Is the effect of job strain on myocardial infarction risk due to interaction between high psychological demands and low decision latitude? Results from Stockholm Heart Epidemiology Program (SHEEP)

Abstract

The objectives are to examine if the excess risk of myocardial infarction from exposure to job strain is due to interaction between high demands and low control and to analyse what role such an interaction has regarding socioeconomic differences in risk of myocardial infarction. The material is a population-based case-referent study having incident first events of myocardial infarction as outcome (SHEEP: Stockholm Heart Epidemiology Program). The analysis is restricted to males 45–64 yr of age with a more detailed analysis confined to those still working at inclusion. In total, 1047 cases and 1450 referents were included in the analysis. Exposure categories of job strain were formed from self reported questionnaire information. The results show that high demands and low decision latitude interact with a synergy index of 7.5 (95% C.I.: 1.8–30.6) providing empirical support for the core mechanism of the job strain model. Manual workers are more susceptible when exposed to job strain and its components and this increased susceptibility explains about 25–50% of the relative excess risk among manual workers. Low decision latitude may also, as a causal link, explain about 30% of the socioeconomic difference in risk of myocardial infarction. The distinction between the interaction and the causal link mechanisms identifies new etiologic questions and intervention alternatives. The specific causes of the increased susceptibility among manual workers to job strain and its components seem to be an interesting and important research question.

Introduction

Since the concept of job strain was introduced in 1979 (Karasek, 1979), it has played an important role in studies on health effects from psychosocial work conditions. Its fruitfulness has several features. Job strain describes important aspects of work organization and work content that are relevant for negotiations at the work site and for interventions from a public health perspective. The concept emphasizes the fundamental interplay between the individual and the environment. It does not rely on passive adaptation to exogenous harmful influences but incorporates the human ability to actively react and to consciously take control. By being based on stress theory it builds bridges between factors in the social environment and biologic reactions in the human body.

Job strain is defined by the demand-control model as the combination of high psychological job demands and low job decision latitudes and the effect on disease risk is assumed to follow from interaction between the two components (Karasek, 1979; Theorell and Karasek, 1996). Several reviews show that many studies now support the notion of causal effects on risk of coronary heart disease from job strain (Schnall et al., 1994; Kristensen, 1995; Theorell and Karasek, 1996). However, despite the fact that interaction is a fundamental part of the job strain theory its nature remains unclear and the empirical evidence supporting this basic mechanism of the job strain theory is scarce (Kasl, 1996). The tradition in earlier research has been merely to examine the disease risk among those with combined exposure. The first objective of this paper is therefore to specify some relevant meanings of interaction, identify empirical criteria by which its presence could be determined and to pursue an empirical test of the interaction hypothesis.

Furthermore, as work environmental factors are presumably important in the causation of prevailing social class differences in risk of coronary heart disease a second objective is to examine the possible contribution of the job strain components and their interaction (Marmot et al., 1997). In this analysis we shall distinguish between two different kinds of mechanisms as they imply different causal processes and different intervention strategies. The two mechanisms are depicted in Fig. 1. They are not mutually exclusive but may operate independently at the same time. In the first type job strain is regarded as a possible causal link for the effect of socioeconomic status implying that the socioeconomic position influences the individual's amount of exposure to job strain. In the second type socioeconomic status instead interacts with job strain possibly through an increased effect in low socioeconomic positions. This situation of synergy could also be described as an increased susceptibility to job strain in low socioeconomic positions. Our questions concern whether job strain (and/or its two components) act as causal links between low socioeconomic status and myocardial infarction, and whether low socioeconomic status increases the effect of job strain (and/or its two components). We shall also examine the quantitative importance of the different mechanisms. The analysis is based on data from SHEEP (Stockholm Heart Epidemiology Program).