Table 4.
Relative risks of incident bladder cancer for current smokers relative to never smokers in previously published studies from United States prospective cohorts*
Author, reference, year |
Cohort | Sex | Years | Mean age |
Never smokers in cohort (N; %) |
Cases in never smokers (N; %) |
Current smokers in cohort (N; %) |
Cases in current smokers (N; %) |
Typical amount of cigarettes smoked per day (among current smokers) |
RR†† (95%CI) for current smoking |
---|---|---|---|---|---|---|---|---|---|---|
Alberg et al,44,2007 | Washington County MD | Men & women | 1963–1978 | 4745 | 11,722; 26% | 20; 22% | 20,037; 44% | 48; 52% | 29% >20 cig/day‡ | 2.7 (1.6–4.7) |
Chyou et al, 46, 1993 | Japanese men in Hawaii | Men | 1965–1991 | 5447 | 2,410; 30% | 17; 18% | 3,495; 44% | 60; 63% | 77% ≥20 cig/day48 | 2.86 (1.67–4.91) |
Mills et al, 49, 1991 | Seventh Day Adventists | Men & Women | 1976–1982 | 5450 | 26,059; 76%‡ | 25; 52% | 1,129; 3%‡ | 4; 8% | 32% ≥25 cig/day‡, ** | 5.67 (1.73–18.61) |
Alberg et al, 44, 2007 | Washington County MD | Men & women | 1975–1994 | 4845 | 15,249; 32% | 40; 23% | 17,006; 35% | 67; 39% | 31% >20 cig/day‡ | 2.6 (1.7–3.9) |
Tripathi et al, 51, 2002 | Iowa Women’s Health Study | Women | 1986–1998 | 6252 | 24,723; 66% | 42; 38% | 5,619; 15% | 45; 41% | 16% >20 cig/day‡, **,53 | 4.23 (2.76–6.70) |
Michaud et al, 54, 2001 | Health Professionals Follow-up Study | Men | 1986–1998 | 5355 | 24,035; 49%56 | 70; 23% | 4,648; 9% | 44; 14% | 33% > 25 cig/day‡,55 | 2.81 (1.85–4.27) |
Cantwell et al, 41, 2006 | Breast Cancer Detection Demonstration Project Follow-up Study | Women | 1987–2000 | 55 | 27,691; 57%‡ | 62; 44% | 7,826; 16%‡ | 30; 21% | 54% >20 cig/day57 | 2.44 (1.56–3.80) |
Summary Estimate† | Men & women | 276 | 298 | 2.94 (2.45–3.54) |
Not all data on this table was available in the original publication which examined the association of smoking and bladder cancer. For publications which lacked some of these variables, we identified other publications from the same cohort containing the desired information; references for these publications are marked where appropriate.
Summary relative risk and 95% confidence intervals are from random effect models. The I2 statistic for heterogeneity across studies was 0.0% and the Cochran Q test p-value for between study heterogeneity was 0.554.
Calculated from person-years in the original publication
Cigarettes smoked per day for both former and current smokers together.
Alberg and Cantwell used Poisson regression models, Chyou, Mills, and Tripathi used Cox proportional hazards regression, and Michaud used logistic regression.