Occupational sectors and cancer mortality in a large administrative cohort

Abstract

Background and aim The International Labor Organization estimates 2.4 million work-related deaths every year. We aimed to investigate the role of the occupational sector in cancer mortality. Methods We used the 2011 census cohort of Rome, including all residents aged 30+ years. We retrieved the information on occupational sectors from the 70s to 2011 for private-sector workers from the National Social Insurance Agency (INPS) database. We classified the occupational sectors into 23 categories and modeled occupational exposure as ever/never been employed in each sector. We followed the subjects from the census reference day (9 October 2011) to 31 December 2019, investigating cancer mortality overall and site-specific cancers (larynx, lung, pleura, kidney, liver). We used Cox regression models to estimate the age and sex-adjusted hazard ratio (HR) with 95% confidence intervals (CI) of being employed in an occupational sector. In addition, we performed the analyses for males and females separately. Results We analyzed a cohort of 910,559 (53% males). We found evidence of a higher cancer mortality risk in workers of the paper and print industry for lung and pleural cancer (HR=1.22 95%CI:1.08-1.38 and HR=2.19 95%CI:1.18-4.05, respectively), food and tobacco for lung cancer (HR=1.34 95%CI:1.15-1.55), building construction industries for liver, lung and pleural cancer (HR=1.44 95%CI:1.21-1.63, HR=1.29 95%CI:1.20-1.39, HR=1.99 95%CI:1.32-3.00, respectively), hotels and restaurants sector for liver and lung cancer (HR=1.30 95%CI:1.05-1.61 and HR=1.27 95%CI:1.15-1.41), hairdressers for liver, larynx and lung cancer (HR=1.65 95%CI:1.30-2.10, HR=1.99 95%CI:1.05-3.80, HR=1.44 95%CI:1.28-1.62), and waste workers for lung cancer (HR=2.04 95%CI:1.34-3.10). The analyses by gender confirmed the results. Conclusion The results confirm the well-known associations between occupational exposures (characteristic of specific sectors) and cancer mortality and highlights sectors with elevated cancer risk. Administrative data might be a valuable tool in occupational epidemiology research.