Healthy Worker Effect

The healthy worker effect is a bias in occupational studies: workers as a group tend to be healthier than the general population, because people who are seriously ill are less likely to be hired or to stay employed. Comparing workers with the general population can therefore make a workplace hazard look smaller than it really is.

Origin

Observers in the 19th century noticed that people in employment had lower death rates than the population at large. The phrase "healthy worker effect" came into common use in occupational epidemiology in the 1970s, as researchers worked out how to correct for it.

Significance in Asbestos Context

Many asbestos studies compare death or cancer rates among workers with rates in the general population, which includes people too ill to work. Because the workers start out healthier, a ratio near 1.0 for some outcome may hide a real excess, and a modest ratio can understate the true risk. A related form, the healthy worker survivor effect, arises when workers who become sick leave exposed jobs early, so long-serving workers look healthier and the link between dose and disease is blurred. The effect is strongest for heart disease and overall death rates and weaker for rare cancers like mesothelioma, which is why mesothelioma excesses remain so striking. In litigation, plaintiffs' experts may point to the healthy worker effect to explain why a study found little or no increased risk, while defense experts may cite those same studies as evidence that an exposure was not harmful. Researchers reduce the bias by comparing workers with other workers, such as unexposed employees in the same plant.

Example Sentence

The plaintiff's epidemiologist argued that the study's overall SMR of 0.9 reflected the healthy worker effect, not a lack of harm from asbestos.

Related Terms

Selection Bias, Cohort Study, standardized mortality ratio (SMR), Standardized Incidence Ratio (SIR), Confounding, Epidemiology

Notes

• A study showing "no excess deaths" among workers can still be consistent with real harm; ask how the comparison group was chosen.
• Studies that compare exposed and unexposed workers in the same industry avoid much of this bias.