Barbara S. McCrady, PhD
J. Scott Tonigan, PhD CHAPTER
70

CHAPTER OUTLINE

Alcoholics Anonymous (AA) is ubiquitous, both in the United States and around the world. In the United States, there are an estimated 58,820 groups and 1,290,716 members (1). Outside of the United States, AA estimates 53,692 groups and 804,357 members. AA also is commonly found in correctional institutions in the United States and Canada, with an estimated 1,555 groups and 38,664 members in these facilities (1). The formal structure of AA is similar across nations, though there is some variability in emphasis on different parts of the AA program, and differences in the demography of membership are apparent, depending on the cultural context in which AA occurs (2). Though most addiction professionals have some familiarity with AA and other self-help groups based on 12-step principles, professionals’ scientific knowledge about AA often is more limited. The past 20 years has witnessed an explosion of research on AA and on treatments designed to facilitate involvement in AA, with more than 300 articles appearing since 2003. Despite earlier skepticism about the possibility of conducting research on AA, researchers have used a range of methodologies, including ethnographic methods, epidemiologic studies, longitudinal studies of treatment-seeking and non–treatment-seeking populations, controlled clinical trials, and meta-analyses, to develop a body of new research about AA that has some coherence, confirms some previous findings and beliefs, and challenges others. This chapter provides a selective review of earlier research on AA and a more comprehensive review from 2002 through 2012. It addresses several major topics, including patterns of utilization of AA, the unique experiences and views of AA among specific population groups, the effectiveness of AA and treatments designed to facilitate AA involvement, and mechanisms of change associated with involvement with AA and other 12-step programs. The chapter concludes with methodologic comments and directions for future research. Research on other 12-step programs for substance use disorders is more limited but is included where relevant data exist.

UTILIZATION OF AA

AA members enter the program by a number of routes, including self-referral or referral by family or friends, referral from treatment centers, or through coercion from the legal system, employers, or the social welfare system.

Population Studies

Population surveys provide information on utilization of AA in the general and alcohol-problem populations. Using data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), Dawson et al. (3) reported that AA attendance among respondents with a history of alcohol dependence was 20.1%. Additional analyses of NESARC data (4) found, among persons who had developed alcohol dependence at least a year before the survey, that 25.5% had sought treatment. Among those seeking assistance, 88.9% attended a 12-step program, including 12.1% who attended only a 12-step program and 66.7% who attended both formal treatment and a 12-step program.

Help-Seeking Populations

A different perspective on the utilization of AA is provided by studies of patterns of help seeking among individuals seeking assistance for an alcohol problem. Dawson et al. (4), also using NESARC data, reported that among individuals with alcohol dependence who sought help, 78.5% had used AA and other 12-step programs (11.7% using only AA, 66.8% using AA in combination with formal treatment), compared to 88.7% who used professional services (21.9% using only professional treatment, 66.8% using treatment in combination with AA). In a 16-year longitudinal study, Timko et al. (5) examined treatment utilization among individuals who first contacted an information and referral center or who underwent alcohol detoxification. One year later, 75% had sought treatment: 18% had attended only AA or another self-help group (24% of help seekers), 25% had sought only outpatient treatment (33% of help seekers), and 32% had sought only inpatient/residential treatment (43% of help seekers). AA involvement was high among treatment seekers, with 66% of outpatients and 68% of inpatients also attending AA. By the time of an 8-year follow-up (6), 17% still had sought no treatment, and 14% had attended only AA. The majority (53%) participated in both formal treatment and AA. Study participants who attended AA (either AA alone or in conjunction with treatment) showed a pattern of remarkably steady and consistent involvement over time: 66 to 91 meetings in the 1st year, 68 to 97 meetings per year in the subsequent 2 years, 63 to 71 meetings per year in years 4 to 8, and 46 to 52 meetings per year in years 9 to 16 (7).

