Psychosocial factors are increasingly recognized as risk indicators for coronary artery disease (CAD) prognosis and they are likely interrelated. The objective of this study is to simultaneously test the relationship between key psychosocial constructs as independent factor scores and recurrent events in CAD patients. There were 1268 CAD outpatients of 97 cardiologists surveyed at two points. Recurrent events or hospitalization in the intervening nine months were reported. Factor analysis of items from the Hospital Anxiety and Depression Scale, Perceived Stress Scale, the ENRICHD Social Support Inventory, and Hostile Attitudes Scale was performed to generate orthogonal factor scores. With adjustment for prognostic variables, logistic regression analysis was performed to examine the relationship between these factor scores and recurrent events. Factor analysis resulted in a six-factor solution: hostility, stress, anxiety, depressive symptoms, support, and resilience. Logistic regression revealed that functional status and anxiety, with a trend for depressive symptoms, were related to experiencing a recurrent event. In this simultaneous test of psychosocial constructs hypothesized to relate to cardiac prognosis, anxiety may be a particularly hazardous psychosocial factor. While replication is warranted, efforts to investigate the potential benefits of screening and to investigate treatments are needed.