Does provider access to technology improve health care? Evidence from a national distribution of phototherapy in Rwanda

Abstract

This study assessed a large-scale national distribution of phototherapy (PT) for infants at risk for severe hyperbilirubinemia. We combined healthcare data for infants with jaundice (using local clinical definitions) with a randomized roll-out of PT devices to estimate the causal effect of the national distribution. Pre-intervention, <3.0% of infants were diagnosed as jaundiced, 41.7% of these infants were not tested for bilirubin, and 17.5% and 34.3% were treated with direct sunlight or standard PT, respectively. We found that providing hospitals with PT devices increased care practices: infants with jaundice were more likely to receive PT [+6.26 percentage points (pp)], and less likely to receive sunlight (-6.96 pp) or standard (irradiance < 30 µW/cm2/nm) PT (-14.0 pp). However, the intervention did not affect the low diagnosis rate. Our findings suggest that complementary investments in improving diagnosis and monitoring of bilirubin levels increases the benefits of expanding provider access to PT devices.