Over the past decade, various models of financing based on performance have been introduced in more than 20 African countries. Urged on largely by pressure to ‘catch up’ on health-related targets in the Millennium Development Goals and to demonstrate efficient use of financial resources in an austere climate, a culture of focusing on and maximising performance is gaining a strong foothold on the continent. But what do we really know about performance-based financing? This paper, the first in a two-part series, provides an overview of the basic concepts behind performance-based financing and the evidence for the effectiveness of performance-based financing models is evaluated. Based on varied experiences from different contexts, considerations to guide the implementation of performance-based financing will be presented in the second part of the series.

Moving from inputs to performance

Traditionally, financing health care in Africa has involved channelling funds to health facilities based either on expenditure trends or on inputs such as number of staff, medication and equipment supplies, and estimated target population.  In the same vein, health care providers have often been paid a fixed salary. These forms of financing health services place little or no risk on the provider or facility and therefore, little incentive to contain cost of care or maximise resource inputs used. More recently, however, local financial constraints and the additional pressure of a global economic recession have contributed to an increased interest in getting the most out of money spent in health care. For Africa, this interest is significant because many countries on the continent are dependent on external funding for a good proportion of their health budgets. For instance, in 2009, external resources for health in Sub-Saharan Africa constituted on average 10.2% of total expenditure. This figure is higher than that for any other region and even hides large outliers, such as Malawi, where the corresponding value for 2009 was as high as 80%.⁽²⁾

Accountability to funding agencies and taxpayers in donor countries has therefore contributed to shaping interest in a model of funding that encourages the maximisation of the use of financial resources for health that could have alternative uses that more directly benefit citizens of these countries.  For many recipient countries, these uses have been narrowed down to targets within the health-related Millennium Development Goals (MDGs), which are broadly recognized as priority health issues in the developing world. There is also substantial experience within the United Kingdom (UK) and the United States in using models that pay for performance as a means of improving quality of care.

Thus, the resulting massive amount of funding from donor countries in recent years, conditioned on implementing financing models that reward performance, led to the widespread adoption and expansion of these models throughout Africa. Performance-based financing models are also encouraged for a number of additional proposed advantages, such as increased provider accountability to patients, perceived positive effects on equity of access through population targeting and strengthening effects on the whole health system.⁽³⁾ However, at the very core of the decision to adopt this approach to financing health is a framework for understanding the behaviour of providers and organisations, as well as their rational response to financial incentives.

Why would we pay for performance?

Performance-based financing (PBF) models are often seen as a means of resolving the issues that arise in health care when a principal delegates duties to an agent. The principal, say the health administrator, and the agent, such as the health care provider, have diverging interests and different levels of knowledge about the production of health care. Thus, in the absence of close supervision, there is little inherent incentive for the agent to act in the interests of the principal. The fact that the administrator may also find it logistically prohibitive to fully monitor the process of care for every patient, means that in the absence of other factors, the medical doctor is unlikely to be incentivised to act in the patient’s best interests if they diverge from doctor’s own. The principal assumption of this model is that, on average, agents will behave rationally and act to maximise their welfare in the face of the incentive structure created by the principal.⁽⁴⁾ PBF approaches further assume that the agent in question is motivated by the desire to maximise financial gain and thus structure the contract such that the agent is incentivised to act in the interests of the principal.⁽⁵⁾ The process of negotiating the performance contract itself allows both the principal and the agent to clarify and align objectives.

Paying the agent for performance therefore gives the agent an incentive to achieve the principal’s objectives, as the agent faces risk if performance targets are not met or production costs are not contained. The amount of risk facing the agent in the contract would depend on the attitudes of both principal and agent towards risk, the degree of control over factors influencing performance and the reliability of measures of performance. The trade-off for the principal is the reduction of oversight of the production process based on the assumption of a better understanding of the health care production process by the agent and an understanding of the need to tailor the process to suit changing population health needs.⁽⁶⁾ In practice, health sector performance contracts are usually mixed with some payments at the margin linked to performance measures and others fixed or linked to inputs.⁽⁷⁾

So what is performance-based financing?

There is considerable controversy over the definition of this term. For the purpose of this discussion though, PBF refers to a form of financing characterised by certain features. First and foremost, PBF is concerned with supply-side incentives, made to facilities, institutions or individuals at various levels of the health system. The cost of these incentives is usually met financially, but could also be met materially. Secondly, PBF is a results-oriented method of financing, as alluded to above, rewarding the attainments of pre-determined ‘performance’ targets defined in terms of quantity and/or quality.⁽⁸⁾

The definition of performance targets is therefore pertinent. In the health (care) production process, inputs into the process are used to deliver outputs which are expected to affect health outcomes for individuals. The impact measures the effect on the health of the population as a function of the product of the average health outcome and the number of people who experienced a given outcome. These different concepts are illustrated in the diagram below. In the strictest of terms, ‘performance’ refers to the production process, say the performance of clinical duties in the administration of care. In practice, however, proxies of performance reflecting the end-product of that process are used. Thus, outputs, which are services or interventions delivered and outcomes, which measure improvement in health outcomes such as deaths averted, are considered reflections of performance.

