Investigating the determinants of health worker performance in Senegal

Lead Research Organisation: London School of Hygiene & Tropical Medicine
Department Name: Public Health and Policy

Abstract

Training, motivating and retaining professional health workers is crucial for the improvement of health outcomes, especially in low and middle-income countries where poor health worker performance has been recognised as one of the key obstacles to achieving better health outcomes. To improve staff motivation and performance, many donors and governments have recently supported the introduction of Performance-Based Financing (PBF) mechanisms, which link part of the health workers' remuneration to performance targets. While promising results have shown that these programmes can improve utilisation of health services and some health outcomes, it is still unclear what specific effects they have on the motivation and behaviours of health workers. Yet it is key to understand better how PBF works for health workers, and in particular whether all aspects of health workers' performance have improved or whether PBF can induce some undesired effects, e.g. a reduction in time spent with each patient.

The aim of the proposed research is to generate new knowledge and understanding of how PBF programmes work and influence health workers' performance. The proposed research takes advantage of a quasi-experimental study introduced by the World Bank in four of the poorest and most rural regions of Senegal in 2014. The project will collect a rich set of primary data in control and PBF facilities to answer several critical questions.

The research will first undertake some interviews to assess whether the extent to which what was planned in the PBF scheme is actually what is happening in practice. This will help understand exactly the nature of the intervention being implemented. The research will then undertake a large survey, collecting information in control and PBF facilities on the working environment and characteristics, behaviours and performance of health care workers. The aim is to determine the effects of PBF, by comparing the performance and behaviours of health workers with and without PBF incentives. Specifically, the survey will explore whether PBF improves the productivity, quality of care provided and attitude towards patients. The data collected will also help determine whether the time and effort spent on each patient increase and on non-incentivised activities decrease as a result of the PBF. Finally, with the information collected, we will be able to identify mechanisms through which the performance of health workers changes.

Technical Summary

In many Low- and Middle-Income Countries were the performance of health workers is low, governments are increasingly seeking to make sure that the way in which health workers are paid encourages them to work more efficiently and effectively. To that aim, interest in payment mechanisms linking part of the remuneration to achieving certain performance targets (often termed Performance-Based Financing or PBF) has recently grown. Many are hopeful that these schemes will improve health outcomes in places where, despite huge investments, they remain low. Despite this enthusiasm, there is limited evidence on the effects of PBF, and no understanding of whether and how financial incentives improve the performance of health care workers.

The research aims to evaluate whether and how PBF contributes to improved health worker performance in Senegal, where it will take advantage of the introduction of a PBF programme in a quasi-experimental manner. Following a process evaluation that will help determine the degree of implementation of the programme, a large cross-sectional survey will be carried out to collect a broad range of health workers' performance outcomes in control and PBF facilities. Comparing these outcomes between health workers working in facilities with and without PBF, we will be able to determine the causal impact of PBF on health workers' performance. As the survey will also collect information on health workers' motivation and working conditions, we will be able to understand some of the causal pathways through which any changes occur.

Findings from the project will not only inform the national discussions around the scale-up of PBF in Senegal, but they will also feed into the policy debates on human resources performance in other countries and on the global health scene. This will be achieved through an active research uptake plan that will include the production of policy briefs, the organisation of workshops and a number of networking activities.

Planned Impact

We anticipate that this research will lead to three types of impact. We highlight below the beneficiaries, explain how they might benefit from this research and provide some details about the activities to be undertaken to ensure they benefit from this research.

1. Contribute to the effectiveness of health care policies in Low- and Middle-Income Countries (LMICs) by feeding into the debates around payment mechanisms of health care workers

Given current interest in PBF, this research should directly feed into the reflections of policy-makers involved in the design of financial incentives and other interventions that aim to shape the behaviours of health workers, in Senegal, in other LMICs, as well as on the global health scene (donors, international organisations, etc.). Several activities are planned to ensure that policy-makers will benefit from this research:
- engagement with key leaders and decision makers in Senegal and other LMICs, facilitated by existing networks of the members of the research teams, for example at through the DfID-funded RESYST Consortium.
- Seminars presenting interim findings to policy-makers to discuss interpretations of results, additional analyses and potential lessons drawn: such seminars will be easy to organise at DfID or with regular interactions with donors the research team has; in Senegal, they will take place during specifically organised workshops.
- Findings will be disseminated through working papers and in a non-technical language through policy briefs and other online resources (e.g. blogs, webinars, etc.) that will be sent to policy makers.

2. Improve the quality and efficiency of health care delivery by enabling a greater understanding amongst managers of the ways in which their services can be rearranged to achieve better outcomes and higher performance
The research offers real potential to affect health outcomes by informing how health care providers work with, adapt to and react to incentives. We will make sure that research participants in Senegal, and in general health workers and managers involved in the delivery of health care in LMICs will benefit from findings of this research through the following activities:
- Workshops presenting interim and final findings to health workers and district managers in Senegal to discuss the results and potential lessons drawn.
- Written briefing material highlighting key findings and their implications for policy and practice to demonstrate the various ways in which teams and organisations are likely to react to incentives. This material will seek to draw lessons for and with health care providers on the ways in which they could change the organisation of care to improve their performance.

