Effect of pay for performance on the quality of antenatal care in Zimbabwe : a controlled before-after study

Abstract

Background: Pay for Performance (P4P) mechanisms to health facilities and providers are currently being tested in several low- and middle-income countries (LMIC) to improve maternal and child health (MCH). Though quality of care is necessary to improve MCH, the prevailing evidence on P4P’s effect on MCH quality is limited. Zimbabwe implemented a P4P program from April 2012 to September 2014 to improve its adverse scenario on MCH indicators. This study explores the effect of the P4P program on the quality of antenatal care (ANC) on the following three dimensions – structural quality, process quality and client satisfaction. Methods: The study design was a controlled before-after implementation evaluation in 16 matched pairs of rural districts. Intervention facilities (n=374) received P4P, while control facilities (n=331) continued with the routine government program. Out of these, a subset of 77 randomly selected health facilities and the ANC clients attending these facilities were surveyed before implementation of P4P (385 clients) and after implementation (374 clients) to measure the impact of the program. The impact of P4P was estimated on an intention-to-treat basis by comparing the difference-in-difference in the indices of impact between the two arms of the study. Multilevel regression analyses were used to account for the hierarchical study design. Results: All dimensions of quality showed significant improvements in the P4P arm – structural index was 0.595 standard deviations (SD) higher (p=0.023), process index 0.556 SD (p=0.001) and client satisfaction index 0.6 SD (p=0.001) higher than the mean of the respective index in the control group. Clients consulting a nurse or a male healthcare provider reported higher process quality and satisfaction, whereas those visiting a nurse midwife had lower process quality and client satisfaction. Women with lower levels of wealth and education reported lower process quality but higher satisfaction in the P4P arm. Discussion: The Zimbabwe P4P showed improvements in the quality of ANC. However, there is a need for further exploration of the contextual factors to understand the mechanisms of these improvements.