Table 2.
Summary odds ratios/relative ratios and 95 % confidence intervals of colorectal, colon and rectal cancer for the highest v. the lowest/non coffee consumption in strata of geographic area, sex and/or type of controls and/or follow-up period
| Colorectal cancer | Colon cancer | Rectal cancer | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of comparisons/studies | No. of cases | OR/RR | 95 % CI | P* | I 2 (%)† | No. of comparisons/studies | No. of cases | OR/RR | 95 % CI | P* | I 2 (%)† | No. of comparisons/studies | No. of cases | OR/RR | 95 %CI | P* | I 2 (%)† | |
| Case–control studies | ||||||||||||||||||
| Overall effect | 37/25 | 15 522 | 0·85 | 0·75, 0·97 | <0·001 | 64 | 17/14 | 9191 | 0·79 | 0·67, 0·95 | <0·001 | 68 | 14/12 | 4996 | 0·95 | 0·79, 1·15 | 0·02 | 53 |
| Geographic area | ||||||||||||||||||
| America | 14/9 | 7847 | 0·90 | 0·77, 1·05 | 0·01 | 58 | 8/6 | 4970 | 0·88 | 0·70, 1·10 | 0·02 | 64 | 4/3 | 2347 | 1·00 | 0·85, 1·18 | 0·34 | 8 |
| Asia | 13/6 | 2062 | 0·75 | 0·49, 1·16 | <0·001 | 80 | 6/5 | 1534 | 0·77 | 0·46, 1·29 | <0·001 | 79 | 6/5 | 905 | 0·77 | 0·45, 1·33 | 0·002 | 77 |
| Europe | 10/10 | 5613 | 0·73 | 0·57, 0·93 | 0·009 | 59 | 3/3 | 2687 | 0·68 | 0·57, 0·81 | 0·33 | 9 | 4/4 | 1744 | 1·03 | 0·84, 1·25 | 0·40 | 0 |
| Type of controls | ||||||||||||||||||
| Hospital | 17/11 | 7768 | 0·78 | 0·63, 0·97 | <0·001 | 74 | 7/6 | 4273 | 0·80 | 0·60, 1·08 | 0·004 | 71 | 6/5 | 2614 | 0·97 | 0·71, 1·33 | 0·009 | 67 |
| Population | 20/14 | 7754 | 0·84 | 0·71, 1·00 | <0·001 | 60 | 10/8 | 4918 | 0·79 | 0·61, 1·01 | 0·002 | 69 | 8/7 | 2382 | 0·95 | 0·82, 1·10 | 0·25 | 24 |
| Sex | ||||||||||||||||||
| Males | 13/11 | 7046 | 0·96 | 0·73, 1·25 | <0·001 | 78 | 9/9 | 4033 | 0·86 | 0·62, 1·18 | <0·001 | 77 | 5/5 | 1754 | 1·13 | 0·69, 1·86 | <0·001 | 80 |
| Females | 12/10 | 5530 | 0·82 | 0·73, 0·93 | 0·11 | 35 | 8/8 | 3350 | 0·80 | 0·63, 1·03 | 0·02 | 56 | 5/5 | 1234 | 0·87 | 0·71, 1·08 | 0·74 | 0 |
| Cohort studies | ||||||||||||||||||
| Overall effect | 27/16 | 10 443 | 0·94 | 0·88, 1·01 | 0·85 | 0 | 16/13 | 4838 | 0·93 | 0·86, 1·01 | 0·51 | 0 | 16/13 | 2740 | 0·98 | 0·88, 1·09 | 0·46 | 0 |
| Geographic area | ||||||||||||||||||
| America | 6/4 | 1853 | 0·92 | 0·82, 1·03 | 0·83 | 0 | 3/3 | 1380 | 0·93 | 0·81, 1·06 | 0·48 | 0 | 3/3 | 327 | 1·17 | 0·71, 1·92 | 0·16 | 42 |
| Asia | 7/4 | 2794 | 0·93 | 0·82, 1·05 | 0·18 | 33 | 6/4 | 1848 | 0·89 | 0·76, 1·03 | 0·16 | 38 | 4/3 | 950 | 1·02 | 0·83, 1·26 | 0·95 | 0 |
| Europe | 14/8 | 5796 | 0·96 | 0·88, 1·06 | 0·94 | 0 | 7/6 | 1610 | 0·98 | 0·86, 1·11 | 0·71 | 0 | 9/7 | 1463 | 0·98 | 0·84, 1·14 | 0·38 | 7 |
| Sex | ||||||||||||||||||
| Males | 11/9 | 3562 | 0·97 | 0·87, 1·08 | 0·97 | 0 | 7/7 | 1543 | 0·99 | 0·85, 1·14 | 0·81 | 0 | 7/7 | 1026 | 1·02 | 0·85, 1·23 | 0·44 | 0 |
| Females | 10/9 | 3707 | 0·93 | 0·81, 1·05 | 0·49 | 3 | 7/7 | 1935 | 0·91 | 0·77, 1·07 | 0·21 | 29 | 7/7 | 939 | 1·02 | 0·83, 1·25 | 0·29 | 19 |
| Follow-up periods | ||||||||||||||||||
| <10 years | 12/7 | 2675 | 0·92 | 0·85, 1·01 | 0·35 | 10 | 6/5 | 1582 | 0·90 | 0·81, 1·00 | 0·30 | 17 | 4/4 | 774 | 0·92 | 0·78, 1·08 | 0·54 | 0 |
| ≥10 years | 15/9 | 7771 | 0·96 | 0·88, 1·05 | 0·98 | 0 | 10/8 | 3256 | 0·98 | 0·87, 1·09 | 0·62 | 0 | 12/9 | 1969 | 1.03 | 0.90, 1.19 | 0.40 | 5 |
RR, relative ratio.
*P value for heterogeneity.
†I 2 is interpreted as the proportion of total variation across studies due to heterogeneity rather than chance.