Total cancer risk: Data about total cancer risk was provided by two cohorts including a total of 6,449 cases. No association was found.
Total cancer mortality: Data about total cancer mortality was provided by one very small cohort. No association was found.
Conclusion: No evidence was found for an association between broccoli and total cancer.
| Author | Cohort name | Subjects | Years of follow-up | Cases | End point | Consumption of | Relative Risk (RR) | Adjustments |
| Wang L (2009) | The Women's Health Study | 38,408 female health professionals aged ≥ 45. (USA) | 11.5 (1992-95 to 2007) | 3,219 | Invasive cancer incidence (excluding nonmelanoma skin cancer) | Broccoli |
RR = 1.05 (0.88-1.25; P = 0.28) for the highest vs lowest quartile of consumption. Amount specific data (servings): < 1/mo: RR = 1. 1-3/mo: RR = 1.00 (0.85-1.19). 1/wk: RR = 1.00 (0.85-1.18). ≥ 2/wk: RR = 1.05 (0.88-1.25). Sensitivity analyses that categorized food intake using other cutoffs yielded similar results (data not shown). | Age, race, total energy, randomized treatment assignment, smoking, alcohol use, physical activity, postmenopausal status, hormone replacement therapy use, multivitamin use, family history of colorectal cancer, ovary cancer, or breast cancer, and intake of fruit and vegetables, fiber, folate, and saturated fat. |
| 20) Takachi R. (2007) | The Japan Public Health Center-based Prospective Study (JPHC). | 77,891 subjects (35,909 men 41,982 and women) aged 45-74. (Japan) | 5.8 (1995-1998 to 2002) | 3,230? (636 gastric, 598 colorectal, 397 lung, 233 breast, 193 liver, 1,173 other) | Risk of cancer | Broccoli | Not significantly associated with risk of total cancer (no data shown). | age, public health center area, BMI, physical activity, smoking, alcohol, energy, screening examination, medication, and daily vitamin supplement use. |
| 1) Colditz GA. (1985) | The Massachusetts Health Care Panel Study. | 1,226 residents 66 years or older. (USA) | 5 (1976-1980) | 42? | Total cancer mortality | Broccoli | RR = 0.8 (0.4-1.6; No P-value) for high vs low consumption using median cut (median cut = < 1 serving/wk). | Age. |