Alcoholics Anonymous (AA)

Recovery Outcome Rates

 

Contemporary Myth and Misinterpretation  January 1, 2008

 

Updated with 2007 Survey Results  October 11, 2008

 

Introduction................................................................................................................................................................................... 1

Foreword....................................................................................................................................................................................... 2

a)      Triennial Surveys of AA Membership............................................................................................................................. 4

AA Membership Survey Pamphlets (P-48) Published................................................................................................ 4

Using Length of Sobriety as a Measurement of “Success”....................................................................................... 5

b)      First Year Retention........................................................................................................................................................... 6

Figure C-1 from the 1990 Summary Report of 1977-89 AA Triennial Membership Surveys.............................. 6

Figure C-1 Average of 5 Surveys..................................................................................................................................... 7

Chart 1: 1st Year Retention per Summary of 1977-89 Surveys................................................................................ 9

Table 1: 1st Year Retention Percentages..................................................................................................................... 9

c)      Growth of Long-Term Sobriety...................................................................................................................................... 11

Chart 2: Growth of Long Term Sobriety Ranges....................................................................................................... 12

Table 2: Long Term Sobriety Ranges Percentages................................................................................................. 12

Chart 3: Growth of Long Term Sobriety Averages.................................................................................................... 13

Table 3: Long Term Sobriety Averages...................................................................................................................... 13

d)      Timeline of Sources and Citations of AA’s 50% + 25% “Success Rates”............................................................ 14

e)      Assessment of Claims of AA Success or Failures and Corroboration................................................................... 21

Assessment of Claims of AA Success:......................................................................................................................... 22

Assessment of Claims of AA Failure:........................................................................................................................... 22

Assessment of Claims of Corroboration and Citations:............................................................................................ 23

Assessment of Claims of Success in this paper:....................................................................................................... 24

f)       Prescreening In Early AA................................................................................................................................................. 25

g)      AA Membership Composition and Growth Trend..................................................................................................... 29

Table 4: AA Membership Growth Trend..................................................................................................................... 30

Table 5: Estimated Counts of AA Groups and Members......................................................................................... 31

 

Introduction

This paper is written for AA members and is intended for internal and public circulation as an item of AA historical and archival research. It is offered to help inform the AA membership and academic researchers of a widely circulated misinterpretation and mischaracterization of AA recovery outcomes.

The fellowship of Alcoholics Anonymous, as a matter of long established principle, policy and practice, does not engage in public debate and seeks to avoid public controversy. The authors of this paper must emphasize that we do not speak for AA. We have a personal interest in the history of AA and consider it imperative to correct historical inaccuracies and propagation of myth.

Arthur S, Arlington, TX

Tom E, Wappingers Falls, NY

 Glenn C, South Bend, IN

The publication of this paper does not imply affiliation with, approval by, or endorsement from, Alcoholics Anonymous or any part of its worldwide organization and service structure. The views expressed in this paper are solely those of the authors.

AA, Alcoholics Anonymous, The Big Book, Box 4-5-9, The Grapevine, GV, Box 1980 and La Viña are registered trademarks of Alcoholics Anonymous World Service (AAWS), Inc and AA Grapevine Inc.

Single-use excerpts from AAWS and AA Grapevine publications are cited in this paper under the good faith and fair use provisions of US copyright statutes for purposes of criticism, comment, scholarship and research.


 

Foreword

“Nothing is more responsible for the good old days than a bad memory”

 - Franklin P Adams (1881-1960)

This paper addresses an erroneous myth that AA is experiencing a 5% (or less) “success rate” today as opposed to either a 50%, 70%, 75%, 80% or 93% (take your pick) “success rate” it once reputedly enjoyed in the 1940s and 1950s. The term “myth” is used to emphasize that the low “success rates” promulgated are a product of imagination, invention and inattention to detail rather than fact-based research.

Also noteworthy in the derivation of the mythical percentages, is the absence of fundamental academic disciplines of methodical research, corroborating verification and factual citation of sources. Regrettably, some of the advocates who are propagating the myth are AA members who purport to be “AA Historians” and appear to be advocating agendas that portray fiction as fact and hearsay as history.

The AA Fellowship has a robust verbal tradition. Much information is passed on by word of mouth. This has both its good and difficult sides. How does one know what is fact versus myth? AA members can sincerely state something they believe is true but is inaccurate - this is the difference between myth and fact. Consequently, much effort has been taken to ensure that the contents of this paper are independently confirmed in reliable written reference sources. Those sources are identified in footnotes or in the body of the narrative.

References have been made on the internet, in publications, in individual AA talks and on TV, that depict AA’s early to mid years, as having had a typical successful recovery rate outcome of 50-75%. These high-end numbers are often followed by a second depiction that contemporary AA has deteriorated to a 10% or 5% or less recovery rate outcome. The two sets of numbers (high versus low percentage) are then cast in an idyllic portrayal of past AA contrasted to a dismal scenario of AA today.

·         Claims of a 10%, 5% or less success rate for contemporary AA are erroneous and rest largely and misguidedly on the misinterpretation of data in a 1989-1990 internal AA General Service Office [1] report on “AA Triennial Membership Surveys.”

·         The assertion of a 50-75% success rate in AA is derived from various AA literature sources and other written sources, but is not explicitly demonstrated except in one instance. That instance pertains to the AA members who had their personal stories printed in the first edition “Big Book.” [2]

The 50-75% success rate number has been cited, without change or challenge, since it first publicly appeared in 1941 and it persists to this date.

In 1989-1990, an internal General Service Office (GSO) report, analyzing 5 prior AA membership surveys, contained a hand-written graph that has been persistently misconstrued to attempt to portray AA’s “success rate” as 5% or less (or AA’s “failure rate” as some cast it, of 95% or more). [3]

The graph has been cited as representing “retention” or “success” rates when in fact it simply illustrates the distribution of the length of time the population participating in the survey sample have been attending AA. Other survey questions asked about sobriety. This was only about attendance in the first year after someone’s first meeting.

It is astonishing that negative projections of AA of such magnitude are casually issued with presumed “certainty” but contain no demonstration of where or when the “certainty” had been established. Extant citations have become so careless and perfunctory that in many cases they merely refer to other erroneous citations in other publications to lend credence to the reference. Erroneous citations are used to support other erroneous citations.


 

Even more unfortunate, concerning statistical confidence and accuracy of the citations, is that none of the authors (or self-proclaimed “AA Historians”) has apparently independently performed a critical, unbiased investigation of the original data or attempted to duplicate the calculations of AA’s reputed “failure statistics” from the basic source data listed in the membership surveys. That basic data, and an explanation of what it signifies, follows this section.

