68351 STUDY ON COMPARATIVE EFFICIENCIES IN VACCINE PROCUREMENT MECHANISMS By: Janet Bumpas, Consultant SASHD-HNP WORLD BANK June 2008 1 Table of Contents 1. Executive Summary ................................................................................................. 5 2. Objective ................................................................................................................... 7 3. Methods ..................................................................................................................... 7 4. Overview ................................................................................................................... 7 5. Vaccine Market Fragility.......................................................................................... 8 5.1. Why did the traditional EPI vaccine market crash? ............................... 9 5.2. What helped the market restabilize?....................................................... 11 6. Paris Declaration Support of Self-Procurement .................................................. 13 7. Procurement systems .............................................................................................. 14 7.1. Options........................................................................................................ 14 7.2. The NRA’s implementation of six critical functions............................. 17 7.3. The country’s ability to forecast demands to provide sufficient quantity of vaccines to avoid shortages and program disruptions.............. 19 7.4. Adequate capacity, knowledge, and processes in place for preparing bidding documents. .......................................................................................... 19 7. 5. Adequate capacity and processes in place for realizing the tender and selecting the supplier........................................................................................ 20 7.6. Adequate processes in place for arranging finances............................. 22 8. Procuring the highest quality vaccines at a reasonable price ............................ 23 9. World Bank procurement standards and UNICEF............................................ 30 10. Conclusion............................................................................................................. 31 Annex 1: UNICEF Procurement Services................................................................. 33 Annex 2: Vaccine Procurement in Bangladesh ........................................................ 37 Annex 3: Vaccine Procurement in Botswana ........................................................... 40 Annex 4: Vaccine Procurement in Nigeria ............................................................... 46 Annex 5: Vaccine Procurement in Pakistan ............................................................. 50 2 List of Tables 1. Number of Vaccine Manufacturers Awarded a Tender by UNICEF by Vaccine..…13 2. 2003/2004 GCC and Member State Vaccine Prices in USD .................................... 15 3. Six Critical Functions for an NRA............................................................................ 16 4. Causes for Vaccine Shortages in Countries…………………………………………21 5. Number of Respondents to Pakistani tenders............................................................ 25 6. Botswana Vaccine Procurement 2008: Price comparison with UNICEF ................ 27 7. UNICEF’s average 2003 prices compared to prices offered to Nigeria in its 2002 tender.............................................................................................................................. 27 8. Government of Pakistan Self-procurement Vaccine Prices, 2008 ............................ 28 9. UNICEF Vaccine Procurement Sources 2000 and 2006 .......................................... 34 10. Vaccine doses procured - Bangladesh...................................................................... 35 11. Reported Immunization Coverage for Bangladesh .................................................. 35 12. Reported Immunization Coverage for Botswana ..................................................... 38 13. NRA Capacity in Botswana ..................................................................................... 41 14. Government of Botswana vaccine prices 2008: Comparison with UNICEF.......... 42 15. Reported Immunization Coverage for Nigeria......................................................... 44 16. UNICEF’s average 2003 prices compared to prices offered to Nigeria in its 2002 tender.............................................................................................................................. 46 17. Accuracy of vaccine forecasting in Nigeria –Forecast and actual doses by year. ... 47 18. Reported Immunization Coverage for Pakistan ....................................................... 48 19. Financing Sources and Procurement Mechanisms for Vaccines in Pakistan........... 49 20. Pakistani NRA status for WHO six critical functions.............................................. 51 21. Number of Respondents to Pakistani tenders………………………………………52 22. Government of Pakistan Self-procurement Vaccine Prices, 2008 .......................... 54 23. Tender Differences Between UNICEF and Government of Pakistan...................... 55 List of Figures 1. Distribution of Vaccine Doses Bought Worldwide in 1999 ....................................... 8 2. Vaccine Quantities Offered To UNICEF .................................................................... 8 3. Measles Vaccine Doses Offered to UNICEF 1992-2006............................................ 9 4. Number of Countries Exercising All Regulatory Functions Relevant to Their Source of Vaccine Procurement…………………………………………………………..…17 5. Countries for which UNICEF procures………………………………………………31 6. UNICEF Annual Vaccine Procurement ……………………………………………..31 3 List of Acronyms Acronym Definition AEFI Adverse Events Following Immunization BCG Bacille Calmette-Guérin (Vaccine against Tuberculosis) CIDA Canadian International Development Agency CMS Central Medical Store CMSD Central Medical Supply Department (Bangladesh) DFID UK Department for International Development DRU Drug Registration Unit (Botswana) DQA Data Quality Audit DTP Diphtheria Tetanus Pertussis EPI Expanded Program of Immunization GAVI Global Alliance for Vaccines and Immunizations GCC Gulf Cooperation Council HepB Hepatitis B HNP Health Nutrition and Population (Bangladesh) HPPP Health and Population Program Project (Bangladesh) IACC Inter-Agency Coordination Committee ICB International Competitive Bidding ICDDR,B International Center for Diarrhoeal Diseases, Bangladesh JICA Japanese International Cooperation Agency LDC Least Developed Countries LGA Local Government Authority (Nigeria) LTA Long Term Agreement NID National Immunization Day NIH National Institute of Health NPHCDA National Primary Health Care Development Agency (Nigeria) NPI National Program on Immunization (Nigeria) NRA National Regulatory Authority OPCS Operations Policy and Country Service OPV Oral Polio Vaccine PAHO Pan American Health Organization PPRA Pakistan Procurement Regulatory Authority PMA Pakistan Medical Association PPADB Public Procurement and Asset Disposal Board (Botswana) SADC Southern African Development Community SCMS Supply Chain Management Services (Botswana) TT Tetanus toxoid UN United Nations UNICEF United Nations Children’s Fund USAID United States Agency for International Development VII Vaccine Independence Initiative VVM Vaccine Vial Monitors WHO World Health Organization 4 1. Executive Summary Vaccinations are amongst the most cost-effective public health interventions. The World Bank is involved in the financing of vaccines through its loans, grants, and IDA buy- down programs. Governments use this funding to procure vaccines in a variety of methods, the most common of which are self-procurement and procurement through UNICEF. Vaccine procurement is a complex issue that interweaves the domains of public health, commodity security, ethics, and procurement. Its cross-disciplinary nature means that neither a straightforward analysis stemming from just one discipline nor a cookie-cutter application of World Bank procurement principles of economy, efficiency, equal opportunity, promoting domestic contracting and transparency will lead to an optimal solution. A more holistic approach is required. The World Bank has therefore historically considered vaccine procurement as “special� and has allowed for exceptions from its guidelines for World Bank funded vaccine procurement. In general, five core principles should apply in some form to World Bank financed vaccine procurement. They are: a) The vaccine market is extremely fragile and so World Bank actions should work to strengthen this market, not weaken it. In 2000-2001, the traditional EPI vaccine market crashed. Two factors restabilized the market. The United Nations Children’s Fund (UNICEF), the largest procurement agent, moved from awarding competitive tenders to one lowest-cost manufacturer to a longer-term, more partnership-oriented approach. The Global Alliance for Vaccines and Immunizations (GAVI) entered the marketplace, providing developing countries with significant funding in 2000. These two factors restabilized the market because it created incentives large enough to keep vaccine manufacturers interested in this market. The World Bank should consider the impacts of World Bank funded procurement decisions on the market and adopt vaccine security friendly policies into its procurement decisions. The World Bank should encourage procurement situations where multiple manufacturers win tenders, where price is important but not the ultimate focus, and where country risk is spread. b) The Paris Declaration states that all multilateral institutions should work to improve country capacity, including procurement. Thus, World Bank supported procurement should contribute towards building national capacity to procure vaccines and not towards dependence. c) Quality procurement systems should be used. The World Bank should consider the country’s National Regulatory Authority (NRA) skills and capacity, the country’s forecasting abilities, the skills and capacities of the procurement cell, and the country’s ability to arrange finances. Other considerations include corruption and local agents’ political influence. Quality procurement should predominate because the worst case scenario would be to procure a low quality vaccine which could harm a healthy child. 5 d) The World Bank funding should be used to procure the highest quality vaccines at the lowest reasonable price. Due to its procurement volumes and relationships with multiple manufacturers, UNICEF usually obtains the lowest prices. A country will most likely not be able to match these prices. Furthermore, UNICEF procurement policies work to keep multiple manufacturers in the marketplace, leading to healthier competition and lower prices for all. e) Procurement must be economical and efficient, provide equal opportunity for all bidders to compete, encourage local contracting, and be transparent. These five core objectives provide a challenge because, while not mutually exclusive, there is clearly some tension amongst them. While some core objectives would promote measured competition, others would promote unfettered International Competitive Bidding (ICB). For example, factors one, three, and four indicate that the World Bank should favor procurement mechanisms such as UNICEF. Factors two and five indicate that the World Bank should favor procurement mechanisms such as government self- procurement. Therefore, a case by case decision is necessary for vaccine procurement. While all factors are important, the World Bank should place high importance on vaccine security, the fragility of the market, the low prices offered by UNICEF procurement channels and the challenges highlighted in the four surveyed countries’ experiences with regards to vaccine procurement. With this in mind, the World Bank should shift from putting the burden of proof on country teams for explaining why it would like to work with UNICEF for procurement to putting the burden of proof on country teams to explain why it would like to work through national procurement systems. If the World Bank determines that a national country’s procurement systems do not have the capacity to act as an effective channel, the World Bank should work to strengthen procurement and quality control institutions. A high priority for technical assistance should go to NRA capacity building and this should be in place before procurement cells are activated. Countries with adequate procurement and quality control capacity have fewer procurement challenges, but will most likely not be able to provide as low-cost of a procurement option as UNICEF is able to. UNICEF will most likely always provide a lower price due to its large volume and greater competition among manufacturers bidding on orders. Furthermore, individual country self-procurement decisions may negatively impact global vaccine security unless countries incorporate these principles into their procurement decisions. Finally, the decision to move from UNICEF procurement to government self-procurement can be driven by local private sector interests and not increased efficiency and effectiveness of procurement. Therefore, caution should still be exerted when moving from UNICEF procurement to self-procurement in countries that have full procurement and quality control capacity. 6 2. Objective We carried out a multi country comparative assessment of vaccine procurement issues with regards to country capacity, agencies used, and procurement methods followed. The study was designed to provide strong operational recommendations for issues such as efficiency, price, quality control, and logistic management. This study is intended mainly for internal purposes to facilitate discussions with Operations Policy and Country Service (OPCS) regarding refining the World Bank’s position on procurement or supply of vaccine through UNICEF. 3. Methods The study was undertaken between November 2007 and May 2008. It included: a) A desk review of vaccine procurement practices in developing countries with a focus on costs, procurement agent, and efficiency and the determinants for these variables. b) A review of vaccine procurement practices in four countries, Bangladesh, Botswana, Nigeria, and Pakistan. This assessment reviewed: ƒ The procurement method, procurement agent, prices and quality obtained and logistic management. ƒ The procurement capacity of public authorities for vaccines and pharmaceuticals including their capacity for pre-qualification. ƒ Reliability of national procurement capacity, including timely planning, handling and distribution of the products. ƒ The supply chain effectiveness with a focus on vaccines. c) A review of the global vaccine procurement situation, including consultation with UNICEF, WHO, and GAVI. 4. Overview Vaccine procurement is a complex issue that interweaves the domains of public health, commodity security, ethics, and procurement. In general, five principles should apply in some form to all vaccine procurement, including if World Bank financed: a) The vaccine market is extremely fragile and so procurement actions should work to strengthen this market, not weaken it; b) The Paris Declaration states that all multilateral institutions should work to improve country capacity, including procurement. Thus, World Bank activities should lead towards building government capacity to procure vaccines and not towards dependence; c) Quality procurement systems should be used. Quality control issues should predominate because the worst case scenario would be to procure a low quality vaccine. Other considerations include the procurement team’s capacity and skills, corruption, and local agents’ political influence; d) The World Bank should try and procure the highest quality vaccines at the lowest reasonable price; 7 e) World Bank procurement standards should apply. Each of these core objectives is discussed in more detail in the following sections. 