median value from national surveys carried out in 23 low- and lower-middle-income countries in each time period. Next steps for an index of essential example updating indicators to match SDG indicator definitions as they become available, are described in health services Table 1.1. Beyond SDG alignment, future work will likely include the following: For the first time, an index of essential health services that is consistent with the definition of SDG indicator Replacing proxy indicators for NCD service coverage. 3.8.1 has been operationalized (9). It must be stressed that Global databases on treatment coverage for the selected tracer indicators are a subset of measurable hypertension and diabetes should be completed in indicators, and not a recommended basket of services. 2018, and these indicators can be used in lieu of current Global, regional and national UHC monitoring frameworks proxy indicators. should continue to develop and report relevant indicators of service coverage (Box 1.4). These efforts could be used Investigating the feasibility of using inpatient and to adapt the index, or provide a fuller picture by reporting outpatient service utilization rates as indicators of on individual indicators. The desire for information to service capacity, access and use. monitor UHC must be balanced against the costs of collecting it. Streamlining household health surveys so Increasing relevance to higher income countries. Many that they cover a wide range of health areas may be more high-income countries are approaching 100% coverage efficient than conducting separate surveys on specific for tracer indicators in the RMNCH and service capacity health topics. This also makes it more straightforward to and access categories. Other tracer indicators, or a characterize time trends and disaggregate a broad set of hybrid method that incorporates avoidable mortality, service coverage indicators for equity analysis (Box 1.5). should be assessed. The current UHC service coverage index has several Expanding the set of tracer indicators with equity limitations, not all of which can be currently addressed information to allow fuller disaggregation of the index. due to data constraints. Some expected changes, for 18 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Box 1.4. WHO Western Pacific Region adaptation of the UHC service coverage index The WHO Regional O ce for the Western Paci c (WPRO) has developed a framework for monitoring progress towards the SDGs and UHC (49–50). It includes a total of 88 indicators under three groups: 27 indicators fall under SDG 3 (Health), 20 are from other SDGs, and 41 are additional indicators of progress towards UHC. The following criteria were used to determine a list of ‘ t for purpose’ indicators. 1. Focus on common health issues and indicators across the Western Paci c Region to allow within-country and cross-country comparisons, mutual learning and sharing of experience. 2. Align the regional-level indicators with existing global collections where possible, to encourage information/data exchange between Member States in the Region. 3. Ensure that, in addition to tracking progress in SDGs and UHC, the indicators can be used to review progress and to support policy and programme development at multiple levels (national, subnational, local) and for di erent population groups. 4. Ensure where possible that information to track progress towards SDGs and UHC is disaggregated by sex, age, socioeconomic status, education, ethnicity and place of residence. 5. Ensure the indicators are theoretically sound and commonly understood. 6. Ensure that the indicators re ect a balance in the selection of targets, not overemphasizing one health condition, but capturing characteristics that re ect the health pro le of country populations. Following this framework, WPRO is in the process of producing a report describing the Region’s current SDG and UHC baseline situation and presenting the results of analyses that countries may consider when incorporating SDG and UHC monitoring into policy and decision-making. The aim is for countries to use this report as a benchmark not only to support their own monitoring e orts and activities, but to assist in the formulation of policies, programmes and practices targeting health system development to reaching UHC. Box 1.5. Tunisia – UHC tailored surveys The Tunisian Health Examination Survey (THES) is speci cally designed to collect data to monitor progress towards UHC using a set of standard modules. The Tunisian Ministry of Health and National Institute of Public Health, in collaboration with the National O ce of Family and Population and the Research Laboratory in Epidemiology and Prevention of Cardiovascular Diseases, designed and executed this survey to generate information to support evidence- informed health policy and strategy development. It is in recognition of a lack of more complete and recent high-quality information about the health of people living in Tunisia, the way they use health services and their household health spending, that the survey was carried out, to inform the national government’s health policy. The survey collected data on adults’ self-reported health status and determinants including risk factors such as alcohol and tobacco use; diagnosed chronic diseases such as heart disease, diabetes and depression and their treatment; health care utilization; responsiveness of health-care services; health insurance coverage; household expenditure including out-of-pocket spending on health (total and disaggregated for di erent types of health services and goods); contributions to insurance schemes and reimbursements; and sources of nance to pay for health (for instance current income, selling assets or borrowing). Women of reproductive age were interviewed about their reproductive health and their children’s immunization coverage. In addition, a suite of biological measurements were taken, including blood glucose, glycosylated haemoglobin (HbA1c), total cholesterol, height, weight, blood pressure and visual acuity. Results from this survey reveal that while 97.2% of children received full vaccination coverage and 86% of mothers had antenatal care, only 42.7% of those with diabetes and only 5.5% of those with depression received treatment in the past two weeks. Among eligible women only 10.8% had been screened for cervical cancer with a Pap smear and only 8.1% had mammography screening for breast cancer. Additionally, of the 26.4% of the population that were hypertensive, 37.9% were aware of their hypertension, 30.7% were on current treatment and only 9.7% were adequately controlled. There were also signi cant di erences between socioeconomic groups and governorates. Awareness and treatment of diabetes, for example, were highest in the Central- East (60.4% both aware and on treatment) and lowest in the North-West (45.2% both aware and on treatment). Regarding nancial protection against catastrophic health expenditures (see Chapter 2), the survey also revealed that 9.1% of the total population incurred catastrophic out-of-pocket spending on health (de ned as out-of-pocket health expenditures exceeding 25% of total household expenditure). Looking at the incidence rate by area of residence, rural populations were slightly more a ected by catastrophic health expenditures (9.9%) compared with urban populations (8.8%). Twelve per cent of respondents also stated that they did not receive health care when they last needed it. While over 40% of these respondents did not seek treatment because they felt they were not sick enough, 16.9% – especially those from the poorer segments of the population – said this was because they could not a ord the cost of the visit and 5% said they could not a ord the cost of the transport. The poor more often reported a ordability as a reason for not accessing care. The THES provides an illustration of the measurement of key markers of progress towards Universal Health Coverage. While immunization and antenatal care coverage are high, the coverage for chronic conditions and cancer screening will need considerable policy interventions and targeted e orts to strengthen programmes to increase public awareness, detection and appropriate management in primary care settings. Additionally, e orts will need to be made to reduce socioeconomic inequalities in coverage of key health interventions and to reduce gaps in the quality of care, costs and work conditions between the public and private health sectors. Finally, a signi cant number of households are not protected from nancial risk due to out-of-pocket spending on health, including the most vulnerable who are choosing to forgo treatment due to una ordable costs. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 19 References 16. Ng M, Fullman N, Dieleman JL, Flaxman AD, Murray CJ, Lim SS. Effective coverage: a metric for monitoring Universal Health Coverage. Plos Med. 2014;11(9):e1001730. 1. The World Bank, World Health Organization. Tracking universal 17. Ramke J, Gilbert CE, Lee AC, Ackland P, Limburg H, Foster A. health coverage: first global monitoring report. Geneva: World Effective cataract surgical coverage: an indicator for measuring Health Organization; 2015. quality-of-care in the context of Universal Health Coverage. PloS One. 2017;12(3):e0172342. 2. Boerma T, AbouZahr C, Evans D, Evans T. Monitoring intervention coverage in the context of universal health coverage. PLoS Med. 18. To err is human: building a safer health system. Institute of 2014;11(9):e1001728. Medicine; Kohn LT, Corrigan JM, Donaldson MS, editors. Washington DC: The National Academies Press; 2000. p. 312. 3. Wehrmeister FC, Restrepo-Mendez MC, Franca GV, Victora CG, Barros AJ. Summary indices for monitoring universal coverage 19. Carinci F, Van Gool K, Mainz J, Veillard J, Pichora EC, Januel JM in maternal and child health care. Bull World Health Organ. et al. Towards actionable international comparisons of health 2016;94(12):903-12. system performance: expert revision of the OECD framework and quality indicators. Int J Qual Health Care. 2015;27(2):137- 4. Collaborators GS. Measuring the health-related Sustainable 46. Development Goals in 188 countries: a baseline analysis from the Global Burden of Disease Study 2015. Lancet. 20. Improving value in health care. Paris: Organisation for Economic 2016;388(10053):1813-50. Co-operation and Development; 2010. 5. Sachs J, Schmidt-Traub G, Kroll C, Durand-Delacre D, Teksoz 21. Rutstein DD, Berenberg W, Chalmers TC, Child 3rd CG, Fishman K. An SDG index and dashboards - global report. New York: AP, Perrin EB et al. Measuring the quality of medical care: a Bertelsmann Stiftung and Sustainable Development Solutions clinical method. New England Journal of Medicine. 1976 Mar Network; 2016. 11;294(11):582-8. 6. Leegwater A, Wong W, Avila C. A concise, health service 22. Health at a Glance 2016. Paris: Organisation for Economic Co- coverage index for monitoring progress towards universal health operation and Development; 2016. coverage. BMC Health Serv Res. 2015;15:230. 23. Nolte E, McKee M. Does healthcare save lives? Avoidable 7. Wagstaff A, Cotlear D, Eozenou PHV, Buisman LR. Measuring mortality revisited. London, UK: Nuffield Trust; 2004. progress towards universal health coverage: with an application to 24 developing countries. Oxford Rev Econ Pol. 2016;32(1):147- 24. Mathers CD, Stevens GA, Boerma T, White RA, Tobias MI. 89. Causes of international increases in older age life expectancy. Lancet. 2015 Feb 13;385(9967):540-8. 8. Wagstaff A, Dmytraczenko T, Almeida G, Buisman L, Hoang- Vu Eozenou P et al. Assessing Latin America’s progress 25. Barber RM, Fullman N, Sorensen RJ, Bollyky T, McKee M, toward achieving universal health coverage. Health Affairs. Nolte E et al. Healthcare Access and Quality Index based on 2015;34(10):1704-12. mortality from causes amenable to personal health care in 195 countries and territories, 1990–2015: a novel analysis from 9. Hogan DR, Stevens GA, Hosseinpoor AR, Boerma T. An index the Global Burden of Disease Study 2015. Lancet. 2017 May of the coverage of essential health services for monitoring 18;390(10091):231-266. Universal Health Coverage within the Sustainable Development Goals. Lancet Global Health. 2017. DOI: http://dx.doi. 26. Das J, Chowdhury A, Hussam R, Banerjee AV. 2016.The impact org/10.1016/S2214-109X(17)30472-2. of training informal health care providers in India: A randomized controlled trial. Science. 2016 Oct 7;354(308). 10. Boerma T, Eozenou P, Evans D, Evans T, Kieny MP, Wagstaff A. Monitoring progress towards universal health coverage at 27. Currie J, Lin W, Meng J. Addressing antibiotic abuse in country and global levels. PLoS Med. 2014;11(9):e1001731. China: an experimental audit study. Journal of Development Economics. 2014 Sept;110:39-51.(https://doi.org/10.1016/j. 11. The World Bank, World Health Organization. Monitoring jdeveco.2014.05.006). progress towards universal health coverage at country and global levels: a framework [joint discussion paper]. Washington 28. Health at a glance 2017. Paris: Organisation for Economic Co- DC, Geneva: The World Bank, World Health Organization; 2013. operation and Development; 2017. 12. The World Bank, World Health Organization. Monitoring 29. Bedoya G, Dolinger A, Rogo K, Mwaura N, Wafula F, Coarasa J progress towards universal health coverage at country and et al. Observations of infection prevention and control practices global levels: framework, measures and targets. Geneva: World in primary health care, Kenya. Bull World Health Organ. 2017 Health Organization; 2014. Jul 1;95(7):503-516. 13. Measurement of trends and equity in coverage of health 30. Zgierska A, Rabago D, Miller MM. Impact of patient satisfaction interventions in the context of universal health coverage: report ratings on physicians and clinical care. Patient Prefer Adherence. of meeting held at the Rockefeller Foundation Center, Bellagio, 2014 Apr 3;8:437–46. September 17-21, 2012. Geneva: World Health Organization; 2012. 31. Gayet-Ageron A, Agoritsas T, Schiesari L, Kolly V, Perneger TV. Barriers to participation in a patient satisfaction survey: who 14. Monitoring universal health coverage: summary report of a are we missing? PLoS One. 2011 Oct 26;6(10):e26852. (http:/ / meeting in Bellagio, 11-13 March 2014. Geneva: World Health www.ncbi.nlm.nih.gov/pmc/articles/PMC3202588/). Organization; 2014. 32. Evans DK, Welander Tärneberg A. Health-care quality and 15. Monitoring universal health coverage: report of a meeting at information failure: evidence from Nigeria. Health Economics. the Rockefeller Center, Bellagio, 16-19 November 2015. Geneva: 2017;1–4. (https://doi.org/10.1002/hec.3611). World Health Organization; 2015. 20 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT 33. Model-based estimates and projections of family planning 43. WHO report on the global tobacco epidemic, 2017: monitoring indicators 2017. New York: United Nations DoEaSA, Population tobacco use and prevention policies. Geneva: World Health Division; 2017 (http://www.un.org/en/development/desa/ Organization; 2017. population/theme/family-planning/cp_model.shtml, accessed 8 Sept 2017). 44. UHC data portal. Geneva: World Health Organization; 2017 (http://apps.who.int/gho/cabinet/uhc.jsp, accessed 12 Feb 34. WHO/RHR global database 2017. Geneva: World Health 2017). Organization; 2017 (http://www.who.int/gho/maternal_health/ who_rhr_anc4_detailed_2017.xls, accessed 11 Jan 2017). 45. WHO global health workforce statistics, 2016 update. Geneva: World Health Organization; 2016 (http://www.who.int/hrh/ 35. World Health Organization, UNICEF. WHO/UNICEF estimates statistics/hwfstats/en/, accessed 18 Aug 2016). of national immunization coverage (WUENIC) 2017. Geneva: World Health Organization; 2017 (http://apps.who.int/ 46. Global Health Observatory data repository: International Health immunization_monitoring /globalsummary/timeseries/ Regulations (2005) monitoring framework. Geneva: World tswucoveragedtp3.html, accessed 7 Sept 2017). Health Organization; 2017 (http://apps.who.int/gho/data/node. main.IHR?lang=en, accessed 3 Oct 2017). 36. Global databases: children under 5 with symptoms of pneumonia taken to a health care provider 2016. New York: UNICEF; 2016 47. Victora CG, Fenn B, Bryce J, Kirkwood BR. Co-coverage (https://data.unicef.org/, accessed 8 Sept 2017). of preventive interventions and implications for child- survival strategies: evidence from national surveys. Lancet. 37. Global tuberculosis report. Geneva: World Health Organization; 2005;366(9495):1460-6. 2016. 48. Hosseinpoor AR, Bergen N, Koller T, Prasad A, Schlotheuber A, 38. People living with HIV receiving ART (%). Geneva: UNAIDS; Valentine N et al. Equity-oriented monitoring in the context of 2017 (http://aidsinfo.unaids.org/, accessed 2 Oct 2017). universal health coverage. PLOS Medicine. 2014;11(9):e1001727 (doi:10.1371/journal.pmed.1001727). 39. World malaria report 2016. Geneva: World Health Organization; 2016. 49. Resolution WPR/RC67/8. Regional action agenda on achieving the Sustainable Development Goals in the Western Pacific. 