The article by Kivimäki and colleagues1 includes several methodological and interpretational errors.
First, we believe the authors should have excluded obesity from their definition of an unhealthy lifestyle. The authors defined an unhealthy lifestyle arbitrarily as a combination of the following four risk factors of coronary artery disease: smoking, heavy drinking, leisure-time physical inactivity and obesity. Although the first three risk factors may be considered lifestyle risk factors, obesity has a multifactorial etiology and cannot be simplistically labelled a lifestyle risk factor. Therefore, the study’s results likely overestimate the effect of an “unhealthy lifestyle” on coronary artery disease.
Second, the authors should have made clear that two of the three behavioural risk factors (heavy drinking and leisure-time physical inactivity) did not offset the impact of job strain on coronary artery disease. Figure 1 in the article by Kivimäki and colleagues1 indicates that there would be no significant reduction in the risk of coronary artery disease from reducing heavy drinking behaviour or physical inactivity among workers with job strain. Rather, it indicates the importance of addressing both job strain and unhealthy behavioural risk factors for prevention of coronary artery disease. Emerging evidence shows that organizational- or task-level interventions for increasing job control and decreasing high job demands can be beneficial for the health of workers and organizations.2,3
Third, the conclusion of the authors (“a healthy lifestyle may substantially reduce disease risk among people with job strain”) is misleading; 84.3% of workers with job strain had none or only one of the four risk factors, and in this group there was no significant effect modification of any one of the four risk factors on the association between job strain and coronary artery disease. Therefore, the conclusion is irrelevant for the majority of the workers with job strain. Job strain, an adverse working condition, may be a structural barrier that keeps workers from adopting and maintaining healthy behaviours as implied in another article4 by the same research group.
Last, throughout the article, the authors equate job strain with perceived work stress or work stressors in general. The authors inappropriately compared the effects of perceived work stress (i.e., how often participants felt stress at work) in the INTERHEART study with job strain in their study.1 Equating job strain with work stressors in general ignores other important work stressors (i.e., poor workplace social support, job insecurity, long work hours, and effort–reward imbalance) that have been associated with cardiovascular disease.5