associated with mortality in the univariate analysis. After adjustment for covariates, rhesus
incompatibility, asphyxia, exposure to hemolytic substances and ABE were predictive of mortality.
Infants who received ET had 69% probability of survival. The discriminatory power of
the final model was also strong (c-statistic: 0.922, 95% CI: 0.893–0.957). High bilirubin levels
(OR:1.11, 95% CI:1.07–1.14), exposure to hemolytic substances (OR:7.73 95% CI:3.16–18.91)
and rhesus incompatibility (OR:6.13 95% CI:2.44–15.43) were predictive of mortality among
infants with concomitant ABE (data not shown). Infants with at least one risk factor for