Urinary bladder cancer risk among motor vehicle drivers: a meta-analysis of the evidence, 1977-2008.

BACKGROUND The aim of the study was to summarize bladder cancer risk in motor vehicle drivers and railroad workers using meta-analysis techniques. METHODS We retrieved all published results (3 cohort studies and 27 case-control studies) during 1977-2008. We assessed the heterogeneity of the results assuming a fixed-effect model. For cohort studies, the observed and the expected number of cases were added, respectively, to yield pooled observed/expected ratio. For case-control studies, we calculated pooled odds ratio (OR) and corresponding 95% confidence interval (CI) as a weighted average of the ORs in each study, by giving a weight proportional to the inverse of the variance of the ORs. RESULTS No overall meta-analysis was performed because of heterogeneity in results. The overall pooled risk among motor vehicle and railroad workers based on all cohort studies was 1.08 (95%: 1.00-1.17). The overall pooled risk among truck drivers was 1.18 (95% CI: 1.09-1.28 based on 18 case-control studies). The stratified analysis by year of publication indicated that pooled risk among truck drivers was 1.20 (95% CI: 1.00-1.40) for the period 1998-2008. The corresponding risk for the period 1977-1987 was 1.30 (95%: 1.16-1.46). The overall pooled risk among bus drivers was 1.23 (95% CI: 1.06-1.44 based on 10 case-control studies). The pooled risk among bus drivers was 1.21 (95% CI: 0.72-2.01) for the period 1998-2008 and the corresponding risk for the period 1977-1987 was 1.30 (95%CI: 1.10-1.53). The pooled risk among railroad workers was 1.20 (95% CI: 1.02-1.41 based on 15 case-control studies). Stratified analysis by year of publication was not statistically significant among railroad workers. CONCLUSION The pooled analysis suggested an increased bladder cancer risk among motor vehicle drivers and railroad workers. However, the risk among these workers is reduced in recent publications compared to the earlier publications.