Sie sind auf Seite 1von 5



Prior Alcoholics Anonymous (AA) Affiliation and the

Acceptability of the Twelve Steps to Patients Entering
UK Statutory Addiction Treatment*
National Addiction Centre/Institute of Psychiatry, South London and Maudsley NHS Trust, Denmark Hill, London, England

ABSTRACT. Objective: The study investigates levels of affiliation with and 26%, respectively). Each of these three AA-attitude groups expressed
AA and beliefs about the organization and its philosophy among a co- greater endorsement of “Personal Responsibility” steps than of “Higher
hort of alcoholics entering a UK (non-AA) alcohol treatment service. Power mediated” steps. Conclusions: Few participants were universally
Method: A total of 150 consecutive admissions (75% men) were inter- negative to AA or the Twelve Steps—most regarded some of the steps
viewed by an independent researcher within 5 days of their entry into a as positive, but many rejected those referring to a Higher Power. Most
residential alcohol treatment unit. Results: Although about three quar- also regarded some aspects of the organization and its philosophy worth-
ters of these patients had previously attended AA meetings, levels of while, with attitudes spread across the continuum of opinion. As AA
affiliation were low, with only 16% having worked any of the Twelve remains one of the most widely sought forms of help for alcohol prob-
Steps. Previous AA attenders were more likely to be older, drinking lems, a clearer understanding is needed of its impact on patients and
greater daily quantities prior to treatment and to have first sought alco- the appropriateness of its integration within substance misuse programs
hol treatment at a younger age. Roughly equal groups expressed “posi- which are not explicitly Twelve Step in orientation. (J. Stud. Alcohol 64:
tive,” “neutral” and “negative” current attitudes towards AA (38%, 36% 257-261, 2003)

T HERE ARE MORE THAN 3,000 AA meetings each

week in the UK, approximately 500 in London (per-
sonal communication with the AA General Service Board).
middle class men and more chronically and severely alco-
holic (Emrick, 1987; Galaif and Sussman, 1995). More
recent research (Humphreys et al., 1995; Room and
AA is built around a program to recovery, the Twelve Steps, Greenfield, 1993), however, has failed to identify clear de-
and is based on mutual help, group affiliation and identifi- mographic or alcohol-related characteristics associated with
cation (Spickard, 1990). It is widely believed that the AA attendance. AA membership surveys have also observed
Twelve-Step model contributes to recovery (Miller and an increase in the percentage of women, people under 30
McCrady, 1993) and many treatment programs incorporate and polydrug users attending meetings (Chappel, 1993;
Twelve-Step principles and recommend AA attendance as Emrick, 1987).
an aftercare resource. Alcoholics Anonymous is the most common form of
AA meetings differ in terms of their membership, social help for those wit∞ alcohol problems (McCrady and Miller,
structure, group dynamics (Montgomery et al., 1993) and 1993) and most U.S. treatments operate within this broad
program implementation (Tonigan et al., 1995), and AA context (Wallace, 1996). However, there is little research
therefore advocates that new members attend different on AA in the UK, and there are concerns about generaliz-
groups until they find one that suits their needs. Early re- ing from American evidence. Little is known about AA’s
search (Leach, 1973; Ogbourne and Glaser, 1981; Trice acceptability and effectiveness in the UK and even less
and Roman, 1970) suggested that AA affiliates were not about its relationship with “statutory” treatment providers.
representative of alcoholics receiving treatment (Emrick, While many drug and alcohol treatment services have links
1987). The AA affiliates were characterized as older, single, with AA groups, their integration and role vary. Some may
incorporate Twelve-Step principles, others recommend AA
attendance as an aftercare resource. The current article ex-
plores the AA experiences and attitudes of a cohort of Na-
Received: June 14, 2001. Revision: April 23, 2002. tional Health Service alcohol treatment seekers. Levels of
*Funding was provided by the Eva and Hans Rausing Trust through the
prior AA attendance and affiliation are described and dif-
charity Action on Addiction.
†Correspondence should be sent to Dr. David Best, Police Complaints ferences between AA attenders and nonattenders examined.
Authority, 10 Great George Street, London SW1P 3AE, England, or via Attitudes towards AA and the Twelve Steps and potential
email to: barriers to AA engagement are also explored.