40. United Nations Children’s Fund, World Health Organization. In: Sixty-seventh session of the Regional Committee, Manila, Progress on drinking water, sanitation and hygiene: 2017 update 10-14 October 2016. Manila: WHO Regional Office for the and SDG baselines. New York, Geneva: UNICEF/WHO; 2017. Western Pacific; 2016 (http://www.wpro.who.int/about/ regional_committee/67/documents/wpr_rc67_8_sdgs.pdf, 41. Zhou B, Bentham J, Di Cesare M, Bixby H, Danaei G, Cowan MJ accessed 21 August 2017). et al. Worldwide trends in blood pressure from 1975 to 2015: a pooled analysis of 1479 population-based measurement studies 50. Resolution WPR/RC66/6 on universal health coverage. In: Sixty- with 19.1 million participants. Lancet. 2017;389(10064):37-55. sixth session of the WHO Regional Committee for the Western Pacific, Guam, United States of Amercica, 12-16 October 2015. 42. Danaei G, Finucane MM, Lu Y, Singh GM, Cowan MJ, Paciorek Manila: WHO Regional Office for the Western Pacific; 2015 CJ et al. National, regional, and global trends in fasting plasma (http://www.wpro.who.int/about/regional_committee/66/ glucose and diabetes prevalence since 1980: systematic analysis documents/wpr_rc66_06_uhc_7sep.pdf?ua=1; accessed 21 of health examination surveys and epidemiological studies August 2017). with 370 country-years and 2.7 million participants. Lancet. 2011;378(9785):31-40. CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 21 CHAPTER 2 FINANCIAL PROTECTION CHAPTER 1. COVERAGE OF ESSENTIAL HEALTH SERVICES 23 Financial protection in health occurs when families who get needed care do not suffer undue financial hardship as a result. This chapter presents various measures of financial protection and their operationalization, and presents data on levels and trends, beginning with the official SDG indicators, and then moving to various non-SDG indicators that are also considered important in monitoring financial protection. Box 2.1 presents key findings. Box 2.1 Financial protection: key findings 808 million people worldwide incur catastrophic health spending defined as out-of-pocket expenditures exceeding 10% of household total consumption or income. In 2010, 808 million people (11.7% of the world’s population) incurred catastrophic spending at the 10% threshold. At the 25% threshold, the gures are 179 million and 2.6%. These thresholds are both part of the o cial catastrophic expenditures SDG indicator 3.8.2, de ned as “the proportion of population with large household expenditures on health as a share of total household expenditure or income”. Latin America and Asia have the highest rates of people with out-of-pocket expenditures exceeding 10% or 25% of household total consumption or income. Latin America and the Caribbean has the highest rate at the 10% threshold (14.8%). Asia has the second-highest rate (12.8%), and is the region where most people facing catastrophic payments are concentrated. Catastrophic payment incidence has been increasing between 2000 and 2010. Both the percentage and the size of the population facing catastrophic payments have increased at all thresholds since 2000. At the 10% threshold, the region with the fastest increase in population facing catastrophic payments is Africa (+5.9% per annum on average) followed by Asia (+3.6% per annum). North America is the only region where both the incidence and the population exposed has decreased (–0.9% per year). 97 million impoverished by out-of-pocket spending at the 2011 PPP $1.90-a-day poverty line. An estimated 97 million people were impoverished by health care expenditures at the $1.90-a-day poverty line in 2010, equivalent to 1.4% of the world’s population. At the 2011 PPP $3.10-a-day poverty line, the gure is 122 million (1.8%). At these two international poverty lines impoverishment rates in upper-middle-income countries and high-income countries are close to or equal to zero. Africa and Asia have the highest impoverishment rates at the 2011 PPP $1.90-a-day poverty line. Africa and Asia have 1.4% and 1.9% rates of impoverishment respectively at the $1.90-a-day poverty line in 2010. These two regions account for 97% of the world’s population impoverished by out-of-pocket health spending. Impoverishing payment incidence has been falling at the 2011 PPP $1.90-a-day poverty line but not at the 2011 PPP $3.10 line. At the $1.90-a-day poverty line, the number and percentage of people globally impoverished fell between 2000 and 2010 from 130 million (2.1%) to 97 million (1.4%). By contrast, at the $3.10-a-day line, the percentage and number of people impoverished increased from 106 million (1.7%) to 122 million (1.8%). Uneven progress across UN regions on impoverishing spending at the 2011 PPP $1.90-a-day and 2011 PPP $3.10-a-day line. Africa has seen reductions at both the $1.90 and $3.10-a-day poverty lines, while Asia saw a marked reduction at the $1.90 line and an increase at the $3.10 line. Measures of nancial protection to out-of-pocket spending. In the case of a household impoverished by out-of-pocket spending, the change in Catastrophic spending on health (SDG and the gap is the amount by which out-of-pocket spending non-SDG indicators) pushes the household below the poverty line. In the case of an already-poor household, the change in the poverty There is no right or wrong approach to measuring gap is equal to the full amount of the household’s out- catastrophic health expenditures. Different studies of-pocket spending. These amounts are then averaged adopt different approaches. Some studies define out-of- across all households to get the overall average change pocket health expenditures as catastrophic when they in the poverty gap due to out-of-pocket health spending. exceed a given percentage (for example, 10% or 25%) of If multiplied by the poverty line, it gives the average per income or consumption (1). This is the approach adopted capita amount by which consumption or income falls short in SDG 3.8.2. Other studies relate health expenditures of the poverty line. not just to income or consumption, but rather to income or consumption less a deduction for necessities, the argument being that this may provide a better measure of a household’s ability or capacity to pay out-of-pocket Operationalizing measures of for health services. These approaches are part of WHO nancial protection regional frameworks to monitor catastrophic expenditures (2,3,7). Studies of catastrophic health spending often report Defining and measuring out-of-pocket the incidence of such spending not only among the sample spending as a whole, but also among different groups especially those defined in terms of consumption or income, for Out-of-pocket payments are those made by people at the example, income ‘quintiles’. Some studies – including time of getting any type of service (preventive, curative, this report – also make use of a summary measure of rehabilitative, palliative or long-term care) provided by inequality known as the ‘concentration index’.1 any type of provider. They include cost-sharing (the part not covered by a third party like an insurer) and informal payments (for example, under-the-table payments), but Impoverishing spending on health (non-SDG they exclude insurance premiums. Out-of-pocket payments indicators) could be financed out of a household’s income, including remittances, its savings, or by borrowing. They exclude any Impoverishment is not an official SDG indicator of reimbursement by a third party, such as the government, universal health coverage per se, but links UHC directly a health insurance fund or a private insurance company. to the first SDG goal, namely to end poverty in all its forms everywhere. Impoverishment is defined as occurring In practice, household surveys that are used to collect when a household’s consumption including out-of-pocket data on out-of-pocket payments for health suffer some spending is more than the poverty line but its consumption shortcomings. In some countries, it is sometimes unclear excluding out-of-pocket spending is less than the poverty whether the payments are net or gross of reimbursement line (1). The idea is that a household that is impoverished from third party payers. Sometimes, recall periods are by out-of-pocket spending was forced by an adverse probably too long (for example, 12 months for medicines) health event to divert spending away from non-medical or too short (for example, one month for inpatient care) budget items such as food, shelter, clothing, etc. to such an to get accurate data. Whenever the recall period is less extent that its spending on these items is reduced below than 12 months, analysts usually have little choice when the level indicated by the poverty line. Impoverishment annualizing the data but to multiply by the relevant number can be computed as the change in poverty headcount (for example, 12 in the case of a one-month recall) but with and without out-of-pocket spending included in this may well not produce an accurate estimate. Most consumption or income. surveys ask about spending on all health care items (pharmaceutical products, hospital services, medical This ‘headcount’ measure does not tell us how far such services and paramedical services), but it is difficult to be households are pushed below the poverty line. Nor does it sure that all surveys are equally comprehensive. Surveys capture the fact that some already-poor households may without a focus on health-seeking behaviour do not have be pushed even further into poverty by their out-of-pocket information on the indirect costs associated with utilization health spending. These two facets of impoverishment can (e.g. transportation costs). Nor do they have information on be captured by the change in the ‘poverty gap’ attributable the opportunity costs of care-seeking (e.g. income losses). These types of costs can represent a substantial burden, but 1 The concentration index is zero if, on balance, catastrophic expenditures are equally are not included in the estimates below. common among rich and poor households; negative if they are more common among poor households; and positive if they are more common among rich households. CHAPTER 2. FINANCIAL PROTECTION 25 Defining and measuring income and Measuring income also has its challenges, especially consumption in countries where a large fraction of the population is not in formal employment. In low-income and lower- Some studies relate out-of-pocket spending to income, middle-income countries, it has traditionally been some to consumption. When income is used, no allowance claimed that measuring consumption is simpler than is made for the fact that households are able to reduce measuring income, while in upper-middle income and the variability of consumption over time – including in high-income countries, it has traditionally been claimed response to health events necessitating out-of-pocket that income can be measured satisfactorily. Recent years payments – by borrowing (or dissaving) and saving. Using have seen changes in these positions, with income (and consumption, which measures what households consume consumption) being estimated in an increasing number rather than what they receive in income, allows for this. of surveys in low-income and lower-middle-income But it leads to the perverse conclusion that a household countries, and the difficulties of measuring income being that borrows to finance health care ends up being increasingly acknowledged in upper-middle income classified as better off than one with a similar income and high-income countries. This report mostly uses that does not need to spend on health care, and does consumption rather than income; in a small number of not therefore borrow to finance it. Consumption gross of countries, however, income is used because consumption borrowing to finance health care may therefore overstate is not available; and for a selection of countries where a household’s living standards, making a household that both are available, results on inequalities in catastrophic is badly off appear to be relatively well off. By contrast, spending are presented using both. income may understate a household’s living standards, making a household that is genuinely relatively well off (in consumption terms) appear to be relatively badly off. Defining and measuring ability to pay The choice between consumption and income matters less Some studies do not relate out-of-pocket spending to a when measuring the incidence of catastrophic spending household’s actual consumption or income but instead than when measuring inequality in catastrophic spending. relate out-of-pocket spending to a household’s consumption Choosing consumption gross of health expenditures or income less an amount of money deemed required for may result in catastrophic spending occurring among necessities, such as food and housing. This adjustment is households that appear to be well off, but may, in reality, argued to better capture a household’s ability or capacity be well off only because they are borrowing to finance to pay for health expenditures. One approach (1) to define their health spending. By contrast, choosing income ability to pay is therefore to deduct actual food expenditure may result in catastrophic spending occurring frequently from consumption, and relate out-of-pocket spending to among households that appear to be badly off, but may, nonfood consumption. Another (2, 3) is to deduct an estimate in reality, be able to consume more than their income by of the amount of money a household requires to meet its borrowing or using savings. Similarly, for the measurement basic food needs. 2 A third approach (4) is to deduct the of impoverishment, choosing income might underestimate prevailing poverty line, essentially an allowance for basic the long-term consequences of health expenditures, while needs. A fourth approach (7) is to deduct an amount of choosing consumption might overstate the degree to money representing the amount a household needs to meet which health expenditures result in impoverishment in the basic food and shelter needs. The main results in the report short term as households might be able to smooth their employ two methods: no deduction for necessities; and a consequences over time by dissaving or borrowing (1, 5). deduction for actual food spending (the first of the deduction approaches). It also discusses preliminary findings for 25 In practice, analysts are not always free to choose European countries conducted by the WHO Regional Office whether to use consumption or income – many surveys for Europe based on the fourth of the deduction approaches. allow only one to be computed. Measuring both has its Since food expenditures do not rise proportionately with challenges. A household’s consumption is often different income, and, in the other approaches, deductions for food, from its expenditures: families may grow vegetables housing and utilities are usually fixed amounts, making these and keep animals so their food consumption exceeds adjustments disproportionately affects households at low their expenditures on food; families may own their home levels of consumption and income. Empirically catastrophic outright so their consumption of ‘housing’ exceeds their spending is usually less concentrated among “the poor” (or expenditures on housing items; and so on. Often attempts more concentrated among “the rich” ) when the budget are made to go beyond expenditure to get a measure of share approach is used (Box 2.2 ). consumption, by capturing the value of home production, and imputing the consumption value of durables including 2 The food allowance is set equal to average food spending among households whose food spending share (as a percentage of total consumption) is in the 45th to 55th percentile housing (6). range, the assumption being that, at least in low- and middle-income countries, the food intake of this group averages 2000 kilocalories. 