Four broad classes of dependent variables (psychological strain, physical illness symptoms, health‐related behaviour and social participation) were associated with eleven categories of stressors and stress moderators from work and family life, using multiple logistic regression analysis for a random sample of 8700 full‐time male and female members of T.C.O., a major Swedish white‐collar labour federation (covering 25 per cent of the Swedish labour force). Our goal was to find broad patterns of associations by comparing relative magnitudes of effects for (a) stressors and stress moderators; (b) work and family activities, and (c) males and females. Fifty per cent of the associations between environmental factors and dependent variables were significant in the predicted direction at the 5 per cent level. However, only 5 per cent of the associations are as strong, for example, as average smoking/heart disease associations. Our primary conclusion is that job factors are the next strongest set of predictors of health and behaviour after age. Job factors are stronger than family factors for both men and women; proportionally increasing the explained variance by over 60 per cent versus roughly 20 per cent for family factors (over the 25 per cent of explanation due to demographic factors). The overall pattern of stressor/outcome associations is quite similar for men and women, although both job/outcome and family burden/outcome associations are stronger for women than for men. We failed to find a clear linkage between particular stressors and particular physical illnesses. Among the job factors, control and work load have the strongest associations; with the former predicting behaviour patterns and job satisfaction (along with social support), and the latter predicting mental strain symptoms. Family problems are associated with increased health risks (stronger for men) and family responsibilities and constraints affect health behaviour (stronger for women). Job satisfaction is the most successfully predicted outcome in the study, and is similarly affected for men and women.

A research project is outlined in which concepts and methods from social psychology and psychophysiology are integrated in the study of human adaptation to underload and overload related to technically advanced work processes. Attempts are made to identify aversive factors in the work process by studying acute stress reactions, e.g., catecholamine excretion, in the course of work and relating these to long term, negative effects on well-being, job satisfaction and health. Data from a pilot study of sawmill workers support the view that machine-paced work characterized by a short work cycle and lack of control over the work process constitutes a threat to health and well being.

This paper discusses research strategies and summarizes pertinent knowledge that has influenced Scandinavian legislation and trade union policies in the field of stress at the workplace. Work stress is shown to be problematic in two ways. First, there may be a direct relation between certain objective working conditions, physiological and psychological stress, and ill health. Second, certain stress conditions may create fatigue and/or passivity and thus make it more difficult for individuals to actively involve themselves in changing those working conditions-including physical and chemical risk factors-that may be detrimental to health. This latter aspect is especially relevant for prevention on the systems level, where one has to rely on collective means for changes in technology and the organization of work.