In our previous blog, we discussed why making every penny count in the public health sector is becoming more important than ever. When money reaches the frontline, it ought to translate into better care for patients. But this translation hinges on worker motivation, often undermined by factors like low pay, lack of recognition, resource shortages, and poor management.
Starting in the 2010s, as governments looked for ways to fix structural problems in public health service delivery – including low responsiveness and inefficiency – performance-based financing (PBF) emerged as a key tool. In essence, PBF is a “mechanism by which health providers are, at least partially, funded on the basis of their performance” (Meessen, Soucat and Sekabaraga, 2011). Several low- and middle-income countries (LMICs) in Africa and Asia have adopted PBF to improve service delivery, particularly for maternal and child health (MCH). In this blog, we synthesize findings from the literature that evaluates the impact of PBF across different contexts.
Performance incentives can enhance use and quality of services, improve outcomes
Argentina’s Plan Nacer provides insurance for MCH services to uninsured families. While funds are allocated to provinces based on enrollment of beneficiaries, the program involves additional performance incentives linked to the use and quality of services and health outcomes – with the objective of raising provider accountability. Analyzing data from seven provinces during 2004-2008, and leveraging the staggered roll-out of the program across regions, Gertler et al. (2014) found that that there was an increase in the number of prenatal visits and likelihood of receiving a tetanus vaccine, and a reduced probability of low birthweight (by 19%) and in-hospital neonatal mortality (by 74%).
As efforts and resources were directed towards beneficiaries, there was a minor “crowding-out” of non-beneficiaries at clinics covered by Plan Nacer: their prenatal care utilization fell slightly but there was no negative impact on birth outcomes. The authors contend that if program funds are invested in equipment, supplies, and system improvement, then improved quality of care can be achieved for both beneficiaries and non-beneficiaries.
The program is found to be highly cost-effective in terms of saving a disability-adjusted life year (DALY). Overall, the study concludes that Plan Nacer is a promising model for emulation in the health sector in the country as well as internationally.
Overcoming inertia to effect positive change
Often, healthcare providers in the public sector are aware of better ways of doing things and of the value of such improvements – but inertia keeps them in business-as-usual mode. In other scenarios, there may be a need to develop, experiment with, and adopt better strategies. Can temporary incentives motivate change?
Celhay et al. (2019) undertake a field experiment in Argentina whereby a randomly chosen set of health clinics are offered a 200% premium if the first prenatal care visit takes place before the 13th week of pregnancy. The clinics are made aware that the higher fee per visit is applicable for a period of eight months only. The clinics responded by developing outreach strategies to reach target groups and to encourage them to begin care early. For example, they coordinated with staff at clinic pharmacies to identify women that were late in picking up contraceptive pills. Furthermore, these practices continued beyond the period of incentive. As a result, the early initiation of prenatal care went up by 34%, with the positive effects persisting for two years after the end of the intervention.
Hence, the temporary incentive helped overcome adjustment costs. Once the healthcare providers learned what works best, they continued to implement those strategies – for their perceived value rather than incentives. And of course, temporary incentives cost far less than permanent ones.
Concerns around equity
The temporary incentives in the study by Celhay et al. (2019) had the intended effects of early initiation of prenatal care. But strangely, birth outcomes showed no improvement. The authors explain that it may have been easier for healthcare providers to find low-risk mothers and to persuade them to begin prenatal care early. Doing so enabled them to achieve the target and earn the incentive. Since low-risk mothers are less likely to benefit from early initiation of care, overall birth outcomes remained unchanged. Those most in need of intervention were left out – but it is not obvious how a different design of the incentive intervention could have addressed this issue.
Lannes et al. (2016) note that PBF is not inherently pro-poor: for most health services, it improves efficiency rather than equity. In Rwanda’s national-level payment for performance (P4P) scheme, the efficiency gains come from improving access to health services for those easiest to reach, generally the relatively more affluent. The researchers advocate for building mechanisms within PBF strategies for targeting vulnerable groups, early on in the design of the program. Further, they put forth that PBF is likely to be more effective if used in synergy with other programs such as selected free healthcare and health insurance.