26 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Box 2.2. Different ways to measure catastrophic spending on health Some studies de ne out-of-pocket health expenditures as catastrophic when they exceed a given percentage (for example, 10% or 25%) of income or consumption. This so-called ‘budget share’ approach is adopted in SDG 3.8.2. Empirically catastrophic spending is usually less concentrated among “the poor” (or more concentrated among “the rich”) when the budget share approach is used. Other studies relate health expenditures to income or consumption less a deduction for necessities rather than to total income or consumption. The argument is that everyone needs to spend at least some minimum amount on basic needs such as food and housing, and these absorb a larger share of a poor household’s consumption or income than that of a rich household. As a result, a poor household may not be able to spend much, if anything, on health care. By contrast, a rich household may spend 10% or 25% of their budget on health care and still have enough resources left over not to experience nancial hardship. There are di erent approaches to deducting expenditures for basic needs. The main di erences between them include: deducting actual spending versus a standard amount; using one item or a basket of items; the method used to derive the standard amount; and treatment of households whose actual spending is below the standard amount. Some studies deduct all of a household’s actual spending on food (1). However, although poor households often devote a higher share of their budget to food, it may not be a su cient proxy for non-discretionary consumption. Also, spending on food re ects preferences, as well as factors linked to health spending: for example, households that spend less on food because they need to spend on health care will appear to have greater capacity to pay than households that spend more on food. A second approach, aimed at addressing the role of preferences in food spending, is to deduct a standard amount from a household’s total resources to represent basic spending on food (2, 3). In practice, it is a partial adjustment to the actual food spending approach because the standard amount is used only for households whose actual food expenditure exceeds the standard amount. For all other households, actual food spending is deducted instead of the higher, standard amount. Both the actual food and the standard food approaches therefore treat households whose actual food spending is below the standard amount in the same way. Nevertheless, with the standard food approach, catastrophic spending may be less concentrated among the rich than with the actual food spending approach. A third approach is to deduct the prevailing poverty line, essentially an allowance for all basic needs (4). Depending on the poverty line used, this is likely to result in a greater concentration of catastrophic spending among the poor than the rich, compared to the budget share approach. It also has the merit of providing a link between catastrophic health expenditures and impoverishment: those with a negative capacity to pay start o below the poverty line, even before paying for health care, and are pushed even further into poverty by any health spending. By contrast, those with catastrophic spending larger than one are pushed into poverty by their health spending. Building on the second and third approaches, in the WHO European Region an amount representing spending on three basic needs (food, housing (rent) and utilities) is deducted consistently for all households (7). As a result, catastrophic expenditure is more likely to be concentrated among the poor with this approach, compared to the budget share approach. It also provides a link between catastrophic health spending and impoverishment (Box 2.7). Poverty lines relative poverty line, set at 50% of median household consumption – which comes closest to the common This report uses two different poverty lines in measuring definition of a poverty line in high-income countries. As impoverishment due to out-of-pocket spending. The first countries and regions assess their own progress towards is the international $1.90-a-day line measured in 2011 UHC, they could also use relevant locally defined poverty PPPs. This is often referred to as the ‘extreme poverty line’ lines (national or regional). which was estimated at $1.25-a-day in 2005 PPPs, and underlies SDG target 1.1 (8) The second is a $3.10-a-day As the rest of this chapter will show, a lot of progress has international poverty line in 2011 PPPs, which updates the been made in monitoring catastrophic and impoverishing $2.00-a-day poverty line in 2005 PPPs commonly used health spending since the 2015 report on universal health for lower-middle-income countries (8).3 coverage (Box 2.3). For global monitoring, this report focuses on international poverty lines. Elsewhere (9), results are reported for a 3 In October 2017, the World Bank revised the $3.10-a-day poverty line to $3.20-a-day. CHAPTER 2. FINANCIAL PROTECTION 27 Box 2.3. Financial protection 2015–2017 monitoring: what has changed? More countries. The 2015 UHC global monitoring report (UHC GMR) analyzed 37 countries representing one sixth of the world’s population. This report analyses 132 countries representing 93% of the world’s population in 2015. More countries with trend data. The 2015 report analysed trend data for 23 countries. This report analyses trend data for 93 countries. Moreover, for many countries, the report uses more than two years’ worth of data. In total, this report uses 553 datapoints for catastrophic spending and 516 for impoverishment. Quality checks and country consultations. The original dataset consisted of 971 household surveys, each of which was analysed. The estimates of per capita consumption and the health budget share were then compared with WHO and World Bank data, and the time-series of the catastrophic and impoverishing estimates were checked manually. Most of the retained datapoints were shared with countries’ nominated focal points through a WHO consultation process. Different catastrophic payment indicators. The 2015 report de ned catastrophic payments as those exceeding 25% of total consumption, 40% of nonfood consumption, and 40% of consumption less a xed expenditure allowance for food. This report presents results for 10% of consumption and 25% of consumption, both of which are o cial SDG indicators, as well as 40% of nonfood consumption. In addition, results are presented for selected European countries for 40% of consumption less a xed allowance for food and housing expenditures. More evidence on inequalities in catastrophic spending. The 2015 report showed inequalities in catastrophic spending incidence de ned as payments exceeding 25% of total consumption across quintiles of total consumption. This report also reports inequalities for 40% of nonfood consumption across quintiles of total consumption and for the SDG indicators across income quintiles for a subsample of the 132 countries. Impoverishment measures based on international poverty lines in 2011 PPPs. The 2015 report used international poverty lines in 2005 PPPs, speci cally the $1.25-a-day extreme poverty line and the $2.00-a-day line of moderate poverty. In addition, it adjusted poverty lines to match the economic level of each country. This report uses international poverty lines in 2011 PPPs as of September 2011. It includes the $1.25-a-day poverty line (now $1.90-a-day) and the updated $2.00-a-day line (now $3.10-a-day). This report does not adjust poverty lines to match the economic level of each country. Different measures of impoverishment. The 2015 report measured the proportion of the population neither pushed into poverty nor further pushed into poverty. This report measures the incidence of impoverishment (the population pushed into poverty) as the di erence in the poverty headcount with and without out-of-pocket spending included in household total consumption or income. This report does not measure the fraction of the population neither pushed into poverty nor further pushed into poverty, but it does assess the contribution of out-of-pocket payments to the depth of poverty. This captures the monetary impact of out-of-pocket expenditures for both households pushed into poverty and those pushed further into poverty due to out-of-pocket health spending. Trends in catastrophic payments and impoverishment. The 2015 report measured progress over time for 23 countries. This report presents annual percentage point changes in incidence for 93 and 84 countries respectively. Global and regional estimates for three years. The 2015 report presented mean and median rates of catastrophic and impoverishing spending in the 37 countries, with di erent countries having di erent survey years. This report estimates global and regional incidence by estimating rates of catastrophic and impoverishing spending for each country in the world for each of three years – 2000, 2005 and 2010. Global Data – and dealing with in the global dataset. The process described hereafter focuses on the assembly of the global database on ‘missing data’ financial protection. Similarly, all results reported here are based on the global dataset, unless otherwise indicated. Household surveys For the assembly of the global dataset, a total of 1,566 potentially suitable household surveys were identified To measure the incidence of catastrophic spending from microdata catalogues and other sources. Of these, and impoverishment, data are needed from nationally 595 were discarded, because they were either inaccessible representative household surveys containing information or lacked key variables for the analysis. The remaining 971 on out-of-pocket health spending, and household datasets were analysed, and estimates of catastrophic consumption or expenditure or income. Availability of spending and impoverishment were obtained, along with data to produce global estimates may not necessarily align ancillary data. These ‘datapoints’ were then subject to with availability of data at national or regional levels. For a quality assurance process (9, 10): datapoints not close this report, WHO and the World Bank have assembled enough to a benchmark value were discarded, as were the largest global database on financial protection to date. datapoints that did not form part of a consistent time Regional and national collaborations are also ongoing but series (Box 2.4). the results of such collaborations have not been included 28 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Box 2.4. Example of data screening process on financial protection The trends in the various series for each country were plotted as in the two examples below, which show catastrophic spending rates (at the 10% threshold) for Lithuania and Mexico. Datapoints that were identi ed as of concern in the quality assurance process were agged. There was a preference for a single series for each country; in some cases, this meant retaining agged datapoints providing they were not too problematic. Mexico Lithuania Incidence of catastrophic health spending – SDG indicator 3.8.2 Incidence of catastrophic health spending – SDG indicator 3.8.2 40 40 30 30 20 20 10 10 0 0 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 WB−HIES LIS SHES WHO−HIES EAPPOV ECAPOV WB−HBS WHO−HBS MCSS WHS Flagged Kept HIES Household income and expenditure survey SHES Flagged Kept LIS Luxembourg income study harmonized version of HIES ECAPOV Ex-post harmonization of household budget surveys (HBS) and Living Standard MCSS Multi-Country Survey Study on Health and Responsiveness (MCSS) Measurement surveys (LSMS), conducted by the World Bank in the Europe and Central SHES Standardized household expenditure survey Asia World Bank region. WHS World Health Survey HBS Household budget surveys SHES Standardized household expenditure survey At the end of this confirmation process, 553 datapoints representing at least 90% of the world population in for catastrophic spending (512 for impoverishment) 2015.4 There are, however, gaps – some countries did not remained from 132 countries or territories (121 in the have a usable survey at all; others only had a pre-2005 case of impoverishment) spanning the period 1984–2015. usable survey. These breakdown across countries as indicated in Fig. 2.1. Overall, the global dataset has information on countries 4 Ninety-three per cent for the analysis of catastrophic health spending and 90% for impoverishment. Fig. 2.1. Data availability for financial protection in the global database Both 1995–2005 and 2006–2015 Only 2006–2015 This map has been produced by WHO. The boundaries, colours Only 1995–2005 or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank Only pre-1995 or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any 0 850 1,700 3,400 Kilometers Dataset(s) analyzed but discarded No datasets identified endorsement or acceptance of such boundaries or frontiers. Dataset(s) inadequate Not applicable Notes: Total number of countries or territories can be split into those without any data identi ed to produce global estimates (61); those with datasets identi ed analyzed but discarded (12); dataset identi ed but found inadequate (10); with datasets available only pre-1995 (2); only for the period 1995–2005 (35); only for the period 2006–2015 (17); for both periods 1995–2005 and 2006–2015 (79). Availability of data to produce global estimates may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. CHAPTER 2. FINANCIAL PROTECTION 29 Missing data, global and regional The same principles are used to construct global incidence estimation estimates for the reference years 2000, 2005 and 2010. For each of these reference years, survey data are included Very few countries had a survey in each of the three years from up to five years before and up to five years after the selected for the global and regional estimation exercise – reference year. The lining up process is described in the 2000, 2005 and 2010. Methods needed to be developed text and illustrated in the figure in Box 2.5, for a reference therefore to get around this ‘missing data’ problem. The year T*, and a +/– five-year window around the reference global estimates for the share of the population facing year. Global estimates are produced for the incidence of catastrophic or impoverishing payments are generated catastrophic and impoverishing health spending but not by ‘lining up’ the underlying survey data into reference for the depth of impoverishment. years. This process is similar to the process used by the World Bank to generate the global poverty estimates (11). Box 2.5. Estimation of country-level data to produce regional and global estimates of catastrophic and impoverishing health spending Lining up survey data into reference years 1. Reference year point. In countries for which there is an observed datapoint in the reference year T* (country 1 and country 8), this point is used. 2. Two points within band. When there are at least two datapoints around the reference year (country 2) and in the window [T*–5; T*+5], linear interpolation is used to project the value of catastrophic payments in the reference year. 3. One point within band. If only one datapoint is available either before (country 3) or after (country 4) the reference year, and within the +/– ve-year window, a multilevel model of the rate of catastrophic payments (impoverishment) is rst estimated using the aggregate share of OOP over total consumption expenditure (and household nal consumption) as explanatory variable. Then the estimated elasticity of catastrophic payments (impoverishment) to the aggregate share of OOP over total consumption (controlling for household nal consumption) is used to project the observed survey point in the reference year. 4. Fitted. For countries with no datapoint in the 10-year window around the reference year (country 5, 6 and 7), the model mentioned in (3) above is used to project the survey point to the reference year, using the share of aggregate OOP over total consumption if the variable is available. If this variable is not available, we use the regional median value of catastrophic (impoverishing) payments instead to impute the datapoint in the reference year. The breakdown of datapoints in each of these four categories is provided in the following table. For example, for the reference year 2010 and the estimation of the incidence of catastrophic health spending, there are a total of 101 countries with at least one datapoint between 2005 and 2015; these countries together represent 86.1% of the world’s population. Categories of datapoints used to construct global estimates of catastrophic and impoverishing health spending [1995–2005] [2000–2010] [2005–2015] Ref. year 2000 Ref. year 2005 Ref. year 2010 Countries Global population Countries Global population Countries Global population (No.) (%) (No.) (%) (No.) (%) Cata. Impov. Cata. Impov. Cata. Impov. Cata. Impov. Cata. Impov. Cata. Impov. Reference year point 27 25 38.4 38.2 36 34 19.9 19.2 54 54 31.4 31.4 Two points within band 19 13 6.6 5.2 29 23 54 52.5 13 13 21.8 21.8 One point within band 61 42 38 34.2 48 43 15.5 13.5 34 27 32.9 29.4 Total observed 107 80 83 77.6 113 100 89.4 85.2 101 94 86.1 82.6 Fitted 15 10 6.9 4.7 11 2 0.8 0.1 23 12 4.1 3.1 Regional median 89 103 10.1 17.1 87 92 9.8 14.5 87 88 9.8 14.1 Cata: catastrophic health spending; Impov: impoverishing health spending. 