TABLE 2.
Quality of evidence of associations between dietary factors and esophageal cancer risk1
| First author (reference) | Dietary factor | Comparison | MA metric | Random effect size (95% CI) | P | I 2, % | Largest study, 95% CI | 95% prediction interval | Egger's P value | Excess significance test | Evidence class2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| O/E | P | |||||||||||
| Miyazaki, 2017 (23) | Alcohol cessation | Quit drinking vs. current drinking | RR | 0.57 (0.27, 1.17) | 0.12 | 36.1 | 0.33–2.6 | NE | NA | 1/0.7 | 0.58 | No |
| Choi, 2013 (24) | Red meat | Highest vs. lowest | RR | 1.28 (1.07, 1.53) | 0.008 | 35.5 | 1.11–1.69 | 0.93–1.75 | NA | 2/1.6 | 0.51 | IV |
| Han, 2013 (25) | Fish | Highest vs. lowest | RR | 0.84 (0.65, 1.09) | 0.19 | 36.1 | 0.72–1.26 | 0.36–1.99 | NA | 1/0.8 | 0.59 | No |
| Salehi, 2013 (26) | Total meat | Highest vs. lowest | RR | 0.92 (0.58, 1.44) | 0.7 | 63.5 | 0.62–0.86 | 0.006–148.5 | NA | 1/1 | NP5 | No |
| White meat | Highest vs lowest | RR | 0.8 (0.64, 1.0) | 0.051 | 0 | 0.65–1.09 | NE | NA | 0/0.6 | NP | No | |
| Jiang, 2016 (27) | Poultry | Highest vs. lowest | RR | 1.03 (0.62, 1.73) | 0.91 | 66.2 | 0.72–1.26 | 0.001–385 | NA | 0/1 | NP | No |
| Vingeliene, 2016 (28) | Citrus fruits | Per 100-g/d increment | RR | 0.9 (0.79, 1.02) | 0.1 | 0 | 0.78–1.25 | 0.76–1.05 | NA | 1/0.7 | 0.53 | No |
| Yan, 2018 (29) | Pickled vegetables | Highest vs. lowest | OR | 1.04 (0.93, 1.15) | 0.53 | 0 | 0.92–1.15 | NE | NA | 0/0.1 | NP | No |
| Zhang, 2020 (30) | Whole grains | Highest vs. lowest | RR | 0.54 (0.38, 0.76) | 0.0004 | 0 | 0.34–0.82 | NE | NA | 2/1.4 | 0.46 | IV |
| Sakai, 2021 (31) | Fruits | Highest vs. lowest | RR | 0.75 (0.6, 0.92) | 0.01 | 80.9 | 0.7–0.91 | 0.39–1.45 | NA | 3/2.8 | 0.62 | IV |
| Vingeliene, 2017 (32) | Vegetables | Per 100-g/d increment | RR | 0.93 (0.86, 1.01) | 0.07 | 33.7 | 0.93–1.1 | 0.77–1.13 | NA | 1/1.8 | NP | No |
| Green leafy vegetables | Per 50-g/d increment | RR | 0.86 (0.78, 0.95) | 0.004 | 0 | 0.74–0.97 | 0.74–0.99 | NA | 1/1.3 | NP | IV | |
| Processed meat | Per 50-g/d increment | RR | 1.35 (1.06, 1.74) | 0.02 | 56.5 | 0.87–1.36 | 0.66–2.78 | NA | 2/2.3 | 0.75 | IV | |
| Coffee | Per 1-cup/d increment | RR | 0.99 (0.94, 1.03) | 0.56 | 30.6 | 0.91–1.05 | 0.92–1.06 | NA | 0/0.3 | NP | No | |
| Alcohol | Per 10-g/d increment | RR | 1.21 (1.13, 1.31) | <10–6 | 95.4 | 1.35–1.6 | 0.93–1.57 | 0.004 | 8/7.8 | 0.61 | II | |
| Li, 2016 (33) | Milk | Highest vs. lowest | RR | 0.8 (0.48, 1.33) | 0.39 | 62.7 | 0.71–1.41 | NE | NA | 0/0.7 | NP | No |
| Yi, 2020 (34) | Green tea | Highest vs. lowest | RR | 0.75 (0.61, 0.93) | 0.007 | 23.4 | 0.51–1.08 | 0.48–1.19 | NA | 1/1.1 | 0.73 | IV |
| He, 2017 (35) | Total fat | Highest vs. lowest | RR | 0.93 (0.64, 1.34) | 0.69 | 0 | 0.6–1.42 | NE | NA | 0/0.1 | NP | No |
