To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Does AA work? That’s (in part) up to you

Clinician support improves Alcoholics Anonymous’ success rate.

Vol. 4, No. 5 / May 2005

Many clinicians refer alcohol-dependent patients to Alcoholics Anonymous, but how effective is AA in reducing drinking? Evidence is elusive—partly because of AA’s tradition of anonymity—but clinician encouragement is among three variables that appear essential to successful AA use.

The pioneer of twelve-step programs (Box 1),1,2 AA is a widely accessible, free adjunct to professional alcohol abuse treatment. This article describes the evidence supporting AA’s efficacy for reducing drinking among persons with alcohol use disorders. We also recommend referral strategies that increase AA participation and discuss special needs of alcohol-dependent patients with comorbid psychiatric disorders.

Box 1

AA: Mutual help for alcohol abstinence

Alcoholics Anonymous (AA) was founded in 1935 in Akron, OH, by Bill Wilson and Robert Smith, MD, two professionals struggling with alcohol dependence. They joined together to help each other stop drinking.

Their success inspired them to help others, and this mutual-aid society grew under the name Alcoholics Anonymous. AA redefined the then-prevalent view of alcohol dependence as a moral failing, instead conceptualizing it as a disease that can be arrested—but not cured—by alcohol abstinence. The only requirements for AA membership are “a desire to stop drinking,” a respect for maintaining anonymity, and a desire to join a fellowship of mutual support for the goal of abstaining from alcohol.

In 1939, Wilson described AA’s theory and fellowship methods and defined its twelve steps of personal recovery,1 referred to in AA as “The Big Book.” Today, the worldwide fellowship has more than 2 million members, with a male:female ratio of 2:1.2


Besides clinician encouragement, two patient variables—severity of alcohol dependence and self-efficacy—have been associated with successful AA use.

Clinician encouragement. When psychiatrists and other clinicians encourage alcohol-dependent patients to attend AA, the rate of AA use increases and drinking decreases.3-7 When clinicians remain interested in alcohol-dependent patients’ AA use—rather than simply recommending AA—patients are more likely to follow through with a referral and to participate long enough to obtain benefit.

Clinicians play an important role in helping patients benefit from AA referral (Box 2).6 Patients who are personally helped to an AA group will rapidly attend, whereas those only given referral information are unlikely to follow up.7 Moreover, clinicians who encourage patients to attend AA and follow up on that attendance have more patients who attend and participate in AA than do less proactive clinicians.8

Box 2

How to help patients benefit from AA

Disseminate information about alcohol dependence self-help groups such as AA

Become knowledgeable about local AA options to facilitate referral and cooperation

Invite AA groups to use your institutional or clinic space to hold groups and meetings

Offer appropriate self-help referrals to family members, such as Al-Anon and Al-Ateen family groups

Try to match patient preferences with local AA groups, such as women’s AA, and young people’s AA meetings

Use AA as an adjunct to professional care, rather than stand-alone treatment

Learn about alternatives to 12-step treatments—such as SMART Recovery (a CBT-based treatment), Secular Organization for Sobriety, or Women for Sobriety—for patients who prefer other self-help options

Source: Adapted from Workgroup on Substance Abuse Self-Help Organizations’ expert consensus statement, reference 6.

Dependence severity. The more severe a patient’s alcohol dependence, the more likely he or she will attend and participate in AA.4,5,9,10 This finding suggests that persons most severely impaired by alcohol dependence are most likely to accept that they need help.

Self-efficacy. Believing that one can abstain from drinking is associated with being able to reduce one’s drinking. Alcohol-dependent patients with self-efficacy are more likely to use AA, and this trait is believed to be a component of the change process associated with reduced drinking.11,12


Randomized, controlled trials (RCTs) may be the “gold standard” for determining any treatment’s efficacy, but constructing an RCT of AA use is complicated. AA does not engage in or support research, which makes AA use difficult to administer as a controlled variable in clinical trials. Also, the variability of AA groups and environments confounds the interpretation of study results and limits their application to populations at large.13