5. Vaccine Market Fragility In 2000, the global vaccine market for low cost traditional EPI vaccines crashed due to manufacturers withdrawing from the market. This crash caused shortages serious enough to jeopardize immunization programs for children in developing countries. To understand the extent of this crash, the Figure 1 United Nations Children’s Fund (UNICEF) Distribution of Vaccine Doses Bought procurement numbers can be used. As Worldwide in 1999 Figure 1 shows, in 1999, UN agencies procured the majority of vaccines globally, especially when the industrialized country markets are removed. Thus the manufacturer’s withdrawal from UNICEF vaccine procurement market can be used as a proxy for the withdrawal from manufacturers from the general developing country vaccine Source; Vaccine Market Characteristics. Advanced market. Figure 2 shows the quantities of Immunization Management. PATH. August 2005. vaccines manufacturers were willing to sell to UNICEF over time. Vaccine market crash 8 Figure 3 shows the market for measles vaccines. The blue horizontal line represents the total number of doses that manufacturers offered for sale to UNICEF. Each vertical line represents the total quantity offered to UNICEF by a single manufacturer. While multiple vaccine manufacturers competed before 2000, most of the vaccine manufacturers left the market, leaving UNICEF increasingly dependent on one manufacturer to supply the world’s required measles doses. Figure 3 Measles vaccine Measles vaccine doses offered to UNICEF 1992-2006 to UNICEF 1992-2006 250 0.18 0.16 200 0.14 Millions of doses USD per dose 0.12 150 0.1 0.08 100 0.06 50 0.04 0.02 0 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source: UNICEF Supply Division. May 2008 5.1. Why did the traditional EPI vaccine market crash? In general, vaccine manufacturing is a high start-up and fixed cost, high risk, and lower sales price proposition. These challenging business economics for vaccines result in limited profit potential and hence limited interest from the private sector. High costs a) Strict technical and regulatory requirements: Vaccines are derived from a biological entity and thus each vaccine differs slightly. In order to produce high quality and consistently safe vaccines, manufacturers must make significant investments in technology, processes, and oversight. The vaccine control process is also much more challenging than for pharmaceuticals. Whereas pharmaceuticals can be chemically analyzed to determine identity, composition, potency, or dosage, vaccines require lot-by-lot release of the product to ensure product consistency.i b) Significant batch failure risk: Vaccine manufacturing is a technically challenging process based on live biologicals. Batch failure rates vary based on product, facility, and year but range from under 5% up to 30%. When a batch fails, the manufacturer must start again, resulting in lost costs and delayed orders.ii 9 c) Limited shelf life: Due to a vaccine’s perishable nature, manufactures cannot stock excess capacity but must sell or dispose of all stock, leaving little margin for error. Low profit margins mean manufacturers must be conservative in production capacity to minimize waste.iii High risk d) Production versus sales timing risk: Stringent safety rules mean that vaccine manufacturers need approximately six to twenty four months to produce a dose, three years to expand existing capacity and five to seven years to build new capacity.iv While production lead times are long, countries will often place orders 3-6 months in advance and provide funding either upon arrival of goods or upon order placement. This timing risk increases the risk to the capital invested and hence its cost. e) Product failure risk: Vaccines are one of the few drugs that are given to healthy children. Thus, the consequences of an adverse reaction are extremely high, often involving significant humanitarian and possibly legal and financial ramifications. Low revenue f) Consolidated buyers reduce sales price: A few large buyers such as the Pan American Health Organization (PAHO), and UNICEF purchase large quantities and can exert downward price pressure on manufacturers. Public sector sales account for about 80% of the total vaccine production volume but less than 20% of the revenue.v This reduces vaccine manufacturer’s profit potential and thus interest in investing in this field. g) Geographic segmentation and diverging markets: The vaccine market started to bifurcate in the1980s. Industrialized countries migrated to new more sophisticated and expensive vaccines which remained unaffordable to developing countries. Until this point, children in both developing and industrialized countries received the same vaccines and manufacturers could recoup costs in developed markets. Between 1998 and 2001, ten of fourteen manufacturers partially or totally stopped production of the traditional less expensive vaccines.vi Other h) Significant economies of scale: In general, industries with high start-up costs and high fixed costs tend to consolidate to few manufacturers. Vaccine production requires advanced technical facilities and highly trained personnel due to its biological nature (more so than pharmaceuticals which are chemical by nature). This leads to high fixed costs: 60% of costs are fixed and 25% are semi-fixed for each batch produced, regardless of how many are produced.vii In these situations, manufacturers tend to lower prices aggressively to increase sales volume because a large sales volume leads to allocating these large fixed costs across a greater number 10 of units. Smaller producers cannot compete and a “winner takes all� effect takes hold. i) Pharmaceutical firm ownership: Global pharmaceutical entities have acquired major vaccine manufacturers. Vaccines must now compete internally against potential blockbuster drugs for money and attention. With lower profit margins and longer production lead times, vaccines do not return as much for the shareholder’s invested capital and hence pharmaceutical firms do not see them as attractive as blockbuster drugs. 5.2. What helped the market restabilize? Following the 2000 vaccine crisis, two moves helped restabilize the traditional EPI vaccine market. The first is a set of actions by UNICEF to improve vaccine security. It met with vaccine manufacturers, donor governments, procurement officers, and other parties to determine why the vaccine manufactures were leaving the market. Vaccine manufacturers reported that UNICEF procured vaccines as if they were commodities with too short timelines and too focused on price.viii As a result, UNICEF changed its procurement practices from a one year Invitation to Bid tender process based on accepting the lowest price to a more flexible procurement instrument that incorporated vaccine security principles. UNICEF transitioned to a three year Long Term Agreement (LTA) with manufactures. The LTAs establish forecasts between UNICEF and a manufacturer on quantities to be produced and procured over a time period. Hard orders are then placed as needed.ix UNICEF also adopted the following procurement principles in order to ensure a healthy market:x a) A healthy industry is vital to ensure an uninterrupted and sustainable supply of vaccines. b) UNICEF procures from multiple suppliers for each vaccine presentation. Multiple suppliers encourage multiple manufacturers to stay in the market. Due to long lead times in setting up vaccine manufacturing facilities and high entrance costs, once a manufacturer leaves a market, that decision tends to be permanent and not easily rectifiable. Vaccine prices tend to decline over time when multiple manufacturers compete. Also, vaccine production is challenging with occasional batch failures. Relationships with multiple manufacturers leave multiple options for procurement when a batch failure occurs.xi c) UNICEF procures from manufacturers in developing countries and industrialized countries. UNICEF prefers to have vaccine manufacturers in multiple countries to reduce risk. The higher cost of industrialized manufacturers could be considered the price of vaccine security. For example, the World Health Organization (WHO) continuously evaluates national NRAs and has in the past partially suspended countries’ NRAs for not meeting safety and quality standards. If UNICEF concentrates all its procurement in a country whose NRA becomes suspended, it 11 opens itself to risk because UNICEF can only procure from WHO pre-qualified manufacturers and one of a manufacturer’s requirement for WHO pre-qualification is to reside in a country with a WHO certified NRA. d) UNICEF pays a price that is affordable to Governments and Donors and a price that reasonably covers manufacturers minimum requirements e) UNICEF provides manufacturers long term accurate forecasting. Manufacturers provide accurate production plans and governments provide accurate immunization forecasts. The forecast provides the foundation of vaccine security because all production quantities and funding requirements are based upon it. It also helps manufacturers address their timing risks. f) As a public buyer, providing grants to manufacturers is not the most effective method of obtaining capacity increases. In the late 1990s, UNICEF gave a grant to a DTP manufacturer to help it build capacity, but found that this had minimal impact in scaling up production. UNICEF has found that offering LTAs is a much more effective mechanism to encourage manufacturers to stay in the market.xii g) The option to quote tiered pricing should be given to manufacturers in accordance with the World Bank classification. Because UNICEF Supply Division mostly procures on behalf of the lowest income countries, it asks manufacturers for their lowest prices. UNICEF does sometimes procure on behalf of middle income countries who need assistance with the procurement function (e.g. Equitorial Guinea) and so will sometimes offer manufacturers the option to give differential pricing for these countries.xiii The second stabilizing factor was the Global Alliance for Vaccines and Immunizations (GAVI)’s entrance into the marketplace in 2000. GAVI has committed $3.5 billion in multi-year grants to more than 70 of the world’s poorest countries for vaccine purchases from 2000-2015.xiv GAVI’s advent provides developing countries the opportunity to start incorporating newer combination (tetravalent and pentavalent) vaccines into their immunization schedules through shared GAVI and national government funding of these vaccines. Without GAVI funding, the countries could not order the vaccines and the manufacturers would not have an incentive to build manufacturing to scale. Many countries prefer combination vaccines because they require fewer inoculations per child and reduce vaccine wastage. On the other hand, they are significantly more expensive than the older single purpose or trivalent vaccines and require some changes to the vaccination schedules. GAVI’s entrance encouraged manufacturers to produce newer vaccines for developing markets through providing a market for these products. When developing countries procure more newer vaccines and fewer older, single-purpose or trivalent vaccines purchases, this impacts global manufacturing decisions. Each vaccine that a manufacturer produces, be it an older EPI vaccine or a newer combination vaccine is a 12 semi-independent endeavor requiring its own production capacity and certification processes. Manufacturers therefore make long-term commitments to a specific vaccine, not just the area of vaccinations. Also, the technology behind the combination vaccines is much more challenging, raising the technical skills and investments that manufacturers must have in order to participate in the vaccine market. In 2000, GAVI started support for HepB, Hib, and Yellow Fever vaccines. In 2008, GAVI is expecting to start support for PCV and Rotavirus.xv UNICEF procures all GAVI-funded vaccines and devices, $135 million in 2006.xvi In December 2005, the GAVI Alliance Board received a detailed analysis of the supply activities to date and recommended that UNICEF continue as procurement provider for the existing combination Table 1: Number of vaccine manufacturers awarded a products.xvii tender by UNICEF by vaccine Number of Number of Number of While these two factors have Vaccine Group Manufacturers Manufacturers Manufacturers awarded in 2001-2003 awarded in 2004-2006 awarded in 2007-2009 significantly restabilized the BCG 5 4 4 vaccine market, it does remain DTP 5 4 3 fragile. The number of Measles TT 5 7 5 4 3 3 (+1) traditional vaccine tOPV 4 5 5 mOPV1 0 1 5 manufacturers has stabilized mOPV3 0 1 3 and more manufacturers are DTP+Hib 0 0 1 DTP-HepB+Hib 1 1 2 pursuing WHO pre- DTP-HepB 1 1 3 qualification for new vaccines. HepB 4 5 6 YF 3 3 3 Table 1 shows that the market Meningitis 1 1 1 remains relatively stable yet MMR 3 3 2 MR 1 1 2 still fragile with few DT/Td 3 2 2 manufacturers supplying each Source: Vaccine Security. UNICEF Vaccine Procurement Overview. Rob vaccine. Matthews, UNICEF Supply Division. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. With UNICEF now procuring such a large percentage of the vaccines for developing countries, concerns have been raised over a monopoly in vaccine procurement going forward. Examining the risk of such a large percentage of global procurement capacity existing in one institution is an important question, but goes far beyond the intended scope of this study. 6. Paris Declaration Support of Self-Procurement The World Bank, along with many donor countries, developing countries and multilateral institutions signed the Paris Declaration on Aid Effectiveness on March 2, 2005. Excerpts from this document that specifically refer to procurement are:xviii Art 17. Using a country’s own institutions and systems, where these provide assurance that aid will be used for agreed purposes, increases aid effectiveness by strengthening the partner country’s sustainable capacity to develop, implement 13 and account for its policies to its citizens and parliament. Country systems and procedures typically include, but are not restricted to, national arrangements and procedures for public financial management, accounting, auditing, procurement, results frameworks and monitoring. Art 28. Partner countries and donors jointly commit to: • Use mutually agreed standards and processes to carry out diagnostics, develop sustainable reforms and monitor implementation. • Commit sufficient resources to support and sustain medium and long-term procurement reforms and capacity development. • Share feedback at the country level on recommended approaches so they can be improved over time. Partner countries commit to take leadership and implement the procurement reform process. Art 30. Donors commit to: • Progressively rely on partner country systems for procurement when the country has implemented mutually agreed standards and processes. • Adopt harmonized approaches when national systems do not meet mutually agreed levels of performance or donors do not use them. As a signatory to the Paris Declaration, the World Bank has a responsibility to work with and strengthen national systems and assist to define measures of standards and accountability for procurement. If a country does not meet these standards, the World Bank should work to bring a country up to these standards. At the same time, donors have a responsibility to ascertain with how items are being procured with their finances. 7. Procurement systems 7.1. Options A government has several options for vaccine procurement: government self- procurement, semi-autonomous or parastatal procurement, pooled procurement, use of UNICEF or PAHO procurement services, procurement through a commercial agent, and procurement with the assistance of another country’s procurement staff. This study does not investigate commercial agent procurement as it is out of scope. Each government must decide the best route for its country based on the vaccine amount and type purchased, cost/quantity factors, foreign currency availability, restrictions on the use of funds, country NRA capability for vaccine quality assessment, corruption levels, and the procurement staff’s experience and skills.xix Pooled procurement: Pooled procurement may offer lower prices than self-procurement, but it must be properly managed to be efficient. The Gulf Cooperation Council (GCC) acts as a pooled procurement agent for six countries in the Middle East. Table 2 on the next page shows that GCC pricing is up to 46% less than what one of its states received 14 from the same producers for the same vaccines. 15 Table 2 2003/2004 GCC and Member State Vaccine Prices in USD Price per dose one member state Vaccine received through local tender GCC price per dose Percent decrease MMR 5.15 2.79 46% IPV 5.36 4,75 13% DTwP 0.27 0.20 25% DT 0.24 0.17 29% TT 0.19 0.14 26% OPV 0.15 0.12 25% Rubella 1.61 1.32 18% Source: Vaccine Market Characteristics. AIM (Advanced Immunization Management). PATH (Program for the Appropriate Use of Technology). August 2005. UN procurement: UNICEF, PAHO, and WHO offer procurement services to many low- and middle-income countries that have the means to pay for their own requirements but lack the skill and/or infrastructure required for obtaining safe and effective vaccines from the international marketplace.xx These UN agencies also offer several other procurement- related services such as technical assistance for forecasting, cold chain management, and quality control. Whether or not a UN agency functions as a procurement agent does not impact receiving these other forms of technical assistance as the technical assistance is part of their mandate. • WHO: WHO can act as a procurement agent for non-standard vaccines. WHO charges a 3% fee and requires payments in advance.xxi • PAHO: PAHO purchases for countries in Latin American and the Caribbean with a revolving fund that allows countries to pay after delivery in local currency. PAHO has one price for all. • UNICEF: UNICEF will procure either with a revolving fund through its Vaccine Independence Initiative (VII) or through direct payments made in advance. UNICEF can accept local currency for vaccine procurement within the program expenditure ceiling. UNICEF utilizes tiered pricing (i.e. manufacturers offer relatively cheaper prices to LDCs). UNICEF awards to multiple manufacturers and focuses not only on price in order to promote healthy markets and vaccine security. UNICEF’s procurement system in itself is a high quality asset that a country cannot easily replicate on its own. More information on UNICEF can be found in Annex 1. Self-procurement and parastatal procurement: This works well in countries with a well functioning NRA and vaccine procurement system, adequate currency to enter directly into international tender markets, a well-trained staff, and few corruption issues. Government procurement staff should know about international and local procurement procedures, product specifications, inspection procedures, regulatory and financial requirements, international shipping and trade documentation, international and local vaccine sources, and price levels.xxii Self-procurement incurs additional costs to support a procurement cell and incremental NRA capacities. When evaluating a country’s national procurement system, the following should be taken into account: 16 a) The country’s NRA’s (National Regulatory Authority) implementation of six critical functions to ensure the quality of vaccines procured. b) The country’s ability to forecast demands to provide sufficient quantity of vaccines to avoid shortages and program disruptions. c) Adequate capacity, knowledge, and processes in place for preparing bidding documents. d) Adequate capacity and processes in place for realizing the tender and selecting the supplier. e) Adequate processes in place for arranging finances. A more in depth analysis of these five areas follows, including references to case studies. Detailed case studies can be found in Annexes 2 through 5. The goal of the case studies is to learn about factors that affect procurement options and to understand how donor funds can best be used in various procurement scenarios. The case studies can also provide some guidance for countries on issues to consider when navigating a fragile vaccine marketplace. Similarly, all these aspects are important for the World Bank to consider when it decides to support vaccine procurement. 