Prospective cohort studies reporting measures of association between intake of fruits and vegetables and diabetes
| Author, year (reference) . | Population . | n . | Age (years), sex . | Exposure measure . | Adjustments . | Follow-up (years) . | Case ascertainment . | Events . | Association (95% CI) . | Notes . |
|---|---|---|---|---|---|---|---|---|---|---|
| Colditz et al., 1992 (5) | U.S. nurses | 84,360 | 34–59, F | 61-item FFQ | Age, BMI, weight change, alcohol energy | 6 | Follow-up questionnaire | 702 diabetes | HR Q5/1 0.76 (0.50–1.16) vegetable intake | Q5/1: ≥2.9 serving/day of vegetables vs. <1.2 |
| Feskens et al., 1995 (6) | Finnish and Dutch | 338 | 70–89, M | Cross-check diet history | Age, cohort, BMI, past BMI, past energy intake | 30 | OGTT | 71 IGT, 26 diabetes | Inverse association of 2-h postload glucose and intake of vegetables and legumes | Multivariate regression predicting 2-h postload glucose |
| Meyer et al., 2000 (7) | Postmenopausal Iowa women | 35,988 | 55–69, F | 127-item FFQ | Age, smoking, total energy, BMI, alcohol, WHR, education, physical activity | 6 | Iowa death register, biennial questionnaire, and NDI | 1,141 diabetes | HR Q5/1 fruit + vegetables 1.05 (0.84–1.31), HR vegetables 1.07 (0.86–1.32), HR fruit 1.14 (0.93–1.39) | Q5/1: >51 servings/week fruit + vegetables vs. <23, Q5/1: >33.5 servings/week vegetable vs. <14, Q 5/1: >19 servings/week fruit vs. <6.25 |
| Ford and Mokdad, 2001 (9) | NHEFS | 9,665 | 25–74, M + F | Single 24-h dietary recall | Age, race, sex, smoking, BMI, alcohol, SBP, lipids, HTN, physical activity | 19 | Follow-up questionnaire, hospital records, death certificates | 1,018 diabetes | HR fruit + vegetable intake 0.73 (0.54–0.98) | ≥5 times/day vs. 0 times/day, no portion size included; also identified a sex interaction |
| Liu et al., 2004 (9) | WHS | 38,018 | ≥45, F | 131-item FFQ | Age, smoking, total energy, alcohol, BMI, physical activity, HTN, hyperlipidemia, FH | 8.8 | Follow-up questionnaire | 1,614 diabetes | HR Q5/1 fruit + vegetables 1.04 (0.87–1.25), HR fruit 0.97 (0.82–1.23), HR vegetables 1.03 (0.86–1.23) | Q5/1: median 10.1 servings/day fruit + vegetables vs. 2.5, Q5/1: median 3.9 servings/day fruit vs. 0.62 |
| Montonen et al., 2005 (10) | Finnish | 4,304 | 40–69, M + F | Diet history | Age, sex, smoking, total energy, BMI, FH, geographic area | 23 | Finnish Social Insurance Institution's national database | 383 diabetes | HR Q4/1 vegetables 0.77 (0.57–1.03), HR green vegetables 0.69 (0.50–0.93), HR fruit 0.82 (0.61–1.11), HR berries 0.63 (0.47–0.85) | Q4/1: >130 g/day vegetables vs. <42, Q4/1: >43 g/day green vegetables vs. <11, Q4/1: >138 g/day fruit vs. <20 |
| Author, year (reference) . | Population . | n . | Age (years), sex . | Exposure measure . | Adjustments . | Follow-up (years) . | Case ascertainment . | Events . | Association (95% CI) . | Notes . |
|---|---|---|---|---|---|---|---|---|---|---|
| Colditz et al., 1992 (5) | U.S. nurses | 84,360 | 34–59, F | 61-item FFQ | Age, BMI, weight change, alcohol energy | 6 | Follow-up questionnaire | 702 diabetes | HR Q5/1 0.76 (0.50–1.16) vegetable intake | Q5/1: ≥2.9 serving/day of vegetables vs. <1.2 |
| Feskens et al., 1995 (6) | Finnish and Dutch | 338 | 70–89, M | Cross-check diet history | Age, cohort, BMI, past BMI, past energy intake | 30 | OGTT | 71 IGT, 26 diabetes | Inverse association of 2-h postload glucose and intake of vegetables and legumes | Multivariate regression predicting 2-h postload glucose |
| Meyer et al., 2000 (7) | Postmenopausal Iowa women | 35,988 | 55–69, F | 127-item FFQ | Age, smoking, total energy, BMI, alcohol, WHR, education, physical activity | 6 | Iowa death register, biennial questionnaire, and NDI | 1,141 diabetes | HR Q5/1 fruit + vegetables 1.05 (0.84–1.31), HR vegetables 1.07 (0.86–1.32), HR fruit 1.14 (0.93–1.39) | Q5/1: >51 servings/week fruit + vegetables vs. <23, Q5/1: >33.5 servings/week vegetable vs. <14, Q 5/1: >19 servings/week fruit vs. <6.25 |
| Ford and Mokdad, 2001 (9) | NHEFS | 9,665 | 25–74, M + F | Single 24-h dietary recall | Age, race, sex, smoking, BMI, alcohol, SBP, lipids, HTN, physical activity | 19 | Follow-up questionnaire, hospital records, death certificates | 1,018 diabetes | HR fruit + vegetable intake 0.73 (0.54–0.98) | ≥5 times/day vs. 0 times/day, no portion size included; also identified a sex interaction |
| Liu et al., 2004 (9) | WHS | 38,018 | ≥45, F | 131-item FFQ | Age, smoking, total energy, alcohol, BMI, physical activity, HTN, hyperlipidemia, FH | 8.8 | Follow-up questionnaire | 1,614 diabetes | HR Q5/1 fruit + vegetables 1.04 (0.87–1.25), HR fruit 0.97 (0.82–1.23), HR vegetables 1.03 (0.86–1.23) | Q5/1: median 10.1 servings/day fruit + vegetables vs. 2.5, Q5/1: median 3.9 servings/day fruit vs. 0.62 |
| Montonen et al., 2005 (10) | Finnish | 4,304 | 40–69, M + F | Diet history | Age, sex, smoking, total energy, BMI, FH, geographic area | 23 | Finnish Social Insurance Institution's national database | 383 diabetes | HR Q4/1 vegetables 0.77 (0.57–1.03), HR green vegetables 0.69 (0.50–0.93), HR fruit 0.82 (0.61–1.11), HR berries 0.63 (0.47–0.85) | Q4/1: >130 g/day vegetables vs. <42, Q4/1: >43 g/day green vegetables vs. <11, Q4/1: >138 g/day fruit vs. <20 |
F, female; FH, family history of diabetes; HTN, hypertension; IGT, impaired glucose tolerance; M, male; NHEFS, First National Health and Nutrition Examination Study Epidemiologic Follow-up Study; OGTT, oral glucose tolerance test; Q, quantile; SBP, systolic blood pressure; WHR, waist-to-hip ratio; WHS, Women's Health Study.