Over the years, the internet has provided an international forum for anyone who can access it. A number of so-called “recovery” or “AA history” or “AA archives” web sites have proliferated. Many teem with personal grievances, screeds, and widely varying (and revisionist) interpretations of AA history and the AA program of recovery. An abundance of academic and medical special interest web sites have materialized as well.

The erroneous 10%, 5% or less success rate myth for contemporary AA has proliferated without as much as a token challenge to its veracity or investigation of its origin. The topic of AA success or failure outcomes suffers from a great deal of anecdotal misinformation, misinterpretation and editorializing.

Discussion, examination and analysis of the topic of AA recovery outcome rates are divided in this paper into two categories of investigation and reporting:

The first category concerns the examination of the contemporary (and quite erroneous) assertion that AA is only achieving a 5% or less success rate. The appalling success rate assertion is false but a segment of AA members not only readily believes it but also attempts to exploit it to support personal agendas. They propagate revisionist AA history and manufacture exaggerated claims of a superior early AA recovery program.

The second category concerns the examination of a popular and much repeated notion in AA of a 50% immediate success rate with about half (or 25%) of the “slippers” returning to successful recovery to produce an overall 75% success rate. This has been the prevailing “best estimate” of AA’s recovery outcomes since the late 1930s. It is denoted in this paper as a “50% + 25%” success rate (for a 75% total success rate).

Based on research discoveries to date, it is believed that the 50% + 25% success rate is in all probability a very reasonable “best estimate” of AA’s success (both early and contemporary).

·         The sole qualification (it is vitally important and often disregarded) is that the 50% + 25% success outcome rates apply only to those prospects who attempt to give AA a serious try (i.e. you get out of AA what you put into it). This rests on the simple, obvious, premise that a remedy cannot be construed as either a “success” or “failure” until it is at least tried and tested.

·         Also of contextual importance, is that the subset of the past and present prospect population falling into this category is estimated to be 20% to 40% (1 or 2 out of 5) of the total prospect population.

The remainder of this paper identifies information sources for the derivation of AA success or failure outcomes and highlights relevant information that has been omitted, invented or misconstrued. The main and initial item of interest and analysis is the grossly misinterpreted 1989-1990 internal GSO report on AA Triennial Membership Surveys.

 


 

a)       Triennial Surveys of AA Membership

In 1968, Alcoholics Anonymous took an inventory of its membership in the form of a survey. Recognizing the need to know more about the Fellowship, a small trial survey was conducted in a few groups by the Regional Trustees to see how members would respond to a voluntary anonymous questionnaire.

It went so well that a committee was set up to conduct a survey of 5% of the registered groups in the United States and Canada. A later pamphlet “The Alcoholics Anonymous Survey” (previously numbered P-38) explained:

It was Dr. John L Norris, nonalcoholic chairman of AA’s Board of Trustees, who first stated the need for more accurate information about AA and its members.

In dealing with the medical and scientific community on the question of alcoholism and its treatment, Dr. Norris found that he could cite numerous examples of how AA works, but that he lacked facts and figures.

He posed his problem at a meeting of the Policy Committee of the Board of Trustees and requested that the Fellowship explore ways and means of providing more accurate information.

Dr. Norris stated that “There were two major reasons for undertaking the survey:”

1.    To enable AA to furnish more accurate data about the Fellowship and its effectiveness to the growing number of professionals - doctors, psychiatrists, social workers, law enforcement officials and others who are working today in the field of alcoholism.

2.    To provide AA with more information about itself so that members can work more effectively in helping the many millions of alcoholics who still suffer throughout the world.

That first survey in 1968 sampled 11,355 AA members in the United States and Canada. It was so well received and useful that the General Service Conference of Alcoholics Anonymous has continued the practice on a regular basis. The “Triennial Survey” has been conducted by AA every three years since the first survey in 1968. The 1996 survey was delayed by one year while the General Service Conference discussed its content. In general, a new edition of AA pamphlet number P-48 has been published the year after each survey to report the results:

AA Membership Survey Pamphlets (P-48) Published

1971 Profile of an AA Meeting

 

Alcoholics Anonymous 1989 Membership Survey

1974 Profile of an AA Meeting

 

Alcoholics Anonymous 1992 Membership Survey

1977 The AA Member

 

Alcoholics Anonymous 1996 Membership Survey

1980 The AA Member

 

Alcoholics Anonymous 1998 Membership Survey

1983 The AA Member

 

Alcoholics Anonymous 2001 Membership Survey

1986 AA Membership Survey

 

Alcoholics Anonymous 2004 Membership Survey

 

 

Alcoholics Anonymous 2007 Membership Survey

The 2004 AA Triennial Membership Survey occurred during the period August 1-14, 2004. Seven hundred AA groups were previously selected at random. As in prior years, survey questionnaires were distributed to the General Service Representatives (GSRs) or group contacts of the selected groups with the assistance of the Area Delegates.

The most recent survey was conducted in the summer of 2007. The results were published in 2008 and are included in this “Updated” paper. The survey was conducted at regularly scheduled AA meetings. The selected groups were specifically asked not to call a special meeting for conducting the membership survey.

All members attending the regular scheduled meeting were asked to complete a questionnaire unless they had previously done so at another meeting. The forms were anonymous and confidential. Completed questionnaires were returned to the Public Information (PI) service desk of the AA General Service Office (GSO).


 

Using Length of Sobriety as a Measurement of “Success”

For it to be possible to compare recovery outcome "success" rates, it is necessary to understand that the meaning of the word “success” is inconsistent among various observers and critics.

If a member has not had any alcohol for the last 5 years, most observers would agree that is a measure of success. It does not matter if the sobriety began the day of the first meeting or some time substantially later. The member has been sober for 5 years and that is what matters.

Now that AA is easy to find almost anywhere, many people visit AA long before entering the worst of the downward spiral of alcoholism. They already know where AA is when they finally need and want it years later. This was discussed in an AA pamphlet (P-38) published in 1970 titled “The Alcoholics Anonymous Survey.” It reported on the first AA Membership Survey in 1968 (note: the tables “3 and 4” referenced below are not in this paper):

LENGTH OF SOBRIETY

TABLE 3 Time Reported Since Last Drink

For most people who come to AA there is a point after which they just don't drink. This point may be at their first AA meeting. It may be a week, month or years after their first meeting. But once having achieved this point, experience has shown that the great majority do not resume drinking. (emphasis added)

While AA has no measure of the number of alcoholics who have gotten sober, returned to normal life without ever drinking and died without having taken a drink, we can assume that this is a normal pattern for those who accept the AA program.