7.2. The NRA’s implementation of six critical functions Many countries often overlook quality control issues for vaccines when considering procurement mechanisms. The UN has mandated the WHO to review country NRAs technical capabilities. WHO has identified the need to have an independent NRA that can perform six critical functions. Table 3 shows that a country should implement two, four, or six of these functions, depending on the procurement vehicle used. Table 3 Six Critical Functions for an NRA Function Definition Required If UN Procures Government Production Procures Local Independent NRA A country must have an independent national x x x regulatory system 1. Licensing A published set of requirements for licensing (of x x x products and manufacturers) 2. Post-Marketing A system to survey vaccine field performance x x x Surveillance (safety and efficacy) and post-marketing surveillance including Adverse Events Following Immunization (AEFI) 3. Lot Release The receiving country’s NRA should conduct lot x x releasing. Lot releasing requires reviewing the summary lot protocols, which describe the production process in detail and must be included by the producer in each lot. This is separate from 17 any testing that happens in vaccine producing countries. 4. Laboratory Access to a laboratory with testing abilities for x x Access for Testing vaccines in order to respond to reported adverse events. If countries do not have their own qualified laboratory, they should obtain access to a laboratory outside the country, such as a WHO reference laboratory. 5. Good Regular manufacturer inspections for Good x Manufacturing Manufacturing Practices Practice Inspections 6. Clinical Evaluation of clinical performance through X Evaluation authorized clinical trials Source: Vaccine Market Characteristics. Advanced Immunization Management. PATH. August 2005. Procuring countries cannot just Figure 4: Number of countries exercising all rely on WHO pre-qualification of regulatory functions relevant to their source of vaccines in the producing country vaccine procurement, Dec.2007 because some functions, such as 90 post-market surveillance, requires 80 70 domestic government oversight. 60 Also, a 2007 WHO review of the 50 77 NRA in vaccine producing 40 51 13 30 countries, shown in Figure 4, 20 31 10 15 revealed that only 70% adequately 0 12 performed the six regulatory UN Agency Procuring Producing functions that WHO deems critical. Achieved Not achieved Furthermore, in-depth assessments of some countries revealed that Source: WHO and prequalification of vaccines. David Wood, WHO. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008 NRAs who do perform these functions may not be performing them to a depth sufficient to assure vaccine quality.xxiii Gaps in a country’s NRA compromise the country’s ability to ensure that it procures and receives only quality vaccines. Quality control at all stages of vaccine procurement and distribution must be monitored. When adverse effects following immunization happens, it is quite serious. Case study: Measles vaccine in India On Wednesday April 23rd 2008, four children died in Tamil Nadu, India within one hour of receiving a measles vaccination at two separate sites. Possible causes include: • Improper vaccination practices: o Measles vaccine comes in a powder form that needs to be reconstituted by mixing with liquid diluents. Once reconstituted it needs to be used immediately because within three to five hours, harmful bacteria can grow. o Measles vaccine is very sensitive to cold chain practices. If not properly stored, it can easily deactivate. 18 • Syringe issues: Non-sterile syringes and needles can also cause bacterial problems. • Vaccine diluent issues: The diluent itself could have been compromised or incorrectly used. • Vaccine issues: A defective vaccine batch could have caused this. The Government of India immediately sent a team to investigate. They autopsied the children and sent vaccine, vial, and diluents samples for testing. The Government of India suspended 45M doses of the measles vaccine made by the manufacturer. On May 5th 2008, the Government of India laboratory results indicated that the vaccine was safe. The Government of India concluded that toxic shock syndrome from improper vaccination practices killed the children.xxiv 7.3. The country’s ability to forecast demands to provide sufficient quantity of vaccines to avoid shortages and program disruptions. Without an accurate forecast, a country cannot hope to have the appropriate amount of vaccines delivered to a country. Forecasting provides the foundation of vaccine procurement. Forecasting is independent of procurement in that the procurement mechanism chosen should not impact how a country forecasts. However, if a country’s forecasts are off, this impacts its procurement decisions, e.g. inadequate forcasting will result in more emergency procurement. Case study: Botswana The Central Medical Store (CMS) in Botswana’s Ministry of Health is responsible for vaccine procurement in Botswana. In the past year, they have had vaccine stock-outs, mostly due to tender processing issues and forecasting issues. The CMS team does the vaccine forecasting without much input from the Expanded Program on Immunizations (EPI) team. CMS has chosen to use a historical forecasting method whereby the previous year’s vaccine quantities used form the foundation of the upcoming year’s forecast. This approach is not in alignment with international best practices because, for example, a stockout would lead to lower forecasted needs in the following years. A much more accurate approach would be to base a forecast on a target population analysis. 7.4. Adequate capacity, knowledge, and processes in place for preparing bidding documents. How a government crafts a tender dictates a large part of the response it will receive from the market. Many governments choose to include WHO pre-qualification as a tender requirement because it provides an indication of quality. This does limit potential respondents - by 2001, 48 countries produced vaccines,xxv but only twenty four manufacturers in fourteen industrialized and six developing countries had achieved WHO pre-qualification.xxvi If governments self-procure vaccines, they must remain vigilant against pressure from private manufacturers and fillers in their countries who are not WHO-prequalified yet would like to participate in a tender and can bear sometimes significant political pressure. 19 Case Study: Pakistan In 2007, the Government of Pakistan switched from UNICEF procurement to self-procurement and launched its first tender. Clause 2.1 of this first tender issued in October 2007 states: “This invitation for bids is open to all original manufacturers within Pakistan and abroad and their authorized agents / importers / suppliers subject to the conditions that: a. In the case of foreign manufacturers, they must offer the product pre-qualified by the World Health Organization (WHO). b. In case of local manufacturer of vaccines, they will qualify only if they are using concentrates from a manufacturer duly pre-qualified by WHO. c. In the case of authorized agents / importers / suppliers, they must quote such products which are pre-qualified by WHO.�xxvii Due to Clause 2.1.b, Amson Vaccines and Pharma Limited, a Pakistani pharmaceutical manufacturing company, qualified for the tender by proposing to import WHO pre-qualified vaccine concentrate and locally add the diluents and do the packaging. This dilution process impacts overall product quality and is an important step. Amson itself is not WHO pre-qualified. Amson ended up winning the tender to deliver 979,500 doses of TT vaccine.xxviii WHO does not consider this product to be WHO pre-qualified.xxix 7. 5. Adequate capacity and processes in place for realizing the tender and selecting the supplier. When a government procures vaccines, they must comply with the wider government procurement rules to ensure public moneys are well spent. Sometimes rules designed to help the overall country can make vaccine procurement more challenging. Case study: Botswana In 2006, Botswana established the Public Procurement and Asset Disposal Board (PPADB) to guide all government procurement. Its new procurement regulations, largely based on World Bank and other international guidelines, have significantly altered vaccine procurement requirements in the past and changes are likely to continue for some time while processes are developed to meet the new regulations. These changes have caused significant tender delays for the Central Medical Store (CMS), the part of the government responsible for vaccine procurement. In fact, CMS could not obtain a positive outcome from the PPADB on documentation submitted and could not launch a tender for vaccines in 2006. They therefore had to conduct direct purchases on an ad hoc basis once CMS stock levels had depleted. This resulted in rationing to the districts and stock out issues. In 2007, CMS launched a temporary "stop-gap" tender to supply vaccines through the end of March 2009.xxx The two areas causing CMS the most problems are: • The 2004 Civil Empowerment Directive states that if a resident of Botswana can respond to a government procurement tender with a product of equal quality and within 10% of international prices, the tender must go to the Botswanan party. CMS does not currently intend to use the Citizens Empowerment Directive because there are no local human vaccine manufacturers. CMS would prefer to work directly with an international manufacturer than with a local agent because the government procurement rules do not give them the authority to monitor and enforce local agent’s cold chain supply.xxxi Due to the Civil Empowerment Directive, CMS has not been able to develop a tender document template that PPADB accepts and so CMS must submit each tender of greater than 100,000 Botswana Pula (approximately $16,000) through the Ministry of Health Clinical Services cost centre and central Ministry of Health finance to the PPADB.xxxii The updated regulations invalidated how CMS had historically pre-qualified all suppliers so they have had to re-qualify all suppliers. CMS estimates that, as of May 2008, they are approximately 80% through this process.xxxiii 20 Realizing a tender is also a time-consuming, paperwork-intensive process that can be subject to corruption, delays, and other challenges. Governments must have adequate capacity and skills in place to successfully manage a tender. Case study: Bangladesh The EPI team in Bangladesh prefers to procure vaccines with UNICEF due to concerns regarding the procurement capacity of the Central Medical Supply Department (CMSD). While Bangladesh’s procurement rules are clearly laid out, their application and interpretation often lacks. Since 2000, both the US government through USAID and the World Bank have supported government procurement, improving the Government of Bangladesh’s ability to procure. However, due to government staff turnover, these gains are not permanent and must be continually reinforced.xxxiv In 2002, the World Bank conducted an assessment of the Government of Bangladesh’s procurement capacities. It found that actual implementation of procurement to be “quite uneven� and “far from satisfactory.� Some unsatisfactory features were:xxxv • Procurement delays - These were the single most serious malady found. It should take the government 180 to 330 days to successfully complete an international tender.xxxvi However, some contracts have been in process for over 1100 days. • Poor advertisement • A short bidding period • Poor specifications • Nondisclosure of selection criteria • Award of contract by lottery • One-sided contract documents • Negotiation with all bidders • Rebidding without adequate grounds • Other miscellaneous irregularities • Possible corruption and outside influence. In 2005, the World Bank commissioned an Independent Procurement Review of World Bank financed projects in Bangladesh where the government was responsible for procurement. This review focused on a total of 75 contracts within four World Bank projects. The review audited 29 contracts valued at $11.3M in the Health and Population Program Project (HPPP) and 6 contracts valued at $6.3M in the HIV/AIDS Prevention Project. The review found that the procurement in the HPPP project was “highly insufficient� while procurement in the HIV/AIDS project was “insufficient.� Specific areas of issue included: • Filing and transparency • Indications of collusion between bidders • Insufficient World Bank supervision • Untimely project execution • Payment delays Government procurement, however, is gradually taking on more functions successfully. For example, UNFPA used to procure $40-$50M of condoms annually for Bangladesh with a 5-6% procurement fee. In 2000-2001, the Government of Bangladesh switched to self-procurement for condoms. As a result, the government now successfully procures condoms without paying up front (UNFPA requirement) and without the procurement fee. It also found that they could procure the condoms for a lower price than UNFPA.xxxvii 21 Different cadres of health workers may differ in motivation. For example, in Ghana, a World Bank process mapping study discovered that the behaviors and motivation driving people working in the EPI program differed significantly from that of essential drugs or family planning. Higher levels of commitment often exist in vaccine procurement as supply chain staff internalize the mission that if they did not do their job right, then a child would die.xxxviii 7.6. Adequate processes in place for arranging finances. Once the government awards a tender, it must release financing. Table 4 shows that globally, the greatest historical reasons for vaccine stockouts are delays in release of funding. Funding issues often have nothing to do with the Ministry of Health or Table 4: Causes for vaccine shortages in procurement capacity or procurement Aggregated causes COUNTRY Forecast & stock management issues No. of Vaccines agent used. These issues can be caused Burundi East Timor 1 1 by a lack of sustained flow of funds for Guinea-Conakry Haiti 1 1 vaccines. Funding issues Bhutan 1 Bosnia-Herzegovina 2 Equatorial Guinea 4 Lao PDR 3 Niger 4 Nigeria 7 Senegal 3 Somalia 2 Swaziland 1 Tanzania 5 Procurement process issues Indonesia 4 Nepal 1 Quality issues Albania 1 Shipping Georgia 1 Madagascar 2 Rwanda 1 Other West Bank & Gaza 1 Source: Vaccine Security. UNICEF Vaccine Procurement Overview. Rob Matthews, UNICEF Supply Division. UNICEF Supply Division Vaccine Manufacturers Meeting. 22 Case study: Nigeria In 2004, Nigeria’s National Population Commission reported that Nigeria’s national coverage for full immunization was less than 13%, one of the lowest in the world. Vaccine stock-outs significantly impeded vaccination campaigns. A 2003 UNICEF-WHO Joint Vaccine Security Mission highlighted that all routine vaccines were out of stock, that the Government of Nigeria had last released funds for vaccine procurement in April 2001, and that no basic EPI vaccines except for OPV had entered the country since Q3 of 2001.xxxix The EPI managers successfully conducted international tenders, but they could not then award a contract to the chosen vendor due to the Ministry of Finance not releasing the funds.xl For example, although the Ministry of Finance released 60% of the initial 2001 budget for vaccines, they released no funds in 2002, and as of March 2003, the funding cycle has only reached the budget approval stage. The Ministry of Finance also released the 2001 funds late (April 2001) forcing the government to buy vaccine on the spot market, at inflated prices, with uncertain availability, and potentially with less shelf life.xli This non-release of funds has multiple impacts ranging from current vaccination campaigns to future campaigns to global vaccine security. When a country issues a tender and initiates negotiations with distributors or agents, manufacturers initiate production plans to meet the demand, which in Nigeria’s case is substantial. (Due to its large size, Nigeria has a significant impact on DTP, HepB, and Yellow Fever Markets). By not continuing to an order, a country loses credibility with manufacturers, which can lead to fewer bidders for future tenders. Also, manufacturers can incorporate risk factor charges going forward in the vaccine prices offered to that country. And from an international supply point of view, this situation could encourage more manufacturers to exit the market, leading to even further global vaccine scarcity and further increases in cost price. 