489

Background. Several recent reports suggest that obesity, or conditions associated with obesity, particularly type 2 diabetes, might be risk factors for NHL, a cancer that has increased dramatically in western countries over the last several decades. Physical inactivity is a risk factor for obesity and type 2 diabetes, but only a single study has evaluated the association of recreational physical activity with risk of NHL. Methods. We evaluated these factors in a population-based case-control study conducted in Detroit, Iowa, Los Angeles, and Seattle from 1998-2000. Incident HIV- NHL cases aged 20-74 years were rapidly reported through the SEER registries in each area (N=1321). Controls were identified through random digit dialing and Medicare files, and were frequency matched to cases on sex, age, race, and study site (N=1057). Risk factor data were collected by in-home interviews and self-administered questionnaires. Frequency and duration of moderate and vigorous physical activity were converted to metabolic equivalents per week. Anthropometric variables were categorized by the quartile distribution among controls. Unconditional logistic regression was used to estimate the odds ratio (OR) and 95% confidence intervals (CI), adjusting for age, gender, race and study site. NHL subtypes (follicular, diffuse, T-cell and other/unknown) were derived from the SEER database, and risk of each subtype was estimated using polychotomous logistic regression. Results. There was no association of weight or body mass index with risk of NHL overall, while there was a weak positive association with height (ORs=1, 1.09, 1.37, 1.38; p-trend=0.06). In subtype analysis, weight and BMI were positively and significantly associated with diffuse NHL, while height was positively and significantly associated with follicular NHL. There was no association of history of hypertension, gallbladder disease, or type 2 diabetes with NHL risk overall or by NHL subtypes, with the exception of a positive association of gallbladder disease with risk of diffuse NHL (OR=1.97, 95% CI 1.16-3.34). Physical activity was inversely associated with risk of NHL (ORs with increasing level: 1, 0.75, 0.71, 0.55, 0.68, p-trend=0.04). Inverse associations with physical activity were seen for each subtype, with the clearest association for diffuse NHL. Conclusion: These data from a population-based case-control study found little evidence for a role of obesity or obesity-related diseases, including type 2 diabetes, in the etiology of NHL overall, although obesity may play a role in diffuse NHL. The weak positive association of height with NHL risk confirms some prior data. The inverse association of recreational physical activity with NHL risk has not been previously reported, and requires confirmation given the potential public health importance of this modifiable risk factor.

[Proc Amer Assoc Cancer Res, Volume 45, 2004]