Thus, AA views success as continuous sobriety - that is, the alcoholic lives a normal life without ever drinking. However, the medical and scientific community often has a less demanding criterion for success. This criterion is frequently one year of total abstinence.

Using that standard, there is apt to be a lot of successful sobriety in any AA meeting anywhere in the United States or Canada.

Of the 11,355 members who filled out questionnaires at AA meetings, 60% reported that they had not had a drink of alcohol for one year or more (emphasis added). This is one indication that AA works. According to the data which follows, it can be assumed that many of the remaining 40% are newcomers who have either not had a drink or who stopped drinking shortly after attending their first meeting.

...

HOW LONG FOR AA TO TAKE

TABLE 4 Length of Time from First Visit to AA to Time of Last Drink

Another indication that AA works well is shown by the fact that a total of 64% of the respondents in the survey reported that they had stopped drinking either immediately after their first meeting (41%) or within the first year (23%). (emphasis added)

A total of 68% of the women in the survey reported that they had stopped drinking within a year of their first meeting as compared with only 63% of the men. At the same time, 74% of those under thirty reported they had stopped drinking within a year of attending their first AA meeting, compared with 63% of those over thirty.

Here it should be noted that, while a number of those surveyed have not yet attained successful sobriety, few AA members would be apt to refer to them as hopeless. Most AA's know at least one member who attended AA meetings year after year with little or no success, but then finally sobered up. Unless the alcoholic dies, the most that the majority of AA members will say is that it hasn't worked so far.

 


 

b)      First Year Retention

 

1989S-12a2

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

Figure C-1 from the 1990 Summary Report of 1977-89 AA Triennial Membership Surveys

The tendency of some observers to offer a pessimistic view of AA today is based on a misreading of the graph shown above. It has been extracted from a summary in a 1990 internal GSO report on AA Triennial Membership Surveys. The misreading of the summary report has been circulated widely. The title of the misquoted graph is "% of those coming to AA within the first year that have remained the indicated number of months.” Identified as Figure C-1, the graph is page 12 of the 1990 internal GSO report.

This hand-written graph is at the center of erroneous assertions that contemporary AA has a 5% success rate outcome. The sequence of percentages at the bottom ends in 5% above month 12 of the x-axis of the graph. This 5% value has been erroneously interpreted as the percentage of candidates who stayed a full year and it is a completely inaccurate interpretation of what the 5% value actually represents. This paper presents considerations and discussion regarding the graph, its source data, composition, and most importantly, how it should be interpreted.

The data plotted in Figure C-1 represent a subset of the overall survey sample populations. The population subset reported a year or less since their first-time-ever attendance in AA. The x-axis of the graph shows intervals of time for the first through 12th month of attendance in AA. The points plotted for the five surveys represent the percentage distribution of the population subset in each of those intervals.

Following are citations and notes of how the narrative in Appendix C of the internal GSO report describes the graph:

APPENDIX C: THE FIRST YEAR

“It is possible to calculate from completed questionnaires, by month, the number of members that have “been around” a given number of months. This relies on the question that determines the month and year that the respondent first came to AA.

The calculation has been performed for the twelve months of the first year for the five surveys, and the results are plotted in Figure C-1. Such results can be interpreted to show the probability that a member will remain in the fellowship a given number of months.

To be more explicit: if all the members who report they have been in the Fellowship for less than a month were present a month later, then the number who report being in AA between one and two months should equal the number that report being in less than a month, subject of course, to month-to-month fluctuations and to any possible seasonal effects. The same should apply to succeeding months.

However, it is observed that there is a steady decline (subject to inevitable fluctuations).”

Note that it states, “Such results can be interpreted to show the probability that a member will remain in the fellowship a given number of months.” That type of probability is not shown in the way the graph in Figure C-1 is presented:

APPENDIX C continued: “This has been the case for each of the five surveys we are reporting on, and the remarkable similarity of results for the surveys is shown in Figure C-1, where all five are plotted on a single scale by taking into account the size of each survey. That figure also tabulates the average over the five surveys and that average strongly suggests that about half those who come to AA are gone within three months. Unfortunately, there seems to be no way in which the reason for departure can be determined.”

For the 1997 survey, the number of respondents who were in each month number of their first-time-ever AA attendance was divided by the total that was in their first year and plotted as a dash-dot line. The same type of data was plotted as a plus-plus line for the 1980 survey. The 1983 survey was plotted as a dot-dot line. The 1986 survey was plotted as a dash-dash line. Lastly, the 1989 survey was plotted as a solid line. The vertical axis is the distribution percentage in that month interval for each of the five surveys.

Figure C-1 Average of 5 Surveys

Distribution %

19%

13%

10%

9%

8%

7%

7%

6%

6%

6%

6%

5%

Month

1

2

3

4

5

6

7

8

9

10

11

12

The “Average of 5 Surveys” is tabulated along the bottom of the chart (and shown in the table above). It shows that 19% of the first-year respondents were in their first month. The third and fourth month numbers are about half of that which is how the report author interprets “that about half those who come to AA are gone within three months.

Chart 1 of this paper (which follows below) is titled “1st Year Retention per Summary of 1977-89 Surveys.” It plots the same information at a scale where the shape of the retention (or attrition) curve is easier to understand.

APPENDIX C continued: “It seems impossible that such a systematic effect could be achieved by any mechanism other than a slow attrition of newcomers during the first year. It is little comfort to suggest that many who leave return later, because those who have done that are already counted in the numbers shown here.”

Many prospects that stop coming after their first few meetings do return to AA later and achieve successful sobriety. A car dealership (or any other retail business) looks at the number of cars they sold. They do not bemoan the many window-shoppers that looked but did not buy anything.

AA co-founder, Bill Wilson, wrote in a 1939 letter, “Here in New York, it was the same story. I went along six months talking to a lot of them before any permanent results were obtained, and at that time, I was laboring under the delusion that I was divinely appointed to save all the rummies in the world!” [4]

APPENDIX C continued: “After the first year, survey results show that attrition continues, but at a much slower rate. During such years, it is likely that many whose activity in AA diminishes remain sober but are no longer adequately represented in the sample.”

Section b) of this paper, titled “First Year Retention” and section c) titled “Growth of Long Term Sobriety” address what all the triennial membership surveys to date show about long-term sobriety and continued AA participation after the first year.

