An alternate characterization of hazard in occupational epidemiology: years of life lost per years worked

Am J Ind Med. 2002 Jul;42(1):1-10. doi: 10.1002/ajim.10082.

Abstract

Background: Standardized mortality ratios (SMRs) and other measures of relative risk by themselves may not suffice as descriptors of occupational hazards for many audiences including decision-makers and those at direct risk from hazardous work. To explore other approaches, we calculated excess years of potential life lost and excess lifetime risk for both lung diseases and fatal injuries in a cohort of uranium miners with historical records of exposure to radon gas.

Methods: We used relatively simple life table (SMR) methods and also analyzed lung cancer mortality with Poisson regression methods permitting control for smoking.

Results: Among uranium miners hired after 1950, whose all-cause SMR was 1.5, 28 percent would experience premature death from lung diseases or injury in a lifetime of uranium mining. On average, each miner lost 1.5 yr of potential life due to mining-related lung cancer, or almost 3 months of life for each year employed in uranium mining. As a consequence of all excess lung disease and injury risks combined, a year of mining was associated with 5.9 months loss of potential life. For each year actually working underground, miners lost more than 8 months of potential life. When controlled for smoking (and healthy worker effect) with Poisson regression, the estimates for radon-related lung cancer effects were slightly larger. Although chronic disease deaths dominated in excess years of life lost (due to radon, silica and possibly other exposures), more years were lost on average per individual injury death (38 yr), than per excess lung cancer (20 yr) or other lung disease death (18 yr). Fatal-injury dominated the potential years of life lost up to about age 40.

Conclusions: Years of life lost per years employed provides another, more intuitive summary of occupational mortality risk.

MeSH terms

  • Accidents, Occupational
  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Cohort Studies
  • Colorado / epidemiology
  • Humans
  • Life Expectancy*
  • Life Tables
  • Lung Neoplasms / etiology*
  • Lung Neoplasms / mortality*
  • Middle Aged
  • Mining*
  • Occupational Exposure*
  • Poisson Distribution
  • Quality-Adjusted Life Years*
  • Radon
  • Regression Analysis
  • Risk
  • Time Factors
  • Uranium*
  • Wounds and Injuries / mortality*

Substances

  • Uranium
  • Radon