In the period 1996-2004, a hospital-based, case-control study on diet and risk of squamous cell carcinoma of the esophagus was conducted in Montevideo, Uruguay. The study included 234 cases and 468 controls, frequency matched on age, sex, and residence. Factor analysis (principal components) was used to identify the major food patterns associated with esophageal cancer. The analysis of food groups led to the identification of four patterns, arbitrarily labeled as traditional, healthy, high-fat, and drinker. The traditional pattern was not associated with risk of esophageal carcinoma, whereas the healthy pattern displayed a strong inverse association with this malignancy [odds ratio (OR) 0.29, 95% confidence interval (CI) 0.17-0.48, P value for trend < 0.0001]. Unexpectedly, the high-fat foods diet was also marginally protective (OR 0.60, 95% CI 0.36-1.01, P value for trend = 0.07). On the other hand, the drinker pattern displayed a twofold increase in risk of esophageal carcinoma (OR 2.03, 95% CI 1.17-3.51, P value for trend = 0.01). The present study suggests that factor analysis might contribute to the elucidation of the contribution of diet on human cancer. Because diet represents a complex set of highly correlated exposures, dietary patterns could be more explicative of the etiology of squamous cell carcinoma of the esophagus than the traditional reductionist approach.