APPENDIX C continued: “It is not part of the survey to attempt to determine the causes of this phenomenon beyond noting that suggestions would include individuals sent against their will, individuals who are not convinced of their alcoholism, individuals who are unable to accept one or another characteristic of the AA program, etc. But it does appear that this result and its implied challenge to AA should be widely understood in the Fellowship.”

Alcoholics Anonymous teaches that it is easier to stay sober by drinking absolutely no alcohol at all. That is pretty much a deal-breaker for a visitor who hoped AA would teach him how to control his drinking. When an alcoholic’s reasons for coming to AA exceed his/her excuses for not coming, then he/she has a chance to recover.

APPENDIX C continued: “Individuals may rebel against this result as contradicting our time-honored statement that “half get sober right away, another 25% eventually make it,” etc. That statement applies to observations made at an earlier time, and there is no reason to doubt that changes in society and in AA since that time could create a different circumstance today. Like other findings of the survey, this may be a challenge to the membership to “change the things we can.””

Section d) of this paper, titled “Timeline of Sources and Citations of AA’s 50% + 25% “Success Rates,” explores the history and important qualifications associated with the 50% + 25% success rate to provide a proper context for its interpretation.

The time-honored 50% + 25% success rate assertion is in the Foreword to the second edition Big Book and states, “Of alcoholics who came to AA and really tried, 50% got sober at once and remained that way; 25% sobered up after some relapses…” [5] The first year retention data in the surveys does not contradict the 50% + 25” success rate.

In the first few weeks or months, the prospective member typically answers two questions: (1) am I an alcoholic and (2) am I really trying? Many people come to AA to find out if they are alcoholics. For some the answer is obvious and easy. Others need to explore the question for a while.

Some AA members try to impress newcomers with rhetoric like, “You wouldn’t be here if you weren’t an alcoholic.” However, in AA each individual makes his or her own diagnosis. It is hoped that they stay long enough to understand what it means to be an alcoholic and make an informed decision.

Meeting attendance is not an absolute requirement for recovery in AA but it certainly helps. Some in isolated places have gotten by with only the Big Book and other literature. Some may work one-on-one with another alcoholic if there are no groups with meetings nearby. But that is rare enough that, for the sake of understanding the survey data, a person who stops coming to meetings in the first few months is not assumed to be “really trying.”

 


 

The “Average of 5 Surveys” noted in Figure C-1 of the 1990 internal GSO report is shown in Table 1 below as the columns “Month” and “Distribution %.” The data are plotted in Chart 1 at a normalized scale where the shape of the retention (or attrition) curve is easier to understand.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

Chart 1: 1st Year Retention per Summary of 1977-89 Surveys

Table 1: 1st Year Retention Percentages

Month

#

Distribution

%

Starting

1st month

Starting

2nd month

Starting

3rd month

Starting

4th month

1

19

100

 

 

 

2

13

68

100

 

 

3

10

53

78

100

 

4

9

47

69

89

100

5

8

42

62

79

90

6

7

39

56

70

83

7

7

36

52

65

77

8

6

34

48

61

72

9

6

32

45

57

68

10

6

30

42

54

64

11

6

28

40

52

60

12

5

26

38

50

56

Notes:

·         The "Distribution %" column of Table 1 is taken from the original “Figure C-1” graph.

·         Total Distribution % would add up to 100% (instead of 103%) if the percentages had not been rounded off.

·         Distribution % values are “normalized” to give various starting months.

·         The resulting curves show retention after various “orientation periods.”

The original Figure C-1 graph data were NOT retention percentages as has been frequently misinterpreted and erroneously reported. The reasoning is as follows:

·         Each Triennial Survey is a cross-sectional study - a snapshot at one point in time. Assume that the same number of new people have been attending their first meetings every month. That is how many will be in their first month when the observation is made.

·         The ratio of the second month people in the survey to the first month people is the retention rate between the first and second months. In that same way, it is possible to find the retention between any two sampled months. If there were perfect retention for all twelve months, then 8.3% of the first year people would be found in each of the twelve months. That is not the case but it shows how a retention calculation can be done.

In the actual data presented:

Month1   = 19       does NOT mean that "81% (i.e. 100% - 19%) dropped out in a month as some sources claim.

Month3   = 10       does NOT mean that "90% (i.e. 100% - 10%) leave within 3 months and

Month12 = 5         does NOT mean that "95% (i.e. 100% - 5%) stop active participation in AA inside of a year.

Instead, what the data does show is that for every 100 people surveyed with under a year since first attendance:

19%       of that population were in their first month

13%       were in their 2nd month

9%                          were in their 4th month

7%                          were in their 6th month

6%                          were in their 8th month, etc

By multiplying everything by a normalizing factor (5.25 in this case) such that it starts at 100%, then a reasonable approximation of retention can be derived. The "Starting 1st month" column in Table 1 is scaled to show retention of a first-time newcomer. The "Starting 4th month” column is scaled to show retention after a recommended 90-day introductory period.

·         What is actually shown in Table 1 is that 56% of those who stay beyond three months are still active in AA at the end of a year. Other Survey results show even better retention rates after the first year.

·         Another important consideration for data interpretation and context is that not everyone who attends an AA meeting is an alcoholic.

Some come to a few meetings because of pressure from home, work, the legal system, treatment facilities, friends and even AA members. Some are not “alcoholic enough” to believe they need help. Moreover, drug addicts with no drinking history are relatively common in contemporary AA. Those people were counted if they were at the meeting on the day of the survey. It is not surprising that some of them do not stay at the time of their first exposure to Alcoholics Anonymous.

The important characteristic of randomization is of question in the sample selection method in early AA membership surveys. The Area Delegate determined the groups within the area to be sampled in those early years. This presented an opportunity for sample populations to be skewed in the selection of groups. On the other hand, there are enough US and Canadian delegate areas that the net results would not be significantly affected.

In later years, the groups to be sampled were randomly selected from the AAWS/GSO list of registered groups. This too would have some opportunity for inadvertent sample bias if there were some consistent characteristic of those attending unregistered groups. Thousands of AA groups in the US and Canada do not register.

The graphs and tables that follow reflect AA membership survey data from 1968 through 2007. The accompanying analysis and commentary offers further accurate and appropriate interpretation and projection of the survey data.

 


 

c)       Growth of Long-Term Sobriety

How long have members of AA been sober?