30 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Levels and trends in catastrophic countries (Fig. 2.2). The mean and median catastrophic spending: the SDG 3.8.2 indicators out-of-pocket payment rates at the 10% threshold are 9.2% and 7.1% (IQR: 10.0). Rates are inevitably lower Cross-country variation in catastrophic at the 25% threshold with mean and median incidence spending rates of 1.8% and 1.0% (IQR: 2.1). Coincidentally, the median incidence at the 25% threshold is the same as that The incidence of catastrophic out-of-pocket payments in reported in the first UHC GMR despite fewer countries (37) the most recent surveys available varies markedly across being used there (Box 2.3). Fig. 2.2 Incidence of catastrophic health spending: SDG indicator 3.8.2, latest year 10% threshold Percent of population* 15.00–44.85 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map 10.00–14.99 and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 6.00–9.99 3.00–5.99 Not applicable 0 850 1,700 3,400 Kilometers * with household expenditures on health exceeding 10% of 0.00–2.99 Data not available total household expenditure or income 25% threshold Percent of population* 3.20–11.52 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map 1.50–3.19 and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.70–1.49 0.20–0.69 Not applicable 0 850 1,700 3,400 Kilometers * with household expenditures on health exceeding 25% of 0.00–0.19 Data not available total household expenditure or income Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. CHAPTER 2. FINANCIAL PROTECTION 31 Global and regional estimates of the threshold, while the region represents a little less than catastrophic spending 60% of the world population. The new data contrasts with earlier studies of catastrophic health spending in Aggregating across countries, it is estimated that in terms of the number of people affected (Box 2.6). In 2010, 808 million people incurred catastrophic spending terms of incidence rates of catastrophic payments, there at the 10% threshold, equivalent to 11.7% of the world’s are substantial variations across UN regions with Latin population in 2010. At the 25% threshold, the figures America and the Caribbean having the highest rate of are 179 million and 2.6% (Table 2.1). In 2010, Asia is the catastrophic spending on health at the 10% threshold region which concentrates most of the population facing (14.8%) in 2010, Asia having the second-highest rate catastrophic payments. Between 66% and 77% of the (12.8%), Northern America having the second-lowest population exposed globally is from Asia, depending on rate (4.6%), and Oceania having the lowest rate (3.9%). Table 2.1. Global and regional trends in catastrophic payments – SDG indicator 3.8.2 10% threshold 2000 2005 2010 % Population Million % Population Million % Population Million Global 9.7% 588.5 11.4% 741.3 11.7% 808.4 Africa 8.7% 70.7 10.3% 94.1 11.4% 118.7 Asia 10.4% 381.6 12.2% 479.2 12.8% 531.1 Latin America and the Caribbean 13.4% 70.5 17.5% 98.3 14.8% 88.3 Northern America and Europe 6.2% 64.6 6.5% 68.6 6.4% 68.8 Northern America 5.5% 17.2 5.3% 17.4 4.6% 15.6 Europe 6.5% 47.4 7.0% 51.2 7.2% 53.2 Oceania 3.5% 1.1 3.4% 1.1 3.9% 1.4 25% threshold 2000 2005 2010 % Population Million % Population Million % Population Million Global 1.9% 112.8 2.4% 154.9 2.6% 179.3 Africa 1.5% 12.3 1.9% 17.7 2.5% 25.6 Asia 2.1% 77.1 2.8% 108.7 3.1% 128.7 Latin America and the Caribbean 2.6% 13.6 3.2% 18.0 2.5% 14.9 Northern America and Europe 0.9% 9.6 1.0% 10.3 0.9% 9.8 Northern America 1.0% 3.1 0.9% 3.0 0.8% 2.6 Europe 0.9% 6.5 1.0% 7.3 1.0% 7.2 Oceania 0.5% 0.1 0.4% 0.1 0.5% 0.2 Notes: The World Bank and WHO estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on data available for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. 32 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Box 2.6. Previous global estimates of catastrophic spending The only global study of catastrophic spending prior to the 2015 UHC GMR was conducted by WHO in 2007, based on 116 surveys covering 89 countries with a median survey year of 1997. The study de ned catastrophic spending as out-of-pocket payments exceeding 40% of total consumption net of an allowance for food expenditure. It reported mean and median rates of catastrophic spending of 2.3% and 1.5% respectively, and concluded that an estimated 150 million people globally incur catastrophic spending annually (2, 10). The new global data in this report has an additional 418 datasets for 44 additional countries, and a median year of 2004. Using a de nition of catastrophic expenditures that comes closest to that used in the 2003 WHO study, i.e. out-of-pocket health expenditures exceeding 40% of nonfood consumption, global incidence is estimated to be 2.7% in 2000 or 166 million people. This increases to 3% in 2005 (193 million people) where it is estimated to have stayed up to 2010; but with a growing population, this translates into an additional 15 million people spending more than 40% of their nonfood consumption out-of-pocket on health. In terms of incidence, the increase is less sharp between 2000 and 2005 with the nonfood measure than with the SDG indicators; the nonfood measure shows no change between 2005 and 2010 while the SDG 3.8.2-10% and SDG 3.8.2-25% indicators show an increase. Trends in catastrophic spending the 10% and 25% thresholds respectively, while the population-weighted figures are 0.45% and 0.22%; the At the 10% threshold, the average annual change in fact the latter are larger than the former means that catastrophic spending incidence ranges from –2.7% per catastrophic payment incidence has been falling more annum in Congo (2005–2011) to 3.3% per annum in slowly or rising more quickly in more populous countries. Armenia (2010–2013) (Fig. 2.3). In 48 (52%) of the 94 countries for which we have two or more years of data, Both the percentage and the size of the population facing the incidence of catastrophic out-of-pocket spending catastrophic payments have increased at all thresholds increased over time. At the 25% threshold, catastrophic since 2000 ( Table 2.1). At the 10% threshold, the region payment incidence increased in 54% of countries. In the with the fastest increase in population facing catastrophic 2015 UHC GMR only 48% of the 23 countries with trend payments is Africa (+5.9% per annum on average) data had an increasing rate. The population-unweighted followed by Asia (+3.6% per annum). Northern America median annual changes in catastrophic out-of-pocket is the only region where both the incidence and the payment rates are 0.03% and 0.01% per annum for population exposed decreased (–0.9% per year). CHAPTER 2. FINANCIAL PROTECTION 33 Fig. 2.3. Annual percentage point change in incidence of catastrophic health spending: SDG indicator 3.8.2 Congo 2005−2011 Guinea 2002−2012 Pakistan 1991−2010 Bolivia (Plurinational State of) 1999−2002 Côte d’Ivoire 1998−2008 Albania 2002−2012 Panama 1997−2008 Guatemala 2000−2014 Burkina Faso 1998−2009 Cameroon 1996−2014 Paraguay 1996−2001 Tajikistan 1999−2007 Uganda 1996−2002 Zambia 2004−2010 Ukraine 2002−2013 Bosnia and Herzegovina 2001−2011 Rwanda 2000−2010 Viet Nam 1992−2014 South Africa 1995−2010 Turkey 2002−2012 Colombia 1997−2010 United Republic of Tanzania 2008−2012 Thailand 1994−2010 Mexico 1984−2012 Croatia 1998−2010 Peru 2000−2015 Mozambique 2002−2008 Slovakia 2004−2010 United States 1995−2013 Ethiopia 1999−2004 Sri Lanka 1995−2009 Cabo Verde 2001−2007 Bangladesh 2000−2010 Ghana 1991−2005 Finland 1998−2010 Malaysia 1993−2004 Kyrgyzstan 2005−2011 Kazakhstan 1996−2013 Madagascar 2001−2005 Greece 1998−2010 Mali 2001−2006 Slovenia 1999−2012 Lao People’s Democratic Republic 2002−2007 Kenya 1997−2005 United Kingdom 1995−2013 Malawi 1997−2010 Israel 1997−2012 Jamaica 1991−2004 Denmark 1997−2010 Italy 2001−2010 Serbia 2003−2010 Niger 2005−2011 Mongolia 2002−2012 Belarus 1998−2012 Norway 1996−1998 Russian Federation 1997−2014 Spain 1985−2010 Indonesia 2001−2015 Czechia 1999−2010 Montenegro 2005−2014 Luxembourg 1998−2010 Poland 1998−2012 Philippines 1997−2015 Switzerland 2000−2004 Hungary 1998−2010 Lithuania 1998−2010 Republic of Korea 1999−2008 Tunisia 1995−2010 Ireland 1999−2010 Azerbaijan 2002−2005 Portugal 1990−2010 Costa Rica 1992−2012 India 2004−2011 Belgium 1997−2010 Romania 1998−2012 Iran (Islamic Republic of) 2005−2013 Latvia 2002−2006 The former Yugoslav Republic of Macedonia 1996−2008 Estonia 1995−2010 Nicaragua 1993−2014 Republic of Moldova 1999−2013 Bulgaria 1997−2010 Argentina 1996−2004 Jordan 2002−2006 China 1995−2007 Egypt 1997−2012 Nepal 1995−2010 Georgia 1997−2013 Yemen 1998−2005 Nigeria 2003−2009 Morocco 1998−2006 Chile 1997−2006 Armenia 2010−2013 −4 −3 −2 −1 0 1 2 3 4 Average annual percentage point change 10% 25% Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on data available for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. 34 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Inequalities in catastrophic spending we measure living standards using consumption [gross of out-of-pocket spending] than if we measure living As already explained (Box 2.2), whether catastrophic standards using [current] income. spending incidence is higher among the poor or rich likely depends in part on (a) whether living standards This is indeed what we find across countries at different are measured using consumption or income, and (b) income levels where we have both consumption and any deduction is made from income or consumption for income in the dataset (Fig. 2.4 presents results for a expenditure on necessities. subset of these countries). If we rank households by income, we find the least well-off fifth of the sample have a The sensitivity of inequality to the choice between income relatively high rate of catastrophic spending. But if we rank and consumption is due to the fact that a low-income households by consumption, we find the least well-off fifth household may appear relatively well off if its living have a relatively low rate. For this subset of countries, the standards are assessed using consumption gross of out- concentration index for catastrophic spending incidence is of-pocket health spending if it borrows (or draws on its always negative when households are ranked by income, savings) to finance its out-of-pocket spending. Because indicating the poor face higher rates, and mostly positive of this, we will likely find out-of-pocket health spending when they are ranked by consumption gross of out-of- more highly concentrated among well-off households pocket expenditures, indicating the well off face higher (or less concentrated among badly off households) if rates. Fig. 2.4. Inequalities in incidence of catastrophic health spending SDG indicator 3.8.2, ranking by consumption vs. income, latest year, selected countries C UMIC1 I C LMIC1 I C LMIC2 I C HIC1 I C UMC2 I C HIC2 I −0.2 0 0.2 0.4 0.6 Incidence of catastrophic spending in unit scale and concentration index values Poorest 20% Richest 20% Population Conc. Index Notes: C: households ranked by consumption; HIC: higher-income country; I: households ranked by income; LMIC: lower-middle-income country; UMIC: upper-middle-income country. Catastrophic health spending is de ned as out-of-pocket expenditures exceeding 10% of household total consumption or income (budget share approach, SDG indicator 3.8.2). The gure shows the incidence of catastrophic expenditures in the unit scale (e.g. 0.2 means 20%) at the national level (population) as well as for the poorest and richest 20%. The gure also shows the concentration index for the incidence of catastrophic expenditure for two HICs,UMICs and LMICs. Source: Calculations by the World Bank from the Luxembourg Income Study harmonized datasets. CHAPTER 2. FINANCIAL PROTECTION 35 Whether or not the incidence of catastrophic spending defined as 10% or 25% of total consumption or income is more concentrated among the poor also depends (i.e. no deduction is made for spending on necessities). on whether deductions for expenditures on necessities The concentration index for the actual food approach are made, and on how these deductions are made, as is negative for 30% of datapoints overall, but these are discussed earlier in Box 2.2. mainly in low- and middle-income countries; it is mainly positive in high-income countries. The concentration Fig. 2.5. presents results for 129 countries and 532 index for the budget share approach is negative in only datapoints. It shows that when ranking households by 17% and 14% of datapoints when the 10% and 25% total consumption or income, the incidence of catastrophic threshold of total consumption respectively are used. The spending defined as 40% of nonfood consumption measure used by WHO/Europe finds that catastrophic (the actual food approach) is more often concentrated incidence is consistently concentrated among the poor among the poor than when catastrophic spending is across the 25 countries included in the study (Box 2.7). Fig. 2.5 Inequalities in incidence of catastrophic spending using the SDG indicator 3.8.2, or the actual nonfood approach, latest year HIC HIC LIC LIC LMIC LMIC UMIC UMIC -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 Distribution of the concentration index, ranking by consumption or income Distribution of the concentration index, ranking by consumption or income ability to pay=40% of non-food ability to pay=40% of non-food ability to pay=10% of total ability to pay=25% of total Global database on nancial protection assembled by WHO and the World Bank. # of countries=129 (532 datapoints). Global database on nancial protection assembled by WHO and the World Bank. # of countries=129 (532 datapoints). # of countries with CI<0 ATP(non−food) =69 (161 points) # of countries with CI<0 ATP(non−food) =69 (161 points) # of countries with CI<0 ATP(tot,10) =44 (91 points) # of countries with CI<0 ATP(tot,25) =39 (73 points) ATP: Ability to pay; CI: concentration index; HIC: higher-income country; LMIC: lower-middle-income country; LIC: low-income country; UMIC: upper-middle-income country. Notes: Catastrophic health spending is de ned as out-of-pocket spending exceeding 10% and 25% of total consumption or income (budget share approach with two thresholds – SDG indicator 3.8.2), as well as out-of-pocket spending exceeding 40% of nonfood consumption (actual food approach). The distribution of the concentration index across 129 countries, latest year is illustrated by the means of box-plots. The 129 countries are grouped according to the World Bank income group of the latest year (HIC, UMIC, LIC LMIC). For each group the median value of the concentration index corresponds to line which divides the box into two parts. The upper limit of the box indicates the value below which fall 75% of the concentration indices (the 75th percentile). The lower limit of the box indicates the value below which the concentration index falls (the 25th percentile). WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. 36 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Levels and trends in catastrophic and Africa and Asia (poorer regions by comparison) spending: non-SDG indicators noticeably darker, reflecting the fact that the nonfood- based measure records a higher incidence of catastrophic Nonfood spending as a measure of ability to spending among the poor than the total consumption pay measure. An estimated 208 million people (3% of the world’s population) incurred catastrophic health spending Setting the catastrophic payment threshold at 40% of using this definition (Table 2.2). The increase in the nonfood consumption gives population-unweighted population facing catastrophic payments at a threshold mean and median catastrophic incidence rates of 2.1% of 40% of nonfood expenditure is driven entirely by Asia and 1.0%. Fig. 2.6 shows the incidence of catastrophic (+2.6% per annum) and Africa (+4.9% per annum). In spending across the world using the nonfood measure, the other regions, the exposed population is either stable with, as in Fig. 2.2, cut-points selected so as to divide (Oceania) or decreasing (Europe, Latin America and the the countries with data into five equal-sized groups. The Caribbean, North America). Americas are noticeably lighter in shade than in Fig. 2.2, Fig. 2.6 Incidence of catastrophic health spending, 40% nonfood consumption, latest year Percent of population* 3.80–12.65 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map 1.40–3,79 and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.50–1.39 0.01–0.49 Not applicable 0 850 1,700 3,400 Kilometers * with household expenditure on health exceeding 40% 0.00 Data not available of nonfood consumption Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. CHAPTER 2. FINANCIAL PROTECTION 37 Table 2.2 Global and regional estimates for catastrophic payments defined using 40% nonfood consumption threshold 2000 2005 2010 % Population Million % Population Million % Population Million Global 2.7% 166.8 3.0% 193.1 3.0% 208.2 Africa 2.6% 21.1 3.1% 28.6 3.3% 34.0 Asia 3.4% 124.3 3.7% 146.4 3.9% 160.8 Latin America and the Caribbean 1.9% 10.2 1.8% 9.8 1.1% 6.8 Northern America and Europe 1.1% 11.2 0.8% 8.2 0.6% 6.4 Northern America 0.5% 1.5 0.4% 1.2 0.3% 1.1 Europe 1.3% 9.7 1.0% 7.0 0.7% 5.3 Oceania 0.3% 0.1 0.2% 0.1 0.2% 0.1 Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. National and regional analyses have a vital role to play options for policy. Box 2.7 illustrates how this is done in in identifying the factors that strengthen and undermine the WHO European region. financial protection and highlighting implications and Box 2.7. Monitoring financial protection in the WHO European Region What is WHO/Europe doing? In 2014 the WHO European Region initiated a multi-year project to generate fresh evidence on nancial protection using a new method of measuring catastrophic and impoverishing health spending and a comprehensive approach to monitoring. The project aims to monitor nancial protection in a way that produces actionable evidence for policy; promotes pro-poor policies to break the link between ill health and poverty; and is relevant to all WHO Member States in the European Region, including high-income countries. How is this approach different? First, the method developed to measure