Table 1. Characteristics of studies included in the meta-analysis.
| Study/year | Country | Study design | Sample size | Source of controls | Median/mean age (years) | % Female | Dietary assessment | DII score comparison | HR or OR (95% CI) | Adjustment for covariates | Follow-up (years) | NOS stars |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Shivappa et al. 2014 [12] | USA | Prospective cohort study | 37,403 (1,329 colon and 325 rectal cancer) | — | 55–69 | 100% | FFQ (121 items) | Quintile 5 vs. 1; > 1·10 vs.< −2.75 | Colorectal cancer 1.20 (1.01–1.43); Colon cancer; 1.19 (0.98–1.45) Rectal cancer; 1.21 (0.81–1.79) | Age, BMI, smoking status, pack-years of smoking, HRT use, education, DM, and total energy intake | 19.6 | 7 |
| Shivappa et al. 2015 [13] | Italy | Case–control | Case:1953; Con:4154 | Hospital based | Case 62; Con:58 | Case:42.4% Con: 50% | FFQ-derived dietary data | Quintile 5 vs. 1; > 1·22 vs.≤ −1.05 | Colorectal cancer 1.55 (1.29–1.85) 1.90 (1.47–2.45) M 1.27 (1.00–1.65) F Colon cancer; 1.39 (1.13–1.71) Rectal cancer; 1.47 (1.14–1.90) | Age, sex, study centre, education,, BMI, alcohol drinking, PA, history of CRC, and energy intake | — | 7 |
| Tabung et al. 2015 [14] | USA | Prospective cohort study | 152,536 (1,559 colon and 361 rectal) | — | 50–79 | 100% | FFQ (122 items) | Quintile 5 vs. 1; > 1·953 vs. < −3.14 | Colorectal cancer 1.22 (1.05–1.43); Colon cancer; 1.23 (1.03–1.46) Rectal cancer; 1.20 (0.84–1.72) | Age, total energy intake, BMI, race/ethnicity, PA, education, smoking, family history of CRC, hypertension, DM, arthritis, history of colonoscopy or occult blood tests, NSAID use, estrogen and/or progesterone use, and different trial arms | 11.3 | 7 |
| Wirth et al. 2015 [15] | USA | Prospective cohort study | 489,422 (6,944 cases) | — | 62.0 ± 5.4 | 40.3% | FFQ (124 items) | Quartile 4 vs. 1; > 3.25 vs. < −0.59 | Colorectal cancer 1.40 (1.28–1.53) 1.44 (1.29–1.61) M 1.12 (0.95–1.31) F Rectal cancer; 0.91 (0.67–1.25) | Age, smoking status, BMI, self-reported diabetes, energy intake, PA, marital status, education. | 9.1 | 8 |
| Zamora-Ros et al.2015 [16] | France | Case–control | Case:424; Con:401 | Hospital based | 65.8 ± 12 | 56% | Dietary history questionnaire | Quartile 4 vs. 1; > 3.05 vs.< −0.73 | Colorectal cancer 1.65 (1.05–2.60); Colon cancer; 2.24 (1.33–3.77); Rectal cancer; 1.12 (0.61–2.06) | Age, sex, total energy intake, BMI, first-degree family history of CRC, PA, tobacco consumption, and medication use | — | 7 |
| Cho et al. 2016 [17] | Korea | Case–control | Case:923; Con:1846 | Health check-up | Case:56.6; Con: 56.1 | 32.3% | Semi-quantitative FFQ (116 items) | Tertile 3 vs. 1; ≥ 1.76 vs. ≤ -0.28 | Colorectal cancer 2.16 (1.71–2.73) 1.72 (1.30–2.28) M 2.50 (1.64–3.82) F Colon cancer; 2.05 (1.53–2.74) Rectal cancer; 2.23 (1.66–3.00) | Age, sex, BMI, education, family history of CRC, PA, and total calorie intake. | — | 7 |
| Shivappa et al. 2017 [18] | Jordan | Case–control | Case:153; Con:202 | Hospital personnel, outpatients, and visitors | Case:51.6; Con: 53.8 | Case:52.3% Con: 49% | FFQ (111 items) | Tertile 3 vs. 1; > 2.18 vs. ≤ −1.38 | Colorectal cancer 2.13 (1.23– 3.72) | Age, sex, education, PA, BMI, smoking, family history of CRC. | — | 6 |
| Harmon et al. 2017 [19] | USA | Prospective cohort study | 190,963 (3372 colon, 981 rectum, and 35 with both) | — | 45–75 | 55% | FFQ(180 items) | Quartile 4 vs. 1; > −0.52 vs.< −3.66 | Colorectal cancer 1.21 (1.11–1.32) 1.28 (1.13–1.45) M 1.16 (1.02–1.33) F Colon cancer; 1.20 (1.09–1.33) | Age, sex, race, DM, asthma, heart attack, use of supplements, smoking, family history of colon cancer, education, use of HRT or aspirin | 20 | 8 |
Abbreviations: M, male; F, female; OR, odds ratio; HR, hazard ratio; CI, confidence interval; BMI, body mass index; CRC, colorectal cancer; DM, diabete mellitus; SBP, systolic blood pressure; DBP, diastolic blood pressure; DII, dietary inflammatory index; FFQ, food frequency questionnaire; HRT, hormone replacement therapy; PA, physical activity.