By Petra Vergeer and Sarah McCune 
To improve health outcomes and to boost the efficiency, equity and quality of their health systems, low- and middle-income countries across the globe pilot various results-based financing (RBF) approaches, such as performance-based financing (PBF) to strengthen the supply of health services. These pilots are now beginning to show exciting results: institutional deliveries have significantly increased, ante- and postnatal coverage has greatly expanded, and many more children have received vaccinations. RBF can improve the efficiency of health systems, too. Indeed, thanks to RBF incentives, more uncomplicated deliveries are occurring at health centers, allowing hospitals to focus their attention and resources on treating on more complicated cases. Quality at health facilities has also improved, and variation in health facility quality has decreased. Based on these results, governments and donors are interested in supporting more RBF pilots or their scale-up.
The World Bank’s Health Results Innovation Trust Fund (HRITF), which is funded by the Governments of the United Kingdom and Norway, has played a pivotal role in supporting the design, implementation and evaluation of 36 RBF pilots in 31 low- and lower-middle income countries. The word “pilot” should not underestimate the scale of these programs. For example, the RBF pilot in Nigeria is operational in three states and covers nine million people, and in Haiti the pilot will be introduced in four departments with a catchment area of two million people. Sustaining or expanding successful results in these pilots can change the lives of many people. Governments and donors are thus looking for opportunities to support RBF on both the supply- and the demand-side.
Using examples from the RBF pilots in the World Bank’s HRITF portfolio, this paper will highlight some of these opportunities. Specifically, it will focus on: i.) Purchasing results of specific RBF indicators in a country; ii.) Purchasing RBF results in a specific geographic area in a country; iii.) Financing technical assistance and support for RBF design and implementation in a country; iv.) Supporting RBF globally through the World Bank’s HRITF. The paper will also discuss the advantages and disadvantages of each opportunity.
Purchasing Specific RBF Indicators in a Country
Different development partners may have different strategic interests and priorities, and purchasing specific RBF services is one option for those interested in supporting the implementation or scale-up of RBF. The national PBF program in Rwanda provides an example of how development partners can support RBF by purchasing specific indicators or services.
Rwanda’s PBF program has had dramatic effects on service delivery and utilization, especially for the country’s poorest populations. The impact evaluation, for example, showed good results related to the quantity and quality of services as compared to a control   
. Not only did the quantity and quality of services increase considerably, a significant effect also occurred on the size and weight of children under five years of age living in the catchment areas of PBF facilities. Additionally, institutional deliveries among the poorest quintile increased from 12.1 percent to 42.7 percent. These gains have contributed to improving health equity across the country. Furthermore, since PBF was implemented, more individuals have sought HIV testing. Indeed, the probability that an individual has been tested for HIV has increased by 6.1 percentage points and has increased by 10.2 percentage points for married couples
When PBF was scaled up nationwide, the Government of Rwanda (GOR) purchased 16 general health services, and PEPFAR purchased 15 related to HIV/AIDS. Interested in supporting PBF in Rwanda, the Global Fund (GFATM) agreed to purchase the reproductive health, HIV/AIDS, and tuberculosis indicators. The Center for Disease Control (CDC) has since purchased indicators, as have Family Health International (FHI), Intra-Health, the Elizabeth Glaser Pediatric AIDS Foundation (EGPAF), Catholic Relief Services (CRS), and Management Sciences for Health (MHS) through PEPFAR. The Belgian Technical Cooperation (BTC) and the German Technical Cooperation (GIZ) have also purchased services.
The GOR, the largest funder, drove the donor harmonization and alignment efforts, which ultimately produced two different funding options. In the first, the CDC and the GFATM transfer funds to the Ministry of Health’s (MOH) bank account and the MOH then pays the health facilities for the particular indicators that the two development partners support. The amount transferred to the MOH is based on a defined PBF budget for purchasing the selected indicators for the fiscal year, and the ministry provides the partners with a financial report at the end of the quarter and at the end of the fiscal year. The second option applies to funds from PEPFAR, which flow through its implementing partners, currently the Rwanda Family Health Project (RFHP) to the health facilities. The MOH sends an invoice for the PEPFAR-funded services to the RFHP after the verification and validation of the results by the district steering committee and the PBF Technical Unit at the ministry. The RFHP then directly transfers the RBF payments into the facilities’ individual bank accounts.
