Family climate and expressed emotion in the course of alcoholism.

Family-based predictors of relapse were examined in 100 alcoholics who participated in a 12-week treatment program with 6-month and 18-month followups. "Expressed Emotion" (EE), or attitudes of relatives toward the alcoholic as measured by the Camberwell Family Interview, scales measuring rejection of the alcoholic by relatives, and self-reports of partner interaction were evaluated as possible predictors of abstinence. During therapy, partnership interactions showed a transient deterioration with increased temporary friction. Based on conservative criteria, the abstinence rate was 40% at 6-month followup and 30% at 18-month followup. An association with the relapse at followup could be obtained for the Patient Rejection Scale (PRS) and, using empirically derived classification rules, for the main three variables of the Camberwell Family Interview (CFI) conducted on admission: "Critical Comments," "Emotional Overinvolvement," and "Warmth." A low number of Critical Comments and a high score in Warmth were associated with a lower risk of relapse; however, contrary to expectations, Emotional Overinvolvement of the significant other was associated with more abstinence. In addition, the number of Critical Comments made by relatives about the alcoholic, a major component of high EE as measured by the CFI, had a statistically significant impact on the "survival function" of abstinence, and thus contributed to the prediction of the course of alcoholism in the expected direction.