Occupational
Workforce reductions may harm the health of those who remain at work Kevin Chen* Kevin Chen Ellen A. Eisen
Background: Survivors of workforce reductions may experience increased workplace demands coupled with reduced control, loss of social connections, survivor’s guilt, strained relationships with employers, and greater perceived job insecurity. Here, we seek to estimate the association between experience of workforce reductions while employed and long-term risk of mortality due to preventable chronic diseases.
Methods: Using data from the UAW-GM autoworkers cohort (n = 36,121), we detected workforce reduction events over the years 1946-1985 by identifying months in which the percent reduction in the number of employees was greater than two standard deviations above the time-series prediction. Then, we estimated adjusted hazard ratios (HR) for mortality due to cardiovascular disease (CVD), cancer, and chronic liver disease in a Cox model, comparing levels of cumulative experience of workforce reduction.
Results: Over the follow-up period 1946-2015, we observed 6,711 deaths due to CVD, 4,286 deaths due to cancer, and 287 deaths due to chronic liver disease. The HRs for cancer and chronic liver disease mortality were monotonically increasing for increasing levels of cumulative experience of workforce reduction. Compared to no experience of workforce reduction, experiencing 1 to 2 reductions was associated with a 1.05 (1.03, 1.08) and 1.11 (0.75, 1.66) times greater hazard of cancer and liver disease mortality, respectively. The HRs associated with 7 or more reductions were 1.54 (1.00, 2.38) and 1.65 (1.16, 2.35) for cancer and liver disease mortality. The HRs for CVD mortality were null.
Conclusion: Workers who survived more workforce reduction events experienced a greater hazard of cancer and chronic liver disease mortality in the UAW-GM cohort. Since the number of workforce reduction events experienced depends on duration of employment, specialized methods to account for the healthy worker survivor effect may yield stronger positive associations, particularly for CVD.