The harmonization and alignment efforts have proven advantageous for several reasons. A major advantage of development partners purchasing different PBF services is that a broader package of services can be supported. This can reduce the possible— though unintended—consequences of focusing on one area at the expense of another, which has often been a criticism of vertical programs. Such considerations are very important in any RBF program. In Rwanda, the number of indicators linked to HIV services was reduced from 15 to ten to better balance the different priorities of the country’s health services. The amount of funding for HIV services, however, has remained constant. Having multiple donors support PBF consequently contributes to health outcome improvements, as well as the harmonization and alignment of aid.
Donor mapping and coordination, moreover, have provided effective coverage for purchasing PBF services as the results of the PBF impact evaluation demonstrate. Duplication of verification and reporting processes have also been avoided because all development partners use the same reporting and billing software through the PBF web-enabled application. Additionally, the harmonization efforts have led to better financial forecasting, which now takes into account the different sources of funding and determines a single unit price of each PBF service. A possible disadvantage to this, however, is that it does not allow for flexibility in filling any financial gaps should a partner withdraw or reduce their funding for an indicator. As the largest funder, however, the GOR has been able to flexibly adapt to the exit and entry of various development partners. Its position as the largest funder, moreover, helps to ensure the financial sustainability of PBF. Nevertheless the government’s desire to have one basket for all PBF funding has not come to fruition, and consequently, it does not have the power to allocate development partner funding but is to align the funding to the donors’ different priority indicators. Having multiple partners, though, has significantly limited the GOR’s transaction costs, like such as those related to verification, capacity building, and administration, among others. The GOR pays 5 percent of the total transaction costs while other development partners pay for the remaining 95 percent, of which 77 percent is contributed by PEPFAR/USAID.
Purchasing RBF for Health Results in a Specific Geographic Area of a Country
The presence of multiple donors in a developing country often necessitates that donors concentrate their work in specific geographic areas. Purchasing health results in a particular geographic area is an option for development partners interested in supporting RBF. Benin’s experience with a joint-basket fund arrangement, Burundi’s experience with virtual pooling system, and Senegal’s with a Bank-managed trust fund provide three different examples of development partners supporting RBF based on geographic distinction.
Burundi: Virtual Pooling System
In Burundi, donor harmonization efforts have facilitated the scale-up of a PBF program that has both improved the health of the country’s women and children and has increased health equity across the country. More women are now giving birth in health facilities and are benefiting from post-natal care. Operational data gathered between June 2010 and 2011 showed that institutional deliveries, for example, increased by 25 percent, and curative care consultations for pregnant women increased by 34.5 percent. Burundi’s PBF program also increased vaccination rates: 10.2 percent more children are now vaccinated against common childhood diseases. Health facility quality scores have also increased across all provinces, and the gap between the worst performing facilities and the best has narrowed. These gains in quality scores signal improvements in health equity, which has especially improved for the traditionally marginalized Batwa people.
In 2006, PBF was introduced and piloted in three provinces, and in 2010, the PBF pilots were scaled up throughout the country. Initially, the PBF pilots were somewhat divergent in their methodology and parameters, and all operated outside of the public system. Furthermore, some health facilities received double payments: first from the government through its free healthcare reimbursement mechanism, and then from the donors financing the PBF pilots. To better streamline the scale-up and to facilitate donor harmonization, new institutional structures were created, and a system of virtual pooling was implemented at the central level.
In addition to the Government of Burundi (which is the largest payer for results) and the World Bank’s IDA and HRITF, other development partners participate in the virtual pooling system. They include, FHI, GAVI, Healthnet-TPO, BTC, European Union, Cordaid, Swiss Development Cooperation, the Groupe de Volontariat Civil (GVC) and the Programme Transitoire de Reconstruction Post Conflit (PTRPC). In the virtual pooling system, each health facility produces a monthly invoice, which is then sent to a provincial verification committee for validation and entry into an online record-keeping and billing system. The national PBF Technical Cell and the General Directorate of Resources at the MOH subsequently verifies the invoices and corrects them if necessary. The online system then automatically distributes the invoices to the various development partners, and the development partners then pay a specified amount to each health facility. If the partners cannot pay due to a lack of budget, the government and the World Bank through IDA and HRITF will fill the gap. These two partners also co-finance areas where there are no other partners involved.
