Objective: Infants born in Maryland between June 1992 and June 1996 were used in a case-control study of nonsyndromic oral clefts to test for effects of maternal smoking and a polymorphic genetic marker at the transforming growth factor alpha (TGFA) locus, both of which have been reported to be risk factors for these common birth defects. Design and Setting: Cases were infants with an oral cleft ascertained through three comprehensive treatment centers, with additional ascertainment through a registry of birth defects maintained by the Maryland Health Department. Controls were healthy infants. Medical history information on infants and mothers were collected, along with DNA samples Patients, Participants: Among 286 cases contacted (72% ascertainment), there were 192 nonsyndromic isolated oral clefts (106 M; 86 F) available for this case-control study. Main Outcome Measures: The largest group of 149 Caucasian nonsyndromic cases and 86 controls was used to test for association with maternal smoking and genotype at the Taq1 polymorphism in TGFA. Results: While this modest sample had limited statistical power to detect gene-environment interaction, there was a significant marginal Increase In risk of having an oral cleft If the mother smoked (odds ratio = 1.75, 95%CI = 1.01 to 3.02). We could not demonstrate statistical interaction between maternal smoking and TGFA genotype in this study, however, and the observed increase in the C2 allele among cases was not statistically significant. Conclusions: We could not confirm either the reported association between oral clefts and TGFA genotype or its interaction with maternal smoking. However, these data do show an increased risk if the mother smoked during pregnancy, and this effect was greatest among infants with a bilateral cleft and no close family history of clefts.
Power and sample size for multivariate logistic modeling of unmatched case-control studies
