Pricing and The Use of Data in RBF: Towards a Higher Return on Investment

Maarten Oranje's picture
March 15, 2017

Results-Based Financing (RBF) is about market mechanisms. We offer the healthcare provider a higher price for a particular service – a subsidy on top of lump sum income and possibly patient fees – and we expect service delivery to go up. A simple supply side truth from your micro economics textbook. But full-fledged RBF is of course about much more than just market mechanisms. It is a systems approach, in which the separation of functions, the autonomy of healthcare facilities and the empowerment of communities are equally – if not even more – important than just the simple payment for verified performance. This becomes clear once we lower the level of subsidies after a few years of RBF implementation. If simple micro economic models were correct, this would inevitably lead to a lower supply of services, as the service would now be less profitable for providers. However, as has become clear from temporarily paying subsidies for prenatal care services during the first semester of pregnancy, in Las Misiones province in Argentina, positive effects could still be detected more than a year, probably even two years, after the incentives had stopped. We see a similar example of price inelasticity for outpatient consultations (OPDs) in a Cordaid RBF project in Uganda. Even though subsidies were halved after two years of RBF implementation, access to care did not suffer.

This particular RBF project started in July 2013, in Busoga Region in Uganda. It encompasses 14 public health facilities in Kamuli District and 6 private-not-for-profit (PNFP) facilities in and around the district. For the level 3 and level 4 facilities, ten quantity indicators are included, with a focus on maternal and neonatal care. The two hospitals (level 5 facilities) get paid for those same ten indicators, plus one additional indicator, i.e. caesarean sections. At the same time, the project aims to contribute to improved access to care, by also incentivizing outpatient consultations for all. As is observed in quite a few RBF projects, this one indicator is consuming quite a large chunk of the total project budget. So, when in mid-2015 we were facing budget constraints, it made sense to reconsider the level of the subsidy for OPDs. Analysing monthly performance data, not only did we find that OPDs accounted for almost half of the total payments to health facilities, we also saw that it was the indicator closest to reaching its monthly target (which is deduced from the one OPD per capita per year rule of thumb). Moreover, we found that growth in performance was steadier than for most indicators. Adding to this a context analysis which showed good improvement on quality of care and strong health staff commitment, we made our decision. We decided not to resort to small budget cuts across the board – which we felt would jeopardize the overall setup and goals of the program – but to make the 20% budget cut needed by halving the subsidy for this one indicator. From now on, facilities would receive 500 shilling per verified OPD, instead of the 1,000 shilling they had received until then.

So, did health facilities complain? Yes, initially they did. And did their performance suffer? No, it did not. To the contrary, if anything, the growth in the number of OPDs only seems to have accelerated since mid-2015. When breaking down these numbers, we found that the acceleration of growth was true both for the PNFP facilities, which could have compensated their lower subsidy income by charging higher user fees, and for the public facilities, which provide services free of charge. Even more, we could see from the quarterly quality evaluations that there had been no negative effect on the average quality scores of facilities, despite the lower income. So what had happened? At our request, our local RBF coordinator conducted another brief context analysis. She looked both at supply side factors, such as price policy of the facilities, as well as health staff attitude and motivation, and at the demand side, where patient satisfaction surveys provided valuable information. From this, we learned that the good results were partly due to emerging competition between facilities. Quality in public facilities had improved so much since the start of the project that people were increasingly visiting those centres for consultation. As a result, PNFP facilities had realized they had to improve their game. Some had first increased their patient fees – to compensate for lower subsidy income – but had seen their numbers go down and had reversed their decision. Others had even lowered their patient fees. All in all, clearly some form of price competition has emerged. So market mechanisms definitely do come into play. But it cannot be stressed enough that these mechanisms only started to work once the system was functioning: once adequate infrastructure was in place, once equipment was available, once facilities had become more autonomous and transparent, once quality of service delivery had improved and communities had become involved.

This example of course in no way implies that lowering the level of subsidies in RBF will by definition not hurt performance. As a matter of fact, effects were very different in another Cordaid RBF project, in DR Congo: when the subsidy for OPDs was lowered, in a few steps, from $ 0, 40 to $ 0, 11, the number of new OPDs decreased by 35% over the course of 18 months. Part of the explanation probably lies in the fact that public facilities do charge patient fees there: an evaluation found that facilities had actually started to compensate for the lower subsidy by charging significantly higher patient fees, having a negative impact on access to healthcare.

I do believe, however, that there is a lesson to be learnt from the Uganda case. While on that occasion, a subsidy was lowered purely for budgetary reasons, the example seems to prove that there is such a thing as an ‘overpriced’ indicator. By which I mean that the subsidy is higher than would be necessary to sustain a solid growth in performance. If we could identify those indicators also in other RBF programs, where we may not be facing imminent budget constraints, then we could decide to allocate part of our resources differently, incentivizing yet other services and thereby making RBF even more cost effective. At the same time, it needs to be stressed that we should remain careful in eventually making such decisions, especially when it concerns preventive services. These services might be more dependent on the incentive given and hence be subject to greater price elasticity. Careful monitoring of the effects of a price change will be critical here.     

Concluding, we observe a huge potential for more continuous and proactive analysis of operational data. That will enable strategic purchasing of health indicators, a more efficient allocation of resources and thus a higher return on investment. This also implies that we may have to become more flexible in our program management and in our contracting. We will have to be prepared to sometimes take decisions which have not yet been underpinned by academic research, but by making full use of available operational data, combined with thorough context analysis.

Figure 1 – Cordaid RBF project in Uganda - number of new OPDs per semester (public / PNFP facilities)


Add comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.