Table 3.
Characteristics of cohort studies of coffee consumption and colorectal, colon and rectal cancer risk
First author (year) | Years of study* | Country | No. of subjects | No. of cases | RR | 95 % CI | Coffee (‘highest’ v. ‘lowest/non’) | Adjustment factors |
---|---|---|---|---|---|---|---|---|
Jacobsen (1986)( 53 ) | 1967–1978 (11·5) | Norway | 16 555 (13 664 M, 2891 F) | 97 CC, 63 RC | CC: 0·54 | 0·22, 1·30 | ‘≤2 cups/d’ v. ‘≥7 cups/d’ | Age, sex, alcohol for CC, residence |
RC: 1·07 | 0·41, 2·79 | |||||||
Wu (1987)( 55 ) | 1981–1985 (4·5) | USA | 11 644 | 146 CRC (58 M, 68 F) | For CRC, M: 1·54 | 0·6, 3·7 | ‘4 cups/d’ v. ‘1cup/d’ | Age |
For CRC, F: 1·17 | 0·4, 3·1 | |||||||
Klatsky (1988)( 54 ) | 1978–1984 | USA | 106 203 | 203 CC, 66 RC | CC: 0·92 | 0·80, 1·06 | Continuous variable (cups/d) | Age, sex, alcohol, smoking, BMI, race, education, serum cholesterol |
RC: 0·84 | 0·66, 1·07 | |||||||
Stensvold (1994)( 37 ) | 1977–1990 (10·1) | Norway | 42 973 (21 735 M, 21 238 F) | 130 CC, 79 RC | For CC, M: 0·90 | 0·35, 2·39 | Continuous variable (cups/d) | Age, smoking, residence |
For CC, F: 0·82 | 0·22, 3·05 | |||||||
For RC, M: 0·66 | 0·18, 2·49 | |||||||
For RC, F: 0·47 | 0·10, 2·19 | |||||||
Hartman (1998)( 19 ) | 1985–1993 (8·0) | Finland | 27 111 M smokers | 106 CC, 79 RC | CC: 0·84 | 0·50, 1·40 | ‘>6 cups/d’ v. ‘≤4 cups/d’ | Age, BMI, Ca, PA, intervention group for both CC and RC, serum cholesterol, tea for RC |
RC: 0·74 | 0·40, 1·36 | |||||||
Michels (2005)( 36a ) | 1980–1998 | USA | 87 794 F | 886 CRC (731 CC, 155 RC) | CRC: 0·94 | 0·63, 1·40 | For CRC and CC: ‘>5 cups/d’ v. ‘never’; for RC: ‘≥4 cups/d’ v. ‘never’ | Age, BMI, height, alcohol, smoking, PA, family history of CRC, sigmoidoscopy, aspirin use, vitamin supplement, energy, red meat, menopausal status, postmenopausal hormone use |
CC: 0·85 | 0·55, 1·32 | |||||||
RC: 1·80 | 0·94, 3·44 | |||||||
Michels (2005)( 36b ) | 1986–1998 | USA | 46 099 M | 552 CRC (446 CC, 106 RC) | CRC: 1·09 | 0·55, 2·17 | For CRC and CC: ‘>5 cups/d’ v. ‘never’; for RC: ‘≥4 cups/d’ v. ‘never’ | Age, BMI, height, alcohol, smoking, PA, family history of CRC, sigmoidoscopy, aspirin use, vitamin supplement, energy, red meat |
CC: 1·39 | 0·69, 2·78 | |||||||
RC: 1·33 | 0·69, 2·56 | |||||||
Larsson (2006)( 21 ) | 1998–2004 | Sweden | 81 922 (45 306 M, 36 616 F) | 723 CRC (469 CC, 256 RC) | CRC: 1·06 | 0·74, 1·52 | ‘≥6 cups/d’ v. ‘<1 cup/d’ | Age, BMI, smoking, PA, family history of CRC and diabetes, aspirin use, multivitamin use, energy, red meat, fruits, vegetables, milk, for women postmenopausal hormone use |
CC: 1·16 | 0·73, 1·85 | |||||||
RC: 0·92 | 0·51, 1·65 | |||||||
