.Anorectal atresia is a congenital anomaly with mostly unknown risk factors. Studies have provided evidence of teratogenic effects of alcohol and tobacco, and animal studies have suggested that caffeine may potentiate their teratogenicity. However, it is unclear how these factors affect the risk of anorectal atresia. We analysed data from maternal telephone interviews in a multistate case-control study with 464 infants with anorectal atresia and 4940 infants with no major birth defects. We used unconditional logistic regression to determine the association of exposure to smoking, environmental tobacco smoke (ETS), alcohol or caffeine with anorectal atresia. Effect modification by caffeine intake was assessed on additive and multiplicative scales. There was no association with alcohol intake in this analysis. However, there was some evidence of an association between anorectal atresia and maternal exposure to tobacco smoke and caffeine. Compared with non-smokers not exposed to ETS, the crude odds ratio (OR) and 95% confidence interval [CI] for cigarette smoking was 1.2 [95% CI 1.0, 1.5]. The association with anorectal atresia for non-smokers exposed to ETS at home and work was OR = 2.3 [95% CI 1.2, 4.1]. Compared with the lowest level of caffeine intake (<10 mg/day), the association for the highest caffeine intake (> or =300 mg/day) was OR = 1.5 [95% CI 1.0, 2.2]. Results did not change after adjustment for covariates. This study found evidence of associations between anorectal atresia and caffeine intake, cigarette smoking and exposure to ETS. Because there are currently few additional data to support these results, further study is needed.