Did PEPFAR funding affect non-HIV health indicators? Longitudinal study of facility-based baby deliveries in Uganda, 2005–10

Abstract Background While PEPFAR investments were associated with HIV-related survival, it has been less clear what the effects of this investment in HIV care have been for health-care use generally, especially in maternal care. Understanding of these effects could have implications for how to organise future global health initiatives for major health problems. Maternal mortality in Uganda remains high (310 per 100 000 livebirths in 2010) and could be reduced by increasing deliveries at health facilities. Methods We visited all of Uganda's 56 districts to collect maternity care data from Health Management Information System records, between the years 2005 and 2010, including monthly reports from all public and most private facilities. Our unit of analysis was district-month. We fitted a negative binomial mixed effects model, with number of infants delivered in health facilities per district-month as outcome, and three categories (tertiles) of population-adjusted annual number of PEPFAR-supported patients on antiretroviral therapy (ART) as input. We adjusted for regional HIV prevalence and proportion of elementary school-aged children attending school. To adjust for secular and seasonal effects, we included control variables for year and month. Random effects for district and an offset for district population were included in the model. Findings People on ART per district-month per 1000 population grew from 1079 in 2005, to 6485 in 2010. Facility deliveries averaged 709 per district-month, growing from 561 in 2005 to 830 in 2010. The average rate of health facility deliveries increased by 4% (incidence rate ratio [IRR] 1·04, 95% CI 1·008–1·068, p=0·012) in districts with medium-level PEPFAR investment in ART care, compared with lowest PEPFAR investment districts. In higher-level investment districts, deliveries increased by 8% (IRR 1·08, 95% CI 1·037–1·126, p=0·0001), relative to low-investment districts. Interpretation PEPFAR investments in ART scale-up in Uganda appear to be associated with small increases in health facility deliveries. Funding This research was funded by the US Centers for Disease Control, Division of Public Health Systems and Workforce Development through a cooperative agreement with the University of Washington and through subcontracts with Makerere University and the Uganda Ministry of Health.