catastrophic out-of-pocket payments builds on the capacity to pay approach used by WHO as part of a broader UHC monitoring agenda and aims to address some of its limitations (Box 2.2) (7, 12). It deducts consistently for all households a standard amount representing spending on three basic needs: food, housing (rent) and utilities. The standard amount is referred to as a basic needs or poverty line. With this approach, the incidence of catastrophic expenditure is more likely to be concentrated among the poor than with the budget share approach or with other capacity to pay approaches (Box 2.2). Second, households are classi ed according to their risk of being impoverished after out-of- pocket payments using the basic needs line. Third, the European Region is working with national experts to produce in-depth, context-speci c analysis of nancial protection over time to enhance policy relevance at country level. In 2018, these country-speci c reports will form the basis for a regional monitoring report that will review trends in the incidence and drivers of nancial hardship over time within countries; trends in inequalities in nancial protection within and across countries; and issues around access, including unmet need for health care. The regional report will also highlight examples of good practice and implications for policy. What are the findings? Combining this method with context-speci c analysis provides rich and actionable evidence for policy. Based on preliminary results for 25 countries in the WHO European Region (7) we nd that: Households in the poorest quintile are most likely to experience catastrophic health spending in all countries. Outpatient medicines are a major driver of catastrophic health spending: in countries with a relatively high incidence of catastrophic health spending, most catastrophic out-of-pocket payments are for outpatient medicines; among poor households, most catastrophic out-of-pocket payments are for outpatient medicines in most countries. Changes in nancial protection can be linked to changes in policy within a country and to policy di erences across countries; in Latvia, for example, the abolition of co-payment exemptions after the economic crisis did not change the overall incidence of catastrophic health spending but did increase nancial hardship for poor households (13). Across countries, the incidence of catastrophic health spending rises steadily as the out-of-pocket share of total spending on health increases, but there are outliers; this highlights the importance of careful policy design, in addition to higher public spending on health, in strengthening nancial protection. Which countries are covered by the WHO/Europe study: Albania, Austria, Croatia, Cyprus, Czechia, Estonia, France, Georgia, Germany, Greece, Hungary, Ireland, Kyrgyzstan, Latvia, Lithuania, Netherlands, Poland, Portugal, Republic of Moldova, Slovakia, Slovenia, Sweden, Turkey, Ukraine and United Kingdom. 38 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Levels and trends in impoverishment The incidence of impoverishing out-of-pocket payments at the $1.90-a-day poverty line in our most recent surveys due to out-of-pocket spending: varies markedly across countries (Fig. 2.7), from 0.0% non-SDG indicators in all high-income countries to 4.5% in a lower- middle- income country. The population-weighted median rates Cross-country variation in impoverishment of impoverishment are 1.86% at the $1.90-a-day line, and 2.44% at the $3.10-a-day line. Fig. 2.7. Incidence of impoverishment due to out-of-pocket health spending – 2011 PPP $1.90-a-day and 2011 PPP $3.10-a-day poverty lines, latest year $1.90 poverty line Percent of population* 1.73–6.15 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 1.28– 1.72 any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.81–1.27 0.55–0.80 0.01–0.54 Not applicable 0 850 1,700 3,400 Kilometers * with impoverishing health spending at the 2011 PPP $1.90-a-day 0.00 Data not available poverty line $3.10 poverty line Percent of population* 1.73–6.15 This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 1.28–1.72 any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.81–1.27 0.55– 0.80 0.01–0.54 Not applicable 0 850 1,700 3,400 Kilometers * with impoverishing health spending at the 2011 PPP $3.10-a-day 0.00 Data not available poverty line Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor impoverishment due to out-of-pocket spending. In particular, the international poverty lines of $1.90-a-day and $3.10-a-day are more appropriate for low-income and lower-middle-income countries, with the $1.90-a-day line being geared to extreme poverty – rarely seen in upper-middle-income and high-income countries. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. CHAPTER 2. FINANCIAL PROTECTION 39 Global and regional estimates of (Fig. 2.8). At the $3.10-a-day poverty line, the figure is impoverishment 122 million (1.8%). Estimates for 2010 vary across UN regions, with Asia and Africa having the highest rates of Aggregating across countries, it is estimated that in impoverishment at the $1.90-a-day poverty line (1.9% and 2010, 97 million people were impoverished by out- 1.4% respectively). These two regions account for 97% of-pocket health spending at the $1.90-a-day poverty of the world’s population impoverished by out-of-pocket line, equivalent to 1.4% of the world’s population health spending at the $1.90-a-day poverty line. Fig. 2.8. Global and regional trends in impoverishment due to out-of-pocket health spending – 2011 PPP $1.90-a-day and 2011 PPP $3.10-a-day poverty lines Africa Asia Europe Latin America and the Caribbean 140 130.4 122.3 120 115.6 115.8 106.1 97 100 80 60 40 20 0 2000 2005 2010 2000 2005 2010 $1.90-a-day poverty line $3.10-a-day poverty line Notes: North America and Oceania not shown in the gure because at both international poverty lines impoverishment rates in these two regions are close to or equal to zero. The WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor impoverishment due to out-of-pocket spending. In particular, the international poverty lines of $1.90-a-day and $3.10-a-day are more appropriate for low-income and lower-middle-income countries, with the $1.90-a-day line being geared to extreme poverty – rarely seen in upper-middle-income and high-income countries. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. 40 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Trends in impoverishment (2003–2007) to 0.2% per annum in Nigeria (2003– 2009) (Fig. 2.9). In 17 (20%) of the countries for which At the $1.90-a-day line, the average annual change in we have two or more years of data, the incidence of impoverishment ranges from 0.6% per annum in Tajikistan impoverishing out-of-pocket spending using the $1.90 Fig. 2.9. Annual percentage point change in incidence of impoverishment due to out-of-pocket health spending – 2011 PPP $1.90-a-day and 2011 PPP $3.10-a-day poverty lines Tajikistan 2003−2007 Congo 2005−2011 Bolivia (Plurinational State of) 2001−2002 Viet Nam 1992−2014 Pakistan 2001−2010 United Republic of Tanzania 2010−2012 Kyrgyzstan 2005−2011 Republic of Moldova 1999−2013 Guinea 2002−2012 Lao People’s Democratic Republic 2002−2007 Cameroon 1996−2014 China 2000−2007 Bangladesh 2000−2010 Zambia 2004−2010 Indonesia 2001−2015 Albania 2002−2012 Mozambique 2002−2008 Peru 2000−2015 Burkina Faso 1998−2009 Côte d’Ivoire 1998−2008 Tunisia 1995−2010 Sri Lanka 1995−2009 Panama 1997−2008 Ethiopia 1999−2004 Jamaica 1991−2004 Madagascar 2001−2005 Romania 1998−2012 Latvia 2002−2006 Ukraine 2002−2013 Cabo Verde 2001−2007 Mongolia 2007−2012 Bosnia and Herzegovina 2001−2011 Egypt 1997−2012 Kazakhstan 1996−2013 India 2004−2011 Armenia 2010−2013 Belarus 1998−2012 Senegal 2001−2011 The former Yugoslav Republic of Macedonia 1996−2008 Uganda 1996−2002 Guatemala 2000−2014 Turkey 2002−2012 Thailand 1994−2010 Malaysia 1993−2004 Montenegro 2005−2014 Russian Federation 1997−2014 Lithuania 1998−2010 Argentina 1996−2004 Slovakia 2004−2010 Canada 1994−2010 United States 1995−2013 Bulgaria 1997−2010 Costa Rica 1992−2012 Poland 1998−2012 Serbia 2003−2010 Hungary 1998−2010 Azerbaijan 2002−2005 Croatia 1998−2010 Czechia 1999−2010 Denmark 1997−2010 Finland 1998−2010 Ireland 1999−2010 Israel 1997−2012 Slovenia 1999−2012 Switzerland 2000−2004 United Kingdom 1995−2013 Republic of Korea 1999−2008 Estonia 1995−2010 Mexico 1984−2012 Rwanda 2000−2010 Kenya 1997−2005 South Africa 2000−2010 Philippines 1997−2015 Morocco 1998−2006 Nepal 1995−2010 Colombia 2008−2010 Nicaragua 1993−2014 Chile 1997−2006 Georgia 1997−2013 Niger 2005−2011 Mali 2001−2006 Paraguay 2000−2001 Nigeria 2003−2009 -0.75 -0.5 -0.25 0 0.25 0.5 $1.90 $3.10 Notes: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor impoverishment due to out-of-pocket spending. In particular, the international poverty lines of $1.90-a-day and $3.10-a-day are more appropriate for low-income and lower-middle-income countries, with the $1.90-a-day line being geared to extreme poverty – rarely seen in upper-middle-income and high-income countries. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. CHAPTER 2. FINANCIAL PROTECTION 41 poverty line increased over time. The figure for the $3.10 Depth of impoverishing health spending line is 29%. The population-weighted median annual changes in impoverishing out-of-pocket payment rates The poverty gap increase attributable to out-of-pocket are 0.02% per annum at the $1.90-a-day line and 0.11% health expenditures, in the most recent surveys available per annum at the $3.10-a-day line. for global monitoring, varies markedly across countries (Fig. 2.10) at the $3.10-a-day poverty line, from 0 cents per At the $1.90-a-day poverty line, the number and capita in international dollars in high-income countries to percentage of people globally impoverished fell between a maximum of 12 cents per capita in low-income countries 2000 and 2010 from 130 million (2.1%) to 97 million at 2011 PPP factors. This amount can be interpreted as (1.4%) (Fig. 2.8). By contrast, at the $3.10-a-day line, the the per capita amount by which on average out-of-pocket percentage and number of people impoverished increased spending pushes or further pushes the household below – from 106 million (1.7%) to 122 million (1.8%). The the poverty line. The population-weighted median of the incidence of impoverishment has evolved differently poverty gap increase attributable to out-of-pocket health across UN regions between 2000 and 2010: Africa has expenditures among the 121 countries, is 1.22 cents per seen reductions at both the $1.90 and $3.10 lines, while capita in 2011 PPPs at the 2011 PPP $1.90-a-day line and Asia saw a marked reduction at the $1.90 line and an 3.74 cents per capita in 2011 PPPs at the $3.10-a-day line. increase at the $3.10 line with 2010 values above 2000 ones. This reflects the rise in living standards in Asia, For the 84 countries for which surveys are available for pushing the population above the higher poverty line and two or more years, the population-weighted median increasing the likelihood of families being pushed across annual changes in the poverty gap increase attributable the poverty line, rather than further below it, through out- to out-of-pocket health expenditures, are -0.12 cents per of-pocket spending on health. annum in 2011 PPPs at the $1.90-a-day line and-0.03 cents Fig. 2.10. Poverty gap due to out-of-pocket health spending expressed in per capita international dollar amounts– 2011 PPP $1.90-a-day and 2011 PPP $3.10-a-day poverty lines $1.90 poverty line Cents of international dollars* This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 2.560–13.943 any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.650–2.559 0.021–0.649 0.010–0.020 Not applicable 0 850 1,700 3,400 Kilometers * Increase in poverty gap at the 2011 PPPs $1.90-a-day poverty line 0.000 Data not available expressed in cents per capita of international dollar 42 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Fig. 2.10. continued $3.10 poverty line Cents of international dollars* This map has been produced by WHO. The boundaries, colours or other designations or denominations used in this map and the publication do not imply, on the part of the World Bank or WHO, any opinion or judgement on the legal status of 2.560–13.943 any country, territory, city or area or of its authorities, or any endorsement or acceptance of such boundaries or frontiers. 0.650–2.559 0.021–0.649 0.010–0.020 Not applicable 0 850 1,700 3,400 Kilometers * Increase in poverty gap at the 2011 PPPs $1.90-a-day poverty line 0.000 Data not available expressed in cents per capita of international dollar Cents of international dollar: international dollar; OOP: out-of-pocket health payments. Notes: The WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor impoverishment due to out-of-pocket spending. In particular, the international poverty lines of $1.90-a-day and $3.10-a-day are more appropriate for low-income and lower-middle-income countries, with the $1.90-a-day line being geared to extreme poverty – rarely seen in upper- middle-income and high-income countries. Global estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. Source: Global database on nancial protection assembled by WHO and the World Bank. per annum in 2011 PPPs at the $3.10-a-day line. Thus, at Going forward the $1.90-a-day poverty line, the incidence and depth of impoverishment have both been falling; in contrast, at the This chapter focuses on financial protection indicators. $3.10-a-day line, the incidence of impoverishment has It demonstrates that it is possible to monitor financial been increasing, but the depth has been falling (albeit protection with standard methods that enable cross- only marginally). country comparisons and are important complementary metrics to national and regional monitoring frameworks A low incidence of catastrophic and/or impoverishing in the SDG era. Going forward the aim is to produced spending on health could result from people being disaggregated indicators of catastrophic spending on protected from financial hardship due to out-of-pocket health by socioeconomic groups and place of residence expenditures but it could also result from people not (urban-rural) and for indicators of impoverishing spending getting the care they need because they cannot access on health use other relevant nationally, regionally, it or because they cannot afford it. This is why financial internationally defined poverty lines than those of extreme protection needs to be discussed jointly with service and moderate poverty. coverage Box 2.8. CHAPTER 2. FINANCIAL PROTECTION 43 Box 2.8 Service coverage and financial protection Universal Health Coverage can be measured through SDG indicators 3.8.1 and 3.8.2 on service coverage and nancial protection. SDG indicator 3.8.1 (service coverage), an index of coverage of essential services, provides an assessment of a country’s overall progress towards providing needed quality essential health services. High values of the index indicate high levels of service coverage. The provision of health services always needs to be nanced. When this is done through out-of-pocket payments, this often causes nancial hardship. SDG indicator 3.8.2 (financial protection) identi es the proportion of the population su ering catastrophic expenditures de ned as the fraction of the population with out-of-pocket spending on health exceeding 10% or 25% of household total expenditure or income. Low incidence of catastrophic expenditures can result from the health nancing system’s capacity to limit out-of-pocket payments, but it can also be due to low levels of service coverage provision. To assess the joint levels of service coverage and nancial protection using the SDG indicators, the sample is restricted to those countries with primary data on catastrophic spending on health for the period 2008-2015, and primary data sources for more than half of the service coverage index components. This yields 76 countries which account for 62% of the world’s population in 2015, and includes 9 countries classi ed as low-income in 2015 (32% of the population living in LICs in 2015), 23 lower-middle-income countries in 2015 (87% of LMIC population), 21 upper-middle-income countries in 2015 (37% of UMIC population) and 23 high-income countries in 2015 (69% of HIC population). The gure below is divided into four zones delimited by the median value of the service coverage index across 183 countries and median incidence of catastrophic spending at the 10% threshold across 132 countries for which there are primary data. Using median incidence of catastrophic spending at the 25% threshold only yield di erent results in a few cases explicitly discussed hereafter but not shown in the gure. Median values are of course in uenced by the composition of the sample so any cross-country comparison is relative without any suggestion that median values identify targets. A total of 22 countries have comparatively high rates of service coverage and nancial protection, which is the aim of UHC (in Z-IV). Another 22 countries have very high values of the service coverage index but their incidence of catastrophic spending is also