Overcoming underestimation of the share of colorectal cancer cases attributable to excess weight: a population-based study.

Abstract

Previous research may have underestimated the relationship between overweight/obesity and colorectal cancer (CRC) risk by overlooking important potential sources of bias. We used data from a large, population-based case-control study encompassing 7098 CRC cases and 5757 age- and sex-matched controls with comprehensive information on risk factors, including self-reported body weight. Multivariate logistic regression was used to assess the associations of BMI with CRC risk before and after considering prediagnostic weight loss, history of lower gastrointestinal endoscopy, and potentially increased CRC risk beneath the overweight threshold (BMI 25 kg/m²). Subsequently, population attributable fractions were calculated. In the standard analysis evaluating the BMI-CRC association, in which none of the three aforementioned factors was considered, the fraction of CRC cases attributable to overweight and obesity was estimated to be 11.5%. This finding is consistent with estimates from previous studies, which mostly did not consider any of the three factors. However, when all three factors were considered in the analysis, a higher BMI was estimated to account for 23.4% of all CRC cases. Careful consideration of important sources of bias suggests that a substantially larger share of the CRC burden may be attributable to excess weight than previously thought.