Summary of the Studies Included in the Meta-Analysis
| Cancer site | No. of Studies | No. of Cases | Study design | Pooled RR [and 95% CI] associated with alcohol intake of | Gender effect (p) | Heterogeneity test (p) | |||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
| Cohort | Case-control | 25 g/day [range] | 50 g/day [range] | 100 g/day [range] | |||||
| Oral cavity | 26 | 7,954 | 1 | 25 | 1.73 [1.67–1.78] | 2.77 [2.67–2.95] | 5.75 [5.22–6.34] | n.s. | < 0.01 |
| Esophagusa | 28 | 7,239 | 1 | 27 | 1.51 [1.48–1.55] | 2.21 [2.11–2.31] | 4.23 [3.91–4.59] | < 0.05 | < 0.01 |
| Males | 18 | 3,310 | 1 | 17 | 1.43 [1.38–1.48] | 1.98 [1.87–2.11] | 3.49 [3.14–3.89] | - | < 0.01 |
| Females | 5 | 304 | 0 | 5 | 1.52 [1.42–1.63] | 2.24 [1.95–2.58] | 4.45 [3.37–5.87] | - | |
| Stomach | 16 | 4,518 | 2 | 14 | 1.07 [1.04–1.10] | 1.15 [1.09–1.22] | 1.32 [1.18–1.49] | n.s. | < 0.01 |
| Small intestine | 2 | 415 | 0 | 2 | - - |
- - |
- - |
- | - |
| Colon | 17 | 5,948 | 4 | 13 | 1.14 [1.07–1.21] | 1.21 [1.11–1.32] | 1.32 [1.16–1.49] | n.s. | < 0.01 |
| Rectum | 16 | 3,872 | 3 | 13 | 1.11 [1.03–1.20] | 1.17 [1.06–1.30] | 1.32 [1.16–1.51] | n.s. | < 0.01 |
| Livera | 19 | 1,961 | 3 | 16 | 1.20 [1.13–1.27] | 1.41 [1.26–1.56] | 1.83 [1.53–2.19] | < 0.05 | < 0.01 |
| Males | 10 | 949 | 2 | 8 | 1.28 [1.13–1.45] | 1.51 [1.27–2.10] | 1.62 [1.18–2.24] | - | < 0.01 |
| Females | 3 | 231 | 1 | 2 | 1.97 [1.30–3.00] | 3.57 [1.56–8.21] | 9.15 [1.73–48.41] | - | |
| Gallbladder | 2 | 81 | 1 | 1 | - - |
- - |
- - |
- | - |
| Pancreas | 17 | 2,524 | 4 | 13 | 0.98 [0.90–1.05] | 1.05 [0.93–1.18] | 1.18 [0.94–1.49] | n.s. | < 0.01 |
| Larynx | 20 | 3,759 | 0 | 20 | 1.35 [1.31–1.40] | 1.83 [1.72–1.95] | 3.24 [2.89–3.65] | n.s. | < 0.01 |
| Lung | 6 | 2,314 | 3 | 3 | 1.02 [1.00–1.04] | 1.04 [1.00–1.08] | 1.08 [1.00–1.18] | n.s. | < 0.01 |
| Melanoma | 2 | 708 | 0 | 2 | - - |
- - |
- - |
- | - |
| Female Breast | 49 | 44,033 | 12 | 37 | 1.31 [1.27–1.36] | 1.67 [1.56–1.78] | 2.71 [2.33–3.08] | - | < 0.01 |
| Cervix | 1 | 242 | - | 1 | - - |
- - |
- - |
- | - |
| Endometrium | 6 | 2,473 | 2 | 4 | 1.05 [0.88–1.24] | 1.09 [0.78–1.54] | 1.20 [0.60–2.37] | - | <0.01 |
| Ovary | 5 | 1,651 | - | 5 | 1.11 [1.00–1.24] | 1.23 [1.01–1.54] | 1.53 [1.03–2.32] | - | n.s. |
| Prostate | 11 | 4,094 | 4 | 7 | 1.05 [1.00–1.08] | 1.09 [1.02–1.17] | 1.19 [1.03–1.37] | - | n.s. |
| Bladder | 11 | 5,997 | 2 | 9 | 1.04 [0.99–1.09] | 1.08 [0.98–1.89] | 1.17 [0.97–1.41] | n.s. | n.s. |
| Kidney | 2 | 921 | 0 | 2 | - - |
- - |
- - |
- | - |
| All sites combined | 8 | 14,495 | 6 | 2 | 1.01 [0.90–1.05] | 1.22 [1.11–1.27] | 1.91 [1.77–2.06] | n.s. | < 0.01 |
| Total* | 229 | 115,199 | 46 | 183 | - | - | - | - | - |
NOTE: Table gives the pooled relative risks (RR) and corresponding 95-percent confidence intervals (95% CI) for three alcohol consumption levels. Those levels specified corresponded to approximately two, four, and eight standard drinks per day, respectively. The effects of gender and differences (i.e., heterogeneity) in the study results for various cancer sites also are presented. The RR indicates the strength of the relationship between alcohol consumption and a given type of cancer. A RR greater than 1.0 means that alcohol consumption at the level indicated increased the risk for that type of cancer. The greater the value over 1.0, the greater the risk. The 95-percent confidence intervals indicate the range of RR that is 95 percent likely to show a true RR. A statistically significant heterogeneity level indicates that results varied greatly among the various studies analyzed and that, therefore, an overall summary of average effect across studies must be interpreted with caution.
The number of individual studies does not add up to the total shown because several studies examined more than one type of cancer.
For this type of cancer, the RRs associated with alcohol consumption for males and females are listed separately because a statistically significant difference existed between the two sexes.
n.s. = not significant