comparatively high (in Z-I). All 23 high-income countries have above median levels of service coverage (in Z-IV or Z-I) but not all of them perform equally well in protecting households from catastrophic expenditures. The proportion of people devoting more than 10% of their household budget to health is above median incidence rates in 39% HICs and 43% at the 25% threshold. Within a country the incidence of catastrophic expenditures at 25% threshold is always lower than its rate at the 10% threshold but across countries a country could be above median with one threshold but not with the other. In 16 countries, many people are incurring high out-of-pocket expenses, yet average service coverage is low (in Z-II). For another 16 countries, the challenge would be to increase service coverage without increasing nancial hardship (in Z-III). Out of 9 low-income countries 7 are in Z-III. In these countries, service coverage is low and that might be why the fraction of the population spending more than 10% or 25% of their budget on health is also low. The 23 LMICs analyzed are mostly characterized by below median levels of service coverage index (in Z-II or Z-III) and upper-middle-income ones by above median levels of service coverage index (in Z-IV or Z-I). In terms of incidence of catastrophic spending on health, the 23 LMICs are mostly above the median incidence rate whereas the evidence is mixed for UMICs with most of them characterized by above median incidence rates at the 10% threshold but below median incidence rates at the 25% threshold. Joint visualization of service coverage index and incidence of catastrophic spending across countries de ned as out-of-pocket expenditures exceeding 10% of household total consumption or income 50 Z−II BELOW median service coverage score Z−I ABOVE median service coverage score 40 & & Incidence of catastrophic health spending (SDG 3.8.2 – 10% threshold, latest year) ABOVE median incidence of ABOVE median incidence of catastrophic spending catastrophic spending 30 20 10 0 Z−III Z−IV −20 −10 BELOW median service coverage score ABOVE median service coverage score & & BELOW median incidence of BELOW median incidence of catastrophic spending catastrophic spending 0 10 20 30 40 50 60 70 80 90 100 Service coverage index Low income Lower middle income Upper middle income High income Notes: total number of countries 76 − 9 LMICs; 23 LMICs; 21 UMICs; 23 HICs Source: WHO−World Bank global database as of December 2017 44 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT References 8. Ferreira FHG. A global count of the extreme poor in 2012: data issues, methodology and initial results. Journal of Economic Inequality. 2016;14(2):141-72. 1. Wagstaff A, van Doorslaer E. Catastrophe and impoverishment 9. Wagstaff A, Flores G, Smitz M-F, Hsu J, Chepynoga K, Eozenou in paying for health care: with applications to Vietnam 1993- P. Progress on impoverishing health spending: results for 122 1998. Health Economics. 2003;12(11):921-34. countries. A retrospective observational study. 2017. DOI: http://dx.doi.org/10.1016/S2214-109X(17)30486-2 2. Xu K, Evans DB, Carrin G, Aguilar-Rivera AM, Musgrove P, Evans T. Protecting households from catastrophic health spending. 10. Wagstaff A, Flores G, Hsu J, Smitz M-F, Chepynoga K, Buisman Health Aff (Millwood). 2007;26(4):972-83. LR et al. Progress on catastrophic health spending: results for 133 countries. A retrospective observational study. Lancet 3. Xu K, Evans DB, Kawabata K, Zeramdini R, Klavus J, Murray Global Health. 2017. DOI: http://dx.doi.org/10.1016/S2214- CJ. Household catastrophic health expenditure: a multicountry 109X(17)30429-1 analysis. Lancet. 2003;362(9378):111-7. 11. A measured approach to ending poverty and boosting shared 4. Wagstaff A, Eozenou P. CATA meets IMPOV: a unified approach prosperity: concepts, data and the twin goals [Policy Research to measuring financial protection in health [Policy Research Report]. Washington DC: The World Bank; 2015. Working Paper Series: 6861]. Washington DC: The World Bank; 2014. 12. Thomson S, Cylus J, Evetovits T. How you measure matters: monitoring financial protection to generate policy-relevant 5 Flores G, Krishnakumar J, O’Donnell O, van Doorslaer E. Coping evidence for universal health coverage – lessons from Europe. with health-care costs: implications for the measurement of Organized session 9 July 2017. In: IHEA 12th World Congress catastrophic expenditures and poverty. Health Econ. 2008 Dec in Health Economics, 8–11 July 2017, Boston, MA, USA. 17(12):1393-412. 13. Taube M, Vaskis E, Nesterenko O. Can people afford to pay for 6. Deaton A, Zaidi S. Guidelines for constructing consumption health care? New evidence from Latvia. Copenhagen: WHO aggregates for welfare analysis. Living Standards Measurement Regional Office for Europe; forthcoming 2018. Study Working Paper, no. 135. Washington DC: The World Bank; 2002. 7. Thomson S, Evetovits T, Cylus J, Jakab M. Monitoring financial protection to assess progress towards universal health coverage in Europe. Public Health Panorama. 2016;2(3):357-66. CHAPTER 2. FINANCIAL PROTECTION 45 ANNEXES Annex 1. UHC indicators (service coverage and nancial protection) by country SDG-UHC indicator 3.8.2, latest year: Annex 1 incidence of catastrophic expenditure (%) SDG-UHC indicator SDG-UHC indicator at 10% of at 25% of 3.8.1: Service Data availability to Availability of 3.8.2, most recent household total household total coverage index, construct SDG-UHC estimates for SDG- available estimate consumption or consumption or Country 2015 3.8.1a UHC indicator 3.8.2 (year) income income Afghanistan 34 high yes 2007 4.84 0.07 Albania 62 low yes 2012 16.72 4.95 Algeria 76 high no – – – Angola 36 medium yes 2008 12.38 4.54 Antigua and Barbuda 75 medium no – – – Argentina 76 high yes 2004 16.90 4.70 Armenia 67 high yes 2013 16.05 4.87 Australia ≥80 high yes 2010 3.71 0.50 Austria ≥80 medium yes 1999 4.31 0.66 Azerbaijan 64 medium yes 2005 8.12 1.10 Bahamas 72 medium no – – – Bahrain 72 medium no – – – Bangladesh 46 high yes 2010 13.57 4.84 Barbados 79 high no – – – Belarus 74 high yes 2012 4.38 0.15 Belgium ≥80 medium yes 2010 11.45 1.39 Belize 61 high no – – – Benin 41 high yes 2003 11.11 0.85 Bhutan 59 high no – – – Bolivia (Plurinational State of) 60 medium yes 2002 8.23 3.20 Bosnia and Herzegovina 57 high yes 2011 8.56 1.27 Botswana 60 medium yes 1993 8.54 1.82 Brazil 77 high yes 2008 25.56 3.46 Brunei Darussalam ≥80 medium no – – – Bulgaria 64 medium yes 2010 12.84 0.76 Burkina Faso 39 high yes 2009 3.52 0.62 Burundi 43 high yes 2006 15.03 4.25 Cabo Verde 62 medium yes 2007 2.05 0.02 Cambodia 55 high yes 2009 19.97 5.64 Cameroon 44 high yes 2014 10.78 2.98 Canada ≥80 low yes 2010 2.64 0.51 Central African Republic 33 medium no – – – Chad 29 medium yes 2003 6.28 0.22 Chile 70 medium yes 2006 33.07 11.52 China 76 medium yes 2007 17.71 4.76 Colombia 76 high yes 2010 16.92 2.82 Comoros 47 high no – – – Congo 38 high yes 2011 1.97 0.36 Costa Rica 75 high yes 2012 10.13 1.81 Côte d'Ivoire 44 medium yes 2008 15.19 3.57 Croatia 69 high yes 2010 2.80 0.26 Cuba 78 high no – – – Cyprus 73 low yes 2010 16.07 1.50 Czechia 73 medium yes 2010 2.22 0.05 Democratic People's Republic of 68 low no – – – Korea Democratic Republic of the Congo 40 medium yes 2004 5.81 0.84 48 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Poverty gap due to out-of-pocket Incidence of impoverishment due to out- health spending expressed in cents of of-pocket health spending (%) international dollars at 2011 PPP factors Availability of estimates on Poverty line: Poverty line: Poverty line: Poverty line: impoverishing at 2011 PPP at 2011 PPP at 2011 PPP at 2011 PPP spending on health $1.90-a-day $3.10-a-day $1.90-a-day $3.10-a-day Country yes 0.58 2.55 0.25 2.16 Afghanistan yes 0.36 1.42 0.12 1.08 Albania no – – – – Algeria yes 2.01 2.55 1.47 4.36 Angola no – – – – Antigua and Barbuda yes 0.24 0.62 0.11 0.63 Argentina yes 0.49 2.57 0.18 1.83 Armenia yes 0.00 0.00 0.00 0.00 Australia no – – – – Austria yes 0.00 0.00 0.00 0.00 Azerbaijan no – – – – Bahamas no – – – – Bahrain yes 4.51 4.08 6.47 12.23 Bangladesh no – – – – Barbados yes 0.00 0.00 0.00 0.00 Belarus yes 0.00 0.00 0.00 0.00 Belgium no – – – – Belize yes 3.09 2.04 2.96 6.12 Benin no – – – – Bhutan yes 0.74 1.67 0.60 1.95 Bolivia (Plurinational State of) yes 0.00 0.03 0.00 0.02 Bosnia and Herzegovina yes 0.87 1.09 1.18 2.42 Botswana yes 1.04 2.01 0.74 2.58 Brazil no – – – – Brunei Darussalam yes 0.00 0.13 0.00 0.11 Bulgaria yes 1.15 0.93 1.55 2.70 Burkina Faso yes 2.05 1.03 3.23 4.97 Burundi yes 0.14 0.26 0.09 0.48 Cabo Verde yes 2.99 6.15 8.19 13.94 Cambodia yes 1.86 1.86 1.15 3.44 Cameroon yes 0.03 0.03 0.12 0.16 Canada no – – – – Central African Republic yes 1.36 0.82 2.03 3.46 Chad yes 0.65 2.59 0.73 2.65 Chile yes 2.13 3.09 1.22 4.49 China yes 0.47 0.91 0.20 0.97 Colombia no – – – – Comoros yes 0.71 1.05 0.70 1.74 Congo yes 0.09 0.43 0.09 0.36 Costa Rica yes 2.98 3.34 2.22 6.02 Côte d'Ivoire yes 0.00 0.00 0.00 0.00 Croatia no – – – – Cuba yes 0.00 0.00 0.00 0.00 Cyprus yes 0.00 0.00 0.00 0.00 Czechia Democratic People's Republic of no – – – – Korea yes 0.12 0.12 1.42 1.59 Democratic Republic of the Congo ANNEXES 49 SDG-UHC indicator 3.8.2, latest year: Annex 1 incidence of catastrophic expenditure (%) SDG-UHC indicator SDG-UHC indicator at 10% of at 25% of 3.8.1: Service Data availability to Availability of 3.8.2, most recent household total household total coverage index, construct SDG-UHC estimates for SDG- available estimate consumption or consumption or Country 2015 3.8.1a UHC indicator 3.8.2 (year) income income Denmark ≥80 medium yes 2010 2.93 0.49 Djibouti 47 medium yes 1996 1.42 0.04 Dominican Republic 74 high yes 2007 17.00 4.36 Ecuador 75 high yes 1998 15.23 3.28 Egypt 68 high yes 2012 26.20 3.90 El Salvador 77 high no – – – Equatorial Guinea 45 medium no – – – Eritrea 38 high no – – – Estonia 76 medium yes 2010 8.79 1.19 Ethiopia 39 high yes 2004 0.82 0.18 Fiji 66 high yes 2002 3.37 0.24 Finland 79 medium yes 2010 6.35 0.97 France ≥80 medium no – – – Gabon 52 medium yes 2005 5.67 0.22 Gambia 46 high no – – – Georgia 66 high yes 2013 29.21 8.98 Germany 79 medium yes 1993 1.41 0.07 Ghana 45 high yes 2005 3.11 0.49 Greece 70 low yes 2010 14.64 1.78 Grenada 72 medium no – – – Guatemala 57 high yes 2014 1.36 0.04 Guinea 35 medium yes 2012 6.97 1.25 Guinea-Bissau 39 medium no – – – Guyana 68 high no – – – Haiti 47 high no – – – Honduras 64 medium yes 1998 3.45 0.43 Hungary 70 medium yes 2010 7.38 0.31 Iceland ≥80 medium yes 1995 6.90 0.94 India 56 high yes 2011 17.33 3.90 Indonesia 49 high yes 2015 3.61 0.41 Iran (Islamic Republic of) 65 high no 2013 15.81 3.76 Iraq 63 medium no – – – Ireland 78 medium yes 2010 6.40 0.69 Israel ≥80 medium yes 2012 6.72 0.95 Italy ≥80 medium yes 2010 9.29 1.08 Jamaica 60 medium yes 2004 10.20 2.88 Japan ≥80 medium yes 2008 6.17 2.01 Jordan 70 medium yes 2006 5.31 0.91 Kazakhstan 71 high yes 2013 1.83 0.08 Kenya 57 high yes 2005 5.83 1.51 Kiribati 40 low no – – – Kuwait 77 medium no – – – Kyrgyzstan 66 high yes 2011 3.54 0.81 Lao People's Democratic Republic 48 high yes 2007 2.98 0.26 Latvia 64 medium yes 2006 10.91 1.83 Lebanon 68 medium yes 1999 44.85 10.03 Lesotho 45 high yes 2002 1.80 0.21 Liberia 34 high yes 2007 7.86 1.60 Libya 63 low no – – – Lithuania 67 medium yes 2010 9.79 1.64 Luxembourg ≥80 medium yes 2010 3.38 0.15 Madagascar 30 medium yes 2005 0.77 0.03 Malawi 44 high yes 2010 1.64 0.10 50 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Poverty gap due to out-of-pocket Incidence of impoverishment due to out- health spending expressed in cents of of-pocket health spending (%) international dollars at 2011 PPP factors Availability of estimates on Poverty line: Poverty line: Poverty line: Poverty line: impoverishing at 2011 PPP at 2011 PPP at 2011 PPP at 2011 PPP spending on health $1.90-a-day $3.10-a-day $1.90-a-day $3.10-a-day Country yes 0.00 0.00 0.00 0.00 Denmark yes 0.13 0.05 0.08 0.35 Djibouti yes 0.41 1.63 0.12 1.36 Dominican Republic no – – – – Ecuador yes 0.12 1.07 0.04 0.62 Egypt no – – – – El Salvador no – – – – Equatorial Guinea no – – – – Eritrea yes 0.00 0.08 0.01 0.06 Estonia yes 0.44 0.35 0.42 0.90 Ethiopia no – – – – Fiji yes 0.00 0.00 0.00 0.00 Finland no – – – – France yes 0.64 1.11 0.20 1.06 Gabon no – – – – Gambia yes 3.07 5.33 2.18 7.28 Georgia yes 0.00 0.00 0.00 0.00 Germany no – – – – Ghana yes 0.00 0.00 0.00 0.00 Greece no – – – – Grenada yes 0.29 0.22 0.03 0.26 Guatemala yes 2.48 1.46 9.07 11.61 Guinea no – – – – Guinea-Bissau no – – – – Guyana no – – – – Haiti no – – – – Honduras yes 0.00 0.03 0.00 0.02 Hungary yes 0.00 0.00 0.00 0.00 Iceland yes 4.16 4.61 2.13 7.69 India yes 0.07 0.66 0.01 0.39 Indonesia no 0.1 0.01 0.17 0.09 Iran (Islamic Republic of) no – – – – Iraq yes 0.00 0.00 0.00 0.00 Ireland yes 0.00 0.00 0.00 0.00 Israel yes 0.00 0.00 0.00 0.00 Italy yes 0.50 1.16 0.24 1.22 Jamaica no – – – – Japan no – – – – Jordan yes 0.00 0.02 0.00 0.01 Kazakhstan yes 1.36 1.61 1.52 3.36 Kenya no – – – – Kiribati no – – – – Kuwait yes 0.33 0.90 0.16 0.91 Kyrgyzstan yes 0.40 0.99 0.18 1.21 Lao People's Democratic Republic yes 0.04 0.11 0.00 0.06 Latvia yes 0.03 0.03 0.01 0.04 Lebanon no – – – – Lesotho yes 2.19 0.62 2.86 4.14 Liberia no – – – – Libya yes 0.00 0.01 0.00 0.00 Lithuania yes 0.00 0.00 0.00 0.00 Luxembourg yes 0.20 0.11 0.60 0.77 Madagascar yes 0.52 0.33 0.86 1.41 Malawi ANNEXES 51 SDG-UHC indicator 3.8.2, latest year: Annex 1 incidence of catastrophic expenditure (%) SDG-UHC indicator SDG-UHC indicator at 10% of at 25% of 3.8.1: Service Data availability to Availability of 3.8.2, most recent household total household total coverage index, construct SDG-UHC estimates for SDG- available estimate consumption or consumption or Country 2015 3.8.1a UHC indicator 3.8.2 (year) income income Malaysia 70 high yes 2004 0.74 0.04 Maldives 55 low yes 2009 20.14 1.61 Mali 32 high yes 2006 3.38 0.09 Malta 79 medium yes 2010 15.93 2.81 Mauritania 33 medium yes 2004 10.54 1.81 Mauritius 64 medium yes 1996 6.79 1.02 Mexico 76 high yes 2012 7.13 1.91 Micronesia (Federated States of) 60 low no – – – Mongolia 63 high yes 2012 2.39 0.46 Montenegro 54 high yes 2014 8.86 0.96 Morocco 65 medium yes 2006 22.00 2.70 Mozambique 42 high yes 2008 1.19 0.31 Myanmar 60 high no – – – Namibia 59 high no – – – Nepal 46 high yes 2010 27.41 3.31 Netherlands ≥80 medium no – – – New Zealand ≥80 medium no – – – Nicaragua 70 medium yes 2014 27.74 8.89 Niger 33 high yes 2011 4.14 0.36 Nigeria 39 high yes 2009 24.77 8.92 Norway ≥80 medium yes 1998 5.09 0.50 Oman 72 medium yes 1999 0.63 0.10 Pakistan 40 high yes 2010 1.03 0.02 Panama 75 high yes 2008 1.41 0.22 Papua New Guinea 41 low no – – – Paraguay 69 medium yes 2001 10.32 2.04 Peru 78 high yes 2015 8.29 1.21 Philippines 58 high yes 2015 6.31 1.41 Poland 75 medium yes 2012 13.93 1.61 Portugal ≥80 medium yes 2010 18.38 3.31 Qatar 77 high no – – – Republic of Korea ≥80 high yes 2008 13.53 4.01 Republic of Moldova 65 high yes 2013 16.05 3.56 Romania 72 medium yes 2012 11.99 2.29 Russian Federation 63 medium yes 2014 4.87 0.60 Rwanda 53 high yes 2010 4.61 0.70 Saint Lucia 69 medium no – – – Saint Vincent and the Grenadines 65 low no – – – Samoa 56 high no – – – Sao Tome and Principe 54 medium yes 2000 10.20 0.96 Saudi Arabia 68 medium no – – – Senegal 41 high yes 2011 3.33 0.19 Serbia 65 high yes 2010 9.04 0.74 Seychelles 68 medium no – – – Sierra Leone 36 high yes 2003 10.42 0.64 Singapore ≥80 medium no – – – Slovakia 76 medium yes 2010 3.77 0.44 Slovenia 78 medium yes 2012 2.90 0.26 Solomon Islands 50 medium no – – – Somalia 22 medium no – – – South Africa 67 medium yes 2010 1.41 0.12 South Sudan 30 medium no – – – Spain 77 medium yes 2010 5.73 1.21 52 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Poverty gap due to out-of-pocket Incidence of impoverishment due to out- health spending expressed in cents of of-pocket health spending (%) international dollars at 2011 PPP factors Availability of estimates on Poverty line: Poverty line: Poverty line: Poverty line: impoverishing at 2011 PPP at 2011 PPP at 2011 PPP at 2011 PPP spending on health $1.90-a-day $3.10-a-day $1.90-a-day $3.10-a-day Country yes 0.09 0.23 0.02 0.13 Malaysia yes 0.52 0.63 0.19 1.29 Maldives yes 1.91 0.95 1.14 2.40 Mali yes 0.00 0.00 0.00 0.00 Malta yes 0.65 1.21 0.30 1.52 Mauritania no – – – – Mauritius yes 0.28 0.69 0.08 0.66 Mexico no – – – – Micronesia (Federated States of) yes 0.02 0.15 0.01 0.10 Mongolia yes 0.00 0.26 0.00 0.04 Montenegro yes 0.63 3.18 0.27 2.42 Morocco yes 0.23 0.12 0.47 0.70 Mozambique no – – – – Myanmar no – – – – Namibia yes 1.85 5.63 1.00 6.11 Nepal no – – – – Netherlands no – – – – New Zealand yes 3.08 5.20 5.35 10.08 Nicaragua yes 1.64 0.99 1.51 3.21 Niger yes 3.72 3.63 3.33 7.87 Nigeria yes 0.00 0.00 0.00 0.00 Norway no – – – – Oman yes 1.00 2.44 0.28 2.53 Pakistan yes 0.04 0.04 0.02 0.09 Panama no – – – – Papua New Guinea yes 0.31 0.51 0.08 0.76 Paraguay yes 0.05 0.20 0.01 0.15 Peru yes 0.83 1.44 0.41 1.86 Philippines yes 0.00 0.09 0.00 0.03 Poland yes 0.00 0.00 0.00 0.00 Portugal no – – – – Qatar yes 0.00 0.04 0.00 0.02 Republic of Korea yes 0.00 0.43 0.01 0.27 Republic of Moldova yes 0.00 0.30 0.00 0.14 