Morrison · Cancer epidemiology 2021 · hospital-based case-control study · n=915

Cruciferous vegetable consumption and pancreatic cancer: A case-control study.

Cited 13 times in the scientific literature.

Level 4 - case-series / case-control

Hospital-based case-control study

PubMed 33714902 · doi:10.1016/j.canep.2021.101924 · record verified 2026-08-30

What was done

Researchers conducted a hospital-based case-control study using historical data (1982–1998) from the Roswell Park Comprehensive Cancer Center, comparing 183 pancreatic cancer cases to 732 cancer-free controls. Dietary intake was assessed via a self-administered 52-item food frequency questionnaire. Multivariable logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) adjusted for age, sex, body mass index (BMI), smoking status, total meat intake, and family history of pancreatic cancer, with subgroup analyses by sex, smoking status, and BMI.

What was found

Subjects consuming >1.5 servings per week (SPW) of raw cruciferous vegetables had a significant 40% lower odds of pancreatic cancer compared to those consuming <0.5 SPW (OR = 0.60, 95% CI: 0.39–0.93). Each additional SPW of total, raw, and cooked cruciferous vegetables was associated with a significant 7% to 15% lower odds of pancreatic cancer, with the strongest association for raw vegetables (OR = 0.85, 95% CI: 0.75–0.95). In subgroup analyses, raw cruciferous vegetable consumption was associated with 50% to 59% lower odds among males, former smokers, and individuals who were overweight.

Why it matters

This study suggests that cruciferous vegetable intake, especially in raw form, is a modifiable dietary factor inversely associated with pancreatic cancer risk, including across higher-risk demographic subgroups.

Limits

The hospital-based case-control design carries inherent risks of selection bias and dietary recall bias. Data rely on a relatively short 52-item questionnaire collected between 1982 and 1998, and the total number of cancer cases was relatively modest (n = 183).

Cited by