Table 2.
Pooled RRs of cancer risk for the highest versus lowest categories of total soy product, tofu and soy milk consumption.
| Characteristic | Studies (n) | RR (95% CI) | Heterogeneity | p-Difference |
|---|---|---|---|---|
| Total soy foods | ||||
| All studies | 28 | 0.69 (0.6, 0.8) | I2 = 82.7%, p < 0.001 | |
| Study design | ||||
| Case–control study | 18 | 0.56 (0.46, 0.69) | I2 = 69.8%, p < 0.001 | p = 0.004 |
| Cohort study | 10 | 0.90 (0.80, 1.01) | I2 = 66.8%, p = 0.001 | |
| Sex | ||||
| Male | 12 | 0.86 (0.74, 1.00) | I2 = 59.0%, p = 0.005 | p = 0.454 |
| Female | 20 | 0.76 (0.65, 0.89) | I2 = 78.3%, p < 0.001 | |
| Cancer type | ||||
| Gastrointestinal cancer | 10 | 0.74 (0.61, 0.89) | I2 = 68.7%, p = 0.001 | |
| Stomach | 5 | 0.63 (0.41, 0.97) | I2 = 81.5%, p < 0.001 | |
| Colorectal | 5 | 0.77 (0.65, 0.91) | I2 = 43.1%, p = 0.134 | |
| Gynecological cancer | 8 | 0.71 (0.54, 0.92) | I2 = 83.3%, p < 0.001 | p = 0.902 |
| Breast | 5 | 0.79 (0.60, 1.03) | I2 = 73.9%, p = 0.004 | |
| Ovarian | 1 | 0.29 (0.20, 0.42) | ||
| Cervical | 1 | 0.80 (0.61, 1.05) | ||
| Endometrial | 1 | 0.84 (0.59, 1.20) | ||
| Prostate cancer | 3 | 0.47 (0.31, 0.71) | I2 = 0%, p = 0.566 | p = 0.163 |
| Lung cancer | 2 | 0.67 (0.52, 0.86) | I2 = 0%, p = 0.564 | p = 0.608 |
| Bladder cancer | 1 | 0.75 (0.40, 1.41) | p = 0.983 | |
| Upper aerodigestive tract cancer | 1 | 0.33 (0.22, 0.49) | p = 0.062 | |
| Multiple myeloma | 1 | 0.10 (0.01, 0.97) | p = 0.257 | |
| Liver cancer | 1 | 1.10 (0.82, 1.47) | p = 0.296 | |
| Geographic location | ||||
| Japan | 10 | 0.89 (0.77, 1.02) | I2 = 65.7%, p = 0.002 | |
| China | 11 | 0.48 (0.34, 0.69) | I2 = 80.8%, p < 0.001 | p = 0.022 |
| Korea | 4 | 0.75 (0.61, 0.92) | I2 = 40.0%, p = 0.172 | p = 0.241 |
| Singapore | 2 | 0.73 (0.57, 0.94) | I2 = 14.1%, p = 0.281 | p = 0.285 |
| USA | 1 | 0.73 (0.51, 1.04) | p = 0.488 | |
| Europe | 1 | 0.52 (0.60, 0.80) | p = 0.160 | |
| Tofu | ||||
| All studies | 26 | 0.78 (0.70, 0.86) | I2 = 47.9%, p = 0.004 | |
| Study design | ||||
| Case–control study | 19 | 0.72 (0.63, 0.83) | I2 = 58.9%, p = 0.001 | p = 0.186 |
| Cohort study | 7 | 0.89 (0.78, 1.01) | I2 = 0%, p = 0.901 | |
| Sex | ||||
| Male | 7 | 0.83 (0.71, 0.98) | I2 = 20.3%, p = 0.268 | p = 0974 |
| Female | 19 | 0.82 (0.74, 0.91) | I2 = 34.7%, p = 0.069 | |
| Cancer type | ||||
| Gastrointestinal cancer | 5 | 0.67 (0.47, 0.96) | I2 = 76.0%, p = 0.002 | |
| Stomach | 4 | 0.56 (0.34, 0.93) | I2 = 79.7%, p = 0.002 | |
