Race/ethnicity, Apgar and infant mortality

Objective. Our general objective of this study is to further assess the predictive validity of Apgar scores on infant mortality using a national-level data set allowing for race/ethnic-specific variation. Method. This analysis is based on the 1989-1991 NCHS Linked Birth/Infant Death files. Multivariate, multinomial logistic regression models were constructed adjusting for maternal behavioral and health risks, socioeconomic and demographic factors, and infant characteristics. Results. After adjusting for maternal risk factors, infants with low Apgar scores experienced 86 times the odds of neonatal death relative to high Apgar score infants while infants with medium Apgar scores experienced 10 times the odds of neonatal death relative to the high Apgar referent. After controlling for maternal risk factors, birth weight and gestational age, the effect of low Apgar scores on neonatal mortality risk relative to high Apgar scores is reduced (from OR = 86.1 to OR = 31.2), but still remains a strong predictor. Fully adjusted, race/ethnic-specific models show that the predictive value of low Apgar on neonatal mortality risk is strongest for Mexican Americans (OR = 44.2) versus Non-Hispanic Whites and Blacks (OR = 36.2 and 20.4 respectively). Conclusion Our results support the conclusion that Apgar scores continue to be strong predictors of infant survival, independent of birth weight, gestational age, and a large number of maternal risk factors. They are also shown to be powerful predictors within different race/ ethnic groups, solidifying their importance as both a diagnostic tool and descriptive health characteristic across various U.S. populations.