Prenatal alcohol exposure and risk of birth defects.

OBJECTIVE

The goal was to examine the associations between dose, pattern, and timing of prenatal alcohol exposure (PAE) and birth defects.

METHODS

Data from a randomly selected, population-based cohort of nonindigenous women who gave birth to a live infant in Western Australia (WA) between 1995 and 1997 (N=4714) were linked to WA Midwives Notification System and WA Birth Defects Registry data. We assessed the associations of PAE before pregnancy, in the first trimester, and in late pregnancy with any birth defect and with birth defects classified as alcohol-related birth defects (ARBDs) by the Institute of Medicine (IOM), by using logistic regression.

RESULTS

The prevalence of birth defects classified as ARBDs by the IOM was low. Compared with abstinence, heavy PAE in the first trimester was associated with increased odds of birth defects classified as ARBDs (adjusted odds ratio: 4.6 [95% confidence interval: 1.5-14.3]), with similar findings after validation through bootstrap analysis. There was no association between low or moderate PAE and birth defects.

CONCLUSIONS

A fourfold increased risk of birth defects classified as ARBDs was observed after heavy PAE in the first trimester. Many individual birth defects included in the IOM classification for ARBDs either were not present in this cohort or were not associated with PAE. Large, population-based studies are needed to strengthen the evidence base for ARBDs.