How incentive payments support Universal Health Coverage, in theory and in practice
What IEG found evaluating the World Bank Group's portfolio of PBF programs.
What is Performance Based Financing (PBF)?
PBF entails the provision of financial incentives to health providers (either a single or network of health facilities and/or health staff) conditioned to a set of measurable performance targets related to the quantity and quality of health services.
They constitute a subset of a broader term Results-Based Financing (RBF) which is defined as "a cash payment or non-monetary transfer made to a national or sub-national government, manager, provider, payer or consumer of health services after predefined results have been attained and verified. Payment is conditional on measurable actions being undertaken." (www.rbfhealth.org). This umbrella term includes both supply and demand-side transfers.
Other terms used in the literature include Pay-for-Performance, Performance-based Payments, and Performance-based Incentives all of which are essentially synonymous with RBF. The proliferation of different names seems to be influenced by the desire to ‘brand’ particular approaches supported by different international community institutions.
In the last two decades an increasing number of developing countries have introduced incentive payments linked to results to widen access to care and to improve the quality and performance of health care services and systems. These incentive schemes, known collectively as Performance-based financing (PBF), play an important role in advancing progress toward universal health coverage, contributing to the achievement of the Sustainable Development Goals.
For over a decade, the World Bank’s strategies have emphasized the importance of paying for results approaches to expanding access to and improving health care services. The World Development Report 2004: Making Services Work for Poor People lists weak incentives for health providers as one of the drivers of poor-quality service, leading to low demand for health services. To address such barriers to service delivery, the report recommends the creation of incentives for better quality service and strengthening accountability. The 2007 Healthy Development: The World Bank Strategy for Health, Nutrition and Population Results report puts greater emphasis on achieving results in the field by increasing the links between health financing and health results.
For our recent Health Services Evaluation, IEG conducted an in-depth analysis of PBF consisting of structured reviews of academic and policy literature, the development of an intervention-specific theory of change, portfolio review analysis, and evidence gap maps of systematic reviews of impact evaluations.
We found that the design of the World Bank’s PBF-related projects included the main components of the intervention theory of change, the quality of their M&E frameworks tend to be better than the rest of the health portfolio, and their development effectiveness was overall positive. However, there are still considerable knowledge gaps in measuring cost-effectiveness and health-workers’ performance, suggesting that PBF interventions need to be further studied in various contexts.
The theoretical underpinning of the PBF theory of change lies in the principal-agent model [i] enriched by a more complex view on human behavior from behavioral economics[ii]. The basic rationale is that health care providers exert more effort when payments are conditioned to the quantity and quality of the health services provided.
IEG’s specific intervention-centric theory of change (pictured below) was based on existing frameworks and relevant literature, and illustrates how desired changes in health outputs and outcomes are expected to occur as a result of PBF, given assumptions and specific contextual factors.
While there is no unique PBF scheme across countries, PBF design has three basic pillars: a defined package of services, the performance payment method (including the measurable targets against which performance payments are made), and verification mechanisms. Decisions about aspects of design (i.e. services to be financially incentivized; type of payment mechanism; performance measures) may not be independent of one another and are not made in isolation from fiscal constraints facing the country, how the health system is financed, as well as the social and cultural and social idiosyncrasies of the country.
For incentive payments to generate a change in health care providers’ motivation there must be:
If health care staff are insufficiently motivated, this would be reflected on what these workers are capable of doing versus what they actually do in practice (also known as the can-do gap). At the same time, the change in the incentive structure for health staff and facility managers impacts the way health facilities are governed (i.e. the way resources are allocated, supervision is enhanced, data is collected and reported), and hence health workers’ behaviors.
PBF schemes do not exist in a vacuum. Rather, they operate within a health system characterized by complexity, and as such, they are just one piece of a bigger puzzle that compromises system-wide factors. PBF programs often call for reform packages that are broader than merely the introduction of a different payment mechanism. Skill-enhancing trainings, investments in infrastructure, and procurement and supply chain systems increase the potential of PBF programs to achieve the desired results. Availability of skilled health workers is essential, because they cannot do better than what they know how to do (known as know-how gap). Providing quality health services in a sustainable way also entails having appropriate equipment and available medicines, so that health workers’ performance is not limited by the installed capacity (known as know-can gap).
Whereas the availability of high-quality services is expected to attract more patients, households’ health-seeking behavior and access may still be limited by demand side barriers (i.e. affordability, geography, and social issues). Still there are different ways in which a PBF program can influence demand for health care (i.e. subsidizing out-of-pocket expenses; differentiating fees by socioeconomic status or remote area; reward outreach activities; provision of in-kind incentives to patients). Understanding the factors that influence the demand for health care and health-seeking behavior of households in a particular context is crucial when designing PBF programs to help the poorest and more vulnerable populations increase their use of health services.
