Interview with a PBF Expert:  Dr György Bèla Fritsche,Senior Health Specialist, HNP Global Practice

    Performance-based financing (PBF) is a form of results-based financing (RBF). In the health sector, PBF refers to programs and projects that pay health facilities directly when specific measurable and verified targets are reached. The PBF approach, which promotes transparency and accountability, has expanded rapidly in lower- and middle-income countries. This growth in health sector PBF has given rise to considerable demand for technical assistance with regard to designing and implementing PBF programs. The Performance-Based Financing Toolkit, by György Bèla Fritsche, Robert Soeters, and Bruno Meessen, draws on the wealth of information that has emerged from pilot projects around the world to enhance our understanding of PBF in the health sector. The book shows how PBF can increase the capability of the poor to access quality health services, as well as motivate health staff and support and strengthen health infrastructure.


    In order to have a deeper understanding of the rich experience with PBF and to share insights with the Community of Practice, Leslie Villegas and Dilshod B. Yusupov from GPOBA interviewed Dr. György Bèla Fritsche. Dr. Fritsche is an MD, and a specialist in tropical medicine. He holds a postgraduate degree in health policy, planning, and financing from the London School of Hygiene and Tropical Medicine. For more than 20 years, he has lived and worked as a practitioner, public health manager, and advisor in Zambia, Senegal, Afghanistan, Kenya, South Sudan, and Rwanda. For ten of those years, he has been involved in the design, implementation, and scale-up of PBF programs in Afghanistan, Rwanda, Burundi, Kyrgyzstan, Nigeria, Djibouti, Lesotho, the Democratic Republic of Congo, and the Republic of Congo. Serving since 2009 as a Senior Health Specialist for the World Bank in Washington, DC, he advises colleagues and governments on the design and implementation of results-based financing programs.



    GPOBA: What has worked in middle-income countries and in low-income countries in terms of design of PBF programs? In terms of sustainability?

    GF: What has worked in LMIC in terms of design of PBF programs are a couple of things. First and foremost taking into account incremental lessons learned over the past 15 years in piloting and scaling up PBF. Second, what has worked is starting small with carefully designed and well implemented pilot programs (say in a few districts), building local capacity and garnering political support on the way. Third, local champions, or change agents are indispensable as drivers of change. These design and implementation lessons are summarized in Chapter 17 of the PBF Toolkit. These lessons are not meant to be cookie cutter – on the contrary. However they are meant as a starting point. I always say: avoid errors made by others, make your own original mistakes. What has worked in terms of sustainability is convincing the host government to finance PBF from the public budget. Rwanda; Burundi; and the Republic of Congo are significant examples.


    GPOBA:  Path dependency states that what has been done in the past in terms of reforms will determine what will likely be done in the future.  How health services have been set-up, financed, and provided in the past will determine to a very large extent the preferences of that country’s health system. It is argued that because of path dependency, it is important to introduce PBF reforms through a pilot before trying to scale up. Influencing path dependency is presented as a key element of PBF reforms.

    Could you provide an example where a PBF pilot has influenced path dependency for better reform outcomes?  Are there cases when the PBF pilot has not been so successful in this regard?


    GF:  In all successfully scaled up PBF projects 3-4 years of piloting preceded scaling-up and this is why starting a small scale pilot is a good practice recommendation. Examples of this are Rwanda (pilots started in 2002 scaled in up 2006); Burundi (pilots started in 2006 scaled up 2010); Nigeria (pilots started 2010/11 scaled up in 2014/15); Cameroon (pilots started 2009 are scaling up now); DRC (pilots since 2001 are current scaling up to approach coverage of 23% of the population); and Zimbabwe (pilot started 2011 is currently being scaled up).

    As to your second question about starting small scale having led to failure, yes indeed we have such examples. And these are a mix of both design and implementation issues. As an example on the implementation side: if the design foresees enhancing health facility autonomy through the creation of bank accounts, and the implementation is hampered because the government feels that it cannot allow this to happen, yes, then we have a clear example of a failure. We also have examples of pilots which are successful, but which had to stop due to a stop to the financing. There are many real and theoretical examples of failures, and these are always a mix of design and implementation issues. How to design a pilot based on best practices can be seen in Chapter 16 of the Toolkit. How to avoid design and implementation failures is described in Chapter 17 of the Toolkit.

    Quite simply said, path dependency is a given fact. To influence path dependency you need a couple of things: you need a window of opportunity such as a real or perceived crises or a new government who really want to change and improve things. You also importantly need change agents or champions in the system that support the reform. Successful reforms are successful on hindsight. So it is perhaps impossible to say prior to the success what the right ingredients are to influence path dependency. The right design and good implementation of a small scale pilot are important as they can showcase success, while at the same time they allow small scale changes (such as opening bank accounts, enhancing autonomy, or enabling providers to purchase drugs from the private sector), which are important for getting results in the first place, to happen. It is easier to break rules on a small scale as it’s less threatening to the established way of doing things. One can invoke a veil of ignorance over small scale reforms which if successful might plant the seeds for large scale reforms in the near future.


