The purpose of this study was to examine wives' anxiety in response to their husbands' first myocardial infarction from the perspective of perceptions and interpretations of their support roles, their husbands' experience, their abilities to act supportively, and how these and other factors contributed to their degree of anxiety. Major study variables were: role clarity (measured by an instrument developed for the study), empathy (measured by the Barrett-Lennard [1978] Relationship Inventory), support role performance (measured by an instrument developed for the study), and anxiety (measured by the State-Trait Anxiety Inventory Form A [Spielberger, Gorsuch, & Lushene, 1970]). Four exogenous variables, husband's condition, previous experience in the support role, selfesteem, and trait anxiety, were used to test alternate hypotheses. Subjects were 82 wives of men admitted to three cardiac care units. Data were collected prior to the husband's hospital discharge and at 3 weeks postdischarge. Data were analyzed with path analysis procedures. Study findings supported two hypotheses, that support role performance has a direct negative effect on anxiety and trait anxiety has a direct positive effect on anxiety. Descriptive data obtained during the postdischarge interview provided documentation of uncertainty as another source of anxiety.