Table 4.
Diabetes medication classes and colorectal cancer risk.
| RR 95% CI | P value for effect | P value for Egger's test | I2 (%) | P value for difference between subgroups | ||
|---|---|---|---|---|---|---|
| Biguanides | ||||||
| All studies (n = 18) | 0.85 (0.78–0.92) | < 0.001 | 0.11 | 97.33 | – | |
| Study design | Cohort (n = 11) | 0.80 (0.71–0.92) | < 0.001 | – | 98.28 | 0.15 |
| Case–control (n = 7) | 0.91 (0.82–1.00) | 0.05 | – | 88.63 | ||
| Risk of bias | Low risk (n = 12) | 0.86 (0.78–0.95) | < 0.001 | – | 98.12 | 0.51 |
| Moderate risk (n = 6) | 0.81 (0.70–0.94) | < 0.001 | – | 78.74 | ||
| Study population | Asian (n = 7) | 0.76 (0.62–0.94) | 0.01 | – | 98.49 | 0.13 |
| Western (n = 11) | 0.90 (0.84–0.96) | < 0.001 | – | 80.05 | ||
| Incretin-based medicines | ||||||
| All studies (n = 5) | 1.03 (0.99–1.06) | 0.11 | 0.34 | < 0.001 | – | |
| Study design | Cohort (n = 4) | 1.01 (0.97–1.06) | 0.58 | – | < 0.001 | 0.34 |
| Case–control (n = 1) | 1.04 (1.00–1.09) | 0.07 | – | – | ||
| Risk of bias | Low risk (n = 4) | 1.03 (1.00–1.06) | 0.10 | – | < 0.001 | 0.36 |
| Moderate risk (n = 1) | 0.06 (0.01–27.59) | 0.36 | – | – | ||
| Study population | Asian (n = 2) | 1.03 (0.99–1.06) | 0.11 | – | < 0.001 | 0.78 |
| Western (n = 3) | 0.99 (0.76–1.28) | 0.93 | – | 19.23 | ||
| Alpha-glucosidase inhibitors | ||||||
| All studies (n = 2) | 1.13 (0.97–1.31) | 0.11 | 0.43 | 93.78 | – | |
| Study design | Cohort (n = 1) | 1.21 (1.17–1.26) | < 0.001 | – | – | < 0.001 |
| Case–control (n = 1) | 1.04 (0.98–1.11) | 0.18 | – | – | ||
| Risk of bias | Low risk (n = 2) | 1.13 (0.97–1.31) | 0.11 | – | 93.78 | – |
| Study population | Asian (n = 2) | 1.13 (0.97–1.31) | 0.11 | – | 93.78 | – |
| Insulin secretagogues | ||||||
| All studies (n = 9) | 1.07 (0.95–1.21) | 0.25 | 0.33 | 94.82 | – | |
| Study design | Cohort (n = 4) | 1.21 (1.04–1.40) | 0.01 | – | 85.06 | 0.04 |
| Case–control (n = 5) | 1.03 (0.97–1.08) | 0.38 | – | 40.29 | ||
| Risk of bias | Low risk (n = 6) | 1.10 (0.97–1.26) | 0.15 | – | 95.75 | 0.22 |
| Moderate risk (n = 3) | 1.00 (0.94–1.07) | 0.96 | – | < 0.001 | ||
| Study population | Asian (n = 4) | 1.13 (0.97–1.32) | 0.13 | – | 97.10 | 0.18 |
| Western (n = 5) | 1.01 (0.95–1.07) | 0.83 | – | < 0.001 | ||
| Thiazolidinediones | ||||||
| All studies (n = 11) | 1.00 (0.96–1.03) | 0.86 | 0.19 | 55.38 | – | |
| Study design | RCT (n = 1) | 0.74 (0.43–1.27) | 0.27 | – | – | 0.50 |
| Cohort (n = 4) | 0.94 (0.76–1.17) | 0.59 | – | 63.45 | ||
| Case–control (n = 6) | 0.99 (0.96–1.03) | 0.72 | – | 52.91 | ||
| Risk of bias | Low risk (n = 8) | 1.01 (0.96–1.06) | 0.74 | – | 60.70 | 0.37 |
| Moderate risk (n = 2) | 0.97 (0.91–1.04) | 0.38 | – | 62.65 | ||
| High risk (n = 1) | 0.74 (0.43–1.27) | 0.27 | – | – | ||
| Study population | Asian (n = 7) | 1.01 (0.98–1.04) | 0.54 | – | 46.96 | 0.08 |
| Western (n = 4) | 0.87 (0.73–1.03) | 0.10 | – | 37.73 | ||
| Insulins | ||||||
| All studies (n = 12) | 1.04 (0.94–1.16) | 0.42 | 0.39 | 95.19 | – | |
| Study design | Cohort (n = 8) | 1.03 (0.86–1.22) | 0.77 | – | 96.81 | 0.75 |
| Case–control (n = 4) | 1.06 (0.97–1.15) | 0.19 | – | 59.90 | ||
| Risk of bias | Low risk (n = 8) | 1.03 (0.91–1.17) | 0.63 | – | 96.58 | 0.54 |
| Moderate risk (n = 4) | 1.15 (0.83–1.61) | 0.41 | – | 87.21 | ||
| Study population | Asian (n = 5) | 1.22 (1.01–1.46) | 0.04 | – | 94.15 | 0.09 |
| Western (n = 7) | 0.96 (0.79–1.18) | 0.72 | – | 95.68 | ||