Romania yes 0.00 0.01 0.01 0.01 Russian Federation yes 0.94 0.39 1.81 2.56 Rwanda no – – – – Saint Lucia no – – – – Saint Vincent and the Grenadines no – – – – Samoa no – – – – Sao Tome and Principe no – – – – Saudi Arabia yes 1.10 1.42 1.24 3.06 Senegal yes 0.05 0.16 0.01 0.14 Serbia no – – – – Seychelles yes 2.56 1.30 3.47 5.78 Sierra Leone no – – – – Singapore yes 0.00 0.02 0.00 0.00 Slovakia yes 0.00 0.00 0.00 0.00 Slovenia no – – – – Solomon Islands no – – – – Somalia yes 0.45 0.61 0.33 0.95 South Africa no – – – – South Sudan yes 0.00 0.00 0.00 0.00 Spain ANNEXES 53 SDG-UHC indicator 3.8.2, latest year: Annex 1 incidence of catastrophic expenditure (%) SDG-UHC indicator SDG-UHC indicator at 10% of at 25% of 3.8.1: Service Data availability to Availability of 3.8.2, most recent household total household total coverage index, construct SDG-UHC estimates for SDG- available estimate consumption or consumption or Country 2015 3.8.1a UHC indicator 3.8.2 (year) income income Sri Lanka 62 medium yes 2009 2.89 0.10 Sudan 43 medium no – – – Suriname 68 high no – – – Swaziland 58 high yes 2009 13.39 2.04 Sweden ≥80 medium yes 1996 5.53 0.69 Switzerland ≥80 medium yes 2004 19.70 6.68 Syrian Arab Republic 60 low no – – – Tajikistan 65 medium yes 2007 11.30 2.72 Thailand 75 high yes 2010 3.38 0.68 The former Yugoslav Republic of 70 medium yes 2008 5.44 0.57 Macedonia Timor-Leste 47 medium yes 2001 2.59 0.00 Togo 42 high yes 2006 10.65 0.02 Tonga 62 medium no – – – Trinidad and Tobago 75 medium no – – – Tunisia 65 high yes 2010 16.69 2.37 Turkey 71 high yes 2012 3.10 0.32 Turkmenistan 67 medium no – – – Uganda 44 high yes 2002 12.01 2.57 Ukraine 63 high yes 2013 7.21 1.07 United Arab Emirates 63 low no – – – United Kingdom ≥80 medium yes 2013 1.64 0.48 United Republic of Tanzania 39 high yes 2012 9.87 2.48 United States of America ≥80 high yes 2013 4.77 0.78 Uruguay 79 high yes 1995 13.87 1.85 Uzbekistan 72 medium no – – – Vanuatu 56 high no – – – Venezuela (Bolivarian Republic of) 73 medium no – – – Viet Nam 73 high yes 2014 9.81 2.07 Yemen 39 high yes 2005 17.06 2.40 Zambia 56 high yes 2010 0.29 0.01 Zimbabwe 55 high no – – – SDG: sustainable development goals; UHC: universal health coverage. a Data availability is classi ed as follows, based on information available in global data bases: high=75% or more of the tracer indicators with primary data since 2010; medium=50% or more (but less than 75%) of the tracer indicators with primary data since 2010; low=less than 50% of tracer indicators with primary data since 2010. ‘Primary data’ refers to original data sources and excludes estimates based on modelling and predictions. 54 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Poverty gap due to out-of-pocket Incidence of impoverishment due to out- health spending expressed in cents of of-pocket health spending (%) international dollars at 2011 PPP factors Availability of estimates on Poverty line: Poverty line: Poverty line: Poverty line: impoverishing at 2011 PPP at 2011 PPP at 2011 PPP at 2011 PPP spending on health $1.90-a-day $3.10-a-day $1.90-a-day $3.10-a-day Country yes 0.05 0.44 0.01 0.29 Sri Lanka no – – – – Sudan no – – – – Suriname yes 1.36 1.29 2.24 3.79 Swaziland yes 0.00 0.00 0.00 0.00 Sweden yes 0.00 0.00 0.00 0.00 Switzerland no – – – – Syrian Arab Republic yes 1.42 3.39 0.53 3.74 Tajikistan yes 0.12 0.34 0.83 1.09 Thailand The former Yugoslav Republic of yes 0.09 0.28 0.09 0.38 Macedonia yes 1.00 0.65 0.80 1.85 Timor-Leste yes 2.54 1.59 2.72 5.05 Togo no – – – – Tonga no – – – – Trinidad and Tobago yes 0.44 1.17 0.19 1.19 Tunisia yes 0.09 0.20 0.00 0.13 Turkey no – – – – Turkmenistan yes 2.68 1.48 3.39 5.71 Uganda yes 0.00 0.02 0.00 0.02 Ukraine no – – – – United Arab Emirates yes 0.00 0.00 0.00 0.00 United Kingdom yes 2.38 1.86 2.26 4.96 United Republic of Tanzania yes 0.00 0.00 0.00 0.00 United States of America yes 0.04 0.27 0.01 0.11 Uruguay no – – – – Uzbekistan no – – – – Vanuatu no – – – – Venezuela (Bolivarian Republic of) no – – – – Viet Nam no – – – – Yemen yes 0.14 0.10 0.16 0.28 Zambia no – – – – Zimbabwe ANNEXES 55 Annex 2. Current values of the UHC index of coverage of essential health services and values of each of the tracer indicators used to calculate the index, by countrya Values are for 2015 unless otherwise noted. Values displayed have not been transformed or rescaled for incorporation into the index calculations. For tracer indicators, data are displayed in bold font if primary data were available since 2010; normal font if primary data were available since 2000; and faded font if estimates for the country were imputed without primary country data. Annex 2 Family planning demand Care- satisifed seeking Insecticide- UHC service with Antenatal behaviour Tuberculosis treated nets coverage modern care, Child for child effective HIV for malaria index Index data methods 4+ visits immunization pneumonia treatment treatment prevention Country (SDG 3.8.1) availabilityb (%) (%)c (DTP3) (%) (%)c (%)d (%) (%)e Afghanistan 34 high 43 18 65 62 51 5 – Albania 62 low 25 67 99 70 67 28 – Algeria 76 high 76 67 95 66 70 65 – Angola 36 medium 25 56 64 39 22 23 40 Antigua and Barbuda 75 medium 80 100 99 88 58 46 – Argentina 76 high 85 90 94 94 45 63 – Armenia 67 high 40 96 94 57 69 29 – Australia ≥80 high 84 95 93 90 69 79 – Austria ≥80 medium 84 97 93 92 64 72 – Azerbaijan 64 medium 32 66 96 33 67 34 – Bahamas 72 medium 83 85 95 77 73 29 – Bahrain 72 medium 59 100 98 90 38 42 – Bangladesh 46 high 74 31 97 42 53 13 – Barbados 79 high 76 98 97 87 87 44 – Belarus 74 high 74 100 99 93 63 42 – Belgium ≥80 medium 90 97 99 91 71 72 – Belize 61 high 68 83 94 67 31 29 – Benin 41 high 24 59 82 23 54 51 69 Bhutan 59 high 83 85 99 74 72 13 – Bolivia (Plurinational State of) 60 medium 52 59 99 62 52 22 – Bosnia and Herzegovina 57 high 27 84 82 87 59 28 – Botswana 60 medium 78 73 95 37 48 77 – Brazil 77 high 88 90 96 50 62 57 – Brunei Darussalam ≥80 medium 83 100 99 86 56 72 – Bulgaria 64 medium 59 88 91 76 83 24 – Burkina Faso 39 high 43 34 91 56 49 55 62 Burundi 43 high 38 49 94 55 46 50 77 Cabo Verde 62 medium 77 72 93 70 34 50 – Cambodia 55 high 59 76 89 69 55 76 – Cameroon 44 high 38 59 84 28 45 30 45 Canada ≥80 low 89 99 91 90 74 72 – Central African Republic 33 medium 37 38 47 30 39 22 49 Chad 29 medium 18 31 46 26 37 54 70 Chile 70 medium 82 97 96 87 51 49 – China 76 medium 95 74 99 79 82 41 – Colombia 76 high 83 89 91 64 61 53 – Comoros 47 high 33 49 91 38 53 32 57 Congo 38 high 37 79 80 28 39 29 36 Costa Rica 75 high 90 90 92 77 71 45 – Côte d'Ivoire 44 medium 34 44 83 38 49 34 73 56 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT International Health Mean Regulations At least fasting Tobacco Hospital core basic Normal plasma non- beds per Physicians Psychiatrists Surgeons capacity sanitation blood glucose smoking 10 000 per 1000 per 100 000 per 100 000 index (%) pressure (%) (mmol/L)f (%) populationc populationc populationc populationc (%)c Country 39 69 5.36 87 5 0.3 0.1 0.9 43 Afghanistan 98 71 5.41 71 28.9 1.3 1.3 5.5 81 Albania 88 75 5.42 85 19 1.2 3.2 20.6 72 Algeria 39 70 5.25 87 8 0.1 <0.05 5.1 18 Angola 88 77 5.65 87 38 1.2 1.1 5.5 87 Antigua and Barbuda 95 77 5.55 77 50 3.8 13.7 18.3 83 Argentina 92 75 5.64 73 41.8 2.8 5.1 63.3 96 Armenia 100 80 5.51 85 37.9 3.5 13.7 20.3 100 Australia 100 79 5.24 69 76.5 5.2 19.7 91.2 87 Austria 89 76 5.72 78 46.9 3.4 3.7 41.3 84 Azerbaijan 92 79 5.82 88 29 2.7 1.1 21 59 Bahamas 100 79 5.79 79 20.3 0.9 4.8 15.1 96 Bahrain 47 75 5.45 77 7.7 0.4 0.1 1.7 85 Bangladesh 97 76 5.9 92 58 1.8 2.5 12.3 90 Barbados 94 73 5.48 71 110.5 4.1 7.7 93.6 90 Belarus 100 83 5.39 71 62.3 3 20.3 50.3 82 Belgium 87 77 5.54 87 13 0.8 0.6 5.6 55 Belize 14 72 5.07 93 5 0.1 0.1 0.8 44 Benin 63 72 4.9 94 17.4 0.3 0.4 0.8 65 Bhutan 53 82 5.4 87 11 0.5 1 17.6 67 Bolivia (Plurinational State of) 95 69 5.57 61 35 1.9 4 7.2 55 Bosnia and Herzegovina 60 71 5.32 80 18 0.4 0.3 1.6 56 Botswana 86 77 5.52 86 22 1.9 3.5 34.4 97 Brazil 96 81 5.33 84 27.4 1.5 4.3 22.5 91 Brunei Darussalam 86 72 5.41 62 68.2 4 7.9 64.1 69 Bulgaria 23 68 5.31 87 4 <0.05 0.1 0.2 50 Burkina Faso 51 71 4.87 87 7.9 0.1 <0.05 0.2 56 Burundi 65 71 6.05 91 21 0.3 1.4 11.5 58 Cabo Verde 49 74 4.73 82 8.3 0.2 0.3 0.8 51 Cambodia 39 75 5.53 87 13 0.1 <0.05 0.4 91 Cameroon 99 87 5.54 85 27 2.5 13.4 21.1 100 Canada 25 69 5.13 87 10 <0.05 <0.05 0.2 24 Central African Republic 10 67 5.3 87 4 <0.05 <0.05 0.1 41 Chad 100 79 5.5 61 22 1 4.7 41 75 Chile 75 81 5.46 75 42 1.5 1.7 21.6 99 China 84 81 5 90 15 1.6 2.5 5.8 85 Colombia 34 72 5.16 86 21.6 0.2 0.1 1.3 29 Comoros 15 74 5.2 76 16 0.1 0.1 0.2 28 Congo 97 81 5.52 88 11.6 2.5 5.2 7.1 83 Costa Rica 30 73 5.39 87 4 0.1 0.1 1.5 87 Côte d'Ivoire ANNEXES 57 Annex 2 Family planning demand Care- satisifed seeking Insecticide- UHC service with Antenatal behaviour Tuberculosis treated nets coverage modern care, Child for child effective HIV for malaria index Index data methods 4+ visits immunization pneumonia treatment treatment prevention Country (SDG 3.8.1) availabilityb (%) (%)c (DTP3) (%) (%)c (%)d (%) (%)e Croatia 69 high 59 93 94 90 62 65 – Cuba 78 high 88 99 99 93 71 69 – Cyprus 73 low 83 97 97 91 51 72 – Czechia 73 medium 83 97 97 88 71 46 – Democratic People's Republic of 68 low 85 94 96 80 73 41 – Korea Democratic Republic of the Congo 40 medium 18 48 81 42 43 32 59 Denmark ≥80 medium 83 97 93 92 49 72 – Djibouti 47 medium 43 23 84 94 65 22 30 Dominican Republic 74 high 84 93 85 73 59 46 – Ecuador 75 high 82 80 78 72 46 50 – Egypt 68 high 80 83 93 68 50 21 – El Salvador 77 high 80 90 91 80 84 45 – Equatorial Guinea 45 medium 25 67 16 54 50 32 26 Eritrea 38 high 27 57 95 45 56 57 19 Estonia 76 medium 77 97 93 89 73 72 – Ethiopia 39 high 58 32 77 29 63 55 61 Fiji 66 high 67 94 99 72 70 31 – Finland 79 medium 88 98 97 92 39 72 – France ≥80 medium 93 99 98 91 67 75 – Gabon 52 medium 37 78 80 68 41 56 11 Gambia 46 high 28 78 97 68 64 24 64 Georgia 66 high 53 87 94 74 66 28 – Germany 79 medium 82 97 95 91 55 72 – Ghana 45 high 43 87 88 56 28 28 66 Greece 70 low 59 97 99 89 71 72 – Grenada 72 medium 79 89 92 82 87 46 – Guatemala 57 high 67 86 74 50 68 36 – Guinea 35 medium 20 57 54 37 46 28 57 Guinea-Bissau 39 medium 40 65 87 34 25 25 78 Guyana 68 high 57 87 95 84 55 56 – Haiti 47 high 49 67 60 38 62 46 – Honduras 64 medium 77 89 97 64 74 48 – Hungary 70 medium 85 88 99 87 69 28 – Iceland ≥80 medium 83 97 92 94 77 72 – India 56 high 72 45 87 77 44 44 – Indonesia 49 high 81 84 78 75 27 10 – Iran (Islamic Republic of) 65 high 76 94 98 76 70 11 – Iraq 63 medium 62 50 58 74 48 43 – Ireland 78 medium 79 97 95 91 49 70 – Israel ≥80 medium 71 97 95 91 77 72 – Italy ≥80 medium 67 87 93 92 79 76 – Jamaica 60 medium 83 86 91 82 14 32 – Japan ≥80 medium 65 97 96 89 46 72 – Jordan 70 medium 62 95 99 77 70 50 – Kazakhstan 71 high 75 95 98 81 80 26 – Kenya 57 high 76 58 89 66 66 57 62 Kiribati 40 low 43 71 78 81 70 41 – Kuwait 77 medium 67 71 99 82 84 71 – Kyrgyzstan 66 high 66 95 97 60 69 28 – Lao People's Democratic Republic 48 high 67 37 89 54 32 35 – Latvia 64 medium 77 88 95 87 72 14 – 58 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT International Health Mean Regulations At least fasting Tobacco Hospital core basic Normal plasma non- beds per Physicians Psychiatrists Surgeons capacity sanitation blood glucose smoking 10 000 per 1000 per 100 000 per 100 000 index (%) pressure (%) (mmol/L)f (%) populationc populationc populationc populationc (%)c Country 98 68 5.37 63 55.6 3.1 15.5 68.2 71 Croatia 91 81 5.67 64 52 7.5 10.3 70.6 99 Cuba 99 80 5.45 64 34.2 2.5 2.7 12.8 62 Cyprus 99 72 5.51 66 64.9 3.7 14.1 73.6 88 Czechia Democratic People's Republic of 77 82 5.13 75 143 2.8 0.5 0.6 73 Korea 20 72 5.06 87 8 0.1 0.1 0.1 75 Democratic Republic of the Congo 100 79 5.34 80 25.3 3.7 17.4 58.7 91 Denmark 51 73 5.4 87 14 0.2 0.1 1.5 46 Djibouti 83 79 5.24 86 16 1.5 1.1 9 70 Dominican Republic 86 82 5.39 93 15 1.7 1.1 34.3 89 Ecuador 93 75 5.1 75 15.6 0.8 0.9 26.8 93 Egypt 91 81 5.61 89 13 1.9 0.5 20.3 94 El Salvador 75 72 5.28 87 21 0.3 0.1 32.7 27 Equatorial Guinea 11 71 5.12 94 7 0.1 <0.05 0.4 71 Eritrea 100 72 5.25 68 49.6 3.3 18.5 82.3 72 Estonia 7 79 4.48 96 3.1 0.1 <0.05 0.4 78 Ethiopia 96 78 5.98 77 23 0.4 0.7 1.8 98 Fiji 99 81 5.5 79 43.5 3.2 23.6 56.4 96 Finland 99 78 5.31 67 64.8 3.2 14.1 29.4 89 France 41 75 5.39 87 13 0.3 0.3 16.4 48 Gabon 42 71 5.56 84 11 0.1 0.2 0.6 33 Gambia 85 74 5.64 70 25.9 4.8 6.5 42.5 81 Georgia 99 80 5.45 69 82.8 4.1 7.5 55.2 99 Germany 14 76 5.49 96 9 0.1 0.1 0.5 69 Ghana 99 81 5.51 56 42.5 6.3 21.9 134.9 76 Greece 78 76 5.67 87 37 0.7 1.9 2.8 66 Grenada 67 79 5.82 87 6 0.9 0.3 1.3 86 Guatemala 22 70 5.3 87 3 0.1 <0.05 0.5 52 Guinea 22 70 5.31 87 10 0.1 <0.05 0.5 50 Guinea-Bissau 86 77 5.68 87 16 0.2 0.5 5.9 81 Guyana 31 76 5.41 87 7 1.2 0.1 1.1 48 Haiti 80 79 5.31 87 7 0.4 0.4 2.6 74 Honduras 98 70 5.4 69 70.4 3.3 4.4 31.9 86 Hungary 99 80 5.47 85 31.7 3.8 25.5 51 84 Iceland 44 74 5.59 88 6.6 0.7 0.3 2.6 94 India 68 76 5.09 61 12.1 0.2 0.3 6.9 96 Indonesia 88 80 5.47 89 15 1.5 1.8 1.6 85 Iran (Islamic Republic of) 86 75 5.78 81 13.8 0.9 0.4 12.6 91 Iraq 92 80 5.38 75 27.6 2.8 6.1 14.5 78 Ireland 100 83 5.58 74 30.9 3.6 6.7 40.4 71 Israel 99 79 5.37 76 34.2 3.9 10.8 20.4 78 Italy 85 78 5.69 83 17 0.4 1.1 3.6 81 Jamaica 100 83 5.31 77 134 2.3 8.4 16.8 100 Japan 97 79 6.25 73 14 2.7 1.3 10.8 97 Jordan 98 73 5.65 74 67.2 3.3 6.3 44.1 78 Kazakhstan 30 71 4.7 89 14 0.2 0.2 0.7 69 Kenya 40 79 6.78 51 18.6 0.2 1.8 3.6 60 Kiribati 100 77 6.06 80 20.4 1.9 3.3 106 86 Kuwait 97 73 5.55 73 45.1 1.9 3.4 32.5 50 Kyrgyzstan 73 75 5.1 70 15 0.2 <0.05 1 74 Lao People's Democratic Republic 93 70 5.42 62 58 3.2 12.1 53 90 Latvia ANNEXES 59 Annex 2 Family planning demand Care- satisifed seeking Insecticide- UHC service with Antenatal behaviour Tuberculosis treated nets coverage modern care, Child for child effective HIV for malaria index Index data methods 4+ visits immunization pneumonia treatment treatment prevention Country (SDG 3.8.1) availabilityb (%) (%)c (DTP3) (%) (%)c (%)d (%) (%)e Lebanon 68 medium 61 71 81 74 66 44 – Lesotho 45 high 76 74 93 63 32 40 – Liberia 34 high 38 78 52 51 31 17 47 Libya 63 low 45 71 97 81 22 43 – Lithuania 67 medium 70 88 93 87 71 20 – Luxembourg ≥80 medium 83 97 99 94 68 72 – Madagascar 30 medium 59 51 69 41 43 4 68 Malawi 44 high 73 51 88 71 40 58 55 Malaysia 70 high 53 80 99 87 68 26 – Maldives 55 low 53 85 99 22 30 13 – Mali 32 high 33 38 64 23 44 32 59 Malta 79 medium 72 97 97 90 66 75 – Mauritania 33 medium 30 48 73 34 38 21 8 Mauritius 64 medium 49 56 97 78 41 39 – Mexico 76 high 83 94 87 73 65 55 – Micronesia (Federated States of) 60 low 59 74 72 65 75 41 – Mongolia 63 high 72 90 99 70 32 33 – Montenegro 54 high 36 87 89 89 55 21 – Morocco 65 medium 78 55 99 70 71 41 – Mozambique 42 high 39 51 80 50 34 44 66 Myanmar 60 high 74 74 89 58 61 47 – Namibia 59 high 77 63 92 68 70 63 – Nepal 46 high 65 60 91 50 69 36 – Netherlands ≥80 medium 87 97 95 91 74 77 – New Zealand ≥80 medium 85 97 92 86 71 72 – Nicaragua 70 medium 89 88 98 58 68 38 – Niger 33 high 41 39 65 59 44 26 39 Nigeria 39 high 33 54 49 35 13 26 45 Norway ≥80 medium 87 97 95 92 73 72 – Oman 72 medium 35 71 99 56 84 43 – Pakistan 40 high 49 37 72 64 59 5 – Panama 75 high 74 88 73 82 63 48 – Papua New Guinea 41 low 48 55 73 63 56 48 – Paraguay 69 medium 81 77 93 74 62 30 – Peru 78 high 64 95 90 60 70 53 – Philippines 58 high 54 84 60 64 78 27 – Poland 75 medium 66 97 98 88 51 72 – Portugal ≥80 medium 83 97 98 91 63 72 – Qatar 77 high 62 85 99 87 74 85 – Republic of Korea ≥80 high 83 98 98 80 76 72 – Republic of Moldova 65 high 63 95 87 79 46 26 – Romania 72 medium 71 76 89 70 74 67 – Russian Federation 63 medium 73 78 97 83 60 28 – Rwanda 53 high 65 44 98 54 72 74 67 Saint Lucia 69 medium 76 90 99 82 86 46 – Saint Vincent and the Grenadines 65 low 81 100 98 81 36 46 – Samoa 56 high 37 73 66 78 68 41 – Sao Tome and Principe 54 medium 52 84 96 69 71 39 – Saudi Arabia 68 medium 45 71 98 82 54 54 – Senegal 41 high 43 47 89 48 55 44 68 Serbia 65 high 36 94 95 90 71 63 – Seychelles 68 medium 39 56 97 83 60 39 – 60 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT International Health Mean Regulations At least fasting Tobacco Hospital core basic Normal plasma non- beds per Physicians Psychiatrists Surgeons capacity sanitation blood glucose smoking 10 000 per 1000 per 100 000 per 100 000 index (%) pressure (%) (mmol/L)f (%) populationc populationc populationc populationc (%)c Country 95 79 5.7 66 28.5 2.4 1.4 45.4 76 Lebanon 44 72 5.5 74 13 <0.05 0.1 0.2 63 Lesotho 17 72 5.31 90 8 <0.05 0.1 0.2 26 Liberia 100 76 5.93 81 37 2.1 1 15.6 65 Libya 94 70 5.5 70 72.8 4.3 16.7 61.2 83 Lithuania 98 78 5.43 76 48.2 2.9 22.5 51.6 89 Luxembourg 10 72 5.12 87 2 0.1 0.1 0.4 29 Madagascar 44 71 5.03 85 13 <0.05 <0.05 0.4 40 Malawi 100 77 5.66 78 18.6 1.3 0.8 6.9 99 Malaysia 96 76 5.14 71 43 1.6 3.7 8.8 60 Maldives 31 68 5.36 88 1 0.1 <0.05 0.7 55 Mali 100 81 5.64 74 46.7 3.9 3.2 43.7 76 Malta 45 68 5.24 87 4 0.1 0.1 1.6 28 Mauritania 93 75 5.6 78 34 1.1 0.8 6.9 68 Mauritius 89 80 5.89 85 15.2 2.4 1 16 96 Mexico 56 75 6.18 75 18.6 0.2 1 10.6 64 Micronesia (Federated States of) 59 71 5.54 74 70 2.9 0.5 14.1 86 Mongolia 96 71 5.34 54 39.6 2.3 8.7 37.6 55 Montenegro 