Risk factors for renal cell carcinoma in the multiethnic cohort study.

PR-16

We examined the independent association of body size, physical activity, cigarette smoking, alcohol consumption, and medical conditions with the risk of renal cell carcinoma (RCC) among 167,200 African-American, Native-Hawaiian, Japanese-American, Latino and white men and women recruited to the prospective Multiethnic Cohort Study in 1993-1996. During a 8.3 year follow-up period, 375 (246 men and 129 women) incident cases of RCC were identified among these participants. Data on known and suspected risk factors for RCC were obtained from the baseline questionnaire. Relative Risk (RR) and 95% confidence interval (CI) were calculated using COX proportional hazard regression and adjusted simultaneously for known or suspected risk factors. The risk of RCC increased with increasing body mass index (BMI) in both men and women with a much stronger trend observed in women (RR=1.07 per unit BMI, p _trend <0.0001) than in men (RR=1.05, p _trend =0.003). The RR associated with being obese (BMI ≥ 30 kg/m²) compared to being lean (BMI < 25 kg/m²) was 1.55 (95% CI 1.08, 2.23) for men and 2.37 (95% CI 1.44, 3.89) for women. Diagnosis of hypertension was associated with RCC; the RR was 1.47 (95% CI 1.13, 1.91) for men and 1.77 (95% CI 1.22, 2.56) for women. Increased total physical activity (as metabolic equivalent, METs) was associated with reduced risk only among women (p _trend =0.013). Cigarette smoking was confirmed to be a risk factor for both men and women. Increased alcohol consumption was significantly associated with lowered risk only among men (p _trend =0.021). Men who drank 1 or more alcoholic drink per day had lower RCC risk than men who did not drink (RR=0.69; 95% CI 0.51, 0.95). Diagnosis of diabetes mellitus or kidney stone was not associated with RCC risk. We found that BMI, smoking, and hypertension were independently associated with RCC risk in both men and women, while the effects of physical activity and alcohol consumption on risk may be gender specific although tests for interactions between sex and these exposures on RCC risk were not significant.