Family history, maternal smoking, and clubfoot: an indication of a gene-environment interaction

Although epidemiologic studies of some birth defects have suggested a gene-smoking interaction, the possibility of this interaction in clubfoot has not been examined. The authors analyzed risk factors among 346 infants with isolated clubfoot and 3,029 infants without defects from the Atlanta Birth Defects Case-Control Study. All infants were born during 1968-1980, and mothers were interviewed in 1982-1983. The authors examined the family history-smoking interaction as an indication of a gene-environment interaction. They defined "smoking" as smoking any time during the first 3 months of pregnancy and "family history" as having a first-degree relative with clubfoot. Conditional logistic regression (matching variables: race, birth hospital, and birth period) was used to obtain effect estimates. The adjusted odds ratios were 1.34 (95% confidence interval (CI): 1.04, 1.72) for smoking only, 6.52 (95% CI: 2.95, 14.41) for family history only, and 20.30 (95% CI: 7.90, 52.17) for a joint exposure of smoking and family history. The effect estimate for the joint exposure was higher than would be expected under either an additive or a multiplicative model of interaction and showed a statistically significant departure from additivity. This study confirms the importance of familial factors and smoking in the etiology of clubfoot and identifies a potentially important interaction.