Mandated Populations

Though there has been considerable controversy about the current criminal justice practice of mandating individuals to attend AA, little research has examined the actual process of criminal justice referral to AA. Speiglman (8) selected four counties in California that varied in the degree to which they used presentencing screening strategies to deal with repeat offenders through the use of driving under the influence (DUI) statutes. Two of the four counties referred cases to AA, referring 37% to 40% of cases. Offenders who were represented by private attorneys were more likely to be referred to AA than those who had public representation. However, among offenders also mandated to parole or to participation in probation-defined treatment, the vast majority (88% to 97%) were required to attend AA. Frequency of attendance also was specified and typically involved two to three required meetings per week. Mandating attendance at AA requires the cooperation of the groups that the offender attends. Information at the AA Web site indicates that

…some groups, with the consent of the prospective member, have the A.A. group secretary sign or initial a slip that has been furnished by the court together with a self-addressed court envelope. The referred person supplies identification and mails the slip back to the court as proof of attendance. Other groups cooperate in different ways. There is no set procedure. The nature and extent of any group’s involvement in this process [of verifying attendance] is entirely up to the individual group. (9)

Patterns of Utilization of AA

Both cross-sectional and longitudinal studies provide information about patterns of utilization of AA. Data from the Epidemiologic Catchment Area Study (10) suggest that individuals who attend AA or other self-help groups make about twice as many visits to meetings as to professional treatment. Alcohol-dependent persons who attend AA averaged 44.8 visits/person/year, or just under one meeting per week. Data from Timko and Moos’ 16-year longitudinal study also showed variability in AA utilization. In the 1st year of help seeking, 24.9% attended AA for more than 26 weeks, 19.1% attended for 9 to 26 weeks, and 14.3% attended for only 1 to 8 weeks (11). Regular attendance was remarkably stable over time—in years 4 to 8, 28% had attended AA for more than 26 weeks. However, infrequent attendance dropped off, with only 8.1% of the sample attending for less than 26 weeks.

More recently, Kaskutas et al. (12) reported on 7-year longitudinal patterns of AA attendance in a treatment-seeking population. The low-attendance group participated in AA during treatment but was attending fewer than five meetings at follow-up points. The medium-attendance group reported attending AA an average of once a week during follow-up. The high-attendance groups attended AA an average of four times per week during the 1st year and then gradually reduced their attendance. Kaskutas et al. also identified a descending-attendance group that had very high attendance initially but then had dropped off sharply in attendance. Abstinence rates were highest for the two groups still attending AA fairly regularly, were somewhat lower for the group that had declined in attendance, and were lowest for the group that discontinued attendance after the first year (13).

Mäkelä (14) studied anniversary announcements published in a Finnish AA newsletter to track AA membership over time. Over 3 consecutive years, he found that the probability of remaining sober and involved with AA was about 67% for those with 1 year of sobriety, 85% for those with 2 to 5 years of sobriety, and 90% for those with more than 5 years of sobriety.

Summary

Data derived from a number of different methodologies converge in suggesting clearly different patterns of involvement with AA—those who initially are actively involved but taper off over time, those with a steady level of involvement, and those who have a more variable or less engaged type of involvement. Data also suggest that consistent involvement is associated with better outcomes.

FACTORS ASSOCIATED WITH SUCCESSFUL AFFILIATION WITH AA

Despite the diversity of the membership of AA, research shows that certain factors are associated with more successful affiliation with AA. Research to identify characteristics of those more likely to affiliate with AA does not imply that individuals without those characteristics will not affiliate. Over time, a body of individual studies has accrued reporting a wide range of characteristics found to be predictive of affiliation with AA, including male gender (15), more serious alcohol problems (5,16–18), greater commitment to abstinence (17), more social support to stop drinking (19), less support from and more stress in marriage/intimate relationships (20), fewer psychological problems such as depression or poor self-esteem (5), use of a more avoidant style for coping with problems (20), and having a greater desire to find meaning in life (21). Most findings, however, are supported by only one recent study, with the exception of severity of alcohol dependence (22,23) and commitment to abstinence, which seem to be stable predictors of affiliation across multiple studies. Findings are contradictory for some variables, such as education, where affiliation is predicted by greater education among Whites but less education among Hispanic Americans, or marital status, where unmarried status predicts affiliation among Hispanics (16), but being married generally is predictive of affiliation in population surveys.

