Send to

Choose Destination
Soc Sci Med. 2015 Feb;126:138-44. doi: 10.1016/j.socscimed.2014.12.021. Epub 2014 Dec 20.

Do time-invariant confounders explain away the association between job stress and workers' mental health? Evidence from Japanese occupational panel data.

Author information

Institute of Economic Research, Hitotsubashi University, 2-1 Naka, Kunitachi-shi, Tokyo 186-8603, Japan. Electronic address:
Department of Public Health, Kitasato University School of Medicine, 1-15-1 Kitasato, Minami, Sagamihara, Kanagawa 252-0374, Japan.
Department of Mental Health, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu, Fukuoka 807-8555, Japan.


It is well known that job stress is negatively related to workers' mental health, but most recent studies have not controlled for unobserved time-invariant confounders. In the current study, we attempted to validate previous observations on the association between job stress and workers' mental health, by removing the effects of unobserved time-invariant confounders. We used data from three to four waves of an occupational Japanese cohort survey, focusing on 31,382 observations of 9741 individuals who participated in at least two consecutive waves. We estimated mean-centered fixed effects models to explain psychological distress in terms of the Kessler 6 (K6) scores (range: 0-24) by eight job stress indicators related to the job demands-control, effort-reward imbalance, and organizational injustice models. Mean-centered fixed effects models reduced the magnitude of the association between jobs stress and K6 scores to 44.8-54.2% of those observed from pooled ordinary least squares. However, the association remained highly significant even after controlling for unobserved time-invariant confounders for all job stress indicators. In addition, alternatively specified models showed the robustness of the results. In all, we concluded that the validity of major job stress models, which link job stress and workers' mental health, was robust, although unobserved time-invariant confounders led to an overestimation of the association.


Fixed effects models; Japan; Job stress; Mental health; Psychological distress

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center