8. Procuring the highest quality vaccines at a reasonable price To compare which procurement channel, UNICEF procurement or government self- procurement, is less expensive, all costs must be considered. This includes: UNICEF procurement: • UNICEF-procured vaccine cost. • UNICEF overhead cost. Government self-procurement: • Government procured vaccine cost. • Government procurement cell. Cost as a percentage of order will vary depending on country size, procurement cell efficiency, and depth of service the procurement cell offers. • The cost of the two incremental NRA critical functions. It is extremely challenging to compare vaccine prices. Vaccine manufacturers offer different prices for each vaccine to each customer and in each contract. Elements affecting price include: xlii • Quality and WHO-certification • The order size • The number of shipments anticipated 23 • When the shipments are needed • Shipping distance • Currency of payment • Payment terms • Product specification including special labeling, packaging and handling requirement • Risk (e.g. risk of non-payment, of currency devaluation, etc) • The number of producers • The market size • Tiered country pricing for vaccines • Patents, intellectual property rights and know-how • Manufacturer inventory positions • Competition in marketplace Final country vaccine prices may also include shipping insurance, inspection, testing, custom duties, clearance fees, procurement agent fees, handling fees, licensing and quality control costs. Most countries waive custom duties and taxes on vaccines. The final factor, competition in the marketplace, is a significant driver of price over the long term. The amount of manufacturer competition for a tender is impacted by two factors: a) Vaccine manufacturer competition: Vaccine prices tend to fall when more manufacturers enter the market and can compete for business. For example,xliii • The world price of plasma-derived hepatitis B vaccine fell from $15-$30 per dose to less than $1 per dose in 1987 when two Korean manufacturers responded to an international tender for hepatitis B. Up to that time, only two firms produced the vaccine. • PAHO’s price for measles and hepatitis B vaccines fell in the 1990s when producers from Asia submitted bids for the first time and were awarded contracts. The manufacturing level of competition is the primary driver of global vaccine prices. The World Bank should therefore focus on increasing competition at the manufacturer level. b) Procurement agent competition: The type of procurement mechanism can impact how many manufacturers respond to a tender. Because the number of manufacturers responding to a tender is the key driver in competition and lower prices, sometimes a sole source procurement agent decision will actually maximize competition. For example, UNICEF has historically worked to maximize the number of manufacturers involved in a tender. A self-procuring country sometimes does not have the size to interest many manufacturers to bid or institutes registration requirements which limit how many manufacturers will respond to a tender and thus have less competition on the manufacturing level. Registration requirements can impact a tender process in the following manners: • European manufacturers: In general, if an EPI tender requires significant 24 investment, European manufacturers will most likely not bid because they are more focused on advanced combination vaccines. With the exception of their commitment to polio eradication, they are moving out of the older EPI vaccines that many developing countries rely on and so will not make significant investments (such as local operational capacity) in a developing country in order to sell these products there. • Developing country manufacturers: These manufacturers tend to focus more on the traditional EPI vaccines and are more willing to meet local registration and operation requirements, depending on country size. • Domestic firms: Domestic firms can either manufacture the vaccine directly (if the firm and the country’s NRA have passed WHO pre-qualification) or can import the vaccine from a WHO pre-qualified international manufacturer. If domestic procurement agents are the prime tender respondents, these agents can be the limiting factor for competition in the country because each domestic firm will usually respond to a tender with just one international manufacturing partner. Case Study: Pakistan Since Pakistan moved from UNICEF procurement to self-procurement, it has launched three tenders. Each has received a small number of respondents, as shown in Table 5. Table 5 Number of Respondents to Pakistani tenders Tender Product Respondents Pass Technical Tender Awarded? Qualifications January 2008 BCG Two One Yes January 2008 TT Two One Yes January 2008 OPV Two Two Yes January 2008 Measles One One Yes March 2008 mOPV1 Three One No March 2008 mOPV3 Two Zero No March 2008 tOPV Four One Yes April 2008 mOPV1 One One Yes April 2008 mOPV3 One One Yes Source: Tender evaluation documentation. Ministry of Health. Government of Pakistan. 2008. Two March tenders had to be relaunched, impacting polio vaccination campaigns. While 71M trivalent OPV doses intended for a June 3rd Sub-National Immunization Day were delivered in March, the mOPV1 and mOPV3 vaccines will not be delivered in time for distribution because the government did not enter into a contract for these until mid-May. Thus, this polio immunization campaign is being delayed. Possible reasons for small number of manufacturers responding to these tenders include: • UNICEF will not respond to an open tender because this could potentially put it in a conflict of interest situation with the vaccine manufactures who supply UNICEF under the Long Term Agreements. UNICEF did notify their manufacturers that Pakistan would be moving to self-procurement and asked them to release quantities dedicated to UNICEF for Pakistan.xliv • Few international vaccine manufacturers can respond to a Pakistani tender due to two tender clauses: o Clause 2.4: “The bidder should have operational office(s) in Pakistan…�xlv Only Glaxo Smith Kline has an operational office in Pakistan. 25 o Clause 3.2: “In case of vaccines, besides WHO pre-qualification, these must: (a) be registered with the Ministry of Health, Government of Pakistan.� xlvi One of the Registration Board’s requirements is, according to Form 5A, “For import purpose a sole agent in Pakistan is required to be nominated by the principal / manufacturer abroad.�xlvii • The Government of Pakistan is restricted from purchasing any products from either Israel or India. Because India is one of the largest suppliers of vaccines, this provides a challenge, but waivers can and have been obtained. • Four domestic firms responded to the tender, each time proposing to import a vaccine from one international manufacturing partner. The four firms are:xlviii o Amson Vaccine & Pharma responded to six tenders. They proposed to import from Japan BCG Lab, Shantha Biotechnics, Novartis, and Panacea Biotec. o Hospital Sales and Services responded to two tenders, always proposing to import from Serum Institute. o Sind Medical Stores responded to three tenders, always proposing to import from Sanofi Pasteur. o Majeed Sons responded to two tenders, always proposing to import from Novartis. In this case, the fact that Amson Vaccines & Pharma has relationships with so many manufacturers but will only propose one manufacturer per tender can limit competition. It is impossible to say which procurement channel offers the lowest vaccine price in every single circumstance because each order receives a different price due to the large number of potential factors impacting price. A 1996 USAID/BASICS/PATH study found that, “in terms of price, UNICEF and PAHO receive substantial discounts from manufacturers since they are high-volume purchasers, but the addition of an administrative fee, e.g. 6-8% for UNICEF, has frequently bought the total cost up to, or beyond, the lowest vaccine prices offered directly to public-sector markets.�xlix In April 2007, after significant criticism over their fees, UNICEF reduced fees globally. UNICEF now charges all Least Developed Countries (LDCs) a handling fee of 4% for EPI vaccines and immunization devices, 3% for new vaccines, and 8% for cold chain equipment. UNICEF charges non-LDCs 4.5% for EPI vaccines and immunization devices, 3.5% for new vaccines, and 8.5% for cold chain equipment.l UNICEF also adds an inspection fee. These fees are standard for countries regardless of order size. A vaccine price less than UNICEF’s prices is rarely seen on the marketplace. Clause 4.33 in UNICEF procurement contracts states: “If at any time during the validity period of the Long Term Agreement [to purchase vaccines], the awarded supplier offers to sell the vaccine at a price lower than the price effective under the Long Term Agreement, the awarded supplier shall offer the same price to UNICEF for the remaining validity period of the Long Term Agreement.� UNICEF has in the past enforced this clause with suppliers but with some exceptions. Cases where UNICEF did not enforce the clause usually revolve around extenuating circumstances with e.g. a vaccine oversupply or a government pressuring a state-owned or a domestic manufacturer to sell it vaccines at below cost.li Three of the countries surveyed offered vaccine pricing comparisons between UNICEF- procured prices and self-procured prices. In all cases, UNICEF procurement channels offered significantly lower prices. 26 Table 6 on the next page shows that Government of Botswana, which procured by itself, obtained pricing anywhere from 25% to 114% higher than the UNICEF World Average prices. This is most likely at least partially due to Botswana’s small purchase volume. 27 Table 6 Botswana vaccine procurement 2008: Price comparison with UNICEF Product Quantity Price Paid Total Unit UNICEF Price Price Amount Currency Price in Price 2008 premium Premium in Dollars1 in World dollars Dollars Average Price2 Polio 8,113 38.00 USD 308,294 0.19 0.15 25% 62,430 BCG 2,600 1.45 EUR 5,858 0.23 n/a3 - - DTP 6,920 1.60 EUR 17,203 0.25 0.18 42% 5,054 DT 3,900 1.25 EUR 7,575 0.19 0.11 77% 3,295 HepB 34,200 3.00 EUR 159,416 0.47 0.22 112% 84,364 Measles 14,100 1.90 EUR 41,625 0.30 0.23 27% 8,915 TT 5,200 1.15 EUR 9,291 0.18 0.08 114% 4,944 Total 75,033 $ 549,262 45% $ 169,002 Source: Interview with Mr. Kgosiemang. Chief Pharmacist. Central Medical Stores. Ministry of Health. Government of Botswana. April 2008. Compared with UNICEF World Average Prices at http://www.unicef.org/supply/index_7991.html Nigeria has a substantial vaccine volume but, as shown in Table 7, received prices ranging from 8% to 248% higher than UNICEF prices. This is most likely due to a lower purchase volume than UNICEF and short purchase timeframes leading to ad hoc tenders on the spot market in 2002. Table 7: UNICEF’s average 2003 prices compared to prices offered to Nigeria in its 2002 tender4 Product Latest Nigeria tender UNICEF World Average Percent increase in price price 2002 (USD/dose) Price 2003 (USD/dose)5 BCG 20 d 0.135 0.063 116% DPT 20 d 0.180 0.085 112% Measles 10 d 0.220 0.131 67% TT 20 d 0.140 0.040 248% YF 10 d 0.430 0.398 8% Source: Personal communication with UNICEF – Copenhagen. April 2008. Table 8 on the next page shows the prices that the Government of Pakistan received on the global market. Their prices are also higher than UNICEF even though Pakistan has a substantial volume. Factors increasing their price include lower purchase volumes than UNICEF, limited competition for tenders, additional cost of local agents, more favorable tender terms, and shorter tender timeframes. Annex 5 details these more fully. 1 Currency converted in May 2008 to USD at rate of 1.55376 2 UNICEF procures from a number of different manufacturers with different prices. The actual price countries receive would vary depending on availability. Price includes 4.5% procurement fee. 3 UNICEF does not buy BCG in the 10 dose vial presentation that Botswana procures. 4 All prices are cost prices, i.e. freight, insurance and clearance cost should be added to both tender prices and UNICEF prices. 5 Includes a 6 % handling fee. As UNICEF procures from a number of different manufacturers with different prices, the actual price countries receive would vary depending on availability. In 2003, UNICEF was in the process of tendering for its 2004-2006 demand and so their manufacturer price increased from 0- 36% starting in 2004. 28 Table 8 Government of Pakistan self-procurement vaccine prices, 2008 UNICEF Cost/ 2008 Increase unit adjusted Increase in price: incl. cost/unit in price: total Date Vaccine Quantity Agent Mnf freight6 7 percent dollars Japan Jan-08 BCG 792,000 Amson 0.161 0.117 38% 35,128 BCG Shantha Jan-08 TT 979,500 Amson 0.061 0.059 4% 2,273 Biotech Sanofi Jan-08 tOPV 2,765,500 Sind 0.198 0.170 16% 77,066 Pastuer Serum Jan-08 Measles 2,644,500 Amson 0.325 0.260 25% 171,139 Insti Mar-08 tOPV 71,111,100 GSK GSK 0.174 0.170 3% 323,943 Sanofi May-08 mOPV1 20,000,000 Sind 0.203 0.164 24% 779,248 Pasteur May-08 mOPV3 20,000,000 GSK GSK 0.223 0.176 27% 950,818 Total 118,292,600 12% 2,339,614 Source: Government of Pakistan tender documentation. May 2008. Please note that the above price analyses do not include the cost of a government procurement cell nor the cost of the two required incremental NRA functions required for government self-procurement. These are traditionally accounted for elsewhere within government financing. A full analysis would include these and increase government self- procurement costs accordingly. 6 Tender costs based in Pakistani rupee and converted at January 2008 rate of 0.01601281, March 2008 rate of 0.016334275 and May 2008 rate of 0.014488633. 7 Based on UNICEF 2008 published World Average Prices with the following adjustments: 1). 10% added to UNICEF price due to the fact that Pakistan prefers not to purchase from Indian manufacturers. UNICEF has historically procured vaccines for Pakistan from European manufacturers who have been 0-50% higher than the UNICEF World Average Price, depending on the order. Pakistan has obtained waivers to procure from India but the majority of their procurement came from Europe. 2). 4.5% UNICEF handling and inspection fee added. 3). 1.832% freight fee added. UNICEF does not include freight. Average UNICEF freight costs varies based on quantity ordered, but do not vary much based on port of origin as most UNICEF vaccines for Pakistan ship from Europe. For the 39 orders placed under the $42M World Bank IDA buy-down fund implemented in July to December 2007, freight charges were 1.826% of vaccine costs. For the 18 orders placed with $21M of the $74M of the second IDA buy-down spent in December 2007 and January 2008, freight were 1.838% of vaccine costs. For this analysis, an average of these freight costs was used. 29 9. World Bank procurement standards and UNICEF Four considerations inform the World Bank guidelines for procurement under IBRD loans and IDA credits:lii ƒ The need for economy and efficiency in project implementation, including the procurement of the goods and works involved. ƒ The Bank’s interest in giving all eligible bidders from developed and developing countries the same information and equal opportunity to compete in providing goods and works financed by the Bank. ƒ The Bank’s interest in encouraging the development of domestic contracting and manufacturing industries in the borrowing country. ƒ The importance of transparency in the procurement process. Historically, UNICEF and the World Bank have had some challenges in partnering for vaccine procurement, mostly focused around the following areas: a) Sole source: The World Bank considers UNICEF as a sole source supplier. World Bank procurement guideline Clause 3.10 indicates that when a UN agency acts as a procurement agent, the UN should follow all Bank procurement guidelines for its procurement which includes open competitive bidding.liii However, while open competition remains the basis for efficient public procurement, the guidelines instruct borrowers to select the most appropriate method for the specific procurement. Clause 3.9 indicates that “procurement directly from specialized agencies of the United Nations, acting as suppliers, pursuant to their own procedures, may be the most appropriate way of procuring … specialized products where the number of suppliers is limited such as for vaccines or drugs.� b) Audit: World Bank procurement guideline 1.11 instructs the Bank “to review the borrower’s procurement procedures, documents, bid evaluations, award recommendations, and contracts to ensure the procurement process is carried out in accordance with the agreed procedures but offers flexibilities in the extent to which the review procedures applies to different categories of goods and works to be financed.� The guidelines also stipulate that the World Bank has the right to inspect the accounts, records, and documents relating to the bid submission and contract performance and to have them audited by auditors appointed by the Bank. UNICEF, as a specialized agency of the United Nations, follows UN regulations which provide internal oversight through UNICEF Financial Rules and Regulations, UNICEF Procurement Procedures, Ethical Code, Contracts Review Committee (independent from contracting staff), and Procurement Training and Certification of contracting staff. External oversight is provided from UN External Audit (French audit in mid-2004 focused on vaccine procurement; no major findings) and UNICEF Internal Audit.liv UN policy does not allow for third party audits. 30 UNICEF procurement contracts with governments stipulate that:lv • “The parties recall that UNICEF’s financial books and records are routinely audited in accordance with the internal and external auditing procedures laid down in UNICEF’s Financial Rules and Regulation.� • “The Parties acknowledge that in accordance with the Financing Agreement, the Government has an obligation to deliver to the Association an annual audit certificate in respect of the funds provided pursuant to that agreement. In order to enable the Government to discharge this obligation, UNICEF will on request respond to all reasonable information requests from the Government and provide all assistance, as may be agreed, between the Parties.