Method dance of 45.3 meetings (range of 1-500); 40 (26.7%) re-

ported that they had never attended AA; 49 (32.7%) had
A total of 150 consecutive admissions to an inpatient attended AA but not in the last year; and 26 (17.3%) had
alcohol treatment service in South London were interviewed first attended in the last year. Those attending in the last
by an independent researcher. Data were collected on de- year (61, 55.5%) had attended a mean of 17.8 meetings. The
mographic profile, drinking and treatment history by means most common route into AA was through treatment services
of the Maudsley Addiction Profile (Marsden et al., 1998), (50, 46.3%) (i.e., via statutory NHS treatment services, gen-
Severity of Alcohol Dependence Questionnaire (SADQ; eral practitioners or specialist addiction services), followed
Stockwell et al., 1979) and the Alcohol Problems Ques- by family and friends (31, 28.7%); the remainder (27, 25.0%)
tionnaire (Drummond, 1990). had first attended of their own accord (valid n = 108).
Previous AA attendance and affiliation were measured
using the AA Involvement Scale (Tonigan et al., 1996a), Affiliation with Alcoholics Anonymous
and a new instrument was used to measure current views
towards AA and the Twelve Steps. Participants were asked Of patients with AA experience, 22 (20.0%) had “ever
to read through the Twelve Steps and indicate their level of considered [themselves] to be a member of AA”; 17 (15.6%)
endorsement of each step. Positive endorsements refer to had worked any steps; 7 (6.4%) had attended 90 meetings
unequivocal acceptance of the step; that is, acceptance with- in 90 days; 19 (17.4%) had ever had a sponsor; 8 (7.3%)
out reservation about either content or language (Best et had celebrated a sobriety birthday and 3 (2.8%) had experi-
al., in press). A modified measure of spiritual orientation, enced a spiritual awakening.
religious faith and involvement was used (Religious Be-
liefs and Behaviour Scale; Connors et al., 1996). Profile of AA attenders

Sample characteristics A backwards elimination logistic regression analysis

found that previous AA attendance was associated with be-
The sample consisted of 112 (74.7%) men and 38 ing older (p < .01), drinking greater daily quantities prior
(25.3%) women, with a mean age of 42.2 years. The ma- to treatment entry (p < .05) and having first sought alcohol
jority (132, 90.4%) were white; 57 (39.0%) were single, 53 treatment at a younger age (p < .01) (χ2 = 30.2, p < .001).
(36.3%), divorced, separated or widowed and 36 (24.7%) All other demographic and alcohol-related measures, in-
living with a partner (valid demographic data n = 146). cluding religious involvement, were not found to be sig-
Most (112, 77.2%) had not been employed in the past nificant predictors of previous AA attendance. Similar
month, although 23 (15.8%) had worked and 6 (4.1%) were proportions of men (77%) and women (63%) had ever at-
retired (valid n = 141). tended AA and had attended in the last year (57% and
When asked to describe current religious beliefs, 49 46%, respectively). Although there were no significant dif-
(32.9%) endorsed “I believe in God, but I am not reli- ferences in levels of lifetime attendance (men: mean of 47.7;
gious”; 40 (26.8%) endorsed “I do not believe in God”; 28 women: mean of 36.6; t = 0.5, p = .6), men reported sig-
(18.8%) endorsed “I believe in God and practice religion”; nificantly higher meeting attendance over the previous year
19 (12.8%) endorsed “I don’t know what to believe about (means of 11.8 and 1.4, respectively; t = 3.6, p < .001).
God”; and y3 (8.7%) endorsed “I believe we can’t really
know about God” (valid n = 149). Positive endorsement of the Twelve Steps
The sample had been drinking for an average of 22.4
After reading through the Twelve Steps, patients endorsed
years and had an average age of 37.4 years when first seek-
a mean total of 5.4 steps. A clear and consistent pattern of
ing alcohol treatment. Of the sample, 99% self-reported
endorsement was evident (Best et al., in press), with the
drinking 34.1 units on a mean of 28.8 days; 128 (86.5%)
majority agreeing with steps that do not explicitly mention
drank on a daily basis. They had a mean Severity of Alco-
God or a Higher Power but encourage acceptance, self-
hol Dependence score of 34.4, and 92 (63.0%) reported a
examination and reparation (grouped as “Personal Respon-
score of 30 or more, denoting severe dependence. A major-
sibility” steps, see Table 1). A considerable proportion
ity of the total sample (116, 78.4%) agreed with the goal of
reported the references to God (80, 54.4%) and the Higher
Power (88, 62.4%) were “off-putting” (the term most com-
Results monly used in pilot interviews for adverse views).
Although the measures of Twelve Step endorsement are
Previous attendance at Alcoholics Anonymous closely correlated, each is predicted to different degrees
and by different variables in stepwise regression analyses
Nearly three quarters of the sample had previously at- using drinking, demographic and AA-related predictor vari-
tended an AA meeting (110, 73.3%) with a mean atten- ables. “Higher Power” endorsement (R2 = 0.25; F = 22.1,