The success of a PBF scheme heavily relies on strong M&E and verification mechanisms, which may also include community client satisfaction surveys and other forms of counter-verification such as tracing patients back in the community. In this regard, PBF schemes provide an opportunity to improve supervision of health facilities and strengthen health management and information and reporting systems to learn about the factors that influence performance, use the results to make adjustments in implementation, and improve the design of future programs.
After developing the Theory of Change, we began the portfolio analysis of World Bank-supported PBF programs. We identified 79 Bank lending operations, of which 48% were closed at the time of the evaluation. This portfolio spanned 51 countries, with 25 of them located in Africa, with the program objective of improving access to health services, with an increasing focus on the quality of those services.
Consistent with the view that PBF cannot be conceived in isolation, Bank projects had an integrated approach to address the main bottlenecks in the capacities of the health system by combining supply and demand side and system wide interventions. In addition to incentivizing health providers through performance payments, projects supported a combination of supply-side interventions (for example, training for health workers, infrastructure and equipment, and essential medicines and medical supplies). This combination of supply-side interventions was expected to improve the quality of services, by addressing motivation; skills and knowledge; and system capacities. Projects have also blended demand-side approaches to tackle access barriers and encourage use of health services (for example, Information Education, and Communications materials (IEC), and, more recently, cash transfers and subsidies for health insurance premiums). Recognizing the necessity of a strong and reliable Health Management Information System (HMIS) to monitor PBF programs, most projects included support for HMIS to strengthen real-time monitoring and evaluation of provider performance, as well as support to bolster the stewardship and regulation functions of the Ministry of Health.
The development effectiveness of World Bank’s projects supporting performance-based payments has been essentially positive. The Bank has made substantial contributions toward improving access to and quality of health care services using performance-based payments.
M&E frameworks of evaluated projects tend to be better than other health projects not supporting PBF interventions, particularly regarding quality measures. Using the Donabedian classification [iii], beyond the commonly used structural quality indicators (e.g. trainings for health workers, infrastructure investments), about half of the projects had process quality indicators measuring compliance with medical protocols and indicating what health providers do to support patients’ health (e.g. proportion of children treated with oral rehydration therapy in case of diarrhea, provision of vitamin A supplementation for children or folic acid for pregnant women.) Process indicators of quality had a relatively high achievement rate (76 percent and higher than access indicators). Few projects have included indicators to measure outcome aspects of quality of care, such as tuberculosis treatment success rate, and prevalence of high blood pressure under control.
These examples of Bank’s M&E indicators are also applicable/suitable for most health projects, and may or may not be used to trigger performance payments under the PBF scheme. Recent projects have made the link between the achievement of the development outcomes of the project and the effectiveness of the PBF scheme more explicit. How? By introducing composite indicators of quality, such as health facility quality scores indexes, in their M&E frameworks to monitor development effectiveness of Bank lending (at the time of the evaluation 15 projects had PBF quality scores indicators). Yet some studies [iv] suggest that quality indicators triggering payments overwhelmingly reflect structural aspects of quality (76%), due to low baseline of in infrastructure and resources capacities, as well as the high costs of data collection health workers knowledge and effort-based indicators.
The body of knowledge on the effectiveness of performance-based payments is growing as shown by an increasing number of rigorous impact evaluations. From 79 projects supporting PBF interventions, 25 planned for an impact evaluation most of which are still ongoing. As of 2016, the Health Results Innovation Trust Fund (HRITF) financed 34 impact evaluations, 28 of them accompany country pilot grant and 6 are stand-alone impact evaluations.