Attempting to pay for outputs of a certain quality, such as guidelines specifying care practices, is particularly complicated as the payment has to cover the cost of following the protocol so as to create incentive to follow the guideline. Paying for impact indicators such as gains in quality adjusted life years or disability adjusted life years, is subject to some controversy, may require a long wait to determine and so is rarely done. Paying for deaths averted from immunisation for measles among under-fives in a community is an example of the rarely feasible impact indicator. The important thing is that one pays for results that are attributable to performance whether they are linked to outputs, outcomes or impact. If possible, a sequence of incentives along a continuum providing rewards for following protocols, accompanied outputs, the resultant outcomes and impacts, promotes performance when results are not solely dependent on the agent’s performance and hence variable.⁽⁹⁾

Figure 1: Overview of performance-based financing methods ⁽¹⁰⁾

Is there evidence for the effectiveness of performance-based financing models?

PBF has been tested at various levels of the health system all over the world. Donors increasingly pay for health targets achieved by recipient countries. National Governments have paid health facilities in the non-profit and private for-profit sectors to deliver health services targeted at poor families, such as in Haiti where non-profit organisations were contracted by the Government to deliver family planning and child health services. National Governments have also restructured payments to local public health authorities to reflect performance.⁽¹¹⁾ Rwanda and Burundi stand out as countries that have led the way in implementing nationwide health sector programmes that pay based on performance. Other countries in Africa that have adopted this model include the Democratic Republic of Congo, Cameroon, Central African Republic, Ghana, Sierra Leone, Tanzania, and Zambia.

Reports from field experience in implementing PBF so far have been promising. Even modest payments have been correlated with considerable improvements in health service utilisation rates, service delivery, supervisory oversight by district health teams, and a team spirit.⁽¹²⁾ In fact, a comparison of facilities which received only financial incentives unattached to indicators of performance indicated poorer performance relative to facilities that received the same amounts but as incentives for performance. Higher incentives, especially for interventions that relied primarily on health care provider behaviour, had larger effects on performance.⁽¹³⁾

PBF has also demonstrated positive effects on human resource management at the macro level. In Rwanda for example, the number of health workers increased by 62% between 2005 and 2008, as the average remuneration increased by 60 to 100% depending on the facility.⁽¹⁴⁾ Other subjectively-assessed widespread positive effects were noted in the health system as a whole. PBF models require a strong management structure, a reliable health information system, mechanisms for accountability, agencies for verification, a strong drug delivery system, and a skilled health workforce. These aspects of health systems were reported to be strengthened as capacity was built to launch the PBF model or as the culture of feedback filtered through the system.⁽¹⁵⁾

Other positive effects attributed to PBF include: increased provider autonomy and innovation within the health care delivery process, which increased efficiency; increased consumer satisfaction; consumer empowerment through involvement in the verification of performance and by choice of provider in models that had some form of internal competition; clarification of population health priorities with targeted funding, which helped reduce inequities in access to health care and in health overall; and decentralisation with the transfer of funds to the frontlines, reducing ‘leakage’ in administrative procedures.⁽¹⁶⁾ In addition, considering the African situation, it is relevant that in Rwanda and the Democratic Republic of Congo, PBF has flourished in fragile state contexts as the vacuum in administrative structures and policy allowed for the establishment of new arrangements to appropriately meet health needs.⁽¹⁷⁾

There is, however, understandable controversy associated with attribution of the abovementioned benefits solely to the introduction of PBF.⁽¹⁸⁾ In Rwanda for example, there are numerous factors that could confound this conclusion, including: the introduction of health insurance schemes, demand-side incentives in the form of conditional cash transfers, and the reduction of user fees, all of which could also have led to increases in utilisation rates for health care and other positive effects.⁽¹⁹⁾ This observation led to calls for thorough evaluations of the PBF model. However, recent rigorous evaluations of the effectiveness of the PBF model have not been as positive as field reports. In a recent Cochrane review by Witter et al. of the evidence on the effectiveness of PBF in low- and middle-income countries, methodological deficiencies in the component studies made it difficult to draw firm conclusions on the overall impact of the PBF model.⁽²⁰⁾ The most rigorous study from Africa, however, demonstrated a seven percentage point increase in institutional deliveries with the implementation of PBF.⁽²¹⁾ In addition, Petersen et al. reported positive effects on quality of care in a systematic review.⁽²²⁾

