Abstract
We investigated the effect of family involvement on abstention of 90 female alcoholics in recovery processes by multivariate analysis. The abstinence rate of the family-involved group was significantly higher than that of the group not involved (p <. 05). The logistic regression model in which family involvement, age, marital status, eligibility for welfare assistance, and history of previous psychiatric inpatient care were used as 5.19; p <. 05; summary estimates of odds ratio=1.65 with a 95% confidence interval of 1.18–6.93).