Occupational
Ionizing Radiation and Cancer Mortality in the INWORKS pooled cohort: Assessing Healthy Worker Survivor Bias Alexander P. Keil* Alexander Keil Keil Keil Keil Keil Keil Keil Keil Keil Keil Keil Keil Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute (NCI), National Institutes of Health (NIH), Department of Health and Human Services (DHHS), Bethesda, Maryland, USA
Ionizing radiation is an established carcinogen. Occupational cohort studies have provided statistically precise estimates of associations between protracted, low dose exposure to ionizing radiation and various health effects. However, uncertainty about whether these statistical estimates of association accurately represent the causal effects of exposure remains due, in part, to potential for healthy worker survivor bias.
Using data from the International Nuclear Workers Study, 1944-2005 (a pooled cohort mortality study of 308,297 radiation-monitored workers employed in France, United Kingdom, and United States) we assessed whether conditions of healthy worker survivor bias are present. Given evidence of potential for this bias, we used g-computation to address healthy worker survivor bias and estimate impacts of hypothetical radiation exposure scenarios on subsequent cancer mortality.
Remaining employed was associated with lower cancer mortality than leaving employment, a key condition of healthy worker survivor bias. We estimate that if all workers had been exposed to 5, 10, or 20 mGy per year while employed approximately 3.2 (90% CI: -1.9, 8.8), 7.6 (90% CI: 0.8, 15), or 17 (90% CI: 6.5, 26) excess cancer deaths per 1,000 workers would be observed by age 90 years as compared to what would have been observed if all workers had been exposed to 0 mGy per year while employed (Figure).
Results suggest that healthy worker survivor bias is present and that standard regression methods may underestimate radiation dose-cancer mortality associations. Our estimates from a population of workers, most of whom were occupationally exposed to low doses of radiation at low dose rates, strengthens understanding of cancer risks associated with protracted low dose exposure to ionizing radiation.