With the discordant findings for research and field-work, controversy has arisen over appropriate research methods for understanding the workings of this intervention, especially since there are still many unanswered questions. Randomised controlled trials though quantitatively rigorous are on their own unlikely to fully capture the contextual factors that influence programme design, implementation and effectiveness. Alternative methods for testing the effectiveness of PBF would include interrupted time-series, randomised trials of pilots before scale-up, controlled before-and-after studies, realist synthesis methods, as well as health system analysis using a complex adaptive system lens and social simulation modelling.⁽²³⁾ Research on performance-based financing is therefore yet to answer many of the questions that practitioners consider important: What effects do contextual factors have on the effectiveness of performance-based financing for different indicators? What is the reliability of various indicators as proxy for long-term impact in population health? What models of verification of performance are cost-effective? What are the effects of performance-based financing on equity of access, financing and health outcomes? What is the comparative cost-effectiveness of different performance-based models of financing?

Conclusion

In spite of the gaps in our current understanding of performance-based financing, there is still considerable optimism among policy makers and health programme managers regarding this model of financing. Gradually health systems all over Africa are making a transition from financing inputs to rewarding results. Performance-based financing potentially provides a means of improving health outcomes and health care quality while increasing the efficiency of resource use. These features make the PBF model a particularly attractive option for cash-strapped African Governments in countries with under-staffed and over-burdened health systems.

Performance-based financing models are, however, not without potential flaws and experience from various contexts points to certain factors as instrumental to successful implementation. It is therefore important to anticipate and check these negative effects while ensuring that the programme components necessary for success are included in PBF models to be implemented.

The second paper in this series will briefly examine factors that should be considered when adopting this model of financing. It is argued that there is still some uncertainty regarding the effectiveness of this model in positively changing provider behaviour, promoting equity, and improving quality of care and health outcomes. Despite this uncertainty however, it is concluded that in the current global financial climate a culture of rewarding efficiency is likely to persist and PBF models of financing, if properly structured, have the potential to strengthen national health systems and improve population health outcomes. 

Written by Adanna Chukwuma (1)

NOTES:

(1) Contact Adanna Chukwuma through Consultancy Africa Intelligence’s Public Health Unit ( public.health@consultancyafrica.com).
(2) ‘World health statistics 2012’, World Health Organization, 2012, http://www.who.int.
(3) Witter, S., et al., 2012. Paying for performance to improve the delivery of health interventions in low- and middle-income countries. Cochrane Database of Systematic Reviews, 2012(2), pp. 1-81.
(4) Ross, S.A., 1973. The economic theory of agency: The principal’s problem. Decision Making Under Uncertainty, 63(2), pp. 134-139.
(5) Eldridge, C. and Palmer, N., 2009. Performance-based payment: Some reflections on the discourse, evidence and unanswered questions. Health Policy and Planning, 24, pp. 160-166.
(6) Savedoff, W.D., ‘Basic economics of results-based financing in health’, Background paper for Results-Based Financing for Health, June 2010, http://www.rbfhealth.org.
(7) Ibid.
(8) Ibid.
(9) Ibid.
(10) Compiled by the author using information available at:Savedoff, W.D., ‘Basic economics of results-based financing in health’, Background paper for Results-Based Financing for Health, June 2010, http://www.rbfhealth.org.
(11) Hecht, R., Batson, A. and Brenzel, L., 2004. Making health care accountable. Finance & Development, March 2004, pp. 16-19, http://www.rbfhealth.org.
(12) Rusa, L., et al., 2009. “Rwanda: Performance-based financing in the public sector”, in Eichler, R. and Levine, R. (eds.). Performance incentives for global health: Potential and pitfalls. Performance-Based Incentives Working Group, Center for Global Development: Washington D.C., http://www.cgdev.org.
(13) Ibid.
(14) Sy, A., Sekabaraga, C. and Soucat, A., 2010. More and better performing health workers: Reforms and the health labour market in Rwanda. African Region Human Development Series. The World Bank: Washington.
(15) Toonen, J., et al., ‘Learning lessons on implementing performance-based financing, from a multi-country evaluation’, KIT (Royal Tropical Institute) in collaboration with Cordaid and WHO, 2009, http://www.kit.nl.
(16) Ireland, M., Paul, E. and Dujardin, B., 2011. Can performance-based financing be used to reform health systems in developing countries? Bulletin of the World Health Organization, 89, pp. 695-698.
(17) Ibid.
(18) Ibid.
(19) Ibid.
(20) Ibid.
(21) Fretheim, A., et al., 2012. Performance-based financing in low- and middle-income countries: Still more questions than answers. Bulletin of the World Health Organization, 90, pp. 559-559.
(22) Petersen, L., et al., 2006. Does pay-for-performance improve quality of health care? Annals of Internal Medicine, 145, pp. 265-272.
(23) Macq, J. and Chiem, J., 2011. Looking at the effects of performance-based financing through a complex adaptive systems lens. Bulletin of the World Health Organization, 89, pp. 699-700.