p < .001) was predicted by religious involvement (t = 5.8, and 39 (26.2%) “positive” (valid n = 149). Demographic
p < .001) and lifetime number of AA meetings attended (t characteristics, alcohol problem severity, religious involve-
= 2.6, p < .05). A smaller proportion of “Personal Respon- ment, meeting attendance, step endorsement and knowledge
sibility” endorsement was explained (R2 = 0.13; F = 9.5, p were entered into a stepwise multiple regression analysis to
< .001) by current AA-attitude (t = 3.1, p < .01) and agree- predict current attitude. Endorsement of “Personal Respon-
ment with the goal of abstinence (t = 2.3, p < .05). sibility” steps (t = 3.3, p < .01) and knowledge of the Twelve
Steps (t = 5.9, p < .001) were significant predictors (R2 =
Current attitude to Alcoholics Anonymous as an organization 0.28; F = 25.7, p < .001).
As shown in Table 2, the three AA-attitude groups dif-
Patients’ attitudes to AA were categorized in three fered significantly with regard to meeting attendance (both
groups: 57 (38.3%) were “negative,” 53 (35.6%) “neutral” lifetime and in the last year), Twelve Step knowledge and

TABLE 1. Positive endorsement of the Twelve Steps by the total sample (N = 150), AA attenders (n = 110) and non-AA attenders
(n = 40)
Total AA Non-AA Difference
sample attenders attenders between AA
Frequency Frequency Frequency vs non-AA
(%) (%) (%) attenders
Endorsement of “Personal
Responsibility” steps
Step 1: We admitted that we were
powerless over alcohol—that our
lives had become unmanageable 127 (84.7) 97 (76.4) 30 (23.6) χ2 = 3.90, p < .05
Step 4: Made a searching and fearless
moral inventory of ourselves 93 (62.0) 69 (74.2) 24 (25.8) χ2 = 0.09, p = .76
Step 8: Made a list of all persons we
had harmed, and became willing to
make amends to them all 96 (64.0) 73 (76.0) 23 (24.0) χ2 = 1.00, p = .32
Step 9: Made direct amends to such
people wherever possible, except
when to do so would injure them or
others 93 (62.0) 68 (73.1) 25 (26.9) χ2 = 1.01, p = .94
Step 10: Continued to take personal
inventory and when we were wrong
promptly admitted it 116 (77.3) 86 (74.1) 30 (25.9) χ2 = 0.17, p = .68
Step 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 54 (36.0) 35 (64.8) 19 (35.2) χ2 = 3.13, p = .08

Mean number (SD) (range = 0-6) 3.9 (1.8) 3.9 (1.7) 3.8 (1.9) t = 0.36, p = .72

Endorsement of “Higher Power

mediated” steps
Step 2: Came to believe that a power
greater than ourselves could restore
us to sanity 47 (31.3) 36 (76.6) 11 (23.4) χ2 = 0.37, p = .54
Step 3: Made a decision to turn our will
and our lives over to the care of God
as we understood Him 32 (21.3) 22 (68.8) 10 (31.3) χ2 = 0.44, p = .51
Step 5: Admitted to God, to ourselves,
and to another human being the exact
nature of our wrongs 49 (32.7) 35 (71.4) 14 (28.6) χ2 = 0.14, p = .71
Step 6: Were entirely ready to have God
remove all these defects of character 27 (18.0) 20 (74.1) 7 (25.9) χ2 = 0.01, p = .92
Step 7: Humbly asked Him to remove
our shortcomings 41 (27.3) 30 (73.2) 11 (26.8) χ2 = 0.001, p = .98
Step 11: We 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 41 (27.3) 31 (75.6) 10 (24.4) χ2 = 0.15, p = .70

