Demand-withdraw couple interaction, disease-model beliefs, and readiness to change problem drinking

Abstract

This study examined relationships among couple demand-withdraw interaction (DWI), alcoholics' and partners' beliefs in the disease model of alcoholism, and alcoholics' readiness to change problem drinking. A secondary purpose was to further investigate the construct validity of the University of Rhode Island Change Assessment Scale (URICA; McConnaughy, Prochaska, & Velicer, 1983), a frequently used measure of readiness to change. A sample of 63 male alcoholics and their female partners received up to 20 sessions of cognitive-behavioral or family-systems therapy for alcoholism as part of a larger treatment project. The three main construct variables were assessed at baseline (T1) and after the first 12 sessions of therapy (T2) by observational ratings (DWI) and self-report questionnaires (disease-model beliefs and readiness to change). Due to a high rate of attrition from measurement at T2, analyses involving data, collected at this point must be interpreted with caution. As predicted, T1 ratings of DWI were negatively associated with alcoholics' readiness to change measured concurrently at T1 and also longitudinally at T2. Also as predicted, this relationship was found primarily with the wife-demand/husband-withdraw DWI role pattern; the opposite husband-demand/wife-withdraw role pattern was largely unrelated to readiness to change. Multiple regression analyses in a panel design indicated no clear causal direction in the lagged correlations between DWI and readiness to change. Although female partners believed more strongly in the disease model of alcoholism than did alcoholics both at T1 and at T2, neither partners' nor alcoholics' disease-model beliefs were related to readiness to change or to DWI. Finally, regarding the construct validity of the URICA, while mean levels of readiness to change increased significantly over the course of treatment, T1 and T2 measures of readiness to change were only mildly---and somewhat inconsistently---associated with treatment outcome. In all, the findings of the current study offer moderate support for the idea that an individual's readiness to change is related to interpersonal factors in his environment but only limited support for the construct validity of the URICA as a measure of readiness to change problem drinking.