||Jansen RJ, Robinson DP, Stolzenberg-Solomon RZ, Bamlet WR, de Andrade M, Oberg AL, Hammer TJ, Rabe KG, Anderson KE, Olson JE, Sinha R, Petersen GM
||OBJECTIVE: Studies on fruit, vegetable, fiber, and grain consumption and pancreatic cancer risk are inconclusive. We used a clinic-based case-control study specifically designed to address limitations of both cohort and case-control studies to examine the relationship. METHODS: Participants were excluded who reported changing their diet within 5Â years prior to study entry. And 384 rapidly ascertained cases and 983 controls (frequency matched on age (Â±5Â years), race, sex, and residence) completed epidemiologic surveys and 144-item food frequency questionnaires. Odds ratios (OR) and 95% confidence intervals were calculated using logistic regression adjusted for age, sex, smoking, body mass index, energy intake, and alcohol consumption. RESULTS: Comparing highest to lowest quintiles, we observed significant inverse associations (ORÂ <Â 0.8) with significant trends (p (trend)Â <Â 0.05) for citrus, melon, and berries, other fruits, dark green vegetables, deep yellow vegetables, tomato, other vegetables, dry bean and pea, insoluble fiber, soluble fiber, whole grains, and orange/grapefruit juice, and an increased association with non-whole grains. Results were similar after adjusting for diabetes or total sugar intake. CONCLUSIONS: We provide evidence that lower consumption of fruits, vegetables, whole grains, and fiber is associated with having pancreatic cancer. This may have a role in developing prevention strategies.