In Indonesia, a broad-based performance incentive did help those most in need. Olken, Onishi and Wong (2014) designed a large-scale field experiment to test the role of community-level financial incentives linked to performance in improving MCH and education. Sub-districts (with about 12 villages each) received annual block grants that could be allocated to support any of the specified indicators of health and education. In a randomly chosen subset of sub-districts, it was said that 20% of the subsequent year’s grant would be awarded to villages based on their relative performance on the targeted indicators. A household survey conducted after 30 months reveals that the performance incentive had a positive impact on eight health indicators such as weight checks and antenatal care. Funds were spent more efficiently and work hours of healthcare providers increased. The effect of incentives was especially pronounced in areas with lower levels of health service delivery at the baseline.
The authors suggest that the equitable, positive results were driven by the reallocation of budgets – the design of incentives was broad enough to allow for budget flexibility and the shuffling of resources to achieve targets. Besides, performance incentives pertained to a small group of close geographical neighbors, which mitigated the risk of bonus money flowing to richer areas.
Yet, no significant difference was found between the treatment and control groups in the long run in terms of health outcomes, with the control group catching up in performance.
How well-intentioned incentives can backfire
To increase utilization of health services, the government in the Democratic Republic of Congo (DRC) designed a payment scheme wherein health workers were provided with incentives to raise service use. Up until this point, a fixed financial payment was made to each health facility, representing about half of its revenue, and determined by size and composition of staff. Following the change, the same total budget was allocated across facilities based on service utilization levels.
In order to assess the effectiveness of this scheme relative to fixed payments, Huillery and Seban (2021) randomly assigned the 96 health areas of Haut-Katanga district to either fee-for-service payments or fixed payments, for a period of about 28 months. They found that, contrary to expectations, service utilization and newborn health outcomes were slightly worse under the incentive scheme. Facility revenue and worker income also declined.
The problem was not lack of worker effort. Providers responded by being more present at facilities, organizing preventive health sessions, conducting outreach activities, and offered lower fees for the targeted services. What was more, they did this for targeted services without neglecting non-targeted services. However, the researchers uncovered a change in the structure of worker motivation, with an increase in the weight on external motives relative to intrinsic motives, and lower job satisfaction.
But why did the extra effort produce counterintuitive results? It turned out that the populations in the incentivized areas perceived a lower benefit from using health services. Perhaps the intense direct selling and reduced fees signalled lower quality of services. It is also possible that the loss of revenue due to reduced fees led to lower quantity and quality of equipment and infrastructure at these facilities, adding to the negative perception. Finally, users may have noticed the change in worker attitudes.
Therefore, there was a mismatch between the incentive design and what workers were actually equipped to do in terms of identifying successful strategies to increase demand for health services.
Increased resources and autonomy are important in and of themselves
So far we see that PBF can improve practices and outcomes in healthcare. When designed right, it can achieve not just efficiency but also equity. But are the benefits worth the additional administrative costs and efforts associated with monitoring and enforcement?
Answering it starts with the landmark trial of this literature. In Rwanda, Basinga et al. (2011) demonstrate that performance-linked incentives lead to higher use and quality of several crucial maternal and child healthcare services, as compared to the alternative of input-based funding. Providers were motivated to translate their knowledge of prenatal care into better practice. Notably, the effects were driven by services for which bigger financial incentives were offered. For example, institutional delivery had a high payment rate; providers leveraged prenatal consultations to encourage women to deliver at facilities, and also partnered with community health workers to promote institutional delivery. Hence, for important services where provider effort can improve outcomes, performance-linked financial incentives – commensurate with the required effort – may be worthwhile.
