Power to the People: Evidence from a Randomized Field Experiment on Community-Based Monitoring in Uganda

Replication paper
Original publication: The Quarterly Journal of Economics
Original researchers: Martina Bjorkman, Jakob Svensson
Replication plan:

Donato's replication plan

Replication researchers: Katherine Donato (with Adrian Garcia Mosqueira)
Current status: Completed Replication Study

The Original Study

This paper investigates whether community-based monitoring of public primary health care providers improves health outcomes and monitoring/accountability of service providers. The intervention consisted of a series of public meetings that culminated in the adoption of a shared action plan by community members and health workers. The study was conducted in nine districts in Uganda as a randomized field experiment among 50 rural public dispensaries and their catchment area, defined as the households within the dispensaries’ 5-kilometre radius area. Information on community monitoring, health workers, health facilities, service delivery and health outcomes was gathered from the dispensaries’ internal records, visual checks and 5,000 surveyed households. The causal impact of the intervention was measured using a regression equation with pre-intervention control variables and district fixed effects, a difference-in-differences (DID) specification where pre and post-intervention data were available, and an instrumental variable (IV) model.

Overall, the intervention was deemed highly successful. The authors found that in treatment communities: a) health facilities were better managed and the customer service was improved; b) health staff’s performance was discussed more frequently in local council meetings and households were more likely to be informed about the roles and responsibilities of the local health governing body; c) the rate of absenteeism of health workers was dropped; d) facilities’ outpatient services and number of deliveries were higher; e) households were less likely to self-treat and rely on traditional healers and more likely to access antenatal care and family planning services; f) newborns and young children were more likely to have received required dose of vaccines and vitamin A supplements; and g) pregnancies, under-five mortality and malnourishment also decreased. 

The Replication

Over the last few decades, reducing child mortality and improving health outcomes have been pressing objectives in sub-Saharan Africa and much of the developing world. While there are a wide range of potentially valuable interventions designed to combat these problems, there is mounting evidence that issues like absenteeism and leakage of public funds can significantly hinder the effectiveness of these interventions. Bjorkman and Svensson (2009) show that a relatively simple intervention – providing community-level health outcome information and guidance on community-based monitoring (CBM) – dramatically improved health outcomes, even rivaling some of the most effective health interventions to date. In this replication paper, we seek to verify the robustness of authors’ results through a pure replication, as well as to expand on their analysis in an effort to gain more insights into the mechanisms that allowed the CBM program to so effectively improve health outcomes.