Pergamon Addictive Behaviors. Vol. 20, No. 3, 395-405, pp. 1995 Copyright 0 1995 Elsevier Science Ltd Printed in the USA. All rights zyxwvutsrqpon reserved 0306-4603/95 $9.50+ .OO 03of&4603(94)ooo80-8 THE RELATIONSHIP OF COUNSELOR AND PEER ALLIANCE TO DRUG USE AND HIV RISK BEHAVIORS IN A SIX-MONTH METHADONE DETOXIFICATION PROGRAM SANDRA L. TUNIS, KEVIN L. DELUCCHI, KIM SCHWARTZ, PETER BANYS, and KAREN L. SEES University of California, San Francisco. and San Francisco Veterans Affairs Medical Center Abstract - The purpose of this study was to examine the relationship of treatment out- comes in opioid detoxification to levels of counselor and peer alliance. Forty-one sub.jects were recruited from a larger, 180-day study of psychosocial treatment. Beginning at day 90. subjects completed monthly measures of alliance. Outcome measures included treatment retention, drug use, and self-reported HIV risk. Measures of alliance were found to be internally consistent and moderately stable over time. During the final 30 days of the metha- done taper, higher levels of both types of alliance were associated with less use of illicit opioids. Alliance with counselor was associated with less frequent needle sharing. For subjects who could be located for 30-day follow-up, greater alliance with peers was associ- ated with more frequent HIV (sexual) risk behaviors. Results suggest that treatment out- come may be improved through approaches that address a patient’s alliance with both counselor and peers. Therapeutic alliance, defined as the commitment and capacity of both treatment provider and patient to engage in the working strategies of treatment, has been a major topic of psychotherapy research (cf. Allen, Deering, Buskir, & Coyne, 1988; Foreman & Marmar, 1985; Marmar, 1990). The development of alliance involves establishing a trusting collaborative relationship (Marmar, Gaston, Gallagher, & Thompson, 1989) and appears more important to the therapeutic process than are factors associated with the qualities of either the therapist or the patient, including the patient’s initial level of symptomatology (cf. Gaston, Marmar, Thompson, & Gallagher, 1988; Hartley, 1985; Millman, 1986). Therapeutic alliance has been shown to predict outcome (i.e., symptom reduction) for individuals participating in several different types of therapy. In a sample of elderly, depressed patients in brief dynamic, cognitive, or behavioral therapy, alli- ance (especially a patient’s commitment and working capacity) was found to be important, particularly during cognitive therapy (Gaston, Marmar, Gallagher. & Thompson, 1989). Despite this work, there are few studies focusing on alliance or on other aspects of the counseling context in substance abuse treatment. McLellan and colleagues (1993) recently reported that the addition of basic counseling to methadone treatment alone significantly improved efficacy in a methadone maintenance program, although the question of the “key ingredient” (e.g., alliance, perceived support, skill of therapist) in the counseling experience was left unexplored. While important differences exist between psychotherapy and drug treatment (e.g., resolution of psychological conflict This pilot study was funded by the San Francisco Treatment Research Unit at the Department of Veterans Affairs Medical Center through grant number IRl8 DA06097 from the National Institute on Drug Abuse (James L. Sorensen, Ph.D., P.I.). We wish to thank Andrew Therber for his valuable assistance in data entry and analysis. Requests for reprints should be sent to Dr. Sandra Tunis. National Council on Crime and Delinquency, 685 Market Street, Suite 620, San Francisco, CA 94105. 395