Mean number (SD) (range = 0-6) 1.6 (2.1) 1.6 (2.1) 1.6 (2.1) t = 0.02, p = .99

Source: Alcoholics Anonymous, 1976, pp. 59-60 (italics in original).


TABLE 2. Step endorsement, meeting attendance and religious involvement reported by the three AA-attitude groups
Negative Neutral Positive
AA attitude AA attitude AA attitude Significance
Mean (SD) Mean (SD) Mean (SD) test
Knowledge of the Twelve Steps (0-12) 0.09 (0.4)a 0.3 (0.9)b 1.7 (2.0)a,b F = 24.9, p < .001
Overall endorsement of the Twelve
Steps (0-12) 4.2 (3.0)a,b 5.9 (3.4)a 6.7 (3.1)b F = 8.2, p < .001
Endorsement of “Higher Power
mediated” steps (0-6) 0.9 (1.6)a,b 2.0 (2.2)a 2.1 (2.2)b F = 5.8, p < .01
Endorsement of “Personal
Responsibility” steps (0-6) 3.3 (1.9)a 3.9 (1.7) 4.6 (1.4)a F = 7.3, p < .01
Lifetime meeting attendance 10.8 (21.7)a 21.8 (70.7)b 81.4 (117.6)a,b F = 11.2, p < .001
Meeting attendance in the last year 0.9 (1.8)a 2.0 (6.9)b 22.9 (35.3)a.b F = 19.5, p < .001
Religious involvement (0-30) 4.5 (4.5) 7.3 (7.2) 6.1 (6.2) F = 2.9, p = .06

Note: Means on the same row having the same subscript letter differ significantly at the p < .05 significance level (Bonferroni
post hoc significance test).

endorsement but not religious involvement. All three groups that drinkers are heavily polarized in AA attitudes. All three
found the “Higher Power mediated” steps less acceptable AA-attitude groups expressed greater endorsement of “Per-
than the “Personal Responsibility” steps. Endorsement of sonal Responsibility” than of “Higher Power mediated”
the “Higher Power mediated” steps was lowest among the steps. The groups were differentiated only in the strength
“negative” group (“neutral” and “positive” groups reported of their approval of “Personal Responsibility” steps and the
similar levels of endorsement). strength of their rejection of “Higher Power mediated” steps.
Those with a negative attitude towards AA may still be
Discussion receptive to some steps, particularly to “Personal Responsi-
bility” steps. Conversely, even those generally positive about
Nearly three quarters of the sample had previous experi- AA may have some reservations, especially about steps that
ence of AA. For many, however, this had been only a fleet- evoke a “Higher Power.”
ing engagement as most of the sample reported low levels Over half the sample found references to God off-put-
of affiliation (exposure was infrequently translated into en- ting, consistent with Room (1998) and Galaif and Sussman
during membership). The current research investigates (1995), who suggest the spiritual emphasis of the Twelve
whether there is a distinctive profile for AA attenders. While Steps might be a barrier to engagement. The “Higher Power”
AA attenders were older than non-AA attenders (as sum- concept may be a barrier that can be overcome in personal
marized by Galaif and Sussman, 1995), other demographics terms, perhaps by its representation as a quality within the
failed to distinguish attenders from nonattenders (support- individual or group. Clinical staff may be able to facilitate
ing Emrick, 1989, and Tonigan et al.’s inpatient finding, a reconceptualization of the “Higher Power,” so that it may
1996b). be tailored to personal beliefs. This problem has been ad-
Furthermore, AA attenders did not differ in religious dressed by adapting the Twelve Steps to downplay refer-
involvement as previously found by Laundergan and ences to a Higher Power in the Seven Points of Links
Kammeier (1978). However, there was a mixed picture with (Kurube, 1992), and, in temperance-based groups in Italy
regard to alcohol problem severity. Ogbourne and Glaser and Croatia, the Twelve Steps play a limited role (Room,
(1981) and Vaillant’s (1983) findings of a positive rela- 1998). Increased understanding of the impact these changes
tionship between AA affiliation and severity are not sup- have on affiliation and abstinence rates, and their pancultural
ported in the current sample. Those who had previously generalizability, would be valuable.
attended AA, however, were more likely to report first seek- It is important to bear in mind the sampling frame when
ing alcohol treatment at an earlier age and drinking larger considering implications. This is a group of self-identified
daily amounts prior to treatment. This suggests likelihood problem drinkers entering treatment in a generic unit with
of attendance might be less predicted by “static” character- no commitment to AA. As a consequence, the sample is
istics, such as gender, and more by historical factors, such not representative of the general drinking population, nor
as age and duration of problem drinking. does it represent “typical” AA recruits. Nonetheless, it is
The observed lack of AA affiliation was not reflected in important to consider the policy and practice issue of
overall AA attitudes. The largest group (38%) expressed whether the appropriate place for AA is within generic al-
“negative” attitudes; however, more than a quarter were cohol treatment services. It is too restrictive to assume that
“positive” and just over a third “neutral.” The presence of all barriers to attendance relate to the Twelve Steps, as
a sizeable nonpolarized group challenges the preconception previous experiences and expectations of AA attendance

