The demands-control-support work stress model and risk of ischemic heart disease: causal inference based on observational epidemiology.

Abstract

Reviews consistently suggest an association between job strain and ischemic heart disease (IHD), but causality remains uncertain. This study aimed to critically assess causal inference using the most informative epidemiological studies. A systematic search in PubMed and Embase up to 15 November 2024 identified observational studies reporting quantitative estimates of associations between job strain (defined by job demands and control) and IHD. Eligible studies were cohort or case-control designs with exposure data obtained independently by medically verified IHD (ICD-8/9: 410-414; ICD-10: I20-I25) and risk estimates adjusted at least for age, sex, and socio-economic status. One estimate per study was included in inverse-variance weighted random-effects meta-analyses. We evaluated main sources of upward and downward bias, potential confounding, and key criteria for causal inference including outcome specificity, exposure-response, and consistency. This review comprised 25 cohort and 1 case-control study (122 risk estimates). The fully adjusted pooled relative risk estimate (RRE) for job strain and all IHD outcomes combined was 1.14 [95% confidence interval (CI) 1.06-1.23; 21 studies]. For myocardial infarction, the RRE was 1.08 (95% CI 1.00-1.15; 11 studies), and, in studies using job-exposure matrices, it was 1.06 (95% CI 0.99-1.13; 7 studies). Strong heterogeneity, small effect sizes, limited exposure-response evidence, net bias in unpredictable directions, and lack of confirmation of findings in studies using alternatives to self-reported exposure assessment preclude causal inference. Evidence for a causal relationship between job strain and IHD is limited. At most, any true effect appears to be small.