The question subsequent studies asked is whether such gains stem from the performance incentive itself, or simply from the additional resources and autonomy that accompany it. Ngo and Bauhoff (2021) re-examined Rwanda’s PBF at the national scale and in the medium-term, comparing against unconditional financing. Using data from Demographic and Health Surveys from Rwanda and several sub-Saharan African countries for the period 2001-2010, it is observed that PBF can indeed have persistent effects for some indicators but unconditional financing is also effective. Increased resources are important in and of themselves.
An experiment in Nigeria seeks to estimate the impact of PBF and additional funding over and above an otherwise identical decentralized financing alternative, in the context of MCH (Khanna et al. 2021). Under both PBF and direct facility financing (DFF), funds are transferred directly to the bank accounts of individual publicly owned health facilities. The facilities are given substantial autonomy in how they use the money, engage community leaders in facility management, and strengthen supervision. Both PBF and DFF were found to have important effects on the coverage and structural quality of MCH services. In particular, PBF is superior to DFF for institutional deliveries and some measures of quality improvement. Both PBF and DFF are significant improvements over business-as-usual.
Like PBF, DFF activates mechanisms of direct funding, autonomy, and supervision, but is simpler and cheaper to implement. The researchers conclude that perhaps PBF and DFF ought to be viewed as complements rather than alternatives. Within DFF, specific indicators that are within the locus of control of the health worker may be chosen for PBF. For example, health workers may not be able to properly respond to incentives for antenatal care on account of demand-side barriers.
When to incentivize the demand side instead of providers
Where the Rwanda study (Basinga et al. 2011) did not find an effect was the use of prenatal care and the timely completion of child immunization schedules. These indicators tend to depend on patients’ healthcare-seeking behavior, with providers having limited control. The authors recommend that in such cases, the incentives may be directed towards patients and/or community health workers rather than providers.
Shapira et al. (2018) test whether demand-side incentives in the form of in-kind transfers to patients or offering rewards to community health worker cooperatives based on service utilization in their communities – layered on top of Rwanda’s national PBF at the facility level – can be effective in increasing utilization of services such as timely antenatal and postnatal care. While they find that patient incentives do work, community worker incentives did not improve the coverage of targeted services, behavior of community health workers, or outcomes at the community level. Also, no synergies were found between patient incentives and community worker incentives.
Measurement and feedback can work too
In the Philippines, Peabody et al. (2011) randomly allocated hospitals to one of two treatment groups or a control group. Across the groups, baseline information was collected on average clinical competence and patient satisfaction, and facility caseload. In the first treatment group, physicians could directly receive extra pay or bonus (a modest 5% of their salary) based on the defined indicators. In the second treatment group, hospitals were given expanded insurance coverage for treatment of common conditions such as pneumonia and diarrhea.
Over the following three years, performance measurement, feedback, and public disclosure were carried out for all hospitals in the study. It is seen that both the treatments resulted in quality improvement of about 10 percentage points, indicating that system-level, indirect incentives may drive individual behavior similar to direct incentives.
More interestingly, by the end of the assessment period, performance improvement occurred in control sites as well. This suggests a lagged dissemination and feedback effect: non-monetary incentives in the form of quality performance feedback can also lead healthcare providers to reflect on their relative performance and adopt improvements.
Concluding thoughts
There is evidence that PBF for healthcare providers can improve the use and quality of maternal and child health services, as well as outcomes – but it is crucial to read the fine print. Performance-linked incentives tend to work best where providers have real control over the outcome being incentivized and understand what is being asked of them. Success is often not PBF acting alone: it may depend on synergy with other programs, such as health insurance, or on pairing PBF with direct facility funding as a complement rather than a substitute. A well-designed scheme can improve both efficiency and equity, but this requires deliberately building in mechanisms to reach the ones most in need, not just rewarding low-hanging fruit. Sometimes simpler, unconditional financing achieves similar gains without the added administrative cost and effort of PBF, and sometimes the lever that needs pulling is on the demand side – patients’ healthcare-seeking behavior – rather than the provider’s. The common thread is fit: incentives work when they are matched to the context.
Drawing on the experiences of deploying PBF to improve the use and quality of MCH services, similar approaches may be worth exploring beyond MCH, for instance, to encourage preventive practices such as screening for non-communicable diseases (NCDs).
