Case: Control Matching Strategies on Associations between Cellular Telephone Use and Glioma Risk in the INTERPHONE Study

Abstract

Case: Control Matching Strategies on Associations between Cellular Telephone Use and Glioma Risk in the INTERPHONE StudyAbstract Number:1680 Michelle Turner*, Siegal Sadetzki, Chelsea Langer, Jordi Figuerola, Bruce Armstrong, Angela Chetrit, Graham Giles, Daniel Krewski, Martine Hours, Mary McBride, Marie-Elise Parent, Lesley Richardson, Jack Siemiatycki, Alistair Woodward, and Elisabeth Cardis Michelle Turner* Centre for Research in Environmental Epidemiology, Spain, E-mail Address: [email protected] Search for more papers by this author , Siegal Sadetzki Chaim Sheba Medical Center, Israel, E-mail Address: [email protected] Search for more papers by this author , Chelsea Langer Centre for Research in Environmental Epidemiology, Spain, E-mail Address: [email protected] Search for more papers by this author , Jordi Figuerola Centre for Research in Environmental Epidemiology, Spain, E-mail Address: [email protected] Search for more papers by this author , Bruce Armstrong The University of Sydney, Australia, E-mail Address: [email protected] Search for more papers by this author , Angela Chetrit Chaim Sheba Medical Centre, Israel, E-mail Address: [email protected] Search for more papers by this author , Graham Giles Cancer Council Victoria, Australia, E-mail Address: [email protected] Search for more papers by this author , Daniel Krewski University of Ottawa, Canada, E-mail Address: [email protected] Search for more papers by this author , Martine Hours Université de Lyon, France, E-mail Address: [email protected] Search for more papers by this author , Mary McBride British Columbia Cancer Agency, Canada, E-mail Address: [email protected] Search for more papers by this author , Marie-Elise Parent Universite du Quebec, Canada, E-mail Address: [email protected] Search for more papers by this author , Lesley Richardson University of Montreal Hospital Research Centre, Canada, E-mail Address: [email protected] Search for more papers by this author , Jack Siemiatycki University of Montreal Hospital Research Centre, Canada, E-mail Address: [email protected] Search for more papers by this author , Alistair Woodward University of Auckland, New Zealand, E-mail Address: [email protected] Search for more papers by this author , and Elisabeth Cardis Centre for Research in Environmental Epidemiology, Spain, E-mail Address: [email protected] Search for more papers by this author AbstractAssociations between cellular telephone use and risk of brain tumors have been examined in a number of epidemiological studies including INTERPHONE. Although results revealed no positive associations between cellular telephone use and glioma risk overall, no exposure response, and no increased risk among long term users, there was some suggestion of an elevated risk among those in the highest decile of cumulative call time (odds ratio (OR) = 1.40, 95% confidence interval (CI) 1.03–1.89). However, there are potential methodological limitations including selection bias, reporting bias, and the fact that participant controls were interviewed later in time than cases during a time period of rapidly increasing cellular telephone use. Further work was conducted in a subset of five INTERPHONE study countries (Australia, Canada, France, Israel, New Zealand) using a post-hoc matching strategy optimizing case to control interview time to investigate the impact of matching and analytical strategy on associations between cellular telephone use and glioma risk with a particular focus on the distribution of time interval between interviews of case and control participants. In comparison with results based on the original INTERPHONE matching, analyses using closely post-hoc matched subjects in time (interviewed within 1 year) showed a small increased risk among long term users (OR 10+ years = 1.21, 95% CI 0.81-1.82) as well as a tendency towards increasing risk in the highest exposure categories, both for cumulative call time (OR 8-9 decile = 1.21, 95% CI 0.84-1.75; 10th decile = 1.47, 95% CI 0.93-2.33) and cumulative number of calls (OR 8-9 decile = 1.15, 95% CI 0.80-1.65; 10th decile = 1.23, 95% CI 0.79-1.93) in conditional logistic regression analyses. Matching time of case and control interviews may be an important factor to consider when exposure patterns are changing rapidly with time.