Measuring health facility readiness and its effects on severe malaria outcomes in Uganda

There is paucity of evidence for the role of health service delivery to the malaria decline in Uganda We developed a methodology to quantify health facility readiness and assessed its role on severe malaria outcomes among lower-level facilities (HCIIIs and HCIIs) in the country. Malaria data was extracted from the Health Management Information System (HMIS). General service and malaria-specific readiness indicators were obtained from the 2013 Uganda service delivery indicator survey. Multiple correspondence analysis (MCA) was used to construct a composite facility readiness score based on multiple factorial axes. Geostatistical models assessed the effect of facility readiness on malaria deaths and severe cases. Malaria readiness was achieved in one-quarter of the facilities. The composite readiness score explained 48% and 46% of the variation in the original indicators compared to 23% and 27%, explained by the first axis alone for HCIIIs and HCIIs, respectively. Mortality rate was 64% (IRR = 0.36, 95% BCI: 0.14-0.61) and 68% (IRR = 0.32, 95% BCI: 0.12-0.54) lower in the medium and high compared to low readiness groups, respectively. A composite readiness index is more informative and consistent than the one based on the first MCA factorial axis. In Uganda, higher facility readiness is associated with a reduced risk of severe malaria outcomes.