Staying on the matter of public finance in the health sector, in the next blog, we examine the politics around “health taxes” as a source of revenue.
REFERENCES
Basinga, P., Gertler, P. J., Binagwaho, A., Soucat, A. L., Sturdy, J., & Vermeersch, C. M. (2011). Effect on maternal and child health services in Rwanda of payment to primary health-care providers for performance: an impact evaluation. The Lancet, 377(9775), 1421–1428.
https://www.sciencedirect.com/science/article/abs/pii/S0140673611601773
Celhay, P. A., Gertler, P. J., Giovagnoli, P., & Vermeersch, C. (2019). Long-run effects of temporary incentives on medical care productivity. American Economic Journal: Applied Economics, 11(3), 92–127.
https://www.nber.org/papers/w21361
Gertler, P., Giovagnoli, P. I., & Martinez, S. (2014). Rewarding provider performance to enable a healthy start to life: Evidence from Argentina’s Plan Nacer. World Bank Policy Research Working Paper No. 6884.
https://openknowledge.worldbank.org/entities/publication/84205239-45c6-542e-83ff-c6d097fe9ffc
Huillery, E., & Seban, J. (2015). Financial incentives are counterproductive in non-profit sectors: Evidence from a health experiment. Working Paper, June 2015. [Later published as: “Financial Incentives, Efforts, and Performances in the Health Sector: Experimental Evidence from the Democratic Republic of Congo.” Economic Development and Cultural Change, 2018.]
https://www.povertyactionlab.org/evaluation/impact-fee-service-schemes-health-service-utilization-democratic-republic-congo
Khanna, M., Loevinsohn, B., Pradhan, E., Fadeyibi, O., McGee, K., Odutolu, O., Fritsche, G. B., Meribole, E., Vermeersch, C. M. J., & Kandpal, E. (2021). Decentralized facility financing versus performance-based payments in primary health care: a large-scale randomized controlled trial in Nigeria. BMC Medicine, 19(1), 224.
https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-021-02092-4
Lannes, L., Meessen, B., Soucat, A., & Basinga, P. (2016). Can performance-based financing help reaching the poor with maternal and child health services? The experience of rural Rwanda. International Journal of Health Planning and Management, 31(3), 309–348.
https://onlinelibrary.wiley.com/doi/10.1002/hpm.2297
Meessen, B., Soucat, A., & Sekabaraga, C. (2011). Performance-based financing: just a donor fad or a catalyst towards comprehensive health-care reform? Bulletin of the World Health Organization, 89(2), 153–156.
https://pmc.ncbi.nlm.nih.gov/articles/PMC3040374/
Ngo, D. K. L., & Bauhoff, S. (2021). The medium-run and scale-up effects of performance-based financing: An extension of Rwanda’s 2006 trial using secondary data. World Development, 139, 105264.
https://www.sciencedirect.com/science/article/abs/pii/S0305750X20303910
Olken, B. A., Onishi, J., & Wong, S. (2014). Should aid reward performance? Evidence from a field experiment on health and education in Indonesia. American Economic Journal: Applied Economics, 6(4), 1–34.
https://www.aeaweb.org/articles?id=10.1257/app.6.4.1
Peabody, J. W., Shimkhada, R., Quimbo, S., Florentino, J., Bacate, M., McCulloch, C., & Solon, O. (2011). Financial incentives and measurement improved physicians’ quality of care in the Philippines. Health Affairs, 30(6), 1216–1223.
https://www.healthaffairs.org/doi/abs/10.1377/hlthaff.2009.0782
Shapira, G., Kalisa, I., Condo, J., Humuza, J., Mugeni, C., Nkunda, D., & Walldorf, J. (2018). Going beyond incentivizing formal health providers: Evidence from the Rwanda Community Performance-Based Financing program. Health Economics, 27(12), 2087–2106.
https://onlinelibrary.wiley.com/doi/abs/10.1002/hec.3822