Even so, AA can be effective as an adjunct to professional treatments—such as detoxification—and as aftercare to maintain reduced drinking:

  • Kelly’s14 comprehensive review and critical analysis of the main studies through 2003 showed a correlation between professional treatment plus AA attendance and improved outcomes.
  • Project MATCH-—a large RCT—also supports AA’s benefits, as reported in smaller, less rigorously controlled studies.4,5

Three meta-analyses—using statistical analyses to pool and integrate data from smaller individual studies of AA use—arrived at somewhat different conclusions:

  • Kownacki and Shadish15 found poorer 12-month drinking outcomes with AA alone, compared with other treatments or no treatment. Their review assessed attendance as a predictor of alcohol use outcomes, and most subjects were attending AA under court orders. Because AA’s philosophy stresses that members must desire to stop drinking (Table 1), these study results may not apply to the voluntary, motivated individuals who usually use AA.
  • Two other meta-analyses9,10 that included nonrandomized studies and RCTs showed that attending AA has modest, favorable effects in reducing drinking and improving psychosocial functioning.

Tonigan et al10 rated 74 studies of alcohol use disorders that measured patients’ affiliation with AA and drinking outcome and/or psychosocial adjustment. Most were cross-sectional inpatient studies; few used collateral interviews or biological measures to confirm self-report data. Most used poor methodology, and their assessments were not psychometrically validated.

Outpatient studies showed positive correlations between AA attendance/involvement and:

  • reduced drinking
  • improved psychological adjustment and improved family relationships.

Outcomes were much more heterogeneous in inpatient samples, and no predictive relationships were seen between AA use and drinking outcome or psychosocial adjustment.

All three meta-analyses excluded studies of alcohol-dependent individuals with co-occurring drug use disorders, leaving unaddressed the effect of adjunctive AA in that population.

Table 1

AA’s Twelve Steps of personal recovery

  1. We admitted we were powerless over alcohol—that our lives had become unmanageable.
  2. Came to believe that a Power greater than ourselves could restore us to sanity.
  3. Made a decision to turn our will and our lives over to the care of God as we understood Him.
  4. Made a searching and fearless moral inventory of ourselves.
  5. Admitted to God, to ourselves, and to another human being the exact nature of our wrongs.
  6. Were entirely ready to have God remove all these defects of character.
  7. Humbly asked Him to remove our shortcomings.
  8. Made a list of all persons we had harmed and became willing to make amends to them all.
  9. Made direct amends to such people wherever possible, except when to do so would injure them or others.
  10. Continued to take personal inventory and when we were wrong promptly admitted it.
  11. Sought through prayer and meditation to improve our conscious contact with God, as we understood Him, praying only for knowledge of His will for us and the power to carry that out.
  12. Having had a spiritual awakening as the result of these Steps, we tried to carry this message to alcoholics and to practice these principles in all our affairs.

Source: Reprinted with permission, Alcoholics Anonymous World Services (AAWS), Inc. 2005. Permission to reprint does not mean that AAWS has reviewed or approved the contents of this publication, or that AA necessarily agrees with the views expressed herein.

Project MATCH. Although not specifically an AA study, Project MATCH16 examined the efficacy of twelve-step, individual therapy in treating alcohol abuse or dependence. This multi-site RCT of 1,726 patients compared 12 sessions of a manual-driven, twelve-step facilitation (TSF) with two other empirically-supported, standardized, individual alcohol dependence treatments:

  • 12 sessions of cognitive-behavioral coping skills (CBT)
  • 4 sessions of motivational enhancement therapy (MET).

TSF encouraged and monitored AA attendance, whereas CBT and MET neither encouraged nor discouraged AA use. Although all three treatment groups were drinking less at 1-and 3-years follow-up, TSF was particularly effective for persons with severe alcohol dependence and low psychiatric severity.4,5,17 After treatment, those in the TSF group participated in AA more than did individuals in the CBT and MET groups.

Project MATCH thus found that a twelve-step individual therapy was at least as effective as CBT and MET in reducing post-treatment drinking and maintaining abstinence.