| Saturated fat | Highest vs. lowest | RR | 1.17 (0.88, 1.56) | 0.27 | 0 | 0.91–1.78 | NE | NA | 0/0.2 | NP | No | |
| Polyunsaturated fat | Highest vs. lowest | RR | 0.97 (0.77, 1.22) | 0.79 | 0 | 0.7–1.19 | NE | NA | 0/0.1 | NP | No | |
| Monounsaturated fat | Highest vs. lowest | RR | 1.28 (0.96, 1.71) | 0.1 | 0 | 0.84–1.66 | NE | NA | 0/0.4 | NP | No | |
| Hong, 2016 (36) | Selenium | Highest vs. lowest | RR | 0.99 (0.55, 1.78) | 0.96 | 32.6 | 0.41–1.4 | NE | NA | 0/0.2 | NP | No |
| Xie, 2016 (37) | Nitrate | Highest vs lowest | RR | 1.06 (0.82, 1.39) | 0.65 | 0 | 0.75–1.61 | 0.73–1.55 | NA | 0/0.3 | NP | No |
| Nitrite | Highest vs. lowest | RR | 1.1 (0.92, 1.32) | 0.31 | 11.5 | 0.84–1.68 | 0.87–1.39 | NA | 0/0.5 | NP | No | |
| Li, 2017 (38) | Calcium | Highest vs. lowest | OR | 0.67 (0.54, 0.84) | 0.0004 | 23.7 | 0.56–1.26 | 0.16–2.84 | NA | 2/1.6 | 0.53 | III |
| Liu, 2017 (39) | Folate | Highest vs. lowest | OR | 0.85 (0.64, 1.14) | 0.27 | 48.6 | 0.61–1.08 | 0.05–15.28 | NA | 0/0.7 | NP | No |
| Ma, 2018 (40) | Iron | Highest vs. lowest | OR | 1.15 (0.89, 1.47) | 0.28 | 56.7 | 0.99–2.17 | 0.56–2.34 | NA | 1/1.6 | NP | No |
| Copper | Per 1-mg/d increment | OR | 1.02 (0.64, 1.62) | 0.93 | 95.7 | 1.2–1.37 | NE | NA | 2/1.4 | 0.5 | No | |
| Zinc | Per 5-mg/d increment | OR | 0.78 (0.71, 0.86) | <0.0001 | 0 | 0.7–0.89 | NE | NA | 2/1.7 | 0.71 | IV | |
| Ma, 2018 (41) | Vitamin B-6 | Highest vs. lowest | OR | 1.02 (0.7, 1.48) | 0.93 | 51.8 | 0.76–1.32 | 0.25–4.2 | NA | 1/0.7 | 0.58 | No |
| Vitamin B-12 | Highest vs. lowest | OR | 1.02 (0.82, 1.25) | 0.89 | 0 | 0.8–1.35 | 0.64–1.61 | NA | 0/0.2 | NP | No | |
| Banda 2020 (42) | Sodium | Highest vs. lowest | OR | 1.04 (0.99, 1.08) | 0.06 | 0 | 0.98–1.07 | 0.81–1.33 | NA | 0/0.6 | NP | No |
MA, meta-analyses; NA, not applicable because number of studies (<10) too small to test for small study effects; NE, not estimated because number of studies (<3) too small to test for 95% prediction interval; NP, not pertinent because estimated number is larger than observed; O/E, observed/expected number of studies with significant results.
Evidence class—class I (convincing): statistical significance at P < 10−6, >1000 cases, the largest component study with a significant effect (P < 0.05), the 95% prediction interval excluded the null, no large heterogeneity (I2 < 50%), no small-study effects (P > 0.10), and no excess significance bias (P > 0.10); class II (highly suggestive): significance at P < 10−6, >1000 cases, the largest component study with a significant effect (P < 0.05); class III (suggestive): statistical significance at P < 10−3, >1000 cases; class IV (weak): the remaining significant associations at P < 0.05; No, nonsignificant association P ≥ 0.05.