Women who have experienced depression are vulnerable to future depressive episodes as well as emotional, cognitive, and interpersonal difficulties. Several theoretical models have explored the link between maladaptive anger expression and women's depression (Jack, 1991; Davanloo, 1980; Cox, Bruckner, & Stabb, 2003), but the existing research examining this relationship has primarily relied on self-report methods. A growing area of research has begun assessing women's communication styles in the context of interactions with their romantic partners. This field of observational research has studied communication patterns in couples with marital conflict, but no studies have yet examined anger expression in women with a history of depression. Therefore, the present study examined anger expression in 26 formerly-depressed (FD) and 30 never-depressed (ND) women and their partners. Couples were a part of a larger community study investigating cognitive and interpersonal factors in depression. Participants completed the Structured Clinical Interview for DSM-IV (SCID; Spitzer et al., 1988) on the telephone to screen for eligibility and determine FD/ND group status. Women and their partners completed self-report measures of depressive symptoms, emotional expression, relationship conflict, and relationship intimacy. Couples also completed an observational interaction task to assess patterns of communication. An observational coding system was developed in collaboration with Deborah Jacobvitz, Ph.D. at the University of Texas at Austin to assess couples' direct anger expression, hostility, and emotional attunement. The results indicated that contrary to predictions, women expressed more direct anger but similar levels of hostility compared to their male partners. Second, findings supported the prediction that self-reported emotional expression would be associated with relationship intimacy and inversely related to relationship conflict. However, the observational patterns of direct anger, hostility, and emotional attunement were not associated with the relationship outcomes. Results also indicated that FD and ND women did not differ in their patterns of direct anger and hostility expression during the observational interaction task. Exploratory analyses found that individuals' behaviors exhibited during the interaction task were consistent with self-reported ratings of these behaviors. Finally, exploratory analyses also indicated that individuals' patterns of behavior such as hostility impacted their partners' perceptions of the quality of their relationships. / text