are likely to color overall beliefs and acceptance. Analysis GALAIF, E.R. AND SUSSMAN, S. For whom does Alcoholics Anonymous
of the qualitative components of the current body of re- work? Int. J. Addict. 30: 161-184, 1995.
HUMPHREYS, K., MOOS, R.H. AND FINNEY, J.W. Two pathways out of drink-
search will examine this issue in greater detail. ing problems without professional treatment. Addict. Behav. 20: 427-
Current AA involvement within statutory alcohol treat- 441, 1995.
ment programs appears largely arbitrary in the UK. If the KURUBE, N. The ideological and organizational development of the Swed-
efficacy of AA is to be optimized, its involvement in treat- ish Links movement. Contemp. Drug Probl. 19: 649-676, 1992.
LAUNDERGAN, J.C. AND KAMMEIER, M.L. Posttreatment Alcoholics Anony-
ment should be implemented in such a way that clients mous Attendance and Treatment Outcome, Center City, MN: Hazelden,
who choose to attend are empowered (Sanchez-Craig, 1990) 1978.
to develop the personal and social skills that are prerequi- LEACH, B. Does Alcoholics Anonymous really work? In: BOURNE, P.G.
site for benefiting from AA meetings (Caldwell and Cutter, AND FOX, R. (Eds.) Alcoholism: Progress in Research and Treatment,

