Low birth weight and Latino ethnicity. Examining the epidemiologic paradox.

OBJECTIVE: To assess the relation between Latino ethnicity, Latino subgroup, and low birth weight (LBW).

DATA SOURCES: From the MEDLINE computer data-base, we used the key words birth weight; infant, LBW; Latinos; Hispanic Americans; Cuban Americans; Mexican Americans; and Puerto Ricans to identify studies that analyzed LBW in Latinos.

STUDY SELECTION: Thirty-two studies, published from 1982 to 1996, that analyzed US Latinos and whites or multiple Latino subgroups, that used the revised definition of LBW (< 2500 g), and had a large sample size (> 10,000) were selected.

DATA EXTRACTION: Two reviewers extracted LBW rates and data on the relation between Institute of Medicine risk factors and LBW by maternal ethnicity and Latino subgroup.

DATA SYNTHESIS: Low-birth-weight rates were similar for Latino (median, 6.2%) and white infants (median, 5.8%). By Latino subgroup, LBW rates were similar for Central/South American, Cuban, Mexican, and white infants. Puerto Rican infants had consistently higher LBW rates (median, 9.1%). Two risk factors-maternal birth-place and gestational weight gain-were identified as confounders of the relation between Latino ethnicity, Latino subgroup, and LBW.

CONCLUSIONS: Low-birth-weight rates of Latinos and whites are similar, consistent with the "epidemiologic paradox" of unexpectedly favorable perinatal outcomes for Latinos. However, this paradoxical relation for all Latinos masks the notably elevated LBW risk among Puerto Ricans. Further study of LBW among Latinos, including cultural factors, is needed.