� c) Liability: When UNICEF acts as a procurement agent or supplier for a country, it does not take legal liability for the products delivered. This rests with the manufacturer from whom UNICEF procured. While UNICEF has in the past worked with the country and the manufacturer to resolve issues, their contracts do not bind UNICEF to do so. d) Payment terms: World Bank guidelines stipulate that payment shall be made upon delivery of goods whereas UNICEF financial regulations require payment in UNICEF’s account before it can place an order. These four challenges have lead to long delays under Bank supported procurement using the services of UNICEF as they require extra clearances and waivers to World Bank procurement guidelines at the highest level in the Bank. The World Bank has waived audit and liability clauses in Agreements between Governments and UNICEF which it finances, as it considers these risks low. At the time of the writing of this report, discussions between UNICEF and the World Bank at the institutional level are ongoing to come to a mutually satisfactory arrangement. 10. Conclusion An analysis of the five core principles that should apply to World Bank financed vaccine procurement decisions does not provide a clear-cut rule for which type of procurement mechanism the World Bank should support in all cases. This is because vaccine procurement is a complex issue that interweaves the domains of public health, market dynamics, commodity security, ethics, and procurement capacity. Its cross-disciplinary nature means that neither a straightforward analysis stemming from just one discipline nor a cookie-cutter application of World Bank procurement principles of economy, efficiency, equal opportunity, promoting domestic contracting and transparency will lead to an optimal solution. In fact, the general application of such procurement principles without considerations for other factors such as vaccine security led to the vaccine market crash in 1999 in the first place. 31 The World Bank has therefore historically considered vaccine procurement as “special� and has waived for it the strict application of its procurement guidelines. Going forward, the World Bank should give highest priority to the considerations of vaccine security and the fragility of the market. When this is combined with the lower total vaccine procurement cost most likely achieved through UNICEF procurement channels and the challenges developing countries face with regards to vaccine procurement, as illustrated by the case studies, the World Bank should shift from putting the burden of proof on country teams for explaining why they would like to work with UNICEF for procurement of vaccines to putting the burden of proof on country teams to explain why they would like to work with country procurement systems. If the World Bank determines that a national country’s procurement systems do not have the capacity to act as an effective agent, the World Bank should work to strengthen procurement and quality control institutions. High priority should be given to technical assistance should for NRA capacity building and this capacity should be in place before national procurement cells are activated and used. Countries with adequate procurement and quality control capacity have fewer procurement challenges, but will most likely not be able to provide as low-cost of a procurement option as UNICEF is able to. UNICEF will most likely always provide a lower price due to its larger volumes and resulting greater competition among manufacturers bidding on UNICEF orders. Furthermore, individual country self- procurement decisions may negatively impact global vaccine security unless countries incorporate these principles into their procurement decisions. Finally, the decision to move from UNICEF procurement to government self-procurement can be driven by local private sector interests and not by concern for increased efficiency and effectiveness of procurement. Therefore, caution should still be exerted when moving from UNICEF procurement to self-procurement. Therefore, caution should still be exerted when moving from UNICEF procurement to self-procurement in countries that have full procurement and quality control capacity. 32 Annex 1: UNICEF Procurement Services UNICEF is not just a procurement agent but is actively involved in almost all facets of vaccinations. Some of UNICEF’s programs regarding vaccines include:lvi • Working with manufacturers to ensure an adequate supply is available. • Working with countries to accurately forecast demand. • Working with countries, donors, and recipient partners to secure funding. • Purchasing and transporting vaccines. • Training health workers. • Galvanizing local communities to bring children to vaccination campaigns. Countries do not have to work with UNICEF as a procurement agent in order to receive UNICEF technical assistance in these areas. In-country UNICEF supply staff work locally with the government during the vaccine procurement process to ensure adequate document preparation and delivery assistance. The local UNICEF office does not receive any of the procurement handling fees directly. Figure 5 shows which countries work with UNICEF for vaccine procurement. Individual countries can significantly impact UNICEF’s procurement. The top ten countries constitute 66% of total demand in terms of dollars. In 2007, the largest countries in terms of UNICEF volume were India, then Pakistan, then Bangladesh. UNICEF uses this capacity to reallocate demand and supply between countries and suppliers to avert crisis and UNICEF Quantity of vaccine UNICEF procures Figure 6: SD Annual procurement value of vaccines, in million USD reduce the impact of demand $700 and supply fluctuations.lvii If a $600 large country drops from Continued scaling up of campaigns, new vaccine introduction and UNICEF procurement, this $500 increased routine coverage limits UNICEF’s ability to $400 manage demand and shift Campaigns + GAVI $300 suppliers. $200 Polio Figure 6 shows how the dollar $100 amounts of UNICEF vaccines $0 purchases increased. In 2007, 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 UNICEF procured vaccines for Source: UNICEF Supply Division. Copenhagen. May 2008. 33 approximately 30% of the world’s children for routine immunizations and when supplemental polio and measles vaccines are included, it procures about 40% of the total vaccines doses or over 3 billion doses, worth approximately $617 million (2.4 billion of those doses were for polio.)lviii UNICEF procures on behalf of 84% of children in low income countries (excluding the self-producing countries of India, China, and Indonesia).lix UNICEF Funding and Tender Process UNICEF invites all manufacturers to apply for multi-year tenders and only WHO pre- qualified manufacturers to compete for single-year tenders. Before a tender, UNICEF weights the importance of the following factors: past on-time delivery performance, price, delivery schedule, future production plans, Vaccine Vial Monitors, shelf-life, ability to maintain buffer stock, timeline for emergency response (packaging), and shipping performance. During a tender, it then chooses a manufacturer for an LTA based upon a quantitative and qualitative review of these factors. For example, in 2003, UNICEF awarded LTAs for HepB. This is a small tender with decreasing annual demand tender due to the rise of combination doses. At the time of the tender, there were five WHO pre-qualified manufacturers, many of whom had production capacities greater than the annual 20-40M dose demand. UNICEF used their methodology to award all five manufacturers, with those offering the lowest price and best historical performance receiving the largest quantities.lx If a manufacturer is not WHO pre-qualified at time of tender, their proposal must include a detailed plan on the timeline to obtain WHO pre-qualification. Upon obtaining WHO pre-qualification, UNICEF will consider re-allocating quantity to the manufacturer if UNICEF is facing a monopoly or near-monopoly situation, there is a lack of performance of current manufacturers or insufficient supply from current manufacturers.lxi UNICEF procures for approximately 100 countries, each one having 3-7 funding sources. UNICEF’s financial rules and regulations require payment in advance. Because UNICEF relies on external funding, it cannot make a firm purchase order without having internal committed commensurate funding in place. In general, 95% of funding for over 2 billion doses of vaccines arrives 1-3 months prior to delivery. UNICEF is cautious to make major changes that will set precedent or expectations that cannot be applied to all vaccines or may not be sustainable.lxii UNICEF designed its tendering process to lessen a manufacturer’s timing risk. For example, UNICEF’s process for the 2007-2009 period was as follows:lxiii • June 2005: UNICEF internal review of tender documents based on vaccine forecasts for 2007-09 period. • December 2005: UNICEF and vaccine manufacturers meet • January 2006: UNICEF issues tender • March 2006: Tender closes • July 2006: Tender awards made 34 Countries’ processes, usually driven by an annual budgeting cycle, are much shorter. For example, Botswana started its April 2006 - March 2007 tendering process in April 2006. Manufacturers who supply UNICEF UNICEF procures from both developed and developing countries. Since the 1990s, new manufacturers from developing countries have started to fill the supply gap in traditional vaccines. Table 9, on the next page, shows that UNICEF increased both the dollar amounts of vaccines purchases from 2000 to 2006 by 538% and increased the percentage of total vaccine purchases from developing countries from 34% in 2000 to 37% by 2006. This represents a $167M increase for manufacturers from developing countries. 35 Table 9 UNICEF Vaccine Procurement Sources 2000 and 2006 Country Company 2006 2000 Change Value Percent Value Percent Developed Countries Australia CSL Limited $378,385 0.08% $285,000 0.20% 133% Belgium GlaxoSmithKline $171,469,155 34.92% $32,397,032 23.14% 529% Canada Intervax $1,122,711 0.23% $2,463,959 1.76% 46% Aventis Pastuer $113,150 0.08% Denmark Statens Serum Institut $4,163,414 0.85% $2,519,205 1.80% 165% France Aventis Pasteur $15,589,249 11.14% Germany Chiron Behring $19,198,668 13.71% Italy Novartis $21,964,330 4.47% Israel Sarel $114,905 0.02% Japan Eisai Co $637,365 0.46% BCG Laboratory $4,058,780 0.83% $1,420,281 1.01% 286% Netherlands NVI $132,090 0.03% Russian State Unitary $264,600 0.19% Federation Enterprise Sweden SBL Vaccin $130,000 0.03% Switzerland Berna Biotech $19,561,981 3.98% Swiss Serum & $830,390 0.59% Vaccine Institute USA Merck $1,261,552 0.26% $136,500 0.10% 924% Other orders less than $100,000 $84,446,073 17.18% $16,308.640 11.65% 518% Subtotal $224,242,398 45.6% $75,855,399 54.18% 296% Developing Countries Brazil FIOTEC $955,500 0.19% China MoH $1,619,996 1.16% Cuba Heber Biotec $240,500 0.05% India Haffkine $620,880 0.13% $9,940,000 7.10% 6% Panacea Biotec $124,281,631 25.31% $20,653,209 14.75% 602% Serum Institute $38,171,454 7.77% $10,995,811 7.85% 351% Shantha Biotechnics $5,667,721 1.15% Indonesia P.T. Bio Farma $3,565,240 0.73% $879,159 0.63% 406% Republic of Cheil Jedang Corp $210,440 0.15% Korea Green Cross Vaccine $2,737,660 1.96% LG Life Sciences $8,598,603 1.75% $157,590 0.11% 5456% Senegal Inst. Pasteur de Dakar $210,000 0.04% $642,096 0.46% 33% Subtotal $182,311,529 37.13% $15,489,480 34.17% 1177% Other orders less than $100,000 $84,446,073 17.18% $16,308.640 11.65% 518% Total $491,000,000 100% $140,000,000 100% 538% Source: UNICEF Supply Division Annual Report 2006 and 2000. 36 Annex 2: Vaccine Procurement in Bangladesh Bangladesh has large vaccination needs with a population of 150 million. In March 2006, Bangladesh had its first polio case since 2000 and responded with significant polio vaccination campaigns, further increasing its vaccine procurement requirements. Table 10, below, shows Bangladesh’s vaccine procurement volume. OPV, measles and TT annual variations follow supply for campaign activities. The 2006 volume represents 8- 10% of the total UNICEF vaccine global purchase volume for 2006.lxiv Table 10 Vaccine doses procured - Bangladesh Vaccine 2001 2002 2003 2004 2005 2006 Total BCG 25,000,000 19,856,000 31,640,000 17,632,000 25,100,000 29,442,000 148,670,000 DTP 12,000,000 18,465,500 18,848,000 19,000,000 20,000,000 19,162,000 107,465,500 HepB 1,001,000 1,681,000 6,083,000 4,777,800 17,518,000 10,468,700 41,529,500 Measles 17,500,000 13,368,000 19,800,000 7,637,600 57,965,000 12,796,000 129,067,400 OPV 95,708,000 76,533,000 123,424,000 3,033,000 24,601,000 249,600,000 572,919,000 TT 21,600,000 37,745,000 16,748,000 19,616,500 45,327,500 21,419,000 162,456,000 Other 15,750 15,750 Total 172,809,000 167,648,500 216,543,000 71,696,900 190,511,500 342,903,450 1,162,123,150 Source: Personal correspondence with Thomas Sorenson, UNICEF Supply Division, Copenhagen. October 2007. Table 11, below, shows immunization coverage rates in Bangladesh. 8 Table 11 WHO-UNICEF Reported Immunization Coverage for Bangladesh 2006 2005 2004 2003 2002 2001 2000 BCG 96 96 96 95 95 94 95 DTP – first dose 96 96 96 93 95 94 93 DTP – third dose 88 88 88 81 83 85 83 HepB – third dose 88 62 10 5 Measles 81 81 81 76 75 77 76 TT – second dose Polio – third dose 88 88 88 82 83 85 83 Government of Bangladesh Reported Immunization Coverage for Bangladesh 2006 2005 2004 2003 2002 2001 2000 BCG 98 96 92 95 95 94 95 DTP – first dose 97 95 97 91 86 87 88 DTP – third dose 84 78 96 72 69 65 68 HepB – third dose 84 84 59 5 Measles 84 78 96 69 65 64 61 TT – second dose 94 88 41 89 89 90 89 Polio – third dose 92 84 96 72 70 66 68 Source: WHO website: http://www.who.int/vaccines/globalsummary/immunization/countryprofileselect.cfm. Accessed April 2008. 8 Government estimates are the official country estimates. UNICEF / WHO estimates are based on government estimates but adjusted to take into account program strength, Independent National Immunization Coverage Survey figures, and population denominator figures. 37 GAVI is currently reviewing the Government of Bangladesh’s application for pentavalent HepB and will make a decision in June 2008. If approved, it will be introduced in 2009. GAVI has an Inter-Agency Coordination Committee (IACC) for vaccines in Bangladesh to improve coordination among partners. It includes UNICEF, WHO, Japanese International Cooperation Agency (JICA), UK Department for International Development (DFID), World Bank, United States Agency for International Development (USAID), Rotary International, Canadian International Development Agency (CIDA), Ministry of Health, Ministry of Finance, Ministry of Forestry, local government agencies, BRAC, and International Center for Diarrhoeal Diseases, Bangladesh (ICDDR,B).lxv Vaccine Procurement UNICEF procured all vaccines in Bangladesh except for a negligible amount used in private hospitals. Therefore, while there are significant EPI skills in Bangladesh, there are very few vaccine procurement skills in Bangladesh. The Government of Bangladesh currently has a multi-year plan for immunization that includes costing for the years 2008-2012. The national budget has a line item for vaccines.lxvi Either World Bank IDA financing or a consortium of donors who pool funds that the World Bank then manages financed $42.8M for vaccines in 2006. The funds passed directly from the World Bank to UNICEF Copenhagen. In 2006, UNICEF charged 6% for their procurement services (fee subsequently lowered to 4%), resulting in $2.57M in calculated procurement fees (or almost $600,000 excluding polio vaccines). The Government of Bangladesh has applied for GAVI assistance for pentavalent vaccines. All parties interviewed in Dhaka feel that UNICEF is doing a very good job with vaccine procurement. The Expanded Program on Immunizations (EPI), the part of the Government of Bangladesh responsible for vaccinations, receives the vaccines on time and the process works smoothly. UNICEF will procure and deliver vaccines in about three months time. EPI prefers that UNICEF manages vaccine procurement as opposed to the Central Medical Supply Department (CMSD) in the Government of Bangladesh because it has concerns regarding CMSD procurement capabilities and has historically been pleased with the service it has received from UNICEF.lxvii UNICEF and EPI have a working relationship with high levels of trust. UNICEF often goes above and beyond what is contractually required. In 2002, the World Bank conducted an assessment of the Government of Bangladesh’s procurement capacities. It found that actual implementation of procurement to be “quite uneven� and “far from satisfactory.� Some unsatisfactory features were:lxviii • Procurement delays - These were the single most serious malady found. It should take the government 180 to 330 days to successfully complete an international tender.lxix However, some contracts have been in process for over 1100 days. 38 • Poor advertisement • A short bidding period • Poor specifications • Nondisclosure of selection criteria • Award of contract by lottery • One-sided contract documents • Negotiation with all bidders • Rebidding without adequate grounds • Other miscellaneous irregularities • Possible corruption and outside influence. In 2005, the World Bank commissioned an Independent Procurement Review of World Bank financed projects in Bangladesh where the government was responsible for procurement. This review focused on a total of 75 contracts within four World Bank projects. The review audited 29 contracts valued at $11.3M in the Health and Population Program Project (HPPP) and 6 contracts valued at $6.3M in the HIV/AIDS Prevention Project. The review found that the procurement in the HPPP project was “highly insufficient� while procurement in the HIV/AIDS project was “insufficient.