Social class differences in mortality and morbidity from a wide range of diseases persist in the USA, UK and other industrialized countries. We wish to determine the extent to which psychosocial factors at work and outside account for these social class differences. The overall aim is to study: a) the effect on health and disease of work environment - psychological workload, control over work pacing and content, opportunity for use of skills, social support at work, b) the effect on health and disease of social supports - both direct and moderating effects, and c) the interaction between these psychosocial factors and other established risk factors in the aetiology of chronic disease. In the British Civil Service there is an unexplained threefold higher mortality from cardiovascular and other disease in the lowest compared to the highest employment grade. In a new study to investigate this, 10,575 male and female civil servants have been studied cross-sectionally, by validated assessments of work characteristics, social supports, and psychiatric symptoms, in addition to detailed information on health behavior, health status and cardiovascular risk factors and disease. Longitudinal data on incidence of cardiovascular and other disease, on sickness absence patterns, on changes in psychiatric and other symptoms, and on mortality rates will be collected by: i) a repeat questionnaire to 10,575 participants; ii) continued collection of sickness absence data; iii) obtaining medical diagnostic data from GPs on long-stay-sickness absences; iv) obtaining death certificates and cancer registrations. A particular feature of the study is that characteristics of each person's job, in addition to their own subjective reports, has been assessed objectively by personnel managers. Longitudinal study will help determine if subjectively experienced, or objectively determined 'job strain' are more important predictors of ill- health. A particular exciting finding to emerge from the cross-sectional data, that will be tested longitudinally, is that low employment status is related to higher thrombotic tendency through an effect of 'low control over work' on fibrinogen levels. This may be an important pathway by which psycho-social factors are related to cardiovascular disease.