The personal characteristic of spirituality, religiosity, or purpose in life has been examined in a series of studies. Professionals and the public alike believe that individuals who are more religious will be more successful in AA because of the intrinsically spiritual nature of the recovery program. An older survey found that program directors in Department of Veterans Affairs (VA) facilities were less likely to refer a patient to AA if the individual was an atheist (24). Winzelberg and Humphreys (25) looked at the relationships among clinician referral to 12-step groups, client religiosity, group attendance, and client outcomes. They too found that professionals were less likely to refer patients to 12-step groups if the patients engaged in fewer religious behaviors. However, though more frequent religious behaviors predicted 12-step meeting attendance, clinician referral to such groups increased attendance regardless of religiosity. They also found that attendance at 12-step groups predicted better outcomes, regardless of religiousness. Tonigan et al. (26) examined religious beliefs and AA affiliation among patients in Project MATCH. They found that clients assigned to twelve-step facilitation (TSF) treatment were most likely to report increased belief in God, but clients who described themselves as atheists or agnostics generally were less likely to attend AA, even if assigned to TSF treatment, compared to clients who were spiritual or religious in their beliefs. Similar to Winzelberg and Humphreys, Tonigan et al. found that AA attendance was positively associated with outcomes regardless of religious beliefs. More recently, Krentzman et al. (23) examined baseline characteristics of individuals who had successfully maintained abstinence for a year or more during a 3-year follow-up period. Those abstinent individuals who considered themselves to be members of AA (compared to non-AA members) were more likely to have been raised in a religious tradition and to endorse a belief in God.

Psychiatric comorbidity is another patient characteristic that could affect AA affiliation. In an earlier study, Tomasson and Vaglum (27) determined that the presence or absence of most comorbid disorders was unrelated to AA attendance in an aftercare sample of alcoholics, though the presence of comorbid disorders was associated with higher rates of professional help seeking. Schizophrenia, however, was the one diagnosis associated with lower rates of attendance. In a recent comprehensive review of the literature on comorbidity and AA involvement, Bogenschutz (28) also concluded that patients with comorbid psychiatric disorders attended AA at about the same rate as other patients, though attendance was lower for those with psychotic disorder diagnoses (29). He also noted the added benefit of 12-step programs that are specialized for those with comorbid disorders and that mechanisms of change associated with success in AA, such as enhanced self-efficacy and greater social support, are similar for those with and without comorbid disorders.

Summary

Data generally support the view of AA as a program that attracts a diverse membership. However, those with more severe drinking problems and those with a greater commitment to change are more likely to affiliate with AA. Patient religiosity affects clinicians’ referrals to AA, and patients with agnostic and atheist beliefs may attend fewer meetings, but patients who go to AA increase their religiosity regardless of their initial beliefs. Patients with comorbid psychiatric disorders also affiliate with AA, though those with schizophrenia are somewhat less likely to attend.

AA AND POPULATION SUBGROUPS

Two contrasting views of AA lead to different predictions about AA and different population subgroups. One perspective suggests that AA is a program of recovery for a person with an alcohol use disorder and that the common experience of alcoholism should supercede superficial individual differences. Because AA groups are autonomous, individual meetings may take on the character of the predominant population in attendance, allowing for meetings that are comfortable for persons of different backgrounds. In the United States, “special interest” groups for certain subpopulations (such as women, gays and lesbians, young people, and certain racial/ethnic groups) are very common (2). An alternative perspective is that, because AA was developed by educated, middle-aged, white, Christian, heterosexual males, its relevance to less-educated, young or older persons, persons of color, non-Christians, gays and lesbians, or women is suspect. AA’s own triennial surveys have found an increase in the proportion of women in AA from about 22% in 1968 to about 35%, leveling off starting in 1989 (30). The average age of AA members responding to AA’s triennial survey has increased to about 49 years of age (30), and the triennial survey data as well as observation of AA meetings reveal a broad diversity among the membership in age, occupational status, and race. Research data about the relevance of AA to various subgroups are limited though increasing.