� Specific areas of issue included: • Filing and transparency • Indications of collusion between bidders • Insufficient World Bank supervision • Untimely project execution • Payment delays Government procurement, however, is gradually taking on more functions successfully. For example, UNFPA used to procure $40-$50M of condoms annually for Bangladesh with a 5-6% procurement fee. In 2000-2001, the Government of Bangladesh switched to self-procurement for condoms. As a result, the government now successfully procures condoms without paying up front (UNFPA requirement) and without the procurement fee. It also found that they could procure the condoms for a lower price than UNFPA.lxx 39 Annex 3: Vaccine Procurement in Botswana Botswana is a middle-income country (2006 GNI per capita of $5,950lxxi) with a small population of approximately 1.7 million and a birth cohort of approximately 43,000.lxxii Botswana self-finances vaccine purchases with no donor assistance. The government self-procures vaccines and then provides vaccines and vaccination services free of charge to the population. The World Bank does not have operations currently in Botswana. Historically, Botswana has prioritized vaccines and has achieved very good coverage rates, as shown in Table 12 below. In 2004 and 2006, the Government of Botswana made changes to their procurement rules which caused significant problems for vaccine procurement. As a result, Botswana had polio, measles, HepB, BCG, diluents and syringes stock-outs in the past year. The immunization coverage rate impact is, as of yet, unknown because 2007 coverage rates has not yet been fully reported. However, in February 2007, a national EPI coverage survey identified that 5040 children were not vaccinated. 70.8% of these children were not vaccinated because the vaccine was not available.lxxiii 9 Table 12 WHO-UNICEF Reported Immunization Coverage for Botswana 2006 2005 2004 2003 2002 2001 2000 BCG 99 99 99 99 99 99 99 DTP – first dose 98 98 98 98 98 98 98 DTP – third dose 97 97 97 97 97 97 97 HepB – third dose 85 85 85 85 85 85 85 Measles 90 90 90 90 90 90 90 TT – second dose 84 83 76 75 71 69 68 Polio – third dose 97 97 97 97 97 97 97 Government of Botswana Reported Immunization Coverage for Botswana 2006 2005 2004 2003 2002 2001 2000 BCG 99 99 93 92 77 72 73 DTP – first dose 99 99 98 99 99 82 9 DTP – third dose 99 99 89 93 87 74 85 HepB – third dose 99 99 79 78 46 64 73 Measles 99 99 86 90 79 77 84 TT – second dose 71 72 55 55 49 46 45 Polio – third dose 99 99 88 91 85 75 85 Source: WHO website: http://www.who.int/vaccines/globalsummary/immunization/countryprofileselect.cfm. Accessed Oct. 2007. Botswana is not procuring tetravalent or pentavalent vaccines. The National Standing Committee on Drugs (NASCOD), comprised of public health professionals and doctors within and outside the Ministry of Health, is responsible for determining which vaccines Botswana should procure. The EPI team in the Ministry of Health supplies evidence to NASCOD as well. EPI does not feel that Botswana has a large enough need to require 9 Government estimates are the official country estimates. UNICEF / WHO estimates are based on government estimates but adjusted to take into account program strength, Independent National Immunization Coverage Survey figures, and population denominator figures. 40 Hib vaccines. EPI is considering doing a study to evaluate the need for pentavalent vaccineslxxiv and is currently not recommending tetravalent vaccines due to the changes this would require in the current vaccination schedule.lxxv Vaccine Procurement Vaccine procurement, as with many health functions in Botswana, is centralized at the federal level. Vaccines and pharmaceuticals are included under the same vote in the government budget and share a line item. The Principle Financial Officer of the MoH indicated a strong reticence to separating the budget line for vaccines or for other health programslxxvi and the Central Medical Stores (CMS) reported that there has always been enough money for all required vaccine purchases, even with escalating antiretroviral purchases.lxxvii Corruption is very low in Botswana and does not play a significant factor in vaccine procurement. CMS, a unit under the control of the Department of Clinical Services within the Ministry of Health, forecasts, procures and distributes all vaccines, pharmaceuticals and medical devices for Botswana. Historically, vaccines were tendered once every two years and supplied on an annual basis. CMS would like to move towards four procurement cycles a year. Once delivered to CMS, CMS distributes the vaccines to all 24 districts based on the pull method, whereby the district submits a requisition to CMS approximately every three months. CMS adjusts the requisition based on past orders and CMS stock availability. The CMS also reports through the Ministry of Health to the Public Procurement and Asset Disposal Board (PPADB). The PPADB, established under the Public Procurement and Asset Disposal Regulations 2006, guides all government procurements. Its new procurement regulations, largely based on World Bank and other international guidelines, have significantly altered vaccine procurement requirements in the past and changes are likely to continue for some time while processes are developed to meet the new regulations. Vaccine Procurement Challenges in Botswana Several vaccine stock-outs have occurred in Botswana in recent years. The main causes are tender delays and improper forecasting but all the following reasons contribute: 1. Tender delays: CMS has had significant difficulties complying with the new PPADB regulations. CMS could not obtain a positive outcome from the PPADB on documentation submitted and could not launch a tender for vaccines in 2006. They therefore had to conduct direct purchases on an ad hoc basis once CMS stock levels had depleted. This resulted in rationing to the districts and stock out issues. In 2007, CMS launched a temporary "stop-gap" tender to supply vaccines through the end of March 2009.lxxviii The two areas causing CMS the most problems are: • The 2004 Civil Empowerment Directive states that if a resident of Botswana can respond to a government procurement tender with a product of equal quality and 41 within 10% of international prices, the tender must go to the Botswanan party. CMS does not currently intend to use the Citizens Empowerment Directive because there are no local human vaccine manufacturers. CMS would prefer to work directly with an international manufacturer than with a local agent because the government procurement rules do not give them the authority to monitor and enforce local agent’s cold chain supply.lxxix International manufacturers have to register in Botswana to qualify for a tender, but do not have to have operational offices. Due to the Civil Empowerment Directive, CMS has not been able to develop a tender document template that PPADB accepts and so CMS must submit each tender of greater than 100,000 Botswana Pula (approximately $16,000) through the Ministry of Health Clinical Services cost centre and central Ministry of Health finance to the PPADB.lxxx • The updated regulations invalidated how CMS had historically pre-qualified all suppliers so they have had to re-qualify all suppliers. CMS estimates that, as of May 2008, they are approximately 80% through this process.lxxxi 2. Forecasting: CMS forecasts vaccine needs every two years based on past consumption. This forecast is then adjusted in the second year based on the consumption from the first year. This process is not in line with international best practices because, for example, a stock out in a district would lead to lower consumption and hence lower future ordering requirements. EPI, a unit within the Public Health Directorate of Ministry of Health, would prefer to use target population as a baseline for forecasting but does not feel it is their position to do the forecast because CMS is responsible for procurement. EPI is not involved in forecasting and must alert CMS if activities are planned which would increase needs. In an effort to improve forecasting and communication, EPI and CMS have bi-monthly meetings regarding stock levels and operational matters. However, communication remains weak and EPI has little involvement in the procurement process. 3. CMS Standard Operating Procedures: CMS does not have documented procurement procedures nor Standard Operating Procedures for its activities. They rely heavily on current staff knowledge which is a challenge due to high staff turnover rates. Supply Chain Management Services (SCMS), led by Booz Allen Hamilton, recently reviewed CMS in order to help develop a Quality Manual for CMS along with appropriate Standard Operating Procedures for its activities. While SCMS focused this review on anti-retroviral procurement, it did cover all CMS operational areas. The SCMS local office indicated that the four largest areas of weakness for CMS werelxxxii: 1) Low collaboration between CMS and other entities procuring health consumables such as laboratory supplies 2) Low communication between CMS and the 24 districts 3) Inadequate procurement procedures, including tendering, and low level of staff training and human resource capacity 4) Inadequate IT systems. 4. Quality Control: While all vaccines supplied to Botswana are required to be WHO pre-qualified, Botswana does not have the current required government capacity to ensure 42 quality. As Botswana self-procures vaccines, it should have four of the six WHO critical functions implemented. Table 13, below, provides an informal summary of the current status of these items. Table 13 NRA Capacity in Botswana Function Status Required? Botswana Capacity Independent The Drug Registration Unit (DRU), a unit of Clinical Y Y NRA Services within the Ministry of Health, acts as Botswana’s National Regulatory Authority. 1. Licensing The DRU licenses pharmaceuticals and vaccines Y Y supplied in Botswana. DRU reviews are submitted to the Drug Advisory Board and the National Standing Committee on Drugs (NASCOD) for technical assessment as well. The same procedure, used for both vaccines and pharmaceuticals, has remained unchanged since 1992. Both WHO and SCMS have reviewed the licensing system in 2007 but the report findings have not yet been made public.lxxxiii 2. Post- DRU is not involved in post marketing surveillance. Y N Marketing Active AEFI testing is not done in Botswana but there is Surveillance adverse drug reporting. Pharmacovigilance, a part of the Health Inspectorate Unit does some random and systematic sampling but notes this is the responsibility of the DRU.lxxxiv 3. Lot Release DRU does not have the capability to do lot by lot Y N release.lxxxv 4. Laboratory The National Quality Control laboratory is currently Y N Access for able to do some quality testing on solid dose form Testing pharmaceuticals but is not equipped to perform testing on biological products and vaccines. CMS receipt procedures involve a physical examination of contents.lxxxvi 5. Tender Specification: CMS works with NASCOD to prepare vaccine tender specifications. Botswana has experienced difficulty in providing tenders with full technical specifications (including strains required and packaging) that do not limiting the possible suppliers too much. Also some physical specifications, particularly packaging, have not been compliant with international recommendations. CMS has also included the compulsory supply of Vaccine Vial Monitors (VVM) with all vaccines however it has not always been supplied.lxxxvii The Government of Botswana recognized that their small size means that it does not get priority attention from manufacturers. 43 6. Vaccine Stock Information Management: Information on vaccine stock levels at the district is weak and an adequate stock management system needs to be developed and implemented. District vaccine requisitions do not include information on usage, wastage or current stock levels so CMS has little visibility into district stock levels when determining how much vaccines to send. The districts also do not report back to CMS if they received the vaccines in good order but the transporters must supply evidence of delivery.lxxxviii 7. Budget Cycle: The budget cycle is challenging for vaccine procurement. Botswana’s fiscal year goes from April to March. The government starts to prepare the budget in January and should allocate funds in April. The vaccine tender process cannot commence until funding is allocated. The PPADB process is challenging and can require 2-3 months to clear a tender. Thus, CMS has limited time to run a tender, receive the vaccines, and then pay the suppliers yet still remain in the fiscal year.lxxxix Furthermore, sometimes the government does not allocate and release budgeted vaccine funds immediately in April.xc Vaccine Pricing Table 14, below, shows the pricing the Government of Botswana received for its most recent tender. Table 14 Government of Botswana vaccine prices 2008: Comparison with UNICEF Product Quantity Manufacturer Total Unit UNICEF Price Total Price Price in Price in 2008 premium Premium in Dollars Dollars World dollars 10 Average Price Polio 8,113 Sanofi Pastuer 308,294 0.19 0.145 25% 62,430 BCG 2,600 Serum Institute India 5,858 0.23 n/a11 - - DTP 6,920 Serum Institute India 17,203 0.25 0.168 42% 5,054 DT 3,900 Serum Institute India 7,575 0.19 0.105 77% 3,295 HepB 34,200 Serum Institute India 159,416 0.47 0.32 112% 84,364 Measles 14,100 Serum Institute India 41,625 0.30 0.222 27% 8,915 TT 5,200 Serum Institute India 9,291 0.18 0.080 114% 4,944 Total 75,033 $ 549,262 45% $ 169,002 Source: Interview with Mr. Kgosiemang. Chief Pharmacist. Central Medical Stores. Ministry of Health. Government of Botswana. April 2008. Other Procurement Possibilities Botswana has investigated pooled procurement via the Southern African Development Community (SADC). However, government interest fell after a comparative price analysis for antiretroviral drugs revealed that Botswana received good prices on their own. 10 Currency converted in May 2008 to USD 11 UNICEF does not buy BCG in the 10 dose vial presentation that Botswana procures. 44 UNICEF and the Government of Botswana have discussed UNICEF assistance for procurement. In 2004, the Government of Botswana and UNICEF signed a Memorandum of Understanding which offers UNICEF’s procurement services for standard and non-standard supplies including vaccines. Although some members of the Government of Botswana expressed a desire to work with UNICEF on all vaccine procurement, the current procurement legislation in Botswana requires payment via a Letter of Credit and UNICEF requires prepayment. There are government flexibilities which allow a waiver for prepayment, but in general the Government of Botswana remains committed to the Letter of Credit method and waivers are time-consuming to acquire.xci UNICEF did procure vaccines for a 2004 measles campaign because Botswana needed large quantities and recognized the price advantage of working with UNICEF. For this procurement, the Ministry of Health received a waiver and did prepayment.xcii 45 Annex 4: Vaccine Procurement in Nigeria Nigeria is one of a very few countries in Africa that fully funds the cost of its EPI vaccines from its own resources. It’s large population, 135M, one of the largest in Africa, means procuring and distributing vaccines is a complex challenge. Nigeria has been making step by step progress and has dramatically improved vaccine procurement over the past five years. In 2004, Nigeria’s National Population Commission reported that Nigeria’s national coverage for full immunization was less than 13%, one of the lowest in the world. Some states in northern Nigeria had coverage rates below 1%, and the average for the whole North West Zone was 4%. Nigeria’s immunization program was also one of the most expensive among developing countries. According to the 2004-2008 Five Year National Strategic Plan, the National Program on Immunization (NPI) budgets $28 per child under one. However, if this budget is set against the internationally accepted denominator of children fully immunized before their first birthday, it amounts to $226 per fully immunized child. In a 2003 National Immunization Coverage Survey, parents and caregivers indicated that “vaccine not available� as by far the most important reason as to why children were not fully immunized.xciii Table 15, below, summarizes vaccine coverage rates from 2000-2006. 12 Table 15 Government of Nigeria Reported Immunization Coverage for Nigeria 2006 2005 2004 2003 2002 2001 2000 BCG 42 49 55 63 34 DTP – first dose 87 45 49 48 DTP – third dose 72 38 38 3 HepB – third dose 36 27 8 Measles 99 38 40 30 TT – second dose 60 52 30 Polio – third dose 67 32 37 38 WHO-UNICEF Reported Immunization Coverage for Nigeria 2006 2005 2004 2003 2002 2001 2000 BCG 69 69 62 55 48 46 45 DTP – first dose 72 72 62 53 43 42 42 DTP – third dose 54 54 44 35 25 24 24 HepB – third dose 41 41 Measles 62 62 53 45 36 35 35 TT – second dose 53 51 51 48 46 44 51 Polio – third dose 61 61 54 46 39 32 26 Source: WHO website: http://www.who.int/vaccines/globalsummary/immunization/countryprofileselect.cfm. Accessed Oct 2007. 12 Government estimates are the official country estimates. UNICEF / WHO estimates are based on government estimates but adjusted to take into account program strength, Independent National Immunization Coverage Survey figures, and population denominator figures. 46 Nigeria bases its national health care system on a three-tier system of primary, secondary and tertiary care provided by the local, state and federal governments respectively. The Federal Ministry of Health has the responsibility to develop policies, strategies, guidelines, plans and programs that provide the overall direction for the national health care delivery system.xciv For vaccines, the federal level is responsible for procurement, cold chain support, policy, and coordination.xcv The State Ministries of Health provide secondary level of care and technical advice as well as supervision to the Local Government Authorities (LGAs). The LGAs are the implementers of the primary health care services including immunization.xcvi For vaccines, the state and LGA levels are responsible for implementation. The National Primary Health Care Development Agency (NPHCDA), a federal level parastatal formed in 1992, is responsible for the delivery of primary health care services, the construction of the new health centers and the establishment and training of local development committees to manage local health care. Recently, the National Program on Immunization (NPI) was rolled into NPHCDA because vaccinations should support primary health care. Currently, Nigeria does not have domestic vaccine production capabilities. Until approximately 1998, the Government of Nigeria used to manufacture Yellow Fever vaccines in Lagos. The Government of Nigeria is currently considering possibilities for pubic private partnerships to manufacture vaccines. Vaccine Procurement Challenges Vaccine procurement has steadily improved over the past decade with the successive removal of the following historical roadblocks: 1. Non-release of funds: issue resolve in 2003. 