by Chaz Bufe, author of Alcoholics Anonymous: Cult or Cure?
AA’s supporters commonly trumpet AA as the best, if not the only, way to deal with alcohol problems. To back their claims, they cite anecdotal evidence and uncontrolled studies; but they ignore the best scientific evidence–the only available controlled studies of AA’s effectiveness–as well as the results of AA’s own internal surveys of its membership.
There have been only two controlled studies (with no-treatment comparison groups) of AA’s effectiveness. Both of these studies indicated that AA attendance is no better than no treatment at all.
The first of these studies was conducted in San Diego in 1964 and 1965, and its subjects were 301 “chronic drunk offenders.”(1) These individuals were assigned as a condition of probation to attend AA, to treatment at a clinic (type of treatment not specified), or to a no-treatment control group. All of the subjects were followed for at least a year after conviction, and the primary outcome measure was the number of rearrests during the year following conviction. The results were that 69% of the group assigned to AA were rearrested within a year, 68% of the clinic-treatment group were rearrested, and 56% of the no-treatment control group were rearrested. Based on these results, the authors concluded: “No statistically significant differences between the three groups were discovered in recidivism rate, in number of subsequent rearrests, or in time elapsed prior to rearrest.”(2)
(As an aside, the results of this study were very close to being “statistically significant.” This normally means–as it did in this study–that there’s a 5% or less likelihood that random chance accounts for an outcome. In the case of this study, there was a 6% or less likelihood that the poor results of the AA group were the result of random chance. Hence, the authors classified the result as “no[t] statistically significant.”)
The second controlled study of AA’s effectiveness was carried out in Kentucky in the mid-1970s, and its subjects were 260 clients “representative of the ‘revolving door’ alcoholic court cases in our cities.”(3) These subjects were divided into five groups: One was assigned to AA; a second was assigned to nonprofessionally led Rational Behavior Therapy; a third was assigned to professionally led Rational Behavior Therapy; a fourth was assigned to professionally led traditional insight (Freudian) therapy; and the fifth group was the no-treatment control group. The individuals in these groups were given an outcome assessment following completion of treatment, and were then reinterviewed three, six, nine, and twelve months later.
The results of this study were as follows: AA had by far the highest dropout rate of any of the treatment groups–68%; the lay RBT group had a 40% dropout rate; the professionally led RBT group had a 42% dropout rate; and the professionally led insight group had a 46% dropout rate.
In terms of drinking behavior, 100% of the lay RBT group reported decreased drinking at the outcome assessment; 92% of the insight group reported decreased drinking; 80% of the professionally led RBT group reported decreased drinking; 67% of the AA attendees reported decreased drinking; and only 50% of the no-treatment controls reported decreased drinking.
But in regard to bingeing behavior, the group assigned to AA did far worse than any of the other groups, including the no-treatment control group. The study’s authors reported: “The mean number of binges was significantly greater (p = .004) (4) for the AA group (2.37 in the past 3 months) in contrast to both the control (0.56) and lay-RBT group (0.26). In this analysis, AA was [over four] times [more] likely to binge than the control [group] and nine times more likely than the lay-RBT group.(5)
It seems likely that the reason for this dismal outcome for the AA group was a direct result of AA’s “one drink, one drunk” dogma, which is drummed into the heads of members at virtually every AA meeting. It seems very likely that this belief all too often becomes a self-fulfilling prophecy, as it apparently did with the AA attendees in this study. Combine this with the million-plus people coerced into AA attendance annually in the U.S., and AA’s astronomically high dropout rate (95%, according to AA’s own triennial surveys), and it’s reasonable to speculate that AA contributes to binge drinking.
The third significant piece of evidence regarding AA’s effectiveness is that provided by AA’s triennial membership surveys. In 1990 or 1991 (there’s no publication date; I obtained the report in 1991), AA produced an analysis of its previous five triennial membership surveys, “Comments on AA’s Triennial Surveys.”(6) This document revealed that 95% of those coming to AA drop out during their first year of attendance. (7) Even if all those who remain in AA stay sober (which is not the case), this is still a poor success rate, even in comparison with the rate of spontaneous remission.