Women

Several controlled and qualitative studies have examined women and AA. Recent studies in the United States (31) and Sweden (32) have found similar rates of AA attendance in men and women, and the 2011 AA triennial survey reports that about 35% of members responding to the survey were women, rates comparable to the rates of women with alcohol dependence in the general population. Likewise, no substantive gender difference was found in 12-step treatment compliance and engagement in Project MATCH (33). However, men and women may have different reasons for affiliating with AA. For example, Kaskutas (34) studied women attending Women for Sobriety (WFS) meetings, approximately 25% of whom also attended AA. The women attended AA for reasons somewhat different from those for attending WFS: AA was cited as the program most crucial to their staying sober, though the fellowship, support, sharing, and spirituality in AA all were cited as important as well. The women perceived WFS as most valuable for the nurturing atmosphere, involvement with an all-women’s program, and exposure to positive female role models.

In a comprehensive review of the scientific literature on women and AA, Ullman et al. (35) examined issues related to help seeking, affiliation, and outcomes for women in AA; potential moderators of women’s affiliation with AA; and issues unique to women. They concluded that women may perceive more barriers to utilizing AA, particularly in terms of access, child care, and a sense of stigma about their alcohol use disorder, but that there was little evidence of gender differences in actual help seeking from or affiliation with AA. Women may use AA differently from men, as they attend more meetings but are less likely to have a sponsor. Bodin (32) also found that women were more likely than men to call other AA members for help, to have experienced a spiritual awakening, and to have read AA literature.

Ullman et al.’s review (35) also found that AA attendance and affiliation generally are related to better outcomes; several studies suggest that AA is more related to positive outcomes in women than men; but some find no differences. Several individual variables may affect the likelihood that women will attend or affiliate with AA. For example, younger women are more likely to attend AA meetings; Black women are more likely to attend than Latino women. Work on sexual minorities is very limited and does not lend itself to clear conclusions although qualitative studies suggest that lesbian women may view AA as heterosexist. There is no overall evidence that mental health problems create unique barriers to AA affiliation for women, with the exception of social phobia. Women with social phobia are less likely to affiliate and in fact appear to have better outcomes with cognitive–behavioral approaches to treatment.

There also may be aspects of AA that make it less appealing to women. In Kaskutas’ (34) research, she found that the women in the sample reported reasons why they did not attend AA, including a feeling that they did not fit in; a perception that AA is too punitive and focused on shame and guilt; disagreement with program principles related to powerlessness, surrender, and reliance on a Higher Power; and a perception that AA is male dominated. A recent survey of 55 women attending AA (36) found that half the women had experienced “thirteenth stepping,” in which they felt sexually targeted by men in the program. Such experience was less common among women who attended at least some women-only AA meetings.

Cultural, Racial, and Ethnic Subgroups

Research on involvement in AA by cultural, racial, and ethnic subgroups is limited. Older data reported by Caetano (16) from a national survey showed that, in general, Hispanics, African Americans, and Whites tended to endorse equally the basic tenets of the disease model. All groups held fairly positive views of AA (meaning that they would be more likely to recommend it than any other treatment modality). Some variability was noted in support for AA, with 97% of Hispanics, 94% of Whites, 87% of African Americans, and 76% of Asian Americans recommending AA as a resource. Recent research (37) utilizing data from three national alcohol surveys showed that Hispanics with lifetime alcohol dependence were less likely to have used AA than White samples (18% of males and 9.7% of females). Within the Hispanic sample, several factors were associated with lower utilization, including being female, older, Spanish speaking, and having fewer social pressures, legal consequences, or dependence symptoms. Tonigan et al. (38) reported on a sub-sample of Project MATCH participants from Albuquerque, NM. They found that though Hispanics attended fewer AA meetings, they reported being equally or more committed to AA than Whites (as evidenced by working the steps, having or being a sponsor, and celebrating AA birthdays) and higher in “God consciousness.” AA involvement predicted better drinking outcomes in both Hispanics and Whites.