An advantage of the virtual pooling system is that it allows each development partner to maintain its own budget and financial procedures when paying a health facility; a long process of fiduciary harmonization with high transaction costs has thus been avoided. Most importantly, however, is that the system has attracted additional development partners, who are interested in supporting RBF in Burundi and continuing the impressive gains that the approach has yielded.
A disadvantage of the virtual pooling system concerns the logistics of health facility payment. Partners may pay at different times (ie, the funds do not all arrive at the same time). Furthermore, the banks used by the health facility do not specify the source of the payment, which makes it difficult for health facility managers to know which partner has or has not paid their invoice. These logistical problems, however, are being addressed.
Benin: Joint-Basket Fund
The PBF pilot in Benin began in 2012 and is relatively new, but showing great promise as a catalyst for ensuring better health outcomes and changing the way the health system operates. Initial results have shown slight increases in performance indicators, and they will likely improve further as the program matures. The government is fully committed to PBF, and it is working with multiple donors to scale it up throughout the country.
The PBF model employed in Benin is identical in all districts—the same indicators and verification mechanisms are used. The Bank, GAVI, and the GFATM are co-financing the purchase of a package of services in 29 districts across the country: the Bank pays for the health results in eight districts, GAVI in two, the GFATM in 19. USAID or the Belgian Technical Corporation (BTC) will possibly pay for the results in the remaining five districts.
Under the direction of a particularly strong minister of health, the MOH has led the donor harmonization efforts. The Bank was the first to support the design and implementation of PBF in the country, including the development of the system for verification and payment. Standardized financial management and procurement rules were developed for the joint-basket fund that the MOH manages and uses to pay health facilities for the number of services they provide and for their quality. Each partner contributes an equal share for the RBF operating costs, such as technical assistance and verification, based on the number of districts in which it works. Importantly, the co-financing arrangement has made it possible to scale-up PBF throughout Benin, and it is now in 85 percent of the country’s districts.
The government was keen to scale up PBF throughout the country, while the development partners were interested in buying health results that were credible, having been verified according to the RBF pilot model. For the development partners, one of the advantages of this co-financing arrangement was reducing the fiduciary risk associated with the implementation of PBF. Partnering with the Bank provided GAVI and the GFATM with strong guarantees on the fiduciary side, which made supporting the expansion of PBF more attractive. A disadvantage of this arrangement, conversely, was the long process of fiduciary harmonization that came with higher transaction costs; for the government, the harmonization and alignment of the funds received actually reduced transaction costs.
Senegal: World Bank-Managed Trust Fund
Although it is still in the early phase (implementation will start in January 2014), the PBF program in Senegal represents an exciting opportunity for the World Bank and USAID to work together to improve the quality and accessibility of health services and to prompt changes in health behaviors and the utilization of services. The World Bank and USAID—though its Government to Government (G2G) initiative— are co-financing a package of services in six regions across the country. Using funding from the International Development Association (IDA), the World Bank is supporting this package in four regions, and USAID is funding it in two. Senegal’s PBF system and its indicators are uniform across all the regions.
The co-financing arrangement represents one of the first instances of donor harmonization in the Senegal’s health sector, and the arrangement is quite straightforward. The Bank has created a trust fund (TF) for USAID’s funding. All funding flows through the Bank, and complies with the institution’s fiduciary, financial management, and procurement rules and procedures. The Ministry of Finance receives the funds from the TF and then distributes them to MOH. The MOH then allocates money to the PBF health facilities based on their achievement of incentivized indicators.
The co-financing arrangement has generally proven beneficial for the Bank, USAID, and the government. For example, transaction costs have been reduced for all partners, especially the costs related to fiduciary rules and financial management. Co-financing has also developed economies of scale around what are typically high-cost inputs, like verification. For all pilots, technical assistance is provided by Abt Associates, a USAID contractor, which has an added benefit: increased funding that can be allotted to PBF payments.
One of the few weaknesses of the co-financing arrangement is the stringent nature of the Bank’s procedures. The technicalities of Senegal’s program have at times also presented challenges. The benefits of the co-financing arrangement have outweighed the negative, however, and other development partners, like Japan, are interested in PBF in Senegal.
Financing Technical Assistance and Support for a Country's RBF Design and Implementation
Development partners can also support RBF by financing technical assistance (TA) and support for RBF design and implementation. The World Bank’s experience with partners in Haiti and Tajikistan demonstrate how these arrangements can be productive.