Mucci (2006)( 17 ) | 1987–2003 | Sweden | 61 467 F | 741 CRC (504 CC, 237 RC) | CRC: 1·0 | 0·7, 1·3 | ‘≥4 cups/d’ v. ‘≤1 cup/d | Age at screening, BMI, education, alcohol, energy, saturated fat, fibre |
CC: 1·1 | 0·8, 1·5 | |||||||
RC: 0·9 | 0·6, 1·4 | |||||||
Oba (2006)( 25 ) | 1993–2000 | Japan | 30 221 (13 894 M, 16 327 F) | 213 CC (111 M, 102 F) | For CC, M: 0·81 | 0·46, 1·42 | ‘≥1 cup/d’ v. ‘<1 cup/month’ | Age, BMI, height, alcohol, smoking, PA, black/green tea |
For CC, F: 0·43 | 0·22, 0·85 | |||||||
Lee (2007)( 22 ) | 1990–2002 (10) | Japan | 96 162 (46 023 M, 50 139 F) | 1163 CRC (763 CC, 400 RC) | For CRC, M: 1·10 | 0·82, 1·47 | ‘≥3 cups/d’ v. ‘almost never’ | Age, BMI, alcohol, smoking, PA, study area, family history of CRC, beef, pork, green vegetables, black/green tea, Chinese tea |
For CRC, F: 0·68 | 0·40, 1·15 | |||||||
Naganuma (2007)( 23 ) | 1990–2001 (11·0) | Japan | 38 701 (18 867 M, 19 834 F) | 457 CRC (281 CC, 180 RC) | CRC: 0·95 | 0·65, 1·39 | ‘≥3 cups/d’ v. ‘never’ | Age, sex, BMI, alcohol, smoking, walking time, family history, education, energy, fruits, vegetables, meat, tea; for F, menopausal status, numbers of pregnancies and deliveries, age at menarche, age at first delivery |
CC: 0·96 | 0·58, 1·59 | |||||||
RC: 0·94 | 0·53, 1·66 | |||||||
Bidel (2010)( 52 ) | 1972–2006 (18) | Finland | 60 041 (29 159 M, 30 882 F) | 538 CRC (304 CC, 234 RC) | CRC: 1·03 | 0·58, 1·83 | ‘≥10 cups/d’ v. ‘0 cup/d’ | Age, sex, study year, education, cigarette smoking, alcohol consumption, leisure-time PA, history of diabetes, tea consumption, BMI |
CC: 0·72 | 0·35, 1·47 | |||||||
RC: 0·99 | 0·71, 5·55 | |||||||
Nilsson (2010)( 24 ) | 1992–2007 (6) | Sweden | 64 603 (32 425 M, 32 178 F) | 321 CRC (174 M, 147 F) | CRC: 1·43 | 0·86, 2·38 | ‘≥4 cups/d’ v. ‘<1 cup/d | Age, sex, BMI, smoking, education, recreational PA |
Peterson (2010)( 32 ) | 1993–2005 (9·8) | Singapore | 61 321 | 591 CC, 370 RC | CC: 0·90 | 0·73, 1·11 | ‘≥2 cups/d’ v. ‘<1 cup/d | Age, sex, dialect group, year of recruitment, education, BMI, smoking, alcohol and tea, PA, history of diabetes, family history of CRC |
RC: 1·06 | 0·81, 1·37 | |||||||
Simons (2010)( 27 ) | 1986–1999 (13·3) | Holland | 120 852 (58 279 M, 62 573 F) | 2899 CRC (1260 M, 939 F)/ 3877 CON | For CRC, M: 1·00For CRC, F: 1·07 | 0·74, 1·360·74, 1·55 | ‘>6 cups/d’ v. ‘≤2 cups/d’ | Age, family history of CRC, PA, smoking, educational, BMI, ethanol, meat, processed meat, folate, vitamin B6, fibre, fluid |
RR, relative ratio; M, males; F, females; CC, colon cancer; RC, rectal cancer; CRC, colorectal cancer; CON, control; PA, physical activity.
*Mean or median duration of follow-up in parentheses.