Kaskutas et al. (39) examined previous self-help group participation among African American and White treatment seekers. African Americans more frequently reported prior Narcotics Anonymous (NA) or Cocaine Anonymous (CA) exposure, with a trend toward more previous AA exposure. Of those who had been exposed to AA in the past, more African Americans (76%) than Whites (55%) said they had gone to AA as a part of prior treatment, whereas more Whites had gone to AA through other referrals or on their own. Active participation was equivalent for both groups, as measured by mean AA affiliation scale scores. However, analysis of the individual items from the affiliation scale revealed differential types of participation. African Americans were more likely than Whites to identify themselves as AA members (64% vs. 54%), to say they had a spiritual awakening through AA (38% vs. 27%), and to have done service at an AA meeting recently (48% vs. 37%). African Americans were less likely than Whites to have a sponsor currently (14% vs. 23%) and less likely to have read AA literature recently (67% vs. 77%). These patterns held true after controlling for prior treatment and exposure to AA during treatment. More recent research on AA utilization in a longitudinal sample (40) found that the association between AA attendance and abstinence was similar between whites and blacks but that this relationship was stronger for whites than blacks. Blacks with alcohol use disorders were more likely than whites to be abstinent without AA, largely due to their higher levels of religiosity and their “drier” social networks.

Research on AA affiliation and Native American populations is very limited. In a study of help seeking among two American Indian reservation populations, Beals et al. (41) found that 38.9% of those with an alcohol or other substance use disorder had sought treatment; within this group, 39% had used a 12-step group. Several factors were correlated with 12-step use, including having an alcohol but not other drug problem, being from a Northern Plains rather than Southwest tribe, having a higher level of education, having a high level of spirituality, and having greater identification with White than Indian culture. Despite the substantial level of AA involvement reported by Beals, a small sample study of Native American participants in Project MATCH had better outcomes with motivational enhancement therapy (MET) than TSF (42).

Age-Specific Groups

There is some hesitancy about involving adolescents in AA because of their developmental status (43) and concern that adolescent substance use disorders are, in some cases, age-limited phenomena. However, several studies have suggested a strong association between AA/NA involvement and abstinence in adolescents, similar to that found in adults. (These studies are reviewed in the section “The Effectiveness of AA and Treatments Based on AA.”) Research to date has focused on adolescents in inpatient treatment programs, arguably the population with the most severe problems. Hohman and LeCroy (44) examined a sample of adolescents who had completed inpatient treatment and reported that about 44% had participated in AA. Kelly et al. (45) found that adolescents were more likely to attend if the AA groups they attended had more age peers. They also found (46) that the severity of the adolescent’s alcohol use disorder was correlated positively with motivation to attend AA. Treatment providers view AA involvement as important for adolescents (47). Adolescents themselves report positively on the group processes in AA such as the sense of universality of experience, received support, and the sense of hope that the program provided them. However, many adolescents report boredom or a sense that the program was not a good fit for them as reasons to discontinue attendance (48).

