Bridging Disciplinary Boundaries (January 11 - 14, 2007)

For more than three decades, substance abuse clinicians and researchers have sought to discern the most effective methods for treating women alcoholics and drug addicts. Although extensive resources have been devoted to predicting outcomes, relatively little attention has been paid to the received content and lived experiences of patients within actual treatment settings. Yet mounting evidence suggests that, within the “black box” of treatment, patients gather information and engage in interactions that allow them to initiate processes of change and lay foundations for new behaviors (Simpson, 2004).

This paper reports findings from a secondary content analysis of one of the few studies that observed residential substance abuse services from “inside the black box.” For six months in 1995-96, a team of ethnographers from the Alcohol Research Group (ARG) in Berkeley conducted an intensive process evaluation of three residential treatment centers in northern California (Kaskutas et al., 1998). Those consisted of a hospital-based, mixed-gender unit (N=144) and two freestanding programs (N=63; N=32) delivering the California social model, a peer-based approach wherein patients are treated in gender-separate settings with minimal professional infrastructure (Borkman, 1984; Shaw & Borkman, 1990).

This secondary analysis examines the ARG data under a fresh lens. Whereas the ARG study omitted coding and analysis by gender, the current study focuses on women's experiences in the medical and social model programs to answer several previously-unasked questions: Were client motivation and engagement similar or different across settings?; How did the content delivered by staff members compare across models?; and finally, Did women clients' relationships with peers and staff vary between settings and/or models?

While the original research utilized grounded theory and ethnographic methods, this study applied a modern derivation of the stages of change theory (Prochaska & DiClemente, 1982) known as the TCU Treatment Process Model (Simpson, 2000; Simpson & Joe, 2004). Using QSR N6 and adopting the TCU framework, the researcher created codes to capture client motivation, program attributes, and patient-staff interactions. Those were analyzed alongside over 300 extant codes from the ARG team's original codebook.

Through multiple iterations of code development and refinement, themes portraying patient readiness and program content gradually emerged. Although program attributes varied little between settings, clients in the social model clustered further along the stages of change continuum than medical model patients did. That is, medical model patients remained primarily in the precontemplation and contemplation stages while social model residents demonstrated more preparation and action processes (Prochaska et al., 1992). Finally, the study found that social model residents reported more meaningful relationships with both staff and peers than did women on the medical model unit.

These findings have implications for social work research and practice with substance-misusing women. First, they buttress previous research that found a correlation between positive therapeutic relationships and patients' treatment success (Luborksky et al., 1997). Next, they illustrate salient themes according to recipients of mixed- and single-gender women's treatment. Finally, the findings suggest directions for practitioners who seek to improve both the content and quality of their therapeutic alliance with clients.