Objective: To investigate the possible differential role of refined and whole grain cereals on the risk of upper digestive and respiratory tract neoplasms. Design: Hospital-based case-control study. Setting: University Hospital of Lausanne, Switzerland. Subjects: A total of 156 incident cases of cancers of the oral cavity and pharynx, 101 of the oesophagus, 40 of the larynx, and 349 control subjects admitted for a wide spectrum of acute non-neoplastic conditions. Intervention: Trained interviewers collected information using a structured and validated questionnaire. Odds ratios (OR) of various cancers for a tertile increment of intake of refined and whole grains were estimated using unconditional multiple logistic regression. Results: Refined grains were directly related to the risk of cancer of the oral cavity and pharynx (OR=1.9 for the highest tertile), oesophagus (OR=3.7) and larynx (OR=4.0). In contrast, whole grain cereals were protective, with OR of 0.6 for oral cavity and pharynx, 0.3 for oesophagus, and 0.7 for larynx. For the three sites combined, the OR for the highest tertile was 5.7 for refined grains and 0.5 for wholegrains. The trends in risk for refined grains were significant for all sites and their combination, and for wholegrain for oesophageal cancer and all sites. Conclusions: Even if inference on causality and the biological interpretation remain open to discussion, the present data indicate and further quantify that refined cereals are an unfavourable, but whole grain ones a favourable indicator of the risk of upper aerodigestive and respiratory tract neoplasms. Sponsorship: Swiss Foundation for Research Against Cancer and Vaud Cantonal League against Cancer. European Journal of Clinical Nutrition (2000) 54, 487-489 | 