Dually Diagnosed Individuals

Research on substance abusers with comorbid psychiatric disorders has been focused primarily on a comparison of the substance use, psychiatric, and other life outcomes of those with multiple diagnoses compared to those with substance use disorders alone. Recent research also supports understanding of those dually diagnosed individuals in AA. In a comprehensive review of the literature on 12-step involvement among dual-diagnosis patients, Aase et al. (49) found that participation in AA and other 12-step programs was associated with positive substance use outcomes for individuals with co-occurring psychiatric diagnoses, although results were more variable for measures of psychological functioning. In addition, they found that both increases in self-efficacy and the social support received in 12-step groups that had a dual focus on recovery and psychiatric disorders mediated the relationship between group involvement and positive outcomes. They also found that 12-step attendance was particularly high for homeless, dually diagnosed individuals. In looking at specific aspects of involvement with AA, Polcin and Zemore (50) found that higher psychiatric severity was associated with lower levels of spirituality and less working the 12 steps or serving as a sponsor.

An important implication of the presence of comorbid disorders is the need for prescription medications. The subject of medication use by AA members is a particularly important one, given that medications play a larger role in mental health treatment today, but there are thoughtful cautions raised in the core AA literature about the use of psychoactive drugs (51). In an anonymous survey, Rychtarik et al. (52) assessed AA members’ attitudes toward the use of medication, either to prevent relapse or to treat other disorders. (Medications included antidepressants, pain medications, anxiolytics, lithium, antipsychotics, naltrexone, and disulfiram.) The majority (53%) of the sample thought that use of medications to prevent relapse was either a good idea or might be a good idea, 17% reported that they did not like the idea of medication and believed the individual should not take it, and 12% said they would recommend that another member discontinue medication use. About 29% said they had been encouraged to stop taking any type of medication, and an additional 20% had heard of others who had been encouraged to discontinue use. Of those who were encouraged to stop medication use, 31% actually stopped.

Swift et al. (53) predicted that persons with more prior AA exposure would be less likely to take naltrexone for alcohol abuse. They reported that in a treatment-seeking sample, willingness to take naltrexone was unrelated to frequency of past AA meetings attended, and, surprisingly, having an AA sponsor was positively related to willingness to take medications, albeit modestly. Studies of AA members’ perceptions and practices support the counterintuitive idea that AA exposure has a negligible effect on the use of medications for an alcohol use disorder. Tonigan and Kelly (54), for example, reported that AA affiliation was unrelated to attitudes about the use of medications for drinking problems. It seems, then, that though negative messages may be voiced in AA about the use of medications, AA exposure in itself does not deter medication compliance.

Gays and Lesbians

Research on the experience of gays and lesbians in relation to AA is quite limited. One ethnographic study (55) recruited lesbians who had been in recovery for at least 1 year. All respondents were familiar with AA; 74% were actively involved. Hall identified three sources of tension for the lesbians in AA. First, they reported a tension between a sense of assimilation and a sense of differentiation. The women said they felt that AA was a program in which people of very different backgrounds could relate because of their common concerns, but at times, they viewed AA as a white, male, heterosexist organization. Second, they said they understood the value of the authority of AA as a prescription for sobriety but at times viewed AA more as a program that provided a set of tools for recovery. The perceived sexist language in the AA literature and the lack of focus on lesbian issues made following the program prescriptively a difficult task. Finally, the women said they experienced tension between the strongly individual focus of AA and their perception of the importance of examining issues in a cultural context.

Summary

It appears that individuals from minority groups (e.g., women, gays and lesbians, racial/ethnic minorities) have a mixed experience in AA, seeing particular value in the support for sobriety but also having a different set of experiences with AA, some of which are somewhat negative.

THE EFFECTIVENESS OF AA AND TREATMENTS BASED ON AA

Answering the apparently simple question “Does AA work?” is a challenge. One approach is to look at the success of AA as an organization: The broad dissemination of the program around the world and the large membership suggest that AA has been enormously successful in attracting persons to AA as a program of recovery. The AA triennial surveys also point to the substantial proportion of abstaining, long-term members, as do Mäkelä’s (14) studies of stability of sobriety in AA. More difficult questions, however, have less clear-cut answers: “Is AA the most effective approach to alcohol dependence?” “Is AA involvement necessary to successful resolution of alcohol problems?” “Does AA lead to better outcomes or is it simply a correlate?” “What are the most effective strategies to engage individuals with AA?” Research to answer these questions has used several different methodologies: (a) randomized clinical trials (RCTs) comparing AA or treatments designed to involve individuals in AA to different forms of alcoholism treatment, (b) naturalistic studies of treatments designed to engage individuals with AA, (c) studies examining the unique contribution of AA to the prediction of outcomes in clinical and nonclinical samples, and (d) studies of effective approaches to engaging patients in AA. Review of these four lines of evidence provides some provocative answers to questions about the relative effectiveness of AA.