3. Build the capacity of future public health professionals and decision makers, especially from LMICs, by promoting a greater understanding of the ways in which health care providers' behaviours can be affected by incentives

The last group of beneficiaries will be future public health professionals and decision-makers in health policy.
The three UK-based members of the research teams contribute to the LSHTM teaching programme on topics of health economics and health services through seminars or lectures provided for different taught modules, online and in London. Teaching material will be developed and used for LSHTM in-house and online courses (case studies, role-play situations) in health economics and health services.

The material developed will then be translated in French and shared with the Senegalese partner to be used in their taught courses. Other African universities will also be able to use the material developed through the Consortium for Health Policy and Systems Analysis in Africa, a partnership of African and European universities, whose purpose is to increase African capacity in the area of high-quality Health Policy and Systems research through the development of teaching material for African educators.

Publications

10 25 50
 
Description We took advantage of a randomized implementation of a Performance-Based Financing (PBF) scheme in Senegal to test whether financial incentives can improve quality of care. Overall, we found low levels of quality of care. In control facilities, providers completed only 30% of an essential checklist of care, and less than half of patients (42%) were managed according to the recommended guidelines. We found no evidence that the financial incentives provided by the PBF scheme improved any of the measures of quality of care, nor that it increased the likelihood that patients would be managed correctly. We ruled out poor clinical knowledge or lack of understanding of the scheme as potential reasons explaining the lack of effect of financial incentives. We found suggestive evidence that in this setting, where patients have limited information about what constitutes good quality of care, financial incentives are unlikely to improve effort through increased accountability.

In addition, we undertook further analyses of the data we collected and found
(1) that patient trust in providers in Senegal is positively associated with continuity of care, provider communication ability, and clinical knowledge.
(2) that overconfident healthcare workers provider lower quality healthcare (they are 26% less likely to manage patients correctly). We found that this may be due to the lower effort exerted during consultations.
(3) that quality of care is higher when providers receive more relevant information from patients, as they do not need to exert much effort during the consultation to extract relevant information.

Overall all of our findings point to failures of providers to exert appropriate effort during consultations. While financial incentives do not seem to provide an adequate response to address this issue, further research is needed to test alternative solutions, such as decision-support tools to help providers diagnose patients correctly.
Exploitation Route Policy-makers currently thinking about strategies to improve quality of care will benefit from the findings of the project in several ways. First, the results caution against the use of Performance-Based Financing reforms as a way to improve quality of care and provider performance. Instead, we point to cognitive biases and the importance to identify new solutions to improve process quality of care during consultations in order to improve quality of care and health outcomes.
Sectors Healthcare

 
Description Findings from the research (some of which are still in the form of working papers) have contributed to informing the debates on Pay-for-Performance (P4P) in Senegal. Our research showed no evidence that P4P improved the quality of care in facilities incentivised to improve the utilisation and quality of services. On the one hand, our preliminary findings (presented to stakeholders) led to a redesign of the program, under the impetus of the World Bank, to change indicators and try to improve the design of the scheme. On the other hand, our findings further fuelled the skepticism of national policy-makers for P4P policies.
First Year Of Impact 2018
Sector Healthcare
Impact Types Policy & public services

 
Description Health Systems Research Initiative
Amount £765,082 (GBP)
Organisation Medical Research Council (MRC) 
Sector Public
Country United Kingdom
Start 04/2020 
End 04/2023
 
Description Senegal collaboration 
Organisation University of California, Davis
Country United States 
Sector Academic/University 
PI Contribution Following some early problems encountered with the originally named local collaborator, we have identified a team of really motivated and committed social scientists from the IPDSR (Institute de Population, Development et Sante de la Reproduction), whose staff has been incredibly helpful and professional in this design phase of the study.
Collaborator Contribution Following local ethical approval, we are now piloting the tools with them and some really good enumerators. Local collaborators have been instrumental in 1)preparing the final submission to the local ethics committee ; 2) presenting and explaining our research project to the local ethics committee ; 3) identifying excellent and knowledgeable enumerators to improve the tools and pilot them ; 4) organise some validation workshops with key stakeholders.
Impact Not yet.
Start Year 2015
 
Description Regional visits and meetings with local stakeholders 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact In December 2015, the PI and project RA carried out an inception visit during which they held various meetings with national and regional stakeholders.
National stakeholders included representatives of the Ministry of Health involved in the implementation of the project studied (financial incentives), as well as donors funding this intervention (World Bank). Additional local meetings were held in Casamance area (the South of the country where part of the study would take place), with district officials as well as field trips to local health care facilities.
Throughout the meetings, the project objectives were presented and discussed to ascertain its feasibility and relevance.
Year(s) Of Engagement Activity 2015
 
Description Workshop with local stakeholders - Validation of tools 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact To ensure that all our tools were adapted to the local context we conducted a validation workshop. The workshop took place in Dakar and lasted 5 days (14th - 18th of March 2016). Eight local experts were invited to participate: one general practitioner, a paediatrician, an immunologist, three nurses and two midwives. All of them had practice in Senegal for more than 5 years. Some of them had also worked in the study areas.
Year(s) Of Engagement Activity 2016