The objective of the new RBF program in Haiti is to increase the access and use of maternal and child health, nutrition and other social services while strengthening the stewardship and management capacity of the health system. The World Bank, through IDA and HRITF funding, is supporting the Ministry of Public Health and Population (MSPP)’s implementation of PBF in a catchment area of 1.9 million people and is planning to scale it up within the next five years. USAID is also implementing a health service-delivery project in accordance with the country’s harmonized PBF approach. The harmonized approach uses one national PBF implementation manual, one set of indicators, the same structure for verification and one monitoring and evaluation system.Some RBF payments for health facilities will be covered by funding from IDA/HRITF and others by USAID, but all will follow the same system.
The Bank and USAID maintain separate fiduciary and procurement procedures, however. For example, the Bank’s funding flows to the contracting unit of the MSPP, which then allocates it to the deputy directorate for service provision to distribute to the RBF health facilities and the hospitals. USAID, conversely, funds contractors that subsequently work with the health facilities and hospitals they support.
Both development partners provide technical assistance to the government on RBF but ensure that the TA work is complementary. The Bank, for example, supports the government in its decision making on how to set fees for the RBF indicators, while USAID has provided support for the development of the national RBF manual. The primary advantage of this arrangement is that there is not a duplication of efforts or processes, which increases efficiency. This arrangement also furthers the harmonization of RBF—as the MSPP has requested. Lastly, it fosters improved communication because the provision of complementary TA requires significant coordination between the two development partners.
The World Bank’s Board has just approved the RBF program in Tajikistan, which aims to improve health equity by increasing the coverage and quality of basic primary health care services in rural health facilities of Tajikistan. PBF will be piloted in eight districts, and will cover 1.86 million people, or 25 percent of the country’s population. Given its scope, the implementation of PBF in Tajikistan has the potential to significantly improve health outcomes and make the country’s health system more efficient and more equitable.
Under the PBF pilot, rural health centers and their subsidiary health houses will be eligible to receive a performance-based payment based on the quality and quantity of the maternal and child health and non-communicable disease services they have delivered over the quarter. These payments, as in other PBF programs, will be supplementary to funds received from the public sector budget, and can be used by the rural health centers and health houses for performances bonuses to staff or for investments in the facility.
The World Bank, through IDA and HRITF, is funding the development and piloting of PBF in Tajikistan, Other development partners, however, are supporting the pilot by financing activities related to the design and implementation of programs. UNICEF, for example, is paying for and is carrying out the independent verification of the PBF program. The organization has also financed the technical assistance to prepare the independent verification of the PBF methodology, and it will contribute the time of the technical staff involved in the verification.
A particular advantage of this arrangement is that it allows the primary funder—in this case, the World Bank—to allot more of its resources to PBF payments for health results. Verification and TA are major contributors to the functioning of RBF, and having a development partner co-finance these inputs can allow for the other parts of the intervention to be better targeted or expanded.
Supporting RBF Design, Implementation and Learning Globally through the Health Results Innovation Trust Fund
Providing support to the HRITF is another option for development partners to co-finance RBF. The HRITF leverages World Bank IDA funding for RBF by awarding grants of up to $20 million, matching the country’s IDA allocation to RBF. The $20 million is used for RBF design and implementation—such as technical assistance and verification— in the 35 programs it currently supports. Because the HRITF grants are linked to IDA, they are part of broader support and policy discussions between the Bank and a country’s government, particularly their Ministries of Finance and Health.
The desire to sustain and expand the successes of RBF has led several governments to start financing RBF through their regular budgets. An example of this comes from Zimbabwe’s PBF program, which has yielded significant changes in the way that the 3.5 million people it covers receive and utilize health services. In only ten months, the coverage for institutional deliveries expanded from 50 to 75 percent, and immunizations almost doubled. Quality of care indicators also increased, and they continue to follow a positive trend. Efficiency gains were also made. The number of “normal” deliveries in health centers has increased, and has decreased in secondary hospitals
. Based on the positive results of the pilot, the government is planning to further scale-up PBF with HRITF funding and will continue to co-finance a package of services. Recently, it created a line item in the budget to fund PBF, and may commit up to five million dollars until October of 2015. The government has also initiated talks with other donors, like UNICEF and the GFTAM, to help support the scale-up.