That is one of the questions the Triennial Surveys can help to answer. Each survey is a snapshot or cross-section of AA membership by sampling those attending meetings in randomly selected groups at about the same time.

Comparing a series of surveys reveals changes in the makeup of the Fellowship. It is not necessary to track individual results for this to be a “longitudinal” study. To the extent that all surveys sample the same population, the comparisons are reasonable.

Each survey pamphlet reported member sobriety in ranges of 0-1 year, 1-5 years and over 5 years. Those were converted to 0-1 year, over 1 year and over 5 years for Chart 2 and Table 2 below. That made the graph easier to read and understand.

Each survey pamphlet also reported the average length of sobriety of members sampled. When the pamphlet said, “Over 4 years,” it was necessary to graph a more specific number.

Trend data for the increased length of sobriety is shown below in Chart 2 “Growth of Long-term Sobriety Ranges” and Table 2 “Long Term Sobriety Ranges Percentages:”

The curve in Chart 3 “Growth of Long Term Sobriety Averages” and Table 3 “Long Term Sobriety Averages” which also follow below give a “best fit” to the statistics as reported.

·         The 2004 Survey showed an increase in the length of sobriety over the 2001 Survey, as has every triennial survey since 1983.

·         As of the 2004 Survey, long-term sobriety was so prevalent that the "Greater Than Five Years" range of previous surveys was subdivided into 2 parts as follows (5-10 Years = 14%) + (>10 Years = 36%) == (> 5 Years = 50%).


 

Trend data for the progressive increased length of sobriety reflected in AA membership surveys:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chart 2: Growth of Long Term Sobriety Ranges

Notes:

·         Published pamphlet data are used for most years but a few of those only gave ranges.

·         When only a range was given, a number in that range was chosen for a smooth curve fit.

·         Values for 1977-1983 are from a 1990 internal AA paper analyzing the 1977-1989 Surveys.

·         1968 and 1971 pamphlets only say that 40% were sober less than 1 year and 60% were sober 1 year or more.

·         The 2004 Survey subdivided the “Over 5 years” category into “5 to 10 years” and “Over 10 years.

Table 2: Long Term Sobriety Ranges Percentages

Year

0-1

1-5

> 5

Pamphlet

Pamphlet

Pamphlet

Source

0-1yr

>1 yr

>5 yr

>10yr

1968

40

 

 

only this

"<1yr = 40%"

">1yr = 60%"

pamphlet

40

60

 

 

1971

40

 

 

only this

"<1yr = 40%"

">1yr = 60%"

pamphlet

40

60

 

 

1974

40

35

25

 

 

 

pamphlet

40

60

25

 

1977

37

39

24

"35-40%"

"35-40%"

"20-30%"

paper

37

63

24

 

1980

36

37

27

"35-40%"

"35-40%"

"20-30%"

paper

36

64

27

 

1983

38

37

25

"35-40%"

"35-40%"

"20-30%"

paper

38

62

25

 

1986

33

38

29

 

 

 

pamphlet

33

67

29

 

1989

34

37

29

 

 

 

pamphlet

34

66

29

 

1992

31

34

35

 

 

 

pamphlet

31

69

35

 

1996

27

28

45

 

 

 

pamphlet

27

73

45

 

1998

27

26

47

 

 

 

pamphlet

27

73

47

 

2001

30

22

48

 

 

 

pamphlet

30

70

48

 

2004

26

24

50

“>5yr” includes

"5-10yr = 14%"

">10yr = 36%"

pamphlet

26

74

50

36

2007

31

24

45

“>5yr” includes

"5-10yr = 12%"

">10yr = 33%"

pamphlet

31

69

45

33

Chart 3 and Table 3 below, offer additional supporting data showing the trend of the average length of sobriety among the population sampled. The table data represent only those who are still attending meetings. Someone who got sober in AA and who is staying sober by some other means, and not attending meetings, would not appear in the survey, nor would someone who died sober. AA does not track individual case histories.

It is very important to keep in mind the reality that attending a few AA meetings does not mean that someone is interested in staying sober. Likewise, not attending AA meetings does not mean that someone is drinking again.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chart 3: Growth of Long Term Sobriety Averages

Notes:

·         Values for 1977-1989 are from the 1990 internal AA paper analyzing the 1977-1989 Surveys.

·         Published pamphlet data are used for 1992-2007.

·         The fractions for “more than” in the pamphlets are chosen to give a smooth curve fit for the graph.

Table 3: Long Term Sobriety Averages

Year

Average

Pamphlet

Source Used

1968

*

not given

*

1971

*

not given

*

1974

*

not given

*

1977

4

not given

paper

1980

4

not given

paper

1983

4

not given

paper

1986

4.3

"52 months"

paper

1989

4.7

"more than 4 years"

paper

1992

5.5

"more than 5 years"

curve fit of pamphlet

1996

6.4

"more than 6 years"

curve fit of pamphlet

1998

7.1

"more than 7 years"

curve fit of pamphlet

2001

7.6

"more than 7 years"

curve fit of pamphlet

2004

8.1

"more than 8 years"

curve fit of pamphlet

2007

8.1

"more than 8 years"

curve fit of pamphlet


 

d)      Timeline of Sources and Citations of AA’s 50% + 25% “Success Rates”

The notion of an overall 75% successful recovery outcome rate in AA owes its durability to anecdotal repetition rather than consistent statistical demonstration. There is very little consistent, verifiable data and record keeping either validating or refuting the claim of any recovery outcome rates for AA.

The only specific population sample identified by AA co-founder Bill W, as achieving a 50% + 25% (or overall 75%) success rate, were the pioneering members who had their personal stories printed in the first edition Big Book. [6] Beyond that, the origin or validation of the percentages is neither explained nor demonstrated.

The information that follows presents a chronological series of references and citations of written works in which recovery outcome rates, and factors influencing the rates, are mentioned.

July 1938 - The earliest archival reference found to a cited success rate of 50% occurs in a letter from Bill W to Dr Richard C Cabot in which Bill W wrote:

We have never developed any accurate statistical information (emphasis added) but I should say that we have dealt with about 200 cases in all, about half of whom seem to have recovered. Doctors tell us that, almost without exception, we have been problem drinkers of a class commonly regarded as hopeless. [7]

August 1938 - The relatively small AA population at the time allowed some measure of tracking success or failure by individual member. In the minutes of the first meeting of the Alcoholic Foundation Board, [8] early New York member Hank P, at the request of Frank Amos, reported a census of the number of alcoholics and prospects in the Fellowship, which then consisted of two groups (Akron, OH, and New York City). The counts Hank P reported were:

Count

Reported

%

41

6

12

10

25

Definite on the ball

Questionables

So difficult practically denied

Definite but out of touch

Prospects

44%

6%

13%

11%

27%

94

Total

 

Out of a population of 94 members and prospects, 51 were considered “definite” recoveries, which comprised a 55% successful outcome. Ten of the 51 “definite” were “out of touch” (which is presumed to mean sober but not attending meetings or in contact with other members). The total number of members reported in August 1938 was four less than that stated when the Big Book was published in April 1939.

