Illness severity scores are essential for NICU outcome comparisons and should predict both morbidity and mortality. Current validated scores include CRIB (clinical risk index for babies < 1500g, 6 items) and SNAP-PE (SNAP perinatal extension, 37 items). We have evaluated the ability of NAPPI-ES(neonatal acute physiology parameters index, extended score)(14 items) to predict morbidity. Outcomes: Data and variables needed to calculate SNAP, SNAP-PE and CRIB were collected from 10,213 infants in 17 Canadian NICUs in 1996. The development of NAPPI-ES and its validation for prediction of neonatal mortality is described elsewhere. Grade 3 IVH, periventricular leucomalacia, intraparenchymal echogenicity or hydrocephalus were classified as adverse cerebral outcomes (n=347). Chronic lung disease (CLD) was defined as oxygen dependency at 28 days (n=436). Logistic regression of the scores against the outcomes was used to calculate probabilities of the outcomes. NAPPI-ES SNAP-PE and CRIB were compared using area under the receiver operating curve (ROC) to assess their predictive ability in infants of all birthweights and in infants under 1500 g (n=2005). Results: Both NAPPI-ES and SNAP-PE are as effective as CRIB in prediction of adverse cerebral outcome and are better predictors of CLD among infants < 1500g(p< 0.0001). NAPPI-ES and SNAP-PE are significantly better than CRIB (p< 0.0001) in predicting morbidity for all birth weights. NAPPI-ES compares well with SNAP-PE in prediction of morbidity among infants of all birth weights. Table

Table 1 No caption available.
Full size table