1998). Although AA might not benefit all problem drink- San Diego, CA: Academic Press, 1973, pp. 245-284.
MCCRADY, B.S. AND MILLER, W.R. (Eds.) Research on Alcoholics Anony-
ers, consideration of the barriers to attendance and engage- mous: Opportunities and Alternatives, New Brunswick, NJ: Rutgers
ment may facilitate integration and implementation so that Center of Alcohol Studies, 1993.
benefits may be extended to a broader group. Thus, those MARSDEN, J., GOSSOP, M., STEWART, D., BEST, D., FARRELL, M., LEHMANN,
who fail to engage as a result of references to God and the P., EDWARDS, C. AND STRANG, J. The Maudsley Addiction Profile (MAP):
A brief instrument for assessing treatment outcome. Addiction 93:
Higher Power may still derive some advantage from an 1857-1868, 1998.
amended version that excludes or reduces the salience of MILLER, W.R. AND MCCRADY, B.S. The importance of research on Alco-
these components. As with other components of treatment, holics Anonymous. In: MCCRADY, B.S. AND MILLER, W.R. (Eds.) Re-
clinicians may benefit from working with a range of “AA- search on Alcoholics Anonymous: Opportunities and Alternatives, New
Brunswick, NJ: Rutgers Center of Alcohol Studies, 1993, pp. 3-11.
style” approaches that seek to match aspects of meetings to
MONTGOMERY, H.A., MILLER, W.R. AND TONIGAN, J.S. Differences among
the beliefs and perceptions of the patient. AA groups: Implications for research. J. Stud. Alcohol 54: 502-504,
Acknowledgments OGBOURNE, A.C. AND GLASER, F.B. Characteristics of affiliates of Alcohol-
ics Anonymous: A review of the literature. J. Stud. Alcohol 42: 661-
675, 1981.
The authors would like to thank the staff and patients of the alcohol ROOM, R. Mutual help movements for alcohol problems in an interna-
unit included in the study. They are also grateful to the Eva and Hans tional perspective. Addict. Res. 6: 131-145, 1998.
Rausing Trust for funding the study and to Action on Addiction for its ROOM. R. AND GREENFIELD, T. Alcoholics Anonymous, other 12-step move-
support. ments and psychotherapy in the U.S. population, 1990. Addiction 88:
555-562, 1993.
References SANCHEZ-CRAIG, M. Brief didactic treatment for alcohol and drug-related
problems: An approach based on client choice. Brit. J. Addict. 85:
169-177, 1990.
ALCOHOLICS ANONYMOUS. Alcoholics Anonymous: The Story of How Many SPICKARD, A., JR. The utilization of self-help groups. In: LERNER, W.D.
Thousands of Men and Women Have Recovered from Alcoholism, AND BARR, M.A. (Eds.) Handbook of Hospital Based Substance Abuse
3rd Edition, New York: Alcoholics Anonymous World Services, 1976. Treatment, New York: Pergamon Press, 1990, pp. 169-183.
MARSHALL, E.J., BEARN, J. AND STRANG, J. Are the twelve steps more The development of a questionnaire to measure severity of alcohol
acceptable to drug users than to drinkers? Europ. Addict. Res., in dependence. Brit. J. Addict. 74: 79-87, 1979.
press. TONIGAN, J.S., ASHCROFT, F. AND MILLER, W.R. AA group dynamics and
CALDWELL, P.E. AND CUTTER, H.S.G. Alcoholics Anonymous affiliation 12-step activity. J. Stud. Alcohol 56: 616-621, 1995.
during early recovery. J. Subst. Abuse Treat. 15: 221-228, 1998. TONIGAN, J.S., CONNORS, G.J. AND MILLER, W.R. Alcoholics Anonymous
CHAPPEL, J.R. Long-term recovery from alcoholism. Psychiat. Clin. No. Involvement (AAI) Scale: Reliability and norms. Psychol. Addict.
Amer. 16: 177-187, 1993. Behav. 10: 75-80, 1996a
CONNORS, G.J., TONIGAN, J.S. AND MILLER, W.R. A measure of religious TONIGAN, J.S., TOSCOVA, R. AND MILLER, W.R. Meta-analysis of the litera-
background and behavior for use in behavior changes research. Psychol. ture on Alcoholics Anonymous: Sample and study characteristics mod-
Addict. Behav. 10: 90-96, 1996. erate findings. J. Stud. Alcohol 57: 65-72, 1996b.
DRUMMOND, D.C. The relationship between alcohol dependence and alco- TRICE, H.M. AND ROMAN, P.M. Sociopsychological predictors of affiliation
hol-related problems in a clinical population. Brit. J. Addict. 85: 357- with Alcoholics Anonymous: A longitudinal study of “treatment suc-
366, 1990. cess.” Social Psychiat. 5: 51-59, 1970.
EMRICK, C.D. Alcoholics Anonymous: Affiliation processes and effective- WALLACE, J. Theory of 12-step oriented treatment. In: ROTGERS, F., KELLER,
ness as treatment. Alcsm Clin. Exp. Res. 11: 416-423, 1987. D.S. AND MORGENSTERN, J. (Eds.) Treating Substance Abusers: Theory
EMRICK, C.D. Alcoholics Anonymous: Membership characteristics and ef- and Techniques, New York: Guilford Press, 1996, pp. 13-36.
fectiveness as treatment. In: GALANTER, M. (Ed.) Recent Developments VAILLANT, G.E. The Natural History of Alcoholism, Cambridge, MA:
in Alcoholism, Vol. 7: Treatment Research, 1989, pp. 37-53. Harvard Univ. Press, 1983.