February 1940 - the first public reference to the 50% + 25% success rate by Bill W occurred in his talk to the dinner guests at the historic Rockefeller dinner at the Union League Club in New York City. He stated:

By the time the book was published last April there were about one hundred of us, the majority of them in the West. Although we have no exact figures, (emphasis added) in counting heads recently, we think it fair to state that of all the people who have been seriously interested (emphasis added) in this thing since the beginning, one-half have had no relapse at all. About 25% are having some trouble, or have had some trouble, but in our judgment will recover. The other 25% we do not know about.

Note the qualification “of all the people who have been seriously interested.” It provides an important context to the claimed success rate to indicate that it did not apply to 100% of everyone who showed up in AA. This key context is often overlooked or omitted.

Consequently, it can fabricate an exaggerated and unrealistic expectation of success. It can also distort the interpretation of past results to yield a false impression that a high percentage level of past AA success was achieved when, in reality, the results achieved were not even close to the percentage level presumed. As will be shown later (circa 1949 in this timeline) it is important to understand that the success rates cited by Bill W applied to a subset of about 1-2 prospects out of five (i.e. 20-40% of the total prospect population). The remaining prospect population (3-4 out of 5, or 60-80%) was described by Bill W to have “quit AA after a brief contact.”

March 1941 - The first nationally published mention of the 50% + 25% success rate occurred in the historic Saturday Evening Post article written by Jack Alexander in which he stated:

One-hundred-per-cent effectiveness with nonpsychotic drinkers who sincerely want to quit (emphasis added) is claimed by the workers of Alcoholics Anonymous. The program will not work, they add, with only those who “want to want to quit” or who want to quit because they are afraid of losing their families or their jobs. The effective desire, they state, must be based upon enlightened self-interest; the applicant must want to get away from liquor to head off incarceration or premature death. He must be fed up with the stark social loneliness which engulfs the uncontrolled drinker and he must want to put some order into his bungled life.

As it is impossible to disqualify all borderline applicants, the working percentage of recovery falls below the 100 per-cent mark. According to AA estimation, 50% of the alcoholics taken in hand recover almost immediately; 25% get well after suffering a relapse or two, and the rest remain doubtful. This rate of success is exceptionally high. Statistics on traditional medical and religious cures are lacking, but it has been informally estimated that they are no more than 2 or 3 per cent effective on run-of-the-mine cases. [9]

Note the qualifier of “according to AA estimation.” It does not state “according to AA records.” Early 1941 membership had grown to an estimated 2,000 members and soared to an estimated 8,000 members by the end of 1941 because of the historic article. [10] The assertion of “One-hundred-per-cent effectiveness with non-psychotic drinkers” is obvious hyperbole but it is also qualified as including only those “drinkers who sincerely want to quit.”

Some early AA groups did try to track membership counts, sober time and slips. However, these limited records are not an adequate basis for claiming a national or regional recovery outcome rate. From 1941 on, the 50% + 25% success formula became an AA mantra and unvarying fixed rate. The percentages are often cited out of context as if they applied to the total prospect population who encountered AA. This is not so. As noted previously, and as will be seen again later, the percentages were qualified to apply to a much smaller subset of the total prospect population.

January 1944 - About 3 years after the Jack Alexander article, Dr Harry Tiebout, in a paper for “The American Journal of Psychiatry” stated a similar, but qualified, assertion about recovery rates in AA from 1935 through 1942: [11]

Statistics at the New York office of the organization read as follows:

5 recovered at the end of the first year,

15 recovered at the end of the second year

40 recovered at the end of the third year

100 recovered at the end of the fourth year

400 recovered at the end of the fifth year

2000 recovered at the end of the sixth year

8000 recovered at the end of the seventh year

Alcoholics Anonymous claims a recovery rate of 75% of those who really try their methods (emphasis added).”

Note the qualification “of those who really try their methods” that the recovery outcome rate of 75% applies to a subset of the total population of prospects.

May 1944 - Further qualifications (regarding context) appear in a reprint of a talk to the Medical Society of the State of New York, in which Bill W stated: [12]

Alcoholics Anonymous is an informal fellowship of about 12,000 former alcoholic men and women who are to be found banded together as groups in about 325 American and Canadian communities, these groups ranging in size from half a dozen to many hundreds in individuals. Our oldest members have been sober eight to nearly ten years.

Of those sincerely willing to stop drinking (emphasis added) about 50% of those has done so at once, 25% after a few relapses, and most of the remainder have improved. It is probable that half of our members, had they not been drinkers, would have appeared in ordinary life to be normal people. The other half would have appeared as more or less pronounced neurotics.

Once again, note the context “Of those sincerely willing to stop drinking.” In other instances where AA contributed information for professional publications, the same traditional formula appeared.

June 1947 - “Survey Midmonthly” magazine contained an article titled “Problem Drinkers” which stated:

Now in its thirteenth year, AA has 1,200 chapters including outposts in Canada and Latin America, and is winning about 1,000 new members a month, according to the February issue of Time. Supported by donations from members, it has no offices, no dues, no big funds. Members are pledged to help all other alcoholics, but give assistance only when called upon. Anonymity is an important rule of the organization, in order that new members may be encouraged to join.

Of its members, some 50 per cent have stopped drinking entirely after joining, 25 per cent have succeeded after one or two slips. By contrast, all but 5 per cent of alcoholics were formerly considered hopeless of cure, according to the Time report. [13]

The success rates cited above are not qualified and give the appearance of applying to the total prospect population.

April 1949 - An issue of “Survey” contained the following excerpts: [14]

One seed was planted in Akron, Ohio by two habitual drunks. One was a doctor, the other a broker, both of some distinction before alcohol addiction began to ruin their careers and threaten to break up their homes. Intelligent men, they fought hard, but without much success, until they managed to get together in a sober interval. Then they decided that one drunk might help another. Out of this idea grew the now famous Alcoholics Anonymous. In a year's time, these two former inebriates had not only achieved continuous sobriety for themselves, but had helped others to the same goal.

