Characteristics of the healthy survivor effect among male and female Hanford workers

Am J Ind Med. 1999 Apr;35(4):343-7. doi: 10.1002/(sici)1097-0274(199904)35:4<343::aid-ajim4>;2-4.


Background: The healthy survivor effect is a selection process whereby healthy workers are selectively retained in the work force while unhealthy workers are removed. Understanding this phenomenon is integral to the accurate assessment of exposure effects in occupational cohorts. To date, scarce information has been published on the descriptive characteristics of the healthy survivor effect.

Methods: Follow-up mortality data on 44,154 employees from the Hanford nuclear facility for the period of 1944-1986 were used to estimate the healthy survivor effect according to frequently measured sociodemographic characteristics.

Results: While Hanford employees did not exhibit a stepwise decline in standardized mortality ratios according to duration of employment, workers in the longest employment duration category demonstrated a substantial survival advantage compared to the rest of the cohort. This effect was present in both males and females, and in all but the following subgroups: males hired at or after age 40, females hired before age 40, and females classified as both professional and nonprofessional.

Conclusion: The findings of the present study suggest that investigators should consider the potential confounding role of the healthy survivor effect when relying on SMRs, or other methods, to assess the adverse health effects of exposure in occupational cohorts. Further studies should be conducted, however, to assess variation in the healthy survivor effect according to sociodemographic characteristics.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Adult
  • Cause of Death
  • Female
  • Healthy Worker Effect*
  • Humans
  • Male
  • Mortality*
  • Occupational Exposure / statistics & numerical data*
  • Power Plants*
  • Radiation, Ionizing*
  • Retrospective Studies
  • Sex Distribution
  • Washington / epidemiology