    GPOBA: How does PBF stimulate pro-poor spending, improved equity in service use, or greater financial protection?

    GF: Well-designed and well-implemented PBF is a health system reform. It is a supply-side intervention (targeting providers), with later approaches integrating demand-side interventions (targeting consumers). There is a long list of examples of how pro-poor spending, improved equity in service use, or greater financial protection can be achieved through PBF design. For an elaborate explanation and description of some of these approaches, see Chapter 5 ‘Equity’ of the PBF Toolkit. In short, stimulating pro-poor spending can be done through service provision of essential health services, both curative and preventive. One can subsidize higher fees for service provision in further away areas (‘geographic equity adjustment’); one can subsidize service provision through private providers in urban areas, where public providers are not able to cater for these services. The former is standard in PBF approaches, the latter is done in countries such as Cameroon and ROC and piloted in Nigeria. One can improve equity by ensuring that the poorest use health services, by targeting the poorest through behavior change communication programs, or targeting with voucher schemes, conditional cash transfer programs or user-fee exemption programs. Greater financial protection can be reached by negotiating downwards user fees of public and private providers and by exempting certain categories of users (i.e. selective groups such as pregnant women or children under five, or the poorest of the poor). The former is done in most well-designed PBF programs, the latter in Burundi, Burkina Faso and ROC.


    GPOBA: How does PBF ensure cost efficiency, especially with regard to program management costs?


    GF: The cost efficiency question has to start with zooming out to see the forest, instead of zooming in on the trees in the forest. The health financing strategy of your country has an aim to move progressively towards universal health care. There are different ways of doing this and in situations of limited fiscal space for health and limited organizational capacity there are limits as to what can be done. If a country decides to embark on a large scale PBF approach as part of its push towards universal health care, then in that case budget is necessary to finance both service provision (through the purchase of a package of minimum and complementary health services), and the transaction costs to ensure reasonable quality verification, counter-verification, technical assistance, coaching and ICT solutions. Transaction costs in large PBF programs are frequently around 20% of total costs. This means that 80% of the budget reaches frontline health providers in cash, and this is very different from other programs that centrally plan and manage health programs. Beyond the cost-efficiency question, there are typical spill-over effects of PBF programs such as enhanced governance, enhanced responsiveness to the population and ability to do strategic purchasing.


    GPOBA:  How does PBF address demand-side barriers?


    GF: Demand side barriers are many. Well designed and well-implemented PBF programs have shown (a) responsiveness of providers to clients (attitudes; presence; opening hours); (b) better quality of services, which is linked to (a); (c) lower cost of use of services by clients; (d) targeting the poorest 20% of the population with exemption mechanisms (e.g. in Burkina Faso; ROC); and (e) targeting the poorest with voucher schemes and behavior change programs in addition to PBF programs, such as in ROC. Frequently the most important barrier is simple presence of quality service provision. And this is a supply side issue. For instance in quite a few LIC countries there is a large amount of unmet need for family planning services. Say 20-30% of all women when asked state that they would like to use family planning services, but they cannot access them, hence it is an ‘unmet need’. High quality growth monitoring and nutritional services are another example. We know the benefits of availability and use of modern FP methods and we know the benefits of high quality growth monitoring services. The former is an important ingredient to get to the demographic dividend, the latter is important to get a quality workforce in the near future. If each woman in Africa has 0.5 less children between now and 2100, there will be 2.5B Africans instead of 4B Africans by 2100 and this means a higher GDP per person and higher quality development in Africa (‘demographic dividend’). If all children are properly monitored for growth and development and nutritional status, especially between the age of 6m and 2 years, there will be much less damage to developing young brains and therefore a more intelligent workforce later on.


    GPOBA: Does PBF strengthen health systems and adopt an integrated approach?  How?


    GF: It depends. And it crucially depends on design and implementation. Absolutely, PBF is meant as a health system strengthening approach with different models. However, in general, PBF is a systems approach and influences financing (by adding more public financing to the frontlines; influencing the amount of out-of-pocket expenditures); provision (supply side incentives; strategic purchasing; quality service provision; public-private mix); stewardship (strengthens governance by better accountability for data, better use of data, incorporating civil society in decision making and asking clients about their opinions related to services rendered) etc. Defragmenting the system is an important characteristic. Integrating vertical programs and leveraging donor financing through harmonization efforts such as in the DRC are very hard but necessary especially when fiscal space for health is limited and fragmented.


    GPOBA: Dr. Fritsche, thank you very much for your time and insights.