Today the organization they started has some 85,000 members in 2,400 chapters throughout the country. The entire membership is composed of alcoholics fighting desperately to help each other stay away from the substance that is poison to them. Many of the members have not touched a drop since they became AA’s. Others have slipped from time to time, but have returned to the organization to keep up the fight. Some former members have sunk back into the mire of perpetual drunkenness. But Alcoholics Anonymous, which maintains that 75 percent of its members have achieved sobriety, is generally conceded to present the most widely successful attempt at alcoholic rehabilitation in this country's history.

Statistics on “success” are unreliable (emphasis added) for a man who is sober today may be drunk tomorrow - even though his sobriety has lasted over a number of years. Nevertheless, doctors, scientists, social workers, clergymen, public health experts, suffering relatives, and others who have had to deal with alcoholics, have watched the AA's achievements with amazement.”

Note the qualification in the Survey article that “Statistics on ‘success’ are unreliable.” The success rate figures also are not qualified. However, AA was not the only one claiming remarkable success rates with alcoholics at the time. In the 1949 Survey article, the author also comments on the Yale Plan Clinics:

... Diagnosis was from the first a major concern of the Yale Plan Clinic, but experience soon demonstrated that if diagnosis was to have any meaning it would have to be followed by therapy and guidance in cases needing treatment not provided by other community services. Since its opening, the clinic has received 1,100 alcoholics, 60 per cent of whom have achieved either complete sobriety or markedly lengthened spacing between their drinking bouts. Referrals between the clinic and Alcoholics Anonymous are commonplace, the clinic getting patients from AA and in turn recommending AA to persons who seem able to benefit from the fellowship program. Many persons are clinic patients and active AA members at the same time.

The Survey article also discussed an alternative to AA whose founder claimed an impressive success rate as well:

... Edward J McGoldrick Jr, director of New York City's Bureau of Alcoholic Therapy, established within the Department of Welfare in 1943, is an individualist among alcoholic therapists for he also holds out against the theory that alcoholism is a disease.

... All therapists at Bridge House are former alcoholics who have been rehabilitated through the McGoldrick method. Though the method differs from procedures of Alcoholics Anonymous, the director goes along with them in the theory that persons who have “hit bottom” as alcoholics themselves can more easily help other alcoholics.

Mr. McGoldrick objects to calling alcoholism a disease on the grounds that it adds to the alcoholic's sense of weakness and helplessness, thus giving him an excuse to go on drinking. He opposes compulsory treatment as useless, for it ignores the ingredient of positive willingness, which he feels, is necessary to reform. Bridge House, with only twenty beds, serves about 350 alcoholics a year, both on a resident and a non-resident basis. Its record of success, using Mr. McGoldrick's measurement of one year of complete sobriety, is 66 percent - a good record but one not affecting some 200,000 alcoholics in New York City who do not reach Bridge House, nor any of the city's alcoholic women. It is, however, a project being watched throughout the country.

November 1949 - In an article in “The American Journal of Psychiatry” Bill W wrote:

Of alcoholics who stay with us and really try (emphasis added) 50% get sober at once and stay that way, 25% do so after some time and the remainder usually show improvement. But many problem drinkers do quit AA after a brief contact, maybe three or four out of five (emphasis added). Some are too psychopathic or damaged. But the majority have powerful rationalizations yet to be broken down. Exactly this does happen provided they get what AA calls “good exposure” on first contact. Alcohol then builds such a hot fire that they are finally driven back to us, often years later.

Note the qualification by Bill W that the success rate applied to a subset of about 1-2 prospects out of five and consisted “Of alcoholics who stay with us and really try.” The remainder of the prospects “maybe three or four out of five” [i.e. 60-80%] “quit AA after a brief contact.”

July 1955 - The Foreword to the second edition Big Book contains the following excerpt:

Of alcoholics who came to AA and really tried (emphasis added) 50% got sober at once and remained that way; 25% sobered up after some relapses, and among the remainder, those who stayed on with AA showed improvement. Other thousands came to a few AA meetings and at first decided they didn't want the program. But a great number of these-about two out of three-began to return as time passed.

Bill W wrote this foreword and again, notes the context “Of alcoholics who came to AA and really tried.” A precise demonstration of who came, who stayed and who returned to AA as of 1955 (AA’s 20th Anniversary) is anyone’s guess and that is all it was, “a best estimate.” However, put something, anything, in the Big Book, especially if it was written by Bill W, and many AA members will interpret it as infallible revelation and beyond refutation.


 

July 1955 - The introduction to the story section in the second edition Big Book states:

When first published in 1939, this book carried twenty-nine stories about alcoholics. To secure maximum identification with the greatest number of readers, the new Second Edition (1955) carries a considerably enlarged story section, as above described. Concerning the original twenty-nine case histories, it is a deep satisfaction to record, as of 1955, that twenty-two have apparently made full recovery from their alcoholism. Of these, fifteen have remained completely sober for an average of seventeen years each, according to our best knowledge and belief. [15]

In the introduction to the Pioneers of AA Section stories, it goes on to state: [16]

Dr Bob and the twelve men and women who here tell their stories were among the early members of AA’s first groups. Though three have passed away of natural causes, all have maintained complete sobriety for periods ranging from fifteen to nineteen years as of this date 1955. Today, hundreds of additional AA members can be found who have had no relapse for at least fifteen years. All of these then are the pioneers of AA. They bear witness that release from alcoholism can really be permanent.

Twenty-two of the 29 stories in the first edition Big Book were dropped for the second edition. This population has also been the subject of an erroneous myth that the stories were dropped because the members returned to drinking. The reality was quite the opposite. The 22 stories were changed simply to reflect a better cross-section of AA membership population at the time (1955).

·         According to Bill W’s introduction to the stories in the second edition Big Book, of the 29 early members, whose stories appeared in the first edition Big Book, 76% (22 of 29) were sober as of AA’s 20th anniversary (1955).

·         Seven of the 29 early members (24%) had returned to drinking but subsequently sobered up again; another seven of the 29 (24%) returned to drinking and did not sober up.

This data appears to be the only population sample explicitly supporting an assertion of the successful 50% + 25% recovery outcome. Its scope would be confined to three groups (Akron, New York and Cleveland) and its population sample (29) would represent around 30% of the original estimate of 100 AA members as of April1939.

