The extent to which the associations between psychosocial factors and coronary heart disease (CHD) are causal has seldom been the subject of systematic enquiry. We are updating our previous systematic review up to 2001. The objective of this study is to assess the relative strength of the epidemiological evidence for causal links between psychosocial factors and CHD incidence among healthy populations, and prognosis among CHD patients. Our methods were to systematically review prospective cohort studies identified through the Science Citation index, which met pre-specified quality criteria. We found that the proportion of etiologic studies reporting a strong or moderate association was: 6/18 for Type A behavior and hostility, 15/22 for depression, 4/8 for anxiety, 10/13 for psychosocial work characteristics and 6/9 for social support. For prognostic studies the proportions were: 2/15 for type A behavior and hostility, 18/34 for depression, 8/18 for anxiety, 2/4 for psychosocial work characteristics and 14/21 for social support. Positive studies were more likely to be cited by other papers than negative studies. We concluded that, based on prospective epidemiological data, there was evidence for an association between depression, social support and psychosocial work characteristics and CHD aetiology and prognosis. Evidence for an effect of anxiety or type A behavior was less consistent. Methods to address bias in the reporting of psychosocial data are required.