There have been many studies of spontaneous remission (sometimes called spontaneous recovery), and one meta-analysis of such studies indicates that between 3.7% and 7.4% of individuals with alcohol problems “spontaneously” recover in any given year.(8) In comparison with this, AA’s annuals 5% retention rate of new attendees is not impressive. And that 5% rate might be optimistic–it was derived from surveys conducted during the 1970s and 1980s, a period of very high growth in AA membership. In contrast, since the mid-1990s, AA’s U.S. membership has been almost flat. According to AA’s own figures, two decades ago it was roughly 1.2 million, and in January 2013 it was approximately 1.3 million.(8). This works out to a minuscule growth rate of under one-half of one percent annually. Even taking into account dropouts with “time” (in this “program for life”) and mortality, this means that AA’s current annual new-member retention rate could well be under 5%.
As for AA being the only way to beat an alcohol problem, it has been known for decades that alcoholism (alcohol dependence–as contrasted with mere alcohol abuse) disappears faster than can be explained by mortality after the age of 40.(9) Also a very large Census Bureau-conducted survey in the early 1990s found that over 70% of the formerly alcohol-dependent individuals surveyed (over 4,500 in all) had recovered without participating in AA or attending treatment of any kind, and that those who had not participated in AA or attended treatment had a higher rate of recovery than those who had.(10)
As well, in contrast to AA and treatment derived from it (the dominant mode of treatment in the U.S.), there are several types of treatment that are well supported by the best available scientific evidence (studies with random assignment of subjects and no-treatment control groups, and/or comparison groups using standard 12-step treatment). Among the best-supported therapies are those known as the community reinforcement approach, social skills training, motivational enhancement, and brief intervention.(11) All of these well-supported therapies are low-cost, cognitive-behavioral approaches in which alcohol abusers are reinforced in the belief that they have power over their own actions, and are responsible for them. (This is in direct contrast to the 12-step approach, which teaches alcohol abusers that they are “powerless.”) Unfortunately, none of these proven-effective, low-cost therapies is in common use in the United States, in which the ineffective 12-step approach dominates.
Finally, over the last third of a century, a number of “alternative” (non-12 step) recovery groups have arisen in the U.S., and many, many individuals have recovered through them. Abstinence-oriented groups include SMART Recovery, Secular Organizations for Sobriety, Lifering, and Women for Sobriety. Moderation-oriented groups include Moderation Management and Harm Reduction Network. Between them, they have hundreds of meetings across the country and all provide help via the Internet.
In sum, those who trumpet AA as the best (or only) way to deal with an alcohol problem do so only by ignoring well supported alternative therapies, the widespread “alternative” self-help groups, the best available scientific evidence, and the evidence generated by AA itself.
1. Ditman, K.S. et al. “A controlled experiment on the use of court probation for drunk arrests.” American Journal of Psychiatry, 124(2), pp. 64-67.
2. Ibid., p. 64.
3. Brandsma, J.M. et al. Outpatient Treatment of Alcoholism: A review and comparative study. Baltimore: University Park Press, 1980.
4. This means there was only a 1 in 250 chance that this was the result of random chance.
5. Op. cit., Brandsma et al., p. 105.
6. “Comments on AA’s Triennial Surveys,” no author listed. New York: Alcoholics Anonymous World Services, n.d. (This document was intended for AA’s internal use only. It’s very crudely produced–mimeographed stapled sheets bound by a staple in one corner–and is not part of AA’s official “conference approved” literature. I first found mention of this document while digging through the reference shelves at AA’s San Francisco intergroup office in 1991, and I–and subsequently other researchers–obtained it by writing to AA’s General Service Office in New York and asking for it.)
7. Ibid., p. 12.
9. Drew, R.D. “Alcoholism as a Self-Limiting Disease.” Quarterly Journal of Studies on Alcohol, 29, pp. 956-967.
10. Dawson, Deborah. “Correlates of past-year status among treated and untreated persons with alcohol dependence: United States, 1992. Alcoholism: Clinical and Experimental Research, Vol. 20, pp. 771-779.
11. See Handbook of Alcoholism Treatment Approaches (3rd Edition), by William Miller and Reid Hester. Pearson, 2002.
- Alcoholics Anonymous does more Harm than Good
- Alternatives to AA — and Why They’re Needed
- Alcoholics Anonymous is Religious, not Spiritual (part 1)
- Alcoholics Anonymous is Religious, not Spiritual (part 2)
- Why is Alcoholics Anonymous Sacroscanct?