Brain cancer mortality in an agricultural region of Rio de Janeiro, Brazil: a population-based age-period-cohort study, 1996-2010

Background: The brain cancer etiology is poorly understood, although several studies indicate that residents of agricultural regions have an increased risk of mortality, the hypothesis being the intense exposure to pesticides. Both the trend and the age-periodcohort effect of brain cancer mortality in Rio de Janeiro, Brazil, were estimated. Methods: This is a descriptive epidemiologic study of mortality from brain cancer in 20-year-old or older residents of an agricultural region (Serrana) and a non-agricultural region (Metropolitan), between 1996 and 2010. The trend analysis was estimated through Joinpoint regression and the individual effects of age-period-cohort were performed using log-linear Poisson regression. Results: The estimated annual percentage change in brain cancer mortality in the Serrana region was 3.8% (CI95% 0.8 and 5.6), in contrast, the Metropolitan region was virtually stable: -0.2 (CI95% -1.2 and 0.7). The age-period-cohort analysis showed an age effect in both regions. In the Serrana region a relevant cohort effect was observed: individuals who were born from 1955/59 to 1985/89 have showed increasing relative risk. On the other hand, an increased RR not was observed in the Metropolitan region. Conclusion: This study showed an increasing trend in time series of brain cancer mortality among adults living in an agricultural area in the state of Rio de Janeiro. The cohort effect seems to be a consequence of changes in the environmental exposure. These findings reinforce the hypothesis that residents of agricultural regions have increased brain cancer mortality.