| Colorectal | 1 | 0.95 (0.73, 1.24) | ||
| Gynecological cancer | 12 | 0.76 (0.66, 0.87) | I2 = 48.7%, p = 0.029 | p = 0898 |
| Breast | 7 | 0.79 (0.66, 0.94) | I2 = 61.6%, p = 0.016 | |
| Endometrial | 3 | 0.77 (0.61, 0.97) | I2 = 0%, p = 0.785 | |
| Cervical | 1 | 0.62 (0.34, 1.14) | ||
| Ovarian | 1 | 0.57 (0.40, 0.81) | ||
| Prostate cancer | 2 | 0.70 (0.43, 1.15) | I2 = 26.0%, p = 0.245 | p = 0.917 |
| Liver cancer | 2 | 0.96 (0.72, 1.28) | I2 = 0%, p = 0.931 | p = 0.465 |
| Lung cancer | 1 | 0.96 (0.70, 1.32) | p = 0.542 | |
| Non-Hodgkin lymphoma | 1 | 1.35 (0.75, 2.44) | p = 0.341 | |
| Upper aerodigestive tract cancer | 1 | 0.49 (0.07, 3.48) | p = 0.874 | |
| Leukemia | 1 | 0.55 (0.34, 0.89) | p = 0.898 | |
| Geographic location | ||||
| Japan | 9 | 0.87 (0.76, 1.00) | I2 = 20.0%, p = 0.265 | |
| USA | 7 | 0.82 (0.71, 0.95) | I2 = 0%, p = 0.863 | p = 0.546 |
| Korea | 6 | 0.58 (0.40, 0.85) | I2 = 80.7%, p < 0.001 | p = 0.169 |
| China | 3 | 0.63 (0.50, 0.80) | I2 = 1.4%, p = 0.363 | p = 0.049 |
| Europe | 1 | 0.89 (0.74, 1.08) | p = 0.876 | |
| Soy milk | ||||
| All studies | 11 | 0.75 (0.60, 0.93) | I2 = 80.6%, p < 0.001 | |
| Study design | ||||
| Case–control study | 8 | 0.65 (0.52, 0.80) | I2 = 72.4%, p = 0.001 | p = 0.031 |
| Cohort study | 3 | 1.10 (0.76, 1.58) | I2 = 72.9%, p = 0.025 | |
| Sex | ||||
| Male | 3 | 0.72 (0.37, 1.41) | I2 = 90.4%, p < 0.001 | p = 0.699 |
| Female | 10 | 0.81 (0.65, 1.00) | I2 = 71.8%, p < 0.001 | |
| Cancer type | ||||
| Gastrointestinal cancer | 2 | 0.58 (0.47, 0.72) | I2 = 0%, p = 0.759 | |
| Stomach | 1 | 0.61 (0.43, 0.87) | ||
| Colorectal | 1 | 0.57 (0.44, 0.73) | ||
| Gynecological cancer | 6 | 0.79 (0.58, 1.06) | I2 = 82.2%, p < 0.001 | p = 0.359 |
| Breast | 3 | 0.91 (0.60, 1.38) | I2 = 80.9%, p = 0.005 | |
| Endometrial | 2 | 0.86 (0.69, 1.06) | I2 = 0%, p = 0.964 | |
| Ovarian | 1 | 0.43 (0.31, 0.6) | ||
| Upper aerodigestive tract cancer | 1 | 0.48 (0.31, 0.74) | p = 0.569 | |
| Lung cancer | 1 | 0.75 (0.53, 1.06) | p = 0.150 | |
| Liver cancer | 1 | 1.31 (0.95, 1.80) | p = 0.433 | |
| Geographic location | ||||
| Japan | 2 | 1.32 (1.05, 1.66) | I2 = 0%, p = 0.922 | |
| China | 3 | 0.57 (0.35, 0.93) | I2 = 84.7%, p = 0.001 | p = 0.068 |
| USA | 2 | 0.69 (0.47, 1.02) | I2 = 41.8%, p = 0.190 | p = 0.074 |
| Korea | 2 | 0.58 (0.47, 0.72) | I2 = 0%, p = 0.759 | p = 0.035 |
| Singapore | 1 | 0.75 (0.53, 1.06) | p = 0.225 | |
| Europe | 1 | 0.94 (0.75, 1.18) | p = 0.284 |