Date this summary statement was written:

Date this evidence summary was written:
November 2012
The community reinforcement approach in addictions: Evidence and implications
for public health

Review on which this evidence summary is based:
Roozen, H.G., Jiska, J.B., van Tulder, M.W., van den Brink, W., De Jong, C.A. & Kerkhof, A.J. (2004). A systematic review of the effectiveness of the
community reinforcement approach in alcohol, cocaine and opioid addiction. Drug and Alcohol Dependence, 74, 1-13.
Review Focus
Adults (16-65 years of age) with alcohol, cocaine and opiate abuse or dependence Community reinforcement approach (CRA), a biopsychosocial multifaceted approach with emphasis on environmental events and social influences Usual care, or CRA plus contingency management, or CRA plus pharmacological support Primary Outcomes: continuous abstinence (urine/blood samples or self reports), abstinence percentages Secondary Outcomes: addiction severity, frequency of substance use, time to relapse Review Quality Rating: 9 (strong) Details on the methodological quality are availabl
Considerations for Public Health Practice
Conclusions from Health Evidence
General Implications
This is a well-done review based on single studies of high methodological quality. Studies offer evidence supporting the • CRA should be used to reduce drinking days with efficacy of CRA with or without medication or contingency alcohol abusers, including both the general management, in various substance-related disorders, including population and homeless populations, and should be alcohol, cocaine and heroin support. There is evidence, based implemented with medication for continuous on a limited number of studies to suggest that: abstinence or reducing number of drinking days in • CRA reduces number of drinking days and is more effective than usual care for the general population and the homeless, • CRA with abstinent-contingent incentives should be and is more effective with medication for number of drinking used to promote cocaine abstinence and should also be considered for opioid detoxification/methadone • CRA with abstinent-contingent incentives is more effective than either usual care or non-contingent incentives for Public health decision makers should be aware that for • CRA with incentives is more effective than usual care in a number of interventions, very limited evidence (e.g. 1 study) is currently available to inform decision making. Evidence and Implications
What’s the evidence?
Implications for practice and policy
1. Alcohol Treatment
1. Alcohol Treatment
Single CRA versus usual care (3 studies, 359 participants)
• Practitioners should consider use of CRA to reduce • Average number of drinking days with CRA was -0.94 the number of drinking days, but should use caution lower than for usual care (CI−1.60 to−0.27) (1 study of and evaluate CRA as an intervention if interested in general population, 1 study in the homeless) • Conflicting evidence with regard to continuous abstinence • Practitioners should consider using CRA to treat in any population and impact is unclear at this time. alcohol abuse in the homeless, while acknowledging CRA with disulfiram versus usual care with disulfiram (3
that positive findings are limited to a single study. studies, 298 participants)
• Practitioners should not rely on CRA with disulfiram • Unclear impact on terms of continuous abstinence or to reduce number of drinking days or increase number of drinking days with CRA plus disulfiram is more effective than usual care plus disulfiram. 2. Cocaine Treatment
2. Cocaine Treatment
CRA with abstinence-contingent ‘incentives’ versus usual
• Practitioners should consider using CRA with care for cocaine treatment (2 studies, 63 participants)
abstinence-contingent incentives to promote cocaine • CRA with ‘incentives’ was more effective with regard to abstinence, rather CRA with non-contingent cocaine abstinence than usual care, both with program incentives, while acknowledging that positive findings duration of 4 weeks or less (RR 3.75, 95% CI 1.79–7.87),
and for program duration between 4 and 16 weeks (RR
• It remains unclear whether CRA alone is more 5.09,95% CI = 1.63–15.86).
CRA with abstinence-contingent ‘incentives’ versus CRA
(non-contingent incentives) (2 RCTs, 110 participants)
• CRA with abstinence-contingent ‘incentives’ was more effective than single CRA (non-contingent incentives)
treatment aimed at cocaine abstinence, for programs with
a duration of 4–16 weeks (Effect Size 1.73, 95% CI 1.04–
2.88).
• Studies in cocaine treatment do not address CRA alone, so it is not clear whether CRA alone is more effective than usual care 3. Opioid Treatment
3. Opioid Treatment
CRA with ‘incentives’ versus usual care in a detoxification
• Practitioners should consider CRA with ‘incentives’ program for opioid treatment (1 RCT, 39 participants)
for opioid detoxification, while acknowledging that • Limited evidence that CRA with ‘incentives’ is more positive findings are limited to a single study. effective than usual care in a detoxification program. Although evidence for opioid treatment did not Contingent vs. non-contingent incentives were not specify whether incentives were contingent vs. non- contingent, practitioners may want to consider that Single CRA versus usual care in a methadone
abstinence-contingent incentives were more effective maintenance program (1 RCT, 180 participants)
• Single CRA is more effective than usual care in a • Practitioners should consider CRA alone for methadone maintenance, while acknowledging that positive findings are limited to a single study. Legend: P – Population; I – Intervention; C – Comparison group; O – Outcomes; CI – Confidence Interval; OR – Odds Ratio; RR – Relative Risk
**For definitions please see the healthevidence.org glossar

Why this issue is of interest to public health in Canada
According to the Centre for Addiction and Mental Health, in 2002, 2.6% of Canadians were dependent on alcohol and fewer than 1% were dependent on
illegal drugs, however more than 25% of all Canadians are considered high-risk drinkers.1 Total direct social costs were approximately $7.5 billion in 2002 for
alcohol use and $3.5 billion for illicit drug use.2 Addictions take the form of psychological or physical dependence.2 Harms of substance use include: injury
while under the influence, mental illness including depression and anxiety, serious medical complications, as well as social and relationship issues. 2
Herie, M., Godden, T., Shenfeld, J., & Kelly, C. (2010) Addiction: An information guide. A guide for people with addiction and their families. Retrieved from Thomas, G., & Davis, C.G. (2007). Comparing the perceived seriousness and actual costs of substance abuse in Canada: Analysis drawn from the 2004 Canadian Addiction Survey. Ottawa: Canadian Centre on Substance Abuse. Retrieved from Other quality reviews on this topic are available on

Suggested citation
Graham, K., & Dobbins, M. (2012). The community reinforcement approach in addictions: Evidence and implications for public health. Hamilton, ON: McMaster University. Retrieved from This evidence summary was written to condense the work of the authors of the review referenced on page one. The intent of this summary is to provide an overview of the findings and implications of the full review. For more information on individual studies included in the review, please see the review itself. The opinion and ideas contained in this document are those of the evidence summary author(s) and healthevidence.org. They do not necessarily reflect or represent the views of the author’s employer or other contracting organizations. Links from this site to other sites are presented as a convenience to healthevidence.org internet users. Healthevidence.org does not endorse nor accept any responsibility for the content found at these sites. Production of this evidence summary has been made possible through a financial contribution from Health Canada to the Canadian Centre on Substance Abuse (CCSA). The views expressed herein do not necessarily represent the views of Health Canada or CCSA

Source: http://www.healthevidence.org/documents/byid/17013/Roozen2004_EvidenceSummary_EN.pdf