Shenyang in Northern China has one of the highest reported prevalence rates of oral clefts in the world. To explore the risk factors for oral clefts in Shenyang, we carried out a population-based case-control study. A total of 360 990 births in 2000 to 2007 were screened for oral clefts; the overall prevalence was 1.76 per 1000. The ratio of cleft lip with or without cleft palate (CL +/- P) : cleft palate only (CP) was 5.60:1. The overall male : female ratio was 2.02:1. CLP and CL were more common in males than in females with a sex ratio (SR) of 2.88:1 and 1.86:1 respectively, whereas CP was more common in females with an SR of 0.71:1. Using a multivariable conditional logistic regression model, 586 oral clefts cases were compared with 1172 control mothers. Maternal factors significantly associated with increased risk for oral clefts were history of a fever or cold (adjusted OR 2.34, 95% CI 1.06, 5.60); use of analgesic and antipyretic drugs (adjusted OR 3.10, 95% CI 1.41, 6.86); poor ventilation during heating (adjusted OR 2.25, 95% CI 1.10, 4.60); and consumption of pickled vegetables >6 per week (adjusted OR 3.86, 95% CI 1.11, 13.47) during pregnancy. Factors which appeared to be protective were meat consumption >or=4 times per week (adjusted OR 0.43, 95% CI 0.28, 0.67); and legume consumption >6 times per week (adjusted OR 0.60, 95% CI 0.41, 0.89). Differences in risk were found between the two most common phenotypes, CL +/- P and CP only. Most of the environmental factors had stronger associations with risk for CL +/- P rather than CP, whereas history of oral clefts, as well as legume consumption, were more strongly associated with the risk for CP than for CL +/- P. The findings suggest that aetiological heterogeneity may exist between CL +/- P and CP.