BACKGROUND.
Although tobacco is the primary etiologic factor for oral precancerous lesions in India, evidence from other sources indicates that diet may modify risk. This case-control study was designed to minimize a variety of biases in its attempt to investigate the relation between diet and oral precancerous lesions.
METHODS.
In a house-to-house survey of 5056 tobacco users in a rural area of Ernakulam district in Kerala, India, 226 individuals (44 females and 182 males) were found to have precancerous lesions (cases), which in 4 cases proved to be cancer.
From among the examinees, an equal number of controls who were free of oral mucosal lesions and were matched to the cases regarding age (Ϯ5 years), gender, ward of residence, and use of tobacco also were enrolled. Dietary data were obtained using a customized interviewer-administered food-frequency questionnaire. All subjects and interviewers were blinded to the disease status of the subject.
RESULTS.
After controlling for tobacco use, intake of fruits, vegetables, and -carotene evinced inverse trends in risk (P Ͻ 0.05), with an average reduction of over 10% per quartile of exposure. Associations with certain micronutrients appeared to differ according to gender, with an apparent 20% reduction in risk per mg of zinc consumed per day among men and the suggestion of an increased risk among those women in the lowest quartile of iron intake (an increase of approximately 2.5-fold) and ascorbic acid intake (an increase of approximately 70% increase) compared with other women (P Ͻ 0.10).