2. Lack of data for forecasting: issue started to resolve in 2005. 3. Distribution at state and Local Government Areas (LGA) level: issue remains a current challenge. 1. Non-release of funds The root cause of the 2002-2003 vaccine shortages was the non-release of funds by the Ministry of Finance. Thus, the EPI managers successfully conducted international tenders, but they could not then award a contract to the chosen vendor due to the Ministry of Finance not releasing the funds.xcvii For example, although the Ministry of Finance released 60% of the initial 2001 budget, they released no funds in 2002, and as of March 2003, the funding cycle has only reached the budget approval stage. The Ministry of Finance also released the 2001 funds late (April 2001) forcing the government to procure vaccine on the spot market at inflated prices, with uncertain availability and potentially with less shelf life.xcviii A 2003 UNICEF-WHO Joint Vaccine Security Mission report highlighted that all routine vaccines were out of stock, that the Government of Nigeria had last released funds for vaccine procurement in April 2001, that no basic EPI vaccines except for OPV had 47 entered the country since Q3 of 2001, and that Nigeria paid more for their vaccines than UNICEF’s prices, as shown in Table 16, below. Table 16: UNICEF’s average 2003 prices compared to prices offered to Nigeria in its 2002 tender13 Product Latest Nigeria tender UNICEF World Average Percent increase in price price 2002 (USD/dose) Price 2003 (USD/dose)14 BCG 20 d 0.135 0.063 116% DPT 20 d 0.180 0.085 112% Measles 10 d 0.220 0.131 67% TT 20 d 0.140 0.040 248% YF 10 d 0.430 0.398 8% Source: Personal communication with UNICEF – Copenhagen. April 2008. Following this assessment, the Government of Nigeria took two steps: • It committed to a timely release of funds • It signed a Procurement Service Agreement with UNICEF which remains in force to this day. UNICEF procurement started in the last quarter of 2003. Vaccine procurement improved remarkably with no report of a stock-out at the national level since the last quarter of 2003. In 2004 and 2005, the Government of Nigeria released 95% and 97% of the provisional plan/budget submitted by UNICEF Supply Division responding to the annual forecast exercise. The 3rd and 4th quarter 2007 funds were received late, however, due to a presidential election and the mergers occurring within the Ministry of Health.xcix Furthermore, part of the funding was used for campaign requirements (meningitis in December 2004 and measles in February 2005) that were not originally included in the budget.c 2. Lack of data for forecasting Nigeria has historically struggled with forecasting. For example, Nigeria submitted only one annual report to WHO on immunization coverage between 1999 and 2005, despite the requirement to do this annually. Within the routine data reporting system there was evidence of systematic falsification of data at the local level in order to meet centrally set targets. Audit teams from EC and the World Bank have noted serious flaws with NPI’s financial reporting.ci Furthermore, UNICEF sometimes could not get stock balance figures from NPI and so assumed zero stock at national level for all vaccines when doing forecasts. 13 All prices are cost prices, i.e. freight, insurance and clearance cost should be added to both tender prices and UNICEF prices. 14 Includes a 6 % handling fee. As UNICEF procures from a number of different manufacturers with different prices, the actual price countries receive would vary depending on availability. In 2003, UNICEF was in the process of tendering for its 2004-2006 demand and so their manufacturer price increased from 0- 36% starting in 2004. 48 Table 17 shows Nigeria’s routine vaccine forecasted and actual order quantities for 2004- 2007. Table 17 Accuracy of vaccine forecasting in Nigeria –Forecast and actual doses by year 2004 2005 2006 2007 BCG Forecast 7,766,060 8,561,080 8,774,800 9,023,140 Actual 6,093,000 4,275,000 7,130,000 4,388,000 Percent Over -22% -50% -19% -51% DTP Forecast 15,532,124 17,122,164 17,549,600 18,046,300 Actual 12,163,000 14,212,000 23,770,000 13,753,000 Percent Over -22% -17% 35% -24% HepB Forecast 12,602,488 17,122,166 17,549,600 18,046,300 Actual 11,428,500 12,792,500 13,130,500 13,287,900 Percent Over -9% -25% -25% -26% Measles Forecast 4,853,788 6,115,056 6,267,700 6,445,100 (routine) Actual 5,112,500 4,559,000 4,368,000 Percent Over 5% -25% -30% Measles Forecast 5,000,000 7,069,560 32,540,400 (campaign) Actual 648,000 46,764,000 39,600,000 Percent Over -87% 561% 22% OPV Forecast 19,415,153 22,829,555 23,399,500 24,061,680 (routine) Actual 4,844,000 7,852,000 13,238,000 5,836,000 Percent Over -75% -66% -43% -76% OPV Forecast 139,370,000 222,840,780 276,000,000 288,000,000 (campaign) Actual 193,246,000 200,404,000 168,324,200 168,161,700 Percent Over 39% -10% -39% -42% TT Forecast 12,943,436 14,222,180 14,624,700 15,038,600 Actual 13,584,000 12,249,000 6,497,000 11,926,050 Percent Over 5% -14% -56% -21% Source: UNICEF Supply Division. May 2008. Today, NPHCDA, UNICEF, and WHO report working together quite well and transparently on vaccine forecasting.cii Forecasting is currently done two quarters in advance and is based on the target population. 3. Distribution at State and Local Government Authority (LGA) level Nigeria uses the “push and pull� vaccine distribution system. Once the Federal Government of Nigeria receives vaccines, it sends them to the central cold store storage facility. From here it goes to one of six NPI-operated zonal cold storage units and then to a 36 state cold storage unit and then to one of 774 LGA storage unit and then to a health facility. The national level "pushes" vaccines and supplies to zonal and to State stores and the LGAs and health facilities "pull" to their levels. Several factors including limited State and LGA program ownership and commitment and limited State and LGA financial and human resources have contributed to the breakdown in the distribution chain at the lower levels and the poor program performance.ciii 49 Annex 5: Vaccine Procurement in Pakistan Pakistan has a large population of 169M and a large polio eradication campaign. The combination of these two requires extensive financial resources for vaccine procurements to ensure coverage. Pakistan has drawn upon multiple financing sources to purchase vaccines from domestic and international manufacturers/agents and used multiple procurement methods, often all at the same time. Table 18, below, summarizes vaccine coverage rates from 2000-2006. 15 Table 18 Government of Pakistan Reported Immunization Coverage for Pakistan 2006 2005 2004 2003 2002 2001 2000 BCG 89 82 80 82 82 93 94 DTP – first dose 90 82 75 77 77 86 88 DTP – third dose 83 80 65 67 68 76 78 HepB – third dose 83 73 65 63 Measles 80 78 67 61 63 75 75 TT – second dose 50 46 45 57 56 51 51 Polio – third dose 83 81 65 69 71 74 80 WHO-UNICEF Reported Immunization Coverage for Nigeria 2006 2005 2004 2003 2002 2001 2000 BCG 89 82 80 82 67 67 66 DTP – first dose 90 82 75 77 77 71 70 DTP – third dose 83 80 65 67 68 63 61 HepB – third dose 83 73 65 63 Measles 80 78 67 61 63 57 56 TT – second dose Polio – third dose 83 81 65 69 71 63 61 Source: WHO website: http://www.who.int/vaccines/globalsummary/immunization/countryprofileselect.cfm. Accessed in May 2008. While there have been no stockouts at the federal level, there have been forecasting issues. For example, differences in census and survey data related to fertility rate have made accurate forecasting a challenge. The Government of Pakistan’s public health system is very decentralized. While the federal level remains responsible for vaccine financing, procurement and policy, the district level is responsible for staff funding and vaccination campaigns. The 124 Pakistani districts also individually forecast needs which the federal level then consolidates into annual forecasts. Although there are no WHO pre-qualified vaccine manufacturers in Pakistan, the following domestic entities are involved in vaccine manufacturing: 15 Government estimates are the official country estimates. UNICEF / WHO estimates are based on government estimates but adjusted to take into account program strength, Independent National Immunization Coverage Survey figures, and population denominator figures. 50 • The National Institute of Health (NIH) has manufactured vaccines since the late 1960s. It currently produces OPV, measles, and TT vaccines based on imported concentrate for a small percentage of Pakistan’s annual requirements. The NIH also manufactures snake venom and anti-rabies vaccines. • Sind Medical Stores produces some vaccines based on imported concentrate and acts as an importing agent. They produce yellow fever for government and army purposes and MMR for the private sector (MMR is not a part of the EPI vaccination schedule). • Amson Pharmaceuticals produces some vaccines based on imported concentrates and act as an importing agent for many other vaccines. History of vaccine financing in Pakistan Pakistan has used multiple financing and procurement mechanisms, as shown in Table 19, below. Table 19 Financing Sources and Procurement Mechanisms for Vaccines in Pakistan Year Financing Vaccines Procurement Amount Percent of Source Method total 2004 GoP routine UNICEF 7,847,000 6% World Bank OPV UNICEF 120,384,000 86% Other donors UNICEF 12,535,000 9% 2005 GoP routine UNICEF 63,000,000 26% World Bank OPV UNICEF 172,774,000 71% UNICEF UNICEF 1,900,000 1% GAVI HepB UNICEF 0% Other donors UNICEF 6,148,300 3% 2006 GoP routine UNICEF 8,100,000 3% World Bank OPV UNICEF 213,710,200 77% GAVI tetravalent UNICEF 0% Other donors UNICEF 56,776,000 20% 2007 GoP routine UNICEF 11,070,000 4% World Bank OPV UNICEF 243,304,000 87% GAVI tetravalent UNICEF 0% Other donors UNICEF 26,092,000 9% 2008 To Date GoP routine + OPV GoP 44,174,982 38% World Bank OPV UNICEF 50,182,000 43% GAVI tetra and penta UNICEF 0% Other donors UNICEF 22,695,000 19% Source: Numbers consolidated from World Bank, UNICEF, and Government of Pakistan. May 2008. The World Bank has provided two rounds of funding through the IDA buy-down program. The first $42M round was approved in 2003 and has been successfully used and converted into grants. The second $74M round was approved in 2006 and is almost finished. It is expected that this one will also successfully be converted into a grant. UNICEF procures these vaccines. 51 DFID has also provided budgetary support to the Government of Pakistan. History of vaccine procurement in Pakistan From 1978 until the mid 1990s, UNICEF procured all vaccines, primarily with USAID funds. In 1995/96, Pakistan self-procured vaccines for one year with very poor results including corruption allegations and vaccines non-shipment issues. In 1997, the Government of Pakistan returned to UNICEF procurement with both government and donor funds financing vaccines. In 2001, the Government of Pakistan took over all vaccine financing except some campaign vaccines. GAVI started providing support for vaccines in Pakistan in 2001. HepB was introduced in 2002 and used until it was replaced with tetravalent in 2006/2007. Pakistan received 26M combination vaccine doses in 2006/2007. The Government of Pakistan contributes $0.3 / dose under the co-financing mechanism. GAVI approved $1.8M for pentavalent vaccine for introduction in Pakistan in July 2008. This agreement runs until 2010. civ In 2004 the Pakistan Procurement Regulatory Authority (PPRA) adopted new procurement regulations based on open competitive bidding procedures and international best practices. The EPI team, concerned about global vaccine market complexities and the influence of local vaccine producers, obtained a waiver allowing them to continue with UNICEF as a procurement agent. In 2006, the Pakistan Medical Association (PMA), encouraged by local manufacturing interests, filed a court case with the High Court in Lahore. They argued that the Government of Pakistan should self-procure vaccines because the current system did not allow the local market to participate, violated PPRA rules, and denied the government tax revenue because UNICEF does not pay income tax. The PMA won this case and the MoH did not appeal to the Supreme Court so starting in June 2007, the Government of Pakistan moved to self-procurement. The Government of Pakistan has done quite well in vaccine procurement. Having not procured vaccines for ten years, it successfully tendered and procured over 118 million doses valued at $44M. EPI launched its first tender for TT, OPV, BCG, and measles vaccines in October 2007. The EPI team required more time than anticipated going forward to launch this tender because it had to develop and obtain PPRA approval for their tender document. In December 2007, EPI announced the tender winners and signed contracts in January 2008. The government has received approximately 70% of the OPV, measles, and BCG vaccine from this tender, but the TT installments have just started arriving due to the tender winner having some difficulties (lacked facilities to process the imported concentrate and issues with customs). 52 Vaccine Procurement Challenges in Pakistan 1. Inadequate quality supervision: When Pakistan moved from UNICEF to self- procurement, the Government of Pakistan became responsible for implementing an additional two critical NRA functions. WHO assessed Pakistan’s NRA in 2003, 2004, 2005, and will complete a further assessment in May 2008. WHO considers a score of 50% or more to be functional. Results are shown in Table 20. Table 20 Pakistani NRA status for WHO six critical functions Function Status WHO Required? Assessment 2003 2004 2005 Independent The Drugs Act of 1976 names the Drugs Y Y Y Y NRA Control Organization, MoH as the NRA for vaccines. It consists of five field offices and the National Control Laboratories. 1. Licensing SRO 691 of the Drug Act of 1976 guides Y 77 61 77 licensing requirements. Guidelines need to be updated to reflect fast-changing science and technology. 2. Post- Post-marketing surveillance is conducted Y 6 19 31 Marketing through the ADR reporting system for drugs. Surveillance EPI is responsible for implementing AEFI tracking. 3. Lot Due to the change of procurement Y 50 62 75 Release mechanisms, NRA has had to develop infrastructure for lot by lot release. 4. The National Control Laboratory has Y 21 50 66 Laboratory implemented most WHO recommendations Access for Testing System Pass No No No Source: National Regulatory System: Review of Vaccines Regulation In Pakistan. World Health Organization. September 2004 and December 2005. Thus, Pakistan’s NRA is considered not functional because it does not meet requirement two, Post-Marketing Surveillance. Both UNICEF procurement and self-procurement require this function, so the change in procurement mechanisms did not involve a change in Pakistan’s NRA status. 2. Lack of capacity in the EPI team for procurement. The Government of Pakistan moved from no vaccine procurement to full procurement very rapidly. The EPI team has not had time to build a procurement cell and so relies on one consultant with minimal 53 previous experience in vaccine procurement. The EPI staff depends heavily on consultants in general which provides continuity challenges. If the Ministry of Health chooses to continue to self-procure, government EPI vaccine procurement capacity must be built. 3. Pressure to include non-WHO pre-qualified manufacturers: Clause 2.1 of this first tender issued in October 2007 states: “This invitation for bids is open to all original manufacturers within Pakistan and abroad and their authorized agents / importers / suppliers subject to the conditions that: a. In the case of foreign manufacturers, they must offer the product pre-qualified by the World Health Organization (WHO). b. In case of local manufacturer of vaccines, they will qualify only if they are using concentrates from a manufacturer duly pre-qualified by WHO. c. In the case of authorized agents / importers / suppliers, they must quote such products which are pre-qualified by WHO.�cv Due to Clause 2.1.b, Amson Vaccines and Pharma Limited, a Pakistani pharmaceutical manufacturing company, qualified for the tender by proposing to import WHO pre- qualified vaccine concentrate from Shanthi in India and locally adding the diluents and packaging. Amson itself in not WHO pre-qualified (Amson is not eligible for WHO pre- qualification because the first requirement is that the NRA of the country in which a manufacturer resides must be WHO certified. Because Pakistan’s NRA has not achieved this, no manufacturer in Pakistan can apply for WHO pre-qualification). 4. Lack of respondents to tenders. Since Pakistan moved from UNICEF procurement to self-procurement, it has launched three tenders. Each has received a small number of respondents, as shown in Table 21. Table 21 Number of Respondents to Pakistani tenders Tender Product Respondents Pass Technical Tender Qualifications Awarded? January 2008 BCG Two One Yes January 2008 TT Two One Yes January 2008 OPV Two Two Yes January 2008 Measles One One Yes March 2008 mOPV1 Three One No March 2008 mOPV3 Two Zero No March 2008 tOPV Four One Yes April 2008 mOPV1 One One Yes April 2008 mOPV3 One One Yes Source: Tender evaluation documentation. Ministry of Health. Government of Pakistan. 