The majority of impact evaluations have focused on the potential of PBF programs to increase access to health care services finding in general a positive impact, yet there are variations across countries and across maternal and child care services. While systematic reviews conclude that the impact of PBF programs on quality of care is still unclear[v], several reviewed programs found improvements in quality measures.[vi]
Still there is a knowledge gap on the cost-effectiveness of PBF programs and therefore it could be a priority for future prospective research. The value for money of PBF programs with respect to more traditional input-based financing is yet to be ascertained as suggested by a recent systematic review which concluded that available studies do not estimate the full economic returns of PBF programs, and that alternative interventions to strengthen the capacities of the healthcare system were not considered.[vii]
Evidence on the effects of PBF on health worker’s performance in low-income settings is still nascent, although crucial to advance our understanding of why incentives works or not. For example, in Zambia and Zimbabwe impact evaluations found that while extrinsic motivation did not crowd-out intrinsic interests and health workers’ satisfaction with their job compensation was higher, job satisfaction was lower due to higher workload.[viii]
Some recent impact evaluations supported by the HRITF advance the unpacking of the “black box” of complex PBF programs into the specific program elements and intermediate effects, for example by disentangling incentives effects from resource and supervision effects (e.g. Cameroon and Zambia).[ix] However, differences between PBF design (de jure) and implementation (de facto) resulted in rather subtle differences in treatment and control groups, reflecting how implementation challenges can affect the effectiveness of the PBF captured by impact evaluations. To learn more about most common factors affecting PBF effectiveness related to design and implementation, health workers’ behavior, and context see Box E.1 p.161 of IEG's evaluation.
The variety of PBF schemes with different program features and varying impacts across countries highlights the importance of continuing to study the effectiveness of RBF programs in new contexts before scaling up the intervention.
Photo: Isatu Bah gets a vaccination for her 5 week old son, Mohamed Kamara at the Ola During Children's Hospital in Freetown, Sierra Leone on February 29, 2016. Photo © Dominic Chavez/World Bank
[i] See Stiglitz JE. 1987. principal and agent (ii). In: Eatwell, J,Milgate,M,Newman P. (eds). TheNewPalgrave: A Dictionary of Economics. Basingstoke: PalgraveMacmillan./ Holmstrom, B. & Milgrom, P. (1991). “Multitask principal‐agent analyses: Incentive contracts, asset ownership, and job Design,” Journal of Law, Economics, & Organization, 7 (Special Issue), 24‐52. / Grossman S.J., Hart O.D. (1992) An Analysis of the Principal-Agent Problem. In: Dionne G., Harrington S.E. (eds) Foundations of Insurance Economics. Huebner International Series on Risk, Insurance and Economic Security, vol 14. Springer, Dordrecht. / Laffont J-J, Martimort D. 2002. The Theory of Incentives: The Principal-Agent Model. Princeton, NJ: Princeton University Press.
[ii] Renmans, D., N. Holvoet, C. Orach, and B. Criel. 2016. “Opening the ‘Black Box’ of Performance-Based Financing in Low- and Lower-Middle-Income Countries: A Review of the Literature.” Health Policy and Planning 31: 1297–1309.
[iii] Donabedian A. 1988. The quality of care: how can it be assessed? JAMA 260: 1743–8.
[iv] Gergen, J., E. Josephson, M. Coe, S. Ski, S. Madhavan, and S. Bauhoffe. 2017. “Quality of Care in Performance-Based Financing: How It Is Incorporated in 32 Programs Across 28 Countries.” Global Health: Science and Practice 5 (1): 90–107.
[v] Witter, S., A. Fretheim, F. L. Kessy, and A. K. Lindahl. 2012. “Paying for Performance to Improve the Delivery of Health Interventions in Low-and Middle-Income Countries.” Cochrane Database of Systematic Reviews 2012 (2): CD007899.
[vi] Chalkley, M., A. Mirelman, L. Siciliani, and M Suhrcke. 2016. “Paying for Performance for Health Care in Low- and Middle-Income Countries: An Economic Perspective.” Center for Health Economics Research Paper 140, University of York, York, U.K.
[vii] Turcotte-Tremblay, A., J. Spagnolo, M. De Allegri, and V. Ridde. 2016. “Does Performance-Based Financing Increase Value for Money in Low- and Middle- Income Countries? A Systematic Review.” Health Economics Review 6: 30.
[viii] Shen, G., H. Nguyen, A. Das, N. Sachingongu, C. Chansa, J. Qamruddin, and J. Friedman. 2017. “Incentives to Change: Effects of Performance-Based Financing on Health Workers in Zambia.” Human Resources for Health 15: 20. / Friedman, J., A. Das, and R. Mutasa. 2016. “Rewarding Provider Performance to Improve Quality and Coverage of Maternal and Child Health Outcomes. Zimbabwe Results-Based Financing Pilot Program Evidence to Inform Policy and Management Decisions.” Report No: 106518-ZW, World Bank, Washington, DC.
[ix] Shen, G., H. Nguyen, A. Das, N. Sachingongu, C. Chansa, J. Qamruddin, and J. Friedman. 2017. “Incentives to Change: Effects of Performance-Based Financing on Health Workers in Zambia.” Human Resources for Health 15: 20. / World Bank 2017. Cameroon Performance-Based Financing Impact Evaluation Report. Washington, DC: World Bank.