Abstract

This cross-sectional study investigates the relationship betwen the psychosocial work environment and cardiovascular disease (CVD) prevalence in a randomly selected, representative sample of 13,779 Swedish male and female workers. It was found that self-reported psychological job demands, work control, and co-worder social support combined greater than multiplicatively in relation to CVD prevalence. An age-adjusted prevalence ratio (PR) of 2.17 (95% CI-1.32, 3.56) was observed among workers with high demands, low control, and low social support compared to a low demand, high control, and high social support reference group. PRs of approximately 2.00 were observed in this group after consecutively controlling for the effects of age together with 11 other potential confounding factors. The magnitude of the age-adjusted PRs was greatest for blue collar males. Due to the cross-sectional nature of the study design, causal inferences cannot be made. The limitations of design and measurement are discussed in the context of the methodological weaknesses of the work stress field. (Am J Public Health 1988; 78:1336-1342.)

Introduction

The role of a stressful work environment in the development of cardiovascular disease (CVD) is a matter of current interest. A model of job stress—the demand control model—has been proposed by Karasek. The model predicts that biologically aversive strain will occur when the psychological demands of the job exceed the resources for control over task content. The research of Karasek, Theorell and colleagues suggests that it is this combination of high demands and low control that produces job strain. Workers in high strain jobs have been shown to have greater risk of developing CVD.

However, important methodological challenges to etiological inference remain in the occupational stress field despite 20 years of research. These include: an over-reliance on cross-sectional as opposed to prospective designs; lack of generalizability due to the frequent restriction of samples to healthy, employed males; lack of valid and reliable measures of chronic disease outcomes; lack of exposure data with stress being evaluated on the basis of only a single measure in time; and incomplete models of the stress process. Investigations using the demand-control model have addressed a number of these problems. However, the model itself has been criticized for not including other, equally important, psychosocial work characteristics.