EVALUATION DESIGN

Since 2010, Burundi implemented its first nationwide Performance-Based Financing (PBF) program in 643 health centers and 54 hospitals. These health facilities were contracted and offered PBF payments based on the quantity and quality of health services achieved. Health facilities considered disadvantaged were offered an additional equity markup. The project’s goal was to increase the utilization of health care services by pregnant mothers and children under five. Although, there was no impact evaluation to assess the effectiveness of the program, performance monitoring data showed positive results on both the quality and the quantity of care provided.

In 2014, additional financing supported the continuation and expansion of the original PBF program, while also adding nutrition and family planning services to the program. These additional services offered an opportunity to design an evaluation, which allowed to measure the effect of the nutritional package services embedded in the PBF program. Specifically, the proposed evaluation seeks to measure the effects of the nutrition related services on primary outcomes such as the prevalence of acute malnutrition and stunting among children. Prevalence of infectious episodes like diarrhea, upper respiratory infections, and fever were also outcomes of interest for the evaluation. Further, the evaluation is assessing whether the introduction of malnutrition prevention and care activities result in better outcomes at the health facility level, in terms of better management of cases (i.e. better recovery raters, shorter periods of treatment).

To evaluate the program, a cluster-randomized controlled trial was designed. Among the eligible health centers providing nutrition services (i.e., health centers treating severe and moderate acute malnutrition) a subset of 90 health centers was randomly selected to participate in the evaluation. The 90 selected health centers were paired on essential parameters of organization and functioning in relation to the outcomes (such as volume of activity, staff structure and training, and percentage of recovery among malnourished children) as measured during the baseline survey. Within each of these pairs of health facilities, one health center was randomly selected to be in the intervention group. 45 health facilities were randomly allocated in the intervention group (health facilities providing nutrition services) and the remaining health facilities formed the control group.

The endline household and health facility surveys are ongoing. The goal is to collect data at the onset and at the end of the program to measure the effectiveness of the program on the aforementioned outcomes. The proposed evaluation study will provide the country to assess the effect of the nutrition services coupled embedded in a PBF program.

FINDINGS

The implementation of the additional features of the project and the impact evaluation are in the second stage of implementation (end line surveys are ongoing). This program has faced delays because of the political and security situation.