Abstract
BACKGROUND: Little is known about occupational class differences in pancreatic cancer survival. METHODS: Using a population-based cancer registry in Japan, 3 578 patients with incident pancreatic cancer (1970-2011) were followed up for 5 years (median follow-up time 0.42 years). We classified patients into four occupational classes based on their longest-held jobs: white-collar (professional and managers), service, blue-collar, and those not actively employed. Using white-collar class as the reference group, hazard ratios (HRs) and 95% confidence intervals (CIs) for overall death were estimated by Cox proportional hazard model. Covariates included age, sex, and year of diagnosis. Prognostic variables (pathology, stage, and treatment) and smoking behaviors were additionally adjusted as possible mediating factors. RESULTS: Overall survival was poor in this population (median, 0.50 and 0.33 years in white-collar and service classes, respectively). Compared with white-collar patients, survival was significantly poorer across all occupational classes, most pronounced in the service worker group: mortality HRs ranged from 1.11 (95% CI 1.00-1.24) in blue-collar workers to 1.24 (95% CI 1.12-1.37) in service workers. Even after controlling for potential mediating factors, service workers showed worse survival. CONCLUSION: We documented occupational class disparities in pancreatic cancer survival in Japan. Even in the setting of lethal prognostic cancer with universal health coverage, high-occupational class groups may enjoy a health advantage.