Selective migration and infant mortality among Puerto Ricans

Objective: This study examines the implications of migration to the United States for infant mortality among Puerto Rican mothers born in Puerto Rico. The roles of selective migration and duration of US residence are assessed. Method: Using survey data collected from mothers of infants sampled from computerized birth and infant death records of six US vital statistics reporting areas and Puerto Rico, we estimate logistic regression models of infant mortality among the sampled infants. These models provide a baseline for comparison with fixed-effects models based on all births within each mother's history. Results: Logistic regression models for sampled infants show that the risk of infant mortality is lower for migrant women than for nonmigrant women in Puerto Rico until the migrants have lived in the United States for a substantial period of time. Fixed-effects models indicate that once unmeasured stable characteristics of the mother are controlled, early migrants do not differ from nonmigrants with respect to the risk of infant death. Both sets of models demonstrate that as mothers' exposure to the US mainland increases, the risk of infant mortality rises. Conclusions: Selective migration plays a role in the relatively low risk of infant mortality among recent Puerto Rican migrants to the United States. Migrants appear to be selected on qualities that contribute to favorable health outcomes for their offspring, but those qualities are later lost with exposure to life in the United States.