Randomized Clinical Trials

RCTs in which persons are randomly assigned to different treatment conditions are considered the most rigorous experimental tests of therapeutic effectiveness. Only three RCTs directly comparing AA alone to different forms of treatment have been reported in the research literature, and no RCT has been reported since 1991 (56). Each of the three RCTs has serious methodologic problems, and all used populations mandated to treatment, so it is difficult to draw specific conclusions about the effectiveness of AA from these studies.

AA and 12-step–oriented treatments have close conceptual links in their adherence to the classic disease concept of alcoholism, emphasis on abstinence, the importance of AA involvement, and working the 12 steps. Differences between AA and 12-step–oriented treatment programs are substantial, however, and the two should not be equated. Several important RCTs of treatments based on 12-step principles have been reported over the past several years. The most prominent and visible RCT, Project MATCH (21,57–59), was designed to study the interactions between specific patient characteristics and one of three structured 12-week outpatient individual treatments: Twelve Step Facilitation (TSF), motivational enhancement therapy (MET), or cognitive–behavioral therapy (CBT). Participants were 1,726 persons with diagnosed alcohol abuse or dependence (952 outpatients and 774 aftercare patients) who were recruited from among 4,481 patients screened at nine participating clinical research units. Participants were assessed thoroughly and then randomly assigned to one of the three treatments. Clinicians were nested within treatments, received extensive training prior to the study, and were carefully supervised throughout. Treatment was delivered over a 3-month period. Individuals assigned to TSF or CBT could receive up to 12 manual-guided treatment sessions, whereas MET participants received up to four treatment sessions over the same 12-week period. All participants were followed up for 15 months from baseline, with research contacts scheduled every 3 months. Participants in the outpatient arm of the study were contacted again 39 months after the initial baseline evaluation, and their functioning during the preceding 3 months was assessed. Though Project MATCH was not designed specifically to study the main effects of the three study treatments, some treatment main effects did emerge. During treatment (58), patients in the outpatient arm of the study were more likely to maintain abstinence or moderate drinking if they received CBT or TSF rather than MET (41% vs. 28%). One year after treatment, patients in the three treatments had comparable outcomes in the percentage of days that they were abstinent and the mean number of drinks consumed per day (21). Two variables favored the TSF treatment: Patients who had participated in TSF treatment were more likely to have maintained continuous abstinence and were less likely to have relapsed to heavy drinking after treatment. At the 3-year follow-up of the outpatient arm of the study, few significant differences among the three treatment conditions were noted, but, as at the 1-year followup, patients who had received the TSF treatment were more likely to have been abstinent during the 3 months prior to the 3-year follow-up. Also, compared to patients who had participated in CBT, TSF subjects had a significantly greater percentage of abstinent days during the preceding 3 months (59). Several significant client–treatment matching effects were found. During treatment, no client–treatment match affected drinking (58). However, during the first year after treatment, patients who had low levels of psychiatric symptoms had more days of abstinence if they had received the TSF rather than the CBT treatment (21). Aftercare patients with higher levels of alcohol dependence also had better outcomes with TSF. In contrast, patients who were low in alcohol dependence had better outcomes with CBT (57). A second important matching finding emerged at 3 years: Outpatients whose social networks were highly supportive of their drinking had better outcomes if they received TSF rather than MET treatment (60).

Stay updated, free articles. Join our Telegram channel

Join