Client Perceptions of Self and Treatment as Predictors of Therapeutic Alliance

Background and Purpose: Research studies demonstrate that therapeutic alliance is a significant contributor to engagement and retention in treatment and that clients' alliance ratings have a strong association with treatment outcomes. Substance abuse treatment research often focuses on alliance as it relates to client outcomes, assessment variables or diagnoses, but not as often on client perspectives of themselves and the treatment process. The purpose of this study was to determine the extent that client ratings on self and treatment measures were significant predictors of reported therapeutic alliance with the treatment counselor. Understanding client perception of these factors may assist in developing methods to promote stronger alliance as well as identify other important factors that influence the development of therapeutic alliance.

Method: This study utilized a cross-sectional design. Participants included a convenience sample of adult outpatient substance abuse treatment clients (n = 3,123) receiving services at thirty-two outpatient clinics during a one week period in 2007. Client surveys were administered at the beginning of the treatment session and included domain subscales from the Client Evaluation of Self in Treatment – CEST (Simpson, 2004) to assess Treatment Motivation (9-item Problem Recognition; 6-item Desire for Help), Social Functioning (8-item Childhood Problems; 8-item Hostility), and Therapeutic Engagement (7-item treatment satisfaction; 12-item treatment participation; 5-item peer support; 9-item social support); a one-item treatment readiness ruler; and the 12-item Working Alliance Inventory Short Form Revised – WAI-SR (Hatcher & Gillaspy, 2006). Multiple regression (entry method) was employed to determine if the CEST subscales and readiness were significant predictors of working alliance ratings.

Results: A significant model emerged (F 9,3,112=244.993; p<=05; R2adj=.413). Significant variables included Treatment Satisfaction (B=.321; p<=05), Peer Support (B=.178; p<=05), Treatment Participation (B=.169; p<=05) and Readiness for Treatment (B=.164; p<=05). No other subscales achieved significance. Treatment Satisfaction accounted for the largest amount of variance in the model, which included more items related to organizational (e.g. session schedules are convenient, program is run well, location is convenient) rather than client or counselor characteristics.

Conclusions and Implications: This study provides insight into client reported factors that contribute to the treatment therapeutic alliance. Findings confirm the importance of client motivation, readiness, and engagement, but also reveal the more unexpected role that organizational factors and support from peers in treatment contribute to the therapeutic alliance. Results suggest it is important for practitioners and researchers to consider factors beyond the client and counselor relationship when assessing therapeutic alliance and its affects on client treatment and outcomes.