Community Reinforcement Approach

CRA Overview

The Community Reinforcement Approach (CRA) is a comprehensive behavioral program for treating substance-abuse problems. It is based on the belief that environmental contingencies can play a powerful role in encouraging or discouraging drinking or drug use. Consequently, it utilizes social, recreational, familial, and vocational reinforcers to assist consumers in the recovery process. Its goal is to make a sober lifestyle more rewarding than the use of substances. Oddly enough, however, while virtually every review of alcohol and drug treatment outcome research lists CRA among approaches with the strongest scientific evidence of efficacy, very few clinicians who treat consumers with addictions are familiar with it.

Scientific Support for CRA

The Community Reinforcement Approach (CRA) is a broad-spectrum behavioral program for treating substance abuse problems that has been empirically supported with inpatients (Azrin, 1976; Hunt & Azrin, 1973), outpatients (Azrin, Sisson, Meyers, & Godley, 1982; Mallams, Godley, Hall, & Meyers, 1982; Meyers & Miller, 2001), and homeless populations (Smith, Meyers, & Delaney, 1998). In addition, three recent meta-analytic reviews cited it as one of the most cost-effective alcohol treatment programs currently available (Finney & Monahan, 1996; Holder, Longbaugh, Miller, & Rubonis, 1991; Miller et al., 1995).

The first study to demonstrate the effectiveness of CRA was conducted more than 25 years ago (Hunt & Azrin, 1973). In this study, with 16 alcohol-dependent inpatients and matched controls, individuals were randomly assigned to either the CRA treatment or a traditional treatment program that focused on the 12 steps of Alcoholics Anonymous (AA). At the 6-month follow-up, participants in the CRA condition significantly outperformed the 12-step group with the former drinking an average of 14% of the follow-up days and the latter drinking 79% of the days. Significant differences in favor of CRA were also found for the number of days institutionalized and employed as well. A second related study of inpatients by Azrin (1976) again contrasted CRA with a 12 step program-this time with a larger sample. And, again, CRA proved to be superior at the 6-month follow-up, with the CRA group averaging 2% of their follow-up time drinking, as compared to 55% for the those participating in the 12-step program. As before, those treated with CRA were more likely also to report fewer days of institutionalization and more days employed. Importantly, the CRA condition's abstention rate at the 2-year follow-up continued to be very high at 90%.

Not only was the next study the first to be conducted with outpatients (n=43), but it also was the first to compare the effects of a disulfiram (Antabuse) compliance program within both CRA and 12-step programs (Azrin, Sisson, Meyers, & Godley, 1982). The disulfiram compliance component involved training a concerned family member or friend to administer the disulfiram to the drinker, and to provide verbal reinforcement. A third condition involved participation in a 12-step program (AA in this case) and a prescription for disulfiram, but it lacked the trained disulfiram monitor that was presumed to be an integral part of the disulfiram compliance protocol. As predicted, the two groups containing the additional disulfiram compliance component reported the most success during the sixth month of follow-up, with the CRA program averaging 97% and the 12-step condition reporting 74% of the days abstinent. It was noteworthy that couples assigned to the disulfiram condition within the 12-step group performed much better than the group's single subjects, even to the point of matching the CRA group's outcome on several variables. In contrast, the 12-step group that received only the disulfiram prescription had an abstinence rate of only 45% of the days.

In another study the effectiveness of CRA was examined within an alcohol-dependent population of homeless individuals (Smith, Meyers, & Delaney, 1998). During this 3-month program all participants were housed in grant-supported apartments. Individuals who were employed at the end of three months were allowed to remain in the apartments for an additional month. Housing privileges were suspended temporarily if random breathalyzer tests detected drinking. In contrast to those in the experimental group who were provided with housing, Standard Treatment group members had access to resources at a large day shelter, which primarily included basic meals, clothing, and showers. The shelter also offered a job program, individual sessions with AA-oriented counselors, and on-site AA meetings.

Participants in the CRA condition were treated in a group therapy format, and two weekly prizes were awarded for good attendance. The focus of most groups was skills training, primarily in the areas of problem solving, communication, and drink-refusal. Additionally, a non-drinking social event was held on Friday evenings, which commonly entailed a group dinner at a local restaurant. Periodically, group sessions were supplemented with relationship counseling or case management meetings. The latter were particularly important for the individuals with dual diagnoses (major mental illness + alcohol dependence) within the sample. At the conclusion of treatment, follow-ups were conducted every 2-3 months for the next year. The results for this trial showed that compared with standard care at the shelter, those treated with CRA showed significantly better outcomes throughout a year of follow-up (Smith et al., 1998).

With respect to the use of CRA with substances other than alcohol, a series of studies have demonstrated that the combination of CRA and contingency management is an excellent program for treating cocaine and methadone-maintained heroin abusers. Contingency management for cocaine abusers entails monitoring cocaine use through frequent urine samples, and rewarding individuals who turned in clean urines with tokens that could be traded for prizes. Studies have ranged from a 2-case design (Budney, Higgins, Delaney, Kent, & Bickel, 1991) to a controlled but nonrandomized trial (Higgins, et al., 1991), and finally to a controlled and randomized experiment (Higgins, et al., 1993). In both controlled trials the CRA plus contingency management group participants decreased their cocaine use significantly more than those in the 12-step (AA philosophy) comparison condition. Figures for the third study showed that the CRA plus contingency management group had fewer program dropouts than the control condition (5% versus 42%) and a greater number of continuous days of cocaine abstinence throughout. Continuous cocaine abstinence was still found 16 weeks into the trial for 42% of the CRA group and only 5% of the 12-step group. CRA has also been extended to the treatment of methadone-maintained heroin addicts, again with an advantage for the CRA-treated clients (Abbott, Weller, Delaney, & Moore, 1998).