84 74 5.58 77 11 0.6 0.5 7.8 95 Morocco 24 71 5.21 83 7 0.1 0.1 0.7 67 Mozambique 65 76 5.02 79 9 0.6 0.3 0.9 86 Myanmar 34 72 5.35 78 27 0.4 0.3 0.8 66 Namibia 46 70 5.44 76 3 0.2 0.2 0.9 72 Nepal 98 81 5.11 74 46.6 3.4 20.1 29.7 94 Netherlands 100 84 5.57 84 28 3 18 18.3 98 New Zealand 76 79 5.33 87 9 0.9 0.9 8.7 76 Nicaragua 13 66 5.2 92 2.8 <0.05 <0.05 0.2 73 Niger 33 76 5.45 94 5 0.4 0.1 1 67 Nigeria 98 80 5.52 79 38.6 4.4 29.7 74.7 98 Norway 99 76 5.71 92 15.8 1.5 2.3 14.2 94 Oman 58 70 5.84 80 6 0.8 0.3 1.3 43 Pakistan 77 80 5.59 94 23 1.6 3.8 17.6 68 Panama 19 75 6.07 63 18.6 0.1 0.1 0.5 64 Papua New Guinea 91 76 5.52 86 13 1.3 2 5.4 82 Paraguay 77 86 4.93 87 16 1.1 0.8 28.4 89 Peru 75 78 5.03 75 5 1.1 0.5 4.3 84 Philippines 98 71 5.15 71 65 2.3 5.1 15.4 74 Poland 99 76 5.28 77 34 4.4 4.5 47.8 95 Portugal 100 79 5.7 86 12 2 3 3.5 97 Qatar 100 88 5.4 76 115.3 2.2 7 62 100 Republic of Korea 78 70 5.48 75 58.3 2.5 5.9 13.8 78 Republic of Moldova 82 70 5.39 70 62.7 2.7 6 40.8 79 Romania 89 73 5.52 59 81.8 3.3 11.1 16.6 81 Russian Federation 62 74 4.93 87 16 0.1 0.1 0.4 41 Rwanda 91 73 5.58 87 13 0.1 1.1 10.9 58 Saint Lucia 87 77 5.67 87 26 0.6 0.9 3.7 35 Saint Vincent and the Grenadines 97 76 6.63 72 18.6 0.5 0.5 2.6 75 Samoa 40 74 5.44 87 29 0.5 0.5 1.5 16 Sao Tome and Principe 100 77 6.59 87 26.5 2.6 2.1 61.6 99 Saudi Arabia 48 70 5.46 91 3 0.1 0.2 0.3 30 Senegal 95 71 5.36 61 56.5 2.5 7.4 43.1 47 Serbia 100 77 5.83 79 36 1 2.1 22.9 82 Seychelles ANNEXES 61 Annex 2 Family planning demand Care- satisifed seeking Insecticide- UHC service with Antenatal behaviour Tuberculosis treated nets coverage modern care, Child for child effective HIV for malaria index Index data methods 4+ visits immunization pneumonia treatment treatment prevention Country (SDG 3.8.1) availabilityb (%) (%)c (DTP3) (%) (%)c (%)d (%) (%)e Sierra Leone 36 high 36 76 86 72 51 21 72 Singapore ≥80 medium 77 97 96 86 68 53 – Slovakia 76 medium 76 97 96 83 77 56 – Slovenia 78 medium 79 97 95 92 67 72 – Solomon Islands 50 medium 56 65 98 73 73 41 – Somalia 22 medium 45 6 42 13 40 10 23 South Africa 67 medium 84 87 75 65 49 49 – South Sudan 30 medium 14 17 31 48 38 9 58 Spain 77 medium 81 97 97 92 42 79 – Sri Lanka 62 medium 74 93 99 58 58 23 – Sudan 43 medium 31 51 93 48 44 8 42 Suriname 68 high 72 67 89 76 62 38 – Swaziland 58 high 79 76 90 58 46 69 – Sweden ≥80 medium 81 97 98 90 77 63 – Switzerland ≥80 medium 87 97 97 92 68 72 – Syrian Arab Republic 60 low 60 64 41 77 56 43 – Tajikistan 65 medium 56 53 96 63 71 22 – Thailand 75 high 91 93 99 83 42 61 – The former Yugoslav Republic of 70 medium 28 94 91 93 87 41 – Macedonia Timor-Leste 47 medium 48 55 76 71 48 41 – Togo 42 high 33 57 88 49 61 39 74 Tonga 62 medium 50 70 78 76 87 41 – Trinidad and Tobago 75 medium 66 100 96 74 56 63 – Tunisia 65 high 75 85 98 60 73 27 – Turkey 71 high 60 89 97 85 76 28 – Turkmenistan 67 medium 74 88 99 51 55 28 – Uganda 44 high 46 48 78 79 40 60 66 Ukraine 63 high 70 87 23 92 53 26 – United Arab Emirates 63 low 60 71 99 88 35 43 – United Kingdom ≥80 medium 93 97 96 89 72 72 – United Republic of Tanzania 39 high 54 43 98 55 33 55 29 United States of America ≥80 high 86 97 95 89 74 72 – Uruguay 79 high 88 96 95 91 65 52 – Uzbekistan 72 medium 84 88 99 68 60 28 – Vanuatu 56 high 59 52 64 72 71 41 – Venezuela (Bolivarian Republic of) 73 medium 82 87 87 72 64 55 – Viet Nam 73 high 77 74 97 81 72 43 – Yemen 39 high 48 25 69 34 52 15 – Zambia 56 high 65 56 90 70 49 64 64 Zimbabwe 55 high 86 70 87 51 58 68 78 DTP3: diphtheria-tetanus-pertussis containing vaccine (third dose); HIV: human immunode ciency virus; UHC: universal health coverage. Notes: a The statistics shown in Annex 2 are based on the evidence available in mid-2017. They have been compiled primarily using publications and databases produced and maintained by WHO or the United Nations groups. Wherever possible, estimates have been computed using standardized categories and methods in order to enhance cross-national comparability. This approach may result, in some cases, in di erences between the estimates presented here and the o cial national statistics prepared and endorsed by individual countries. It is important to stress that these estimates are also subject to uncertainty, especially for countries with weak statistical and health information systems where the quality of underlying empirical data is limited. More details on the indicators and estimates presented here are available at the WHO UHC data portal: http://apps.who.int/gho/cabinet/uhc.jsp b Data availability is classi ed as follows, based on information available in global data bases: high=75% or more of the tracer indicators with primary data since 2010; medium=50% or more (but less than 75%) of the tracer indicators with primary data since 2010; low=less than 50% of tracer indicators with primary data since 2010. ‘Primary data’ refers to original data sources and excludes estimates based on modelling and predictions. c The most recent year of data available was used. d Estimates of the percentage of cases treated are for 2014, while estimates of cases detected are for 2015. e Only pertains to countries with highly endemic malaria. f Estimates are for 2008. 62 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT International Health Mean Regulations At least fasting Tobacco Hospital core basic Normal plasma non- beds per Physicians Psychiatrists Surgeons capacity sanitation blood glucose smoking 10 000 per 1000 per 100 000 per 100 000 index (%) pressure (%) (mmol/L)f (%) populationc populationc populationc populationc (%)c Country 14 70 5.41 74 4 <0.05 <0.05 0.1 64 Sierra Leone 100 85 5.3 83 24 3.4 13.7 102.3 99 Singapore 99 71 5.45 70 57.5 3.4 11.5 18.5 96 Slovakia 99 70 5.42 77 45.5 2.8 10.2 36.3 75 Slovenia 31 78 6.26 75 14 0.2 0.2 0.9 57 Solomon Islands 16 67 5.17 87 8.7 <0.05 <0.05 0.1 6 Somalia 73 73 5.71 79 28 0.8 0.4 6.4 100 South Africa 10 72 5.3 87 8.7 0.1 <0.05 1.2 50 South Sudan 100 81 5.63 70 29.7 3.8 8.1 23.1 90 Spain 94 78 5.38 86 35.1 0.7 0.4 0.6 71 Sri Lanka 35 70 5.24 87 8.2 3.1 0.1 0.8 71 Sudan 79 78 5.75 74 31 0.8 1.5 13.7 71 Suriname 58 71 5.52 91 21 0.1 0.1 4 51 Swaziland 99 81 5.36 81 25.9 4.1 18.3 26.1 92 Sweden 100 82 5.39 74 46.8 4.1 41.4 50.4 91 Switzerland 93 76 5.79 81 15 1.5 0.3 3 63 Syrian Arab Republic 96 74 5.48 69 47.6 1.7 2.2 15.8 89 Tajikistan 95 78 5.15 79 21 0.4 0.9 6.3 98 Thailand The former Yugoslav Republic of 91 72 5.41 69 44.3 2.8 10 36.9 89 Macedonia 44 73 4.98 57 59 0.1 0.3 1.6 66 Timor-Leste 14 71 5.32 92 7 0.1 <0.05 0.3 69 Togo 94 76 6.31 72 18.6 0.6 1 2.8 74 Tonga 92 74 5.74 87 30 1.2 3.1 18.2 70 Trinidad and Tobago 93 77 5.75 67 22.9 1.3 2.6 7.3 65 Tunisia 96 80 5.49 72 26.6 1.7 1.5 8.3 78 Turkey 97 75 5.58 69 73.6 2.3 6.3 42.4 84 Turkmenistan 19 73 5.22 90 5 0.1 <0.05 0.6 73 Uganda 96 73 5.47 69 88 3 10.1 72.5 97 Ukraine 100 80 6.04 81 11.5 1.6 0.1 11 91 United Arab Emirates 99 85 5.38 77 27.6 2.8 14.6 34.1 89 United Kingdom 24 73 5.3 85 7 <0.05 <0.05 0.2 67 United Republic of Tanzania 100 87 5.71 78 29 2.6 12.4 36.7 97 United States of America 96 79 5.51 82 28 3.9 16.9 11.7 83 Uruguay 100 75 5.71 87 39.9 2.5 1.9 26 83 Uzbekistan 54 76 5.38 81 18.6 0.2 0.4 1.9 43 Vanuatu 95 81 5.55 87 8 1.9 1.1 11.1 90 Venezuela (Bolivarian Republic of) 78 77 4.7 76 25.6 1.2 0.9 3.3 99 Viet Nam 60 69 5.59 81 7.1 0.3 0.2 0.4 46 Yemen 31 73 5.16 86 20 0.2 0.1 0.7 92 Zambia 39 72 5.37 84 17 0.1 0.1 0.5 68 Zimbabwe ANNEXES 63 Annex 3. List of countries by United Nations regions AFRICA Northern Africa Algeria, Egypt, Libya, Morocco, Sudan, Tunisia Sub-Saharan Africa Angola, Benin, Botswana, Burkina Faso, Burundi, Cabo Verde, Cameroon, Central African Republic, Chad, Comoros, Congo, Côte d’Ivoire, Democratic Republic of the Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, Somalia, South Africa, South Sudan, Swaziland, Togo, Uganda, United Republic of Tanzania, Zambia, Zimbabwe ASIA Central Asia Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan Eastern Asia China, Democratic People’s Republic of Korea, Japan, Mongolia, Republic of Korea South-eastern Asia Brunei Darussalam, Cambodia, Indonesia, Lao People’s Democratic Republic, Malaysia, Myanmar, Philippines, Singapore, Thailand, Timor-Leste, Viet Nam Southern Asia Afghanistan, Bangladesh, Bhutan, India, Iran (Islamic Republic of), Maldives, Nepal, Pakistan, Sri Lanka Western Asia Armenia, Azerbaijan, Bahrain, Cyprus, Georgia, Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, Turkey, United Arab Emirates, Yemen EUROPE Albania, Austria, Belarus, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Montenegro, Netherlands, Norway, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, The former Yugoslav Republic of Macedonia, Ukraine, United Kingdom of Great Britain and Northern Ireland LATIN AMERICA AND THE CARIBBEAN Antigua and Barbuda, Argentina, Bahamas, Barbados, Belize, Bolivia (Plurinational State of), Brazil, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Saint Lucia, Saint Vincent and the Grenadines, Suriname, Trinidad and Tobago, Uruguay, Venezuela (Bolivarian Republic of) NORTHERN AMERICA Canada, United States of America OCEANIA Australia, Fiji, Kiribati, Micronesia (Federated States of), New Zealand, Papua New Guinea, Samoa, Solomon Islands, Tonga, Vanuatu 64 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Annex 4. UHC service coverage index by WHO and World Bank regions, 2015 WHO regions UHC service coverage index World Bank regions UHC service coverage index Global 64 Global 64 African Region 44 East Asia & Paci c 72 Region of the Americas 78 Europe & Central Asia 73 South-East Asia Region 55 Latin America & Caribbean 75 European Region 73 Middle East & North Africa 65 Eastern Mediterranean Region 53 North America ≥80 Western Paci c Region 75 South Asia 53 Sub-Saharan Africa 42 ANNEXES 65 Annex 5. Financial protection indicators by WHO and World Bank regions Table 1. Incidence of catastrophic health spending SDG indicator 3.8.2: 10% threshold 10% threshold 2000 2005 2010 WHO regions % Population Million % Population Million % Population Million Global 9.7 588.5 11.4 741.3 11.7 808.4 African Region 8.6 56.8 9.6 72.0 10.3 88.1 Region of the Americas 10.5 87.3 13.0 115.3 11.1 103.5 South-East Asia Region 10.7 168.4 11.1 188.3 12.8 233.0 European Region 6.6 56.9 6.9 60.5 7.0 62.2 Eastern Mediterranean Region 7.6 35.8 8.7 45.9 9.5 55.5 Western Paci c Region 10.9 182.1 14.9 258 14.8 264.7 10% threshold 2000 2005 2010 World Bank regions % Population Million % Population Million % Population Million Global 9.7 588.5 11.4 741.3 11.7 808.4 East Asia & Paci c 9.6 194.7 12.9 271.3 12.9 280.9 Europe & Central Asia 6.5 56.5 6.9 60.0 7.0 61.8 Latin America & Caribbean 13.4 70.5 17.5 98.3 14.8 88.3 Middle East & North Africa 8.3 26.3 11.5 40.1 13.4 52.2 North America 5.5 17.2 5.3 17.4 4.6 15.6 South Asia 12.0 166.1 12.0 181.7 13.5 220.6 Sub-Saharan Africa 8.6 57.2 9.6 72.6 10.3 89.0 Note: Catastrophic health spending is de ned as out-of-pocket expenditures exceeding 10% of household total consumption or income. This de nition with this threshold corresponds to SDG indicator 3.8.2, de ned as “the proportion of population with large household expenditures on health as a share of total household expenditure or income”. Source: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability, which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. These estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. 66 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Table 2. Incidence of catastrophic health spending SDG indicator 3.8.2: 25% threshold 25% threshold 2000 2005 2010 WHO regions % Population Million % Population Million % Population Million Global 1.9% 112.8 2.4% 154.9 2.6% 179.3 African Region 1.6% 10.8 2.1% 15.6 2.6% 21.9 Region of the Americas 2.0% 16.6 2.4% 20.9 1.9% 17.5 South-East Asia Region 2.0% 30.8 2.1% 35.1 2.9% 51.8 European Region 1.0% 8.4 1.0% 9.1 1.0% 8.9 Eastern Mediterranean Region 1.0% 4.8 1.1% 5.9 1.4% 8.4 Western Paci c Region 2.5% 41.1 3.9% 68.0 3.9% 70.6 25% threshold 2000 2005 2010 World Bank regions % Population Million % Population Million % Population Million Global 1.9% 112.8 2.4% 154.9 2.6% 179.3 East Asia & Paci c 2.2% 43.6 3.3% 70.4 3.4% 73.2 Europe & Central Asia 1.0% 8.4 1.0% 9.0 1.0% 8.9 Latin America & Caribbean 2.6% 13.6 3.2% 18.0 2.5% 14.9 Middle East & North Africa 1.4% 4.5 1.7% 5.9 2.2% 8.4 North America 1.0% 3.1 0.9% 3.0 0.8% 2.6 South Asia 2.1% 28.8 2.2% 33.0 3.0% 49.4 Sub-Saharan Africa 1.6% 10.8 2.1% 15.7 2.5% 22.0 Note: Catastrophic health spending is de ned as out-of-pocket expenditures exceeding 10% of household total consumption or income. This de nition with this threshold also corresponds to SDG indicator 3.8.2, de ned as “the proportion of population with large household expenditures on health as a share of total household expenditure or income”. Source: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability, which may not correspond to the methods used at regional and/or national level to monitor catastrophic spending on health. These estimates are based on a data availability for global monitoring, which may not necessarily align with availability of data at national or regional levels. ANNEXES 67 Table 3. Incidence of impoverishing health spending at the 2011 PPP $1.90-a-day poverty line $1.90-a-day 2000 2005 2010 WHO regions % Population Million % Population Million % Population Million Global 2.1 130.4 1.8 115.6 1.4 97.0 African Region 2.3 15.4 1.6 11.7 1.7 14.2 Region of the Americas 0.4 3.3 0.5 4.1 0.3 2.8 South-East Asia Region 3.9 61.9 3.3 56.6 3.1 56.8 European Region 0.2 2.0 0.1 0.9 0.1 0.7 Eastern Mediterranean Region 1.4 6.7 0.9 4.7 0.5 3.2 Western Paci c Region 2.4 40.9 2.2 37.4 1.1 19.4 $1.90-a-day 2000 2005 2010 World Bank regions % Population Million % Population Million % Population Million Global 2.1 130.4 1.8 115.6 1.4 97.0 East Asia & Paci c 2.2 45 1.9 40.4 1.0 20.9 Europe & Central Asia 0.2 2.0 0.1 0.9 0.1 0.7 Latin America & Caribbean 0.6 3.3 0.7 4.1 0.5 2.8 Middle East & North Africa 0.7 2.2 0.5 1.8 0.3 1.3 North America 0.0 0.0 0.0 0.0 0.0 0.0 South Asia 4.5 62.4 3.7 56.5 3.5 57.1 Sub-Saharan Africa 2.3 15.5 1.6 11.8 1.6 14.2 Note: Impoverishing spending on health occurs when a household is forced by an adverse health event to divert spending away from nonmedical budget items such as food, shelter, clothing to such an extent that its spending on these items is reduced below the level indicated by the poverty line. Indicators of impoverishing spending on health are not part of the o cial SDG indicator of universal health coverage per se, but link UHC directly to the rst SDG goal, namely to end poverty in all its forms everywhere. Source: WHO and the World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability which may not correspond to the methods used at regional and/or national level to monitor impoverishing pending on health. These estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. 68 TRACKING UNIVERSAL HEALTH COVERAGE: 2017 GLOBAL MONITORING REPORT Table 4. Incidence of impoverishing health spending at the 2011 PPP $3.10-a-day poverty line $3.10-a-day 2000 2005 2010 WHO regions % Population Million % Population Million % Population Million Global 1.7 106.1 1.8 115.8 1.8 122.3 African Region 2.1 13.9 1.4 10.1 1.5 12.5 Region of the Americas 0.9 7.9 1.0 8.8 0.7 6.2 South-East Asia Region 2.4 37.8 2.3 39.1 3.4 61.3 European Region 0.6 4.9 0.2 2.1 0.2 1.5 Eastern Mediterranean Region 1.7 8.0 1.7 9.1 1.3 7.7 Western Paci c Region 2.0 33.4 2.7 46.4 1.8 32.9 $3.10-a-day 2000 2005 2010 World Bank regions % Population Million % Population Million % Population Million Global 1.7 106.1 1.8 115.8 1.8 122.3 East Asia & Paci c 1.9 38.1 2.4 50.0 1.7 37.4 Europe & Central Asia 0.6 4.9 0.2 2.1 0.2 1.5 Latin America & Caribbean 1.5 7.9 1.6 8.8 1.0 6.2 Middle East & North Africa 1.3 4.1 1.1 4.0 0.9 3.3 North America 0.0 0.0 0.0 0.0 0.0 0.0 South Asia 2.7 37.1 2.7 40.7 3.8 61.3 Sub-Saharan Africa 2.1 13.9 1.4 10.2 1.4 12.6 Note: Impoverishing spending on health occurs when a household is forced by an adverse health event to divert spending away from nonmedical budget items such as food, shelter, clothing etc. to such an extent that its spending on these items is reduced below the level indicated by the poverty line. Indicators of impoverishing spending on health are not part of the o cial SDG indicator of universal health coverage per se, but link UHC directly to the rst SDG goal, namely to end poverty in all its forms everywhere. Source: WHO and World Bank estimated values are based on standard de nitions and methods to ensure cross-country comparability, which may not correspond to the methods used at regional and/or national level to monitor impoverishing spending on health. These estimates are based on a data availability for global monitoring which may not necessarily align with availability of data at national or regional levels. ANNEXES 69 Photo credits Cover © WHO/SEARO/Sanjit Das Page1 © Olja Latinovic/World Bank Page23 © 2013 Anil Gulati, Courtesy of Photoshare Page 47 © WHO/Alison Clements-Hunt ISBN 978 92 4 151355 5 http://www.who.int/healthinfo/universal_health_coverage/report/2017/en/ http://www.worldbank.org/health