Real-world efficacy. A naturalistic study that resembled Project MATCH enrolled 3,018 male military veterans with substance use disorders across 15 program sites.18 Drug dependence was not excluded, and 51% of the men had co-occurring alcohol and drug dependence.

Participants received inpatient detoxification, followed by 21 to 28 days of intensive twelve-step treatment, CBT, or both. Alcohol and drug abuse declined equally in all three groups, and subjects were referred for outpatient aftercare and self-help groups. After 1 year, involvement in a self-help group predicted better outcome, regardless of the initial treatment.19

Unlike earlier studies,15,20,21 this trial found that individuals with co-occurring psychiatric disorders and those legally mandated to get treatment did as well at 1-year follow-up as those without these variables. It provides additional evidence that twelve-step treatments can reduce substance use across varied populations, including patients with co-occurring alcohol and drug dependence.


Twelve-step programs appear most effective for individuals who actively participate. This may seem obvious, but most studies of 12-step treatments have monitored meeting attendance rather than engagement. Most studies that separate these variables report that active participation—not passive attendance—correlates with reduced substance use.

Montgomery et al22 followed 66 alcohol-dependent individuals in a 12-step oriented, 28-day residential treatment program and investigated:

  • relationships between AA attendance and participation
  • drinking outcomes over 31 weeks after treatment.

Although AA attendance did not predict outcomes, active involvement in the 12 steps predicted reduced drinking and more-favorable perceptions about life having purpose or meaning (as opposed to despair that life has become meaningless as a consequence of alcohol dependence).

Similarly, Project MATCH participants involved in AA during the first 6 months after treatment had more-frequent abstinent days in the 7 to 12 months after treatment. AA “involvement” included identifying oneself as an AA member, working the steps, having an AA sponsor, and celebrating sobriety milestones.12

The National Drug Abuse Collaborative Cocaine Treatment Study is the most detailed analysis of self-help use. This multi-site RCT by Weiss and colleagues enrolled 487 individuals for behavioral treatment of cocaine dependence.23

Twelve-step attendance did not predict substance use outcomes, but active participation (such as making coffee for a meeting or reading AA literature) in a given month predicted decreased substance use in the following month. Subjects whose participation increased over the first 3 months showed reduced drug use in the following 3 months. Interestingly, those who participated without attending meetings regularly (such as by reading AA literature or calling a sponsor) benefited as much as those who attended meetings more regularly.

Summary. Actively engaging in 12-step treatment—as measured by identifying with the fellowship and following the steps—appears more important to success than simply attending meetings.


To encourage engagement when referring, try to match patients to AA groups attended by persons with whom they feel comfortable (Table 2).24 Adolescents, for example, tend to have more difficulty engaging and remaining in AA than do adults.25 One remedy may be to recommend a Young Person’s AA group composed primarily of adolescents and young adults.

Although AA embraces a spiritual approach to recovery, a person can benefit from participating without having a specific religious affiliation or spiritual beliefs.12,26 The emphasis on spirituality and a “higher power” varies from one AA group to another as well as from region to region. For patients who are uncomfortable with AA’s religiosity, other self-help options for alcohol dependence include SMART Recovery (a CBT-based treatment) and Secular Organization for Sobriety (see Related resources).

Table 2

Matching patients to AA groups:
6 variables to consider



Socioeconomic status

Match by group location


Women-only groups


Young people’s AA

Religious content

Beginners’ groups and speaker or topic discussion groups have less-spiritual focus, whereas Step groups and Sunday meetings have more spiritual focus

Smoking status

Most groups are smoke-free

Drug of choice

Consider AA, Narcotics Anonymous, or a combination of both

Source: Reference 29.

Psychiatric comorbidity. Patients with co-occurring psychopathology often worry that they will not be accepted into their local AA groups. Although AA’s policy is to remain neutral to an individual’s mental health issues, groups vary in their knowledge about and acceptance of co-occurring psychiatric illnesses and medications prescribed for these disorders. Encourage individuals with co-occurring psychiatric disorders to sample a number of local groups to find a good “fit.”

Did you miss this content?
What do >700 letters to a mass murderer tell us about the people who wrote them?