MacArthur SES & Health Network
MacArthur SES & Health Network

printable version

Evidence from studies using the effort-reward imbalance model

Compared with the demand-control model fewer studies have been conducted on the adverse health effect of effort-reward imbalance at work, but those that have suggest that the model provides a fruitful framework for examining work stress and its contribution to the development of disease. So far, six studies have reported findings with partial or full confirmation of the model's basic assumption. An overview of the studies is given in Table 3, which unlike Table 2 includes other outcomes in addition to CHD, and obviously does not use a quality filter. Concerning the study design, three studies are prospective. These are the Whitehall II study mentioned above (Bosma et al, 1998); a German blue-collar study covering some 2,000 person years (Siegrist et al, 1990), and a Swedish cohort study of some 5,720 healthy employed men and women (Peter et al, 1998a). Two studies are cross-sectional: a study of 1,337 male and female transport workers (Peter et al, 1998b), and a study of 179 male middle managers (Siegrist et al, 1997). Furthermore, a follow-up study of 106 coronary patients who underwent percutaneous transluminal coronary angioplasty was conducted to explore the role of effort-reward imbalance in predicting coronary restenosis (Joksimovic et al, 1998). In addition a small-scale intervention study was performed to test the feasibility of a theory-based programme of stress reduction at work (Aust et al, 1997). In terms of health outcome measures, the majority of studies focused on cardiovascular risk factors or documented CHD. Other health indicators were psychiatric disorders, subjective health, reported symptoms, and sickness absence.

With regard to future incident CHD, effort-reward imbalance at work was associated with a two to six fold elevated relative risk compared to those who were free from chronic work stress (Bosma et al, 1998, Siegrist et al, 1990). This excess risk could not be explained by established biomedical and behavioral risk factors as these variables were taken into account in multivariate statistical analysis. Yet, additional evidence derived from cross-sectional investigations shows that chronic work stress in terms of effort-reward imbalance is associated with elevated risks of exhibiting high blood pressure, high level of atherogenic blood lipids or, in one study, elevated fibrinogen. Depending on the sample size, population characteristics and cardiovascular risk factor under study, respective odds ratios varied from 1.3 to 5.8 (Peter et al, 1998a, Siegrist 1997).

The adverse effects on health produced by high cost/low gain conditions at work are not restricted to cardiovascular health. In the Whitehall II study, the relative risk of exhibiting new psychiatric disorder, as assessed by the General Health Questionnaire, was 2.6 in men and 1.7 in women suffering from effort-reward imbalance at work (Stansfeld et al, 1999). Similarly, effort-reward imbalance predicted poor physical, psychological and social functioning after adjustment for the potential confounding effects of age, employment grade, baseline ill health, and negative affectivity in the same data set (Stansfeld et al, 1998). Another study found elevated reports of musculo-skeletal and gastro-intestinal symptoms, fatigue and sleep disturbances among bus and subway drivers who suffered from effort-reward imbalance at work (Peter et al, 1998b). In the middle managers study mentioned above, conditions of low occupational reward only, in the absence of signs of high effort (indicative of a passive state of coping associated with withdrawal behaviour), predicted short-term and long-term sickness absence (Peter and Siegrist 1998). Finally, the probability of experiencing a coronary restenosis was significantly increased in those treated coronary patients who exhibited a high level of work-related overcommitment (high intrinsic effort at work) (Joksimovic et al, 1998). Moreover, in the subsample of coronary patients who were still economically active, the ratio between high effort and low reward at work predicted restenosis after adjusting for relevant clinical variables. In conclusion, the results presented above support the effort-reward imbalance model as a distinct work-related psychosocial risk condition that can potentially provide a scientifically grounded basis for health promotion measures at work.

Comparison of models in predicting future CHD

In the Whitehall II study, a first attempt to compare the two models with respect to the prediction of future reports of CHD has been made. The results show that both effort-reward imbalance and low job control were independently related to CHD outcomes. There was a two-fold higher risk of developing new CHD when each model was controlled for the other and for potential confounders. These findings suggest the potential advantages in devising a job stress model that combines both personal and environmental factors to help explain differences in CHD and other diseases (Bosma et al, 1998).

close window


UCSF Home About UCSF Search UCSF UCSF Medical Center