2008. The delays caused by relaunching the tender are impacting polio vaccination campaigns. While 71M trivalent OPV doses intended for a June 3rd Sub-National Immunization Day were delivered in March, the mOPV1 and mOPV3 vaccines will not be delivered in time 54 for distribution because the government did not enter into a contract for these until mid- May. Thus, this polio immunization campaign is being delayed. Possible reasons for small number of manufacturers responding to these tenders include: • UNICEF will not respond to an open tender because this could potentially put it in a conflict of interest situation with the vaccine manufactures who supply UNICEF under the Long Term Agreements. UNICEF did notify their manufacturers that Pakistan would be moving to self-procurement and asked them to release quantities dedicated to UNICEF for Pakistan.cvi • Few international vaccine manufacturers can respond to a Pakistani tender due to two tender clauses: o Clause 2.4: “The bidder should have operational office(s) in Pakistan…�cvii Only Glaxo Smith Kline has an operational office in Pakistan. o Clause 3.2: “In case of vaccines, besides WHO pre-qualification, these must: (a) be registered with the Ministry of Health, Government of Pakistan.� cviii One of the Registration Board’s requirements is, according to Form 5A, “For import purpose a sole agent in Pakistan is required to be nominated by the principal / manufacturer abroad.�cix • The Government of Pakistan is restricted from purchasing any products from either Israel or India. Because India is one of the largest suppliers of vaccines, this provides a challenge, but waivers can and have been obtained. • Four domestic firms responded to the tender, each time proposing to import a vaccine from one international manufacturing partner. The four firms are:cx o Amson Vaccine & Pharma responded to six tenders. They proposed to import from Japan BCG Lab, Shantha Biotechnics, Novartis, and Panacea Biotec. o Hospital Sales and Services responded to two tenders, always proposing to import from Serum Institute. o Sind Medical Stores responded to three tenders, always proposing to import from Sanofi Pasteur. o Majeed Sons responded to two tenders, always proposing to import from Novartis. In this case, the fact that Amson Vaccines & Pharma has relationships with so many manufacturers but will only propose one manufacturer per tender limits the competition. 5. Time allotted to complete vaccine tender. Vaccine procurement usually entails long lead times yet the EPI team must respect annual budget cycles. EPI did not release the first tender until October 2007 due to the time required to secure PPRA approval. Thus, the procurement time available was compressed further. Without a buffer stock, Pakistan is in a vulnerable procurement position. Furthermore, local agents report challenges in meeting the short tender timeframes. Short tender timeframes results in higher final prices because the local agent must pay more for concentrate supplies in a tight vaccine marketplace. 55 6. Higher cost: Table 22 shows that the Government of Pakistan paid 4% to 38% more than it would have if it had used UNICEF procurement. This amounts to over $2.3M in the first six months of 2008. Table 22 Government of Pakistan self-procurement vaccine prices, 2008 Cost/ UNICEF unit 2008 Increase incl. adjusted Increase in price: freight cost/unit in price: total 16 17 Date Vaccine Quantity Agent Mnf percent dollars Japan Jan-08 BCG 792,000 Amson 0.161 0.117 38% 35,128 BCG Shantha Jan-08 TT 979,500 Amson 0.061 0.059 4% 2,273 Biotech Sanofi Jan-08 tOPV 2,765,500 Sind 0.198 0.170 16% 77,066 Pastuer Serum Jan-08 Measles 2,644,500 Amson 0.325 0.260 25% 171,139 Insti Mar-08 tOPV 71,111,100 GSK GSK 0.174 0.170 3% 323,943 Sanofi May-08 mOPV1 20,000,000 Sind 0.203 0.164 24% 779,248 Pasteur May-08 mOPV3 20,000,000 GSK GSK 0.223 0.176 27% 950,818 Total 118,292,600 12% 2,339,614 Source: Tender data from Government of Pakistan. May 2008. Causes for the price increase include: • Pakistan orders smaller quantities than UNICEF. • Pakistan works with local agents who add cost to their tender. • Limited respondents to tender did not offer Pakistan full competition. • Shorter tender timeframes which cause local agents to pay higher price for imported concentrates • The tender requirements that EPI proposed differ significantly from UNICEF’s standard tenders, as shown below in Table 23. In general, the tender is much more 16 Tender costs based in Pakistani rupee and converted at January 2008 rate of 0.01601281, March 2008 rate of 0.016334275 and May 2008 rate of 0.014488633. 17 Based on UNICEF 2008 published World Average Prices with the following adjustments: 1). 10% added to UNICEF price due to the fact that Pakistan prefers not to purchase from Indian manufacturers and, if they had stayed with UNICEF, would have procured from European manufacturers. European manufacturers tend to be 10-15% higher than the World Average Price, depending on the product. 2). 4.5% UNICEF handling and inspection fee added. 3). 1.832% freight fee added. UNICEF does not include freight. Average UNICEF freight costs varies based on quantity ordered, but do not vary much based on port of origin as most UNICEF vaccines for Pakistan ship from Europe.17 For the 39 orders placed under the $42M World Bank IDA buy-down fund implemented in July to December 2007, freight charges were 1.826% of vaccine costs. For the 18 orders placed with $21M of the $74M of the second IDA buy-down spent in December 2007 and January 2008, freight were 1.838% of vaccine costs.For this analysis, an average of these freight costs was used. 56 favorable to the Government of Pakistan but does require incremental government infrastructure and pose some vaccine security challenges. Table 23 Tender differences between UNICEF and Government of Pakistan UNICEF tender Government of Pakistan tender Contract Terms • Responsibility for problems with vaccine Manufacturer from whom Tender winner UNICEF procured vaccine – not UNICEF • Freight Not included Included in price • Insurance in transit Included Included in price • Handling fee Yes No • Penalty for late delivery No Yes • Performance Security Bond: Tender No Yes awardee gives 10% performance guarantee in advance Payment Requirements • Prepayment Yes No • Currency of payment Dollars / Rupees Rupees • Currency risk GoP Tender winner Vaccine Security • Awardees per bid Multiple One Taxes • Custom Duties No custom duties are required on vaccines • Company income tax No Yes Quality Insurance • Responsible party UNICEF GoP • NRA lab testing required No Yes • Manufacturer registered in Pakistan No Yes • Government must do lot by lot inspection No Yes Legal Jurisdiction • Governed by UN Laws of Pakistan Source: Meeting with Government of Pakistan, Ministry of Health EPI team. Dr. Hussain Baksh Memon, National EPI Program Manager. Mr. Younis Khokar, EPI Procurement. May 2008. Compared with UNICEF procurement documentation. 57 Endnotes i Vaccines and Drugs: Characteristics of their use to meet public health goals. World Bank. Julie Milstien, Amie Batson, and Albert Wertheimer. March 2005. ii Interview with Pascal Perrin. Sanofi Pasteur. May 2008. iii The Vaccine Industry: Does it Need A Shot in the Arm? National Health Policy Forum. January 25, 2006. iv Vaccines: Handled with care. UNICEF. April 2004. v Vaccines and Drugs: Characteristics of their use to meet public health goals. World Bank. Julie Milstien, Amie Batson, and Albert Wertheimer. March 2005. vi UNICEF website. Accessed October 2007. vii The Vaccine Industry: Does it Need A Shot in the Arm? National Health Policy Forum. January 25, 2006. viii Procurement Tender and Overview Process. UNICEF Supply Division. Dec 15, 2005. ix ibid. x Vaccine Security. UNICEF Vaccine Procurement Overview. Rob Matthews, UNICEF Supply Division. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. xi Personal communication with UNICEF Supply Division, Copenhagen. April 2008. xii Interview with UNICEF Supply Division. Copenhagen. May 2008. xiii ibid. xiv GAVI Alliance website. http://www.gavialliance.org. Last accessed May 2008. xv GAVI Vaccine Investment Strategy (VIS) & WHO Categorization of Vaccine- Preventable Diseases Project (Vaccine Prioritisation Project). Andrew Jones, GAVI and Dr. Osman David Mansor, UNICEF. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. xvi UNICEF website. xvii GAVI Alliance Support for Country Vaccine Procurement to Date. http://www.gavialliance.org/resources/Supply_Background.doc. Last accessed May 2008. xviii Paris Declaration on Aid Effectiveness. High Level Forum. Paris, France. March 2, 2005. http://www1.worldbank.org/harmonization/Paris/FINALPARISDECLARATION.pdf xix Procurement of vaccines for public sector programs – A reference manual. WHO. WHO/V&B/99.12. xx Vaccine Procurement and Self-Sufficiency in Developing Countries. Dian Woodle. PATH (Program for Appropriate Technology in Health). Health Policy and Planning. 2000. xxi Procurement of vaccines for public sector programs – A reference manual. WHO. WHO/V&B/99.12. xxii ibid. 58 xxiii WHO website: http://www.who.int/immunization_standards/national_regulatory_authorities%20/role/en/ index.html/ Accessed October 2007. xxiv Government Lab Gives Clean Chit to Tamil Nadu Measles Vaccine. Yahoo! News India. May 15, 2007. xxv Intercountry Meeting on Vaccine Procurement For Self-Procuring Countries (Central and Eastern Europe, Turkey, and NIS). Report on WHO meeting. Copenhagen, Denmark. 3-4 September 2002. http://www.euro.who.int/document/e78698.pdf xxvi WHO and prequalification of vaccines. David Wood, WHO. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. xxvii Bidding Documents for Procurement of EPI Vaccines, Syringes, & Safety Boxes for Financial Year 2007-08. Expanded Programme on Immunization, Ministry of Health, Government of Pakistan. Octoberr 2007. xxviii Correspondance with Pakistan EPI Department. May 2008. xxix Interview with Dr. Khalif Bile, Country Representative, WHO Pakistan. May 2008. xxx Botswana Vaccine Procurement Assessment . Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. xxxi Interview with Dr. Buzan. Chief Pharmacist. Ministry of Health. Government of Botswana. April 2008. xxxii ibid. xxxiii Interview with Mr. Kgosiemang. Chief Pharmacist. Central Medical Stores. Ministry of Health. Government of Botswana. April 2008. xxxiv Interview with Mohammed Ali, Bangladesh Bank, September 2007. xxxv Bangladesh Country Procurement Assessment Report. World Bank. May 11, 2002. xxxvi Interview with Mohammed Ali, Bangladesh Bank, September 2007. xxxvii ibid. xxxviii Personal communication with Sangeeta Raja, Senior Health Specialist, The World Bank. June 2008. xxxix Personal communication with UNICEF Copenhagen. April 2008. xl ibid. xli The State of Routing Immunization Services in Nigeria and Reasons for Current Problems. Feliden Battersby Health Systems Analysts for DFID. June 2005. xlii Procurement of vaccines for public sector programs – A reference manual. WHO. WHO/V&B/99.12. xliii Vaccine Market Characteristics. AIM (Advanced Immunization Management). PATH (Program for the Appropriate Use of Technology). August 2005. xliv Interview with UNICEF-Copenhagen. Copenhagen. April 2008. xlv Bidding Documents for Procurement of EPI Vaccines, Syringes, & Safety Boxes for Financial Year 2007-08. Expanded Programme on Immunization, Ministry of Health, Government of Pakistan. Octoberr 2007. xlvi ibid xlvii Form 5A: Application Form For Registration of an Imported Drug. Website for Government of Pakistan. http://www.dcomoh.gov.pk/downloads/. Last accessed May 2008. 59 xlviii Tender evaluation documentation. Ministry of Health. Government of Pakistan. 2008. xlix Vaccine procurement and self-sufficiency in developing countries. Dian Woodle. PATH. Oxford University Press. 2000. l UNICEF website: http://www.unicef.org/supply/files/PS_ExDir_External.pdf. Last accessed May 2008. li Personal communication with UNICEF Supply Division, Copenhagen. April 2008. lii Guidelines Procurement Under IBRD Loans and IDA Credits. World Bank. May 2004. liii Guidelines: Selection and Employment of Consultants by World Bank Borrowers. World Bank. May 2004. Revised October 2006. liv Procurement Tender and Overview Process. UNICEF Supply Division. Dec 15, 2005. lv Agreement for Procurement Services between the Government of the Islamic Republic of Pakistan by and through the Ministry of Health and the United Nation’s Children’s Fund for the Procurement of Oral Polio Vaccine through the World Bank Second Partnership for Polio Eradication Project. July 4, 2007. lvi UNICEF website. Accessed October 2007. lvii Vaccine Security. UNICEF Vaccine Procurement Overview. Rob Matthews, UNICEF Supply Division. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. lviii UNICEF Supply Division 2006 Annual Report. http://www.unicef.org/supply/files/SD_Annual_report2006_web.pdf lix Vaccine Security. UNICEF Vaccine Procurement Overview. Rob Matthews, UNICEF Supply Division. UNICEF Supply Division Vaccine Manufacturers Meeting. April 3-4, 2008. lx ibid. lxi Procurement Tender and Overview Process. UNICEF Supply Division. Dec 15, 2005. lxii ibid. lxiii Interview with Robert Matthews. UNICEF Supply Division. Copenhagen. May 2008. lxiv Personal correspondence with Thomas Sorenson, UNICEF Supply Division, Copenhagen. October 2007. lxv Interview with Ranjana Kumar, GAVI. October 2007. lxvi World Health Organization data at: http://www.who.int/vaccines/globalsummary/immunization/countryprofileresult.cfm?C=' BGD'. Last accessed in December 2007. lxvii Interview with Dr. Abdul Qadir Miah, Program Manager, Immunization, Government of Bangladesh. September 2007. lxviii Bangladesh Country Procurement Assessment Report. World Bank. May 11, 2002. lxix Interview with Mohammed Ali, Bangladesh Bank, September 2007. lxx ibid. lxxi World Bank data. http://devdata.worldbank.org/AAG/bwa_aag.pdf. Last accessed May 2008. lxxii Botswana EPI Coverage Survey 2007. Expanded Programme on Immunization. Ministry of Health. Government of Botswana. 60 lxxiii ibid. lxxiv Interview with Dr. Akim, WHO Botswana. April 2008. lxxv Botswana Vaccine Procurement Assessment . Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. lxxvi ibid. lxxvii Interview with Mr. Kgosiemang. Chief Pharmacist. Central Medical Stores. Ministry of Health. Government of Botswana. April 2008. lxxviii Botswana Vaccine Procurement Assessment . Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. lxxix Interview with Dr. Buzan. Chief Pharmacist. Ministry of Health. Government of Botswana. April 2008. lxxx ibid. lxxxi Interview with Mr. Kgosiemang. Chief Pharmacist. Central Medical Stores. Ministry of Health. Government of Botswana. April 2008. lxxxii Botswana Vaccine Procurement Assessment . Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. lxxxiii ibid. lxxxiv Interview with Mrs. Tebogo. Pharmacovigilance. Health Inspectorate Unit. Ministry of Health. Government of Botswana. April 2008. lxxxv Interview with Drug Registration Unit. Ministry of Health. Government of Botswana. April 2008. lxxxvi Botswana Vaccine Procurement Assessment. Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. lxxxvii ibid. lxxxviii ibid. lxxxix Interview with Dr. Akim, WHO Botswana. April 2008. xc Botswana Vaccine Procurement Assessment . Sarah Schmitt and Pa Lamin Beyai. WHO Staff. October 2007. xci Interview with Dr. Buzan. Chief Pharmacist. Ministry of Health. Government of Botswana. April 2008. xcii Interview with Dr. Ochola-Odongo. UNICEF Botswana. April 2008. xciii The State of Routing Immunization Services in Nigeria and Reasons for Current Problems. Feliden Battersby Health Systems Analysts for DFID. June 2005. xciv Global Alliance for Vaccines and Immunizations (GAVI) Data Quality Audit – Nigeria. 16th October – 6th November 2006. xcv Personal communication with UNICEF/Nigeria. April, 2008. xcvi Global Alliance for Vaccines and Immunizations (GAVI) Data Quality Audit – Nigeria. 16th October – 6th November 2006. xcvii ibid. xcviii The State of Routing Immunization Services in Nigeria and Reasons for Current Problems. Feliden Battersby Health Systems Analysts for DFID. June 2005. xcix Personal communication with UNICEF-Nigeria. April 2008. c Joint WHO-UNICEF Mission report on Vaccine Security in Nigeria. Dr Rose M. J. Macauley, WHO-AFRO. Mr Souleymane Kone, WHO, Geneva. Dr Celestino Costa, 61 UNICEF-WCARO. Mr. Thomas Sorensen, UNICEF Supply Division, Copenhagen. 24 October to 4th November 2005. ci The State of Routing Immunization Services in Nigeria and Reasons for Current Problems. Feliden Battersby Health Systems Analysts for DFID. June 2005. cii Personal communication with UNICEF-Nigeria. April 2008. ciii Joint WHO-UNICEF Mission report on Vaccine Security in Nigeria. Dr Rose M. J. Macauley, WHO-AFRO. Mr Souleymane Kone, WHO, Geneva. Dr Celestino Costa, UNICEF-WCARO. Mr. Thomas Sorensen, UNICEF Supply Division, Copenhagen. 24 October to 4th November 2005. civ GAVI Alliance Annual Progress Report 2007. Submitted by the Government of Pakistan. May 2008. cv Bidding Documents for Procurement of EPI Vaccines, Syringes, & Safety Boxes for Financial Year 2007-08. Expanded Programme on Immunization, Ministry of Health, Government of Pakistan. Octoberr 2007. cvi Interview with UNICEF-Copenhagen. Copenhagen. April 2008. cvii Bidding Documents for Procurement of EPI Vaccines, Syringes, & Safety Boxes for Financial Year 2007-08. Expanded Programme on Immunization, Ministry of Health, Government of Pakistan. Octoberr 2007. cviii ibid cix Form 5A: Application Form For Registration of an Imported Drug. Website for Government of Pakistan. http://www.dcomoh.gov.pk/downloads/. Last accessed May 2008. cx Tender evaluation documentation. Ministry of Health. Government of Pakistan. 2008. 62