The trust fund supports not only the design and implementation of RBF programs in low- and lower-middle income countries but also plays a vital role in learning about RBF. The HRITF first plays this role for new countries that are considering the appropriateness of RBF. Indeed, such countries have benefited from knowledge and learning grants, which provide funding to support technical dialogue and learning on RBF through, for example, study tours to RBF projects in other countries, feasibility studies and RBF technical workshops for policy makers and stakeholders. This enables policy makers to make an informed decision as to whether RBF is a suitable approach to pilot in their country. For countries that have decided to pilot RBF, the HRITF ensures that a rigorous impact evaluation is embedded in each pilot. These evaluations will evaluate the effectiveness of the approach in the country.
The HRITF also promotes learning from successful (and unsuccessful) RBF experiences from around the world by providing financial support to RBF program evaluation efforts in middle-income countries. A case in point is the recent evidence from Argentina’s innovative RBF program, Plan Nacer, which provides access to an explicit package of maternal and child health services for two million low-income pregnant women, mothers, and children under 6 years of age who lack formal health coverage. The results from Plan Nacer’s impact evaluation (IE), funded in part by the HRITF, shows that birth weights have improved and neonatal mortality has decreased. For example, incorporating clinics into Plan Nacer reduced the probability of low birth weight by 23 percent for beneficiaries. Birth records from larger maternities show that Plan Nacer reduced the probability of in-hospital neonatal death by 74 percent for Plan beneficiaries. Better prenatal care, which prevented low birth weight and better post-natal care contributed to the reduction in neonatal mortality. There was an increase in the use as well as the quality of prenatal care services, which was measured by the number of prenatal care visits and the probability of receiving a tetanus vaccine
The HRITF is also supports global learning through a variety of knowledge- and lesson-sharing events. For example, the annual RBF Results and IE training brings countries implementing RBF together to share lessons and discuss the challenges they may face in implementing RBF and their corresponding IEs. The upcoming workshop will be held in Argentina in March 2014 and will enable participants to see how a RBF program can work in a more complex health system. These workshops allow country teams to review available RBF data and evidence and to reflect on possible solutions to further improve the implementation, evaluation, and results of RBF. The workshop is also an opportunity for countries that are at the beginning stages of implementation to hear lessons on RBF or IE design from countries that are more advanced in the process. These lessons can help these newer countries to determine what may be appropriate for their context.
This paper highlights the multiple opportunities that governments and development partners have for supporting RBF at the country and global levels, and it discusses the advantages and disadvantages of each opportunity. For development partners with strategic interests, purchasing specific indicators or services—such as in Rwanda— may be the most appropriate means of supporting RBF in a county. Other development partners may want to support RBF in a specific geographic area through different financial mechanisms, like Benin’s joint-basket fund, Burundi’s virtual-pooling arrangement, or Senegal’s trust fund. Financing technical assistance and supporting RBF design and implementation, as in Tajikistan and Haiti, is another option for development partners to jointly support RBF. Finally, development partners can support RBF globally through the World Bank’s HRITF. All of these options can help to scale-up and sustain RBF programs—programs that have the potential to improve health outcomes and promote positive changes within an existing health system.
The authors wish to thank Alain-Desire Karibwami, Cedric Ndizeye , Christophe Lemiere, Gyorgy Fritsche, Mirja Sjoblom, Moulay Driss Zine Eddine El Idrissi, Ronald Upenyu Mutasa, Sunil Rajkumar, and Wezi Marianne Msisha for their valuable contributions on the different experiences.
Basinga, P. et al. 2011 “Effect on maternal and child health services in Rwanda of payment to primary health-care providers for performance: an impact evaluation,” The Lancet, Vol 377, pp. 1421-28.
De Walque, D. et al 2013 “Using Provider Performance Incentives to Increase HIV Testing and Counseling Services in Rwanda.” Impact Evaluation Series No. 84. World Bank Policy and Research Working Paper 6364.
Gertler, P. and Vermeersch C. 2012 “Using Performance Incentives to Improve Health Outcomes” World Bank Policy Research Working Paper No. 6100
De Walque, D. et al 2013 “Using Provider Performance Incentives to Increase HIV Testing and Counseling Services in Rwanda.” Impact Evaluation Series No. 84. World Bank Policy and Research Working Paper 6364.
Health Results Innovation Trust Fund. 2013. “Using Results-Based Financing to Achieve Maternal & Child Health” Progress Report.