April 1958 - Bill W addressed the New York City Medical Society on Alcoholism [17] on AA’s early history:

Our next need was publicity, and it was forthcoming. Fulton Oursler, the noted editor and writer, printed a piece in Liberty magazine about us in 1939. The following year, John D Rockefeller, Jr, gave AA a dinner which was widely publicized. The next year, 1941, there was a feature article in the Saturday Evening Post. This story alone brought us thousands of new people. As we gained size, we also gained in effectiveness. The recovery rate went way up.

Of all those who really tried AA (emphasis added) a large per cent made it at once, others finally made it; and still others, if they stayed with us, were definitely improved. Our high recovery rate has since held, even with those who first wrote their stories in the original edition of ‘Alcoholics Anonymous.’ In fact, 75 per cent of these finally achieved sobriety. Only 25 per cent died or went mad. Most of those still alive have now been sober for an average of twenty years.

In our early days, and since, we have found that great numbers of alcoholics approach us and then turn away-maybe three out of five, today (emphasis added). But we have happily learned that the majority of them later return, provided they are not too psychopathic or too brain-damaged. Once they have learned from the lips of other alcoholics that they are beset by an often fatal malady, their further drinking only turns up the screw. Eventually they are forced back into AA; they must do or die. Sometimes this happens years after the first exposure. The ultimate recovery rate in AA is therefore a lot higher than we at first thought it could be.

In Bill W’s 1958 address, the 50% number has been replaced by the term “a large percent.” Note also Bill’s comment about prospects coming to AA and then turning away. The 75% recovery rate assertion is confined to the AA pioneers who had their stories printed in the first edition Big Book.

Perhaps more telling was Bill’s comment that “In our early days, and since, [i.e. up to 1958] we have found that great numbers of alcoholics approach us and then turn away-maybe three out of five, today” [i.e. 60%]. At this point in AA’s history, there was no way to either confirm or refute the success rates quoted by Bill W. Records simply did not exist.

April 1960 - In a presentation to the General Service Conference, Bill W commented:

What does the future hold for us? What, may its perils be like? To put it negatively, how can we strengthen ourselves against the time of peril? To put it positively, how can we increase our communication with the sea of drunks around us? I realize that while I have been saying nice things about us, congratulating us upon our successes so far, we ought never neglect to soberly reflect that this Society has made only the smallest scratch on the total problem of alcoholism … I took note ... that in this generation which has seen AA come alive, this period of 25 years, a vast procession of the world's drunks has passed in front of us and over the precipice. Worldwide, there would appear to have been something like 25 million of them. And out of this stream of despair, illness, misery and death, we have fished out just one in a hundred in the last 25 years.

This statement by Bill W sometimes is misinterpreted as meaning that 25 million alcoholics came to AA but only one in a hundred stayed. Careful reading of the colorful wording shows that the procession of drunks “passed in front of us.” He does not say that they “passed through us.” This is about the “worldwide” problem of alcoholism during its first twenty-five years when Alcoholics Anonymous was almost unknown outside of the United States and Canada and was far from being “available everywhere” even within the United States.

This is not a pessimistic view of the lack of AA success. Rather, it is a challenge for the AA of the future to expand its reach to alcoholics everywhere it possibly can. A visionary Bill W was calling for outreach to more alcoholics and saw that AA was achieving only a small fraction of what was possible. He knows the product is good but needs to be available in more places.

It is possible to quote “one in a hundred” out of context and call it a 1% success rate. Instead, it describes a miracle. In the twenty-five years after two men in Akron, OH met and stopped drinking, their methods helped 1% of all of the alcoholics in the whole world. Later in 1960, in the book “AA Today” [18] Bill W presented the same theme with less ambiguity. He stated, “The rest are still out of reach,” that is because they either don’t know about Alcoholics Anonymous or else, literally, AA has not reached to where they live yet. Those millions of alcoholics around the world are not going to “hear about us and come around” on their own.

During the last twenty-five years, it is quite certain that twenty-five million men and women throughout the world have suffered alcoholism. Nearly all of these are now sick, mad or dead.

AA has brought recovery to something like two hundred and fifty thousand. The rest are still out of reach or else gone beyond recall. An even larger generation of drunks is right now in the making. Facing the enormity of this situation, shall any of us sit comfortably and say, “Well, people, here we are. We hope you hear about us and come around. Then maybe we can give you a hand.”

Of course, we shall do nothing of the sort. We know that we are going to open, wider and wider, every conceivable means and channel through which these kinfolk of ours may be reached. As the inheritors of such a tradition of service, how many of us could ever say “Let George do that Twelfth Step job; he likes to work with drunks anyhow. Besides, I’m busy?” Surely, there could not be many! Complacency would be impossible.

Our next great area of future responsibility may be this one: I’m thinking about the total problem of alcohol and about all of those who still must suffer the appalling consequences of alcoholism. Their number is astronomical.”

Bill W further stated that other “enterprises - public, state and private - have been springing up everywhere in recognition of the fact that alcoholism is a top-priority of health.” He chided AA members for their resentment or suspicion of these agencies and suggested that further cooperation with the friends of AA is necessary.

Could not still more friendly and widespread cooperation with outside agencies finally lead us to countless alcoholics who will otherwise be lost? Maybe we are beginning to stand in our own light. Perhaps we are blocking a communication that has a tremendous potential.

Shouldn’t we therefore have a fresh look at this?

Nearly all of us, when we think about it, agree that we are a long, long way from being anywhere near grown-up, from almost any point of view. We can clearly see that our job as individuals and as a fellowship is to keep right on growing by the constant use of our Twelve Steps.

1968 - In the first AA Membership Survey pamphlet distributed by AAWS/GSO (titled “The Alcoholics Anonymous Survey - 11,355 AA members answer questions about themselves”) [19] page 7 contains the following excerpts:

However, it is now possible for AA members not only to check over the following questionnaire but to measure their own experiences in AA against those of a broad sampling of the membership. This is one result of the survey taken among more than 10,000 AA members attending meetings throughout North America during the summer of 1968.

Up to now, the only broad gauge of measurement and comparison available to all members has been the rule-of-thumb estimate developed in the early years of the Fellowship. This surprisingly accurate estimate held that of all those who seriously attend AA (emphasis added) 50% stop drinking immediately or within the first few weeks; an additional 25% eventually stop drinking and the remaining 25% don’t seem to be able to make it for one reason or another.

Other surveys have been taken over the years. Among them, one in Texas traced the experience of the members of one -group. Another, more elaborate survey was conducted among members of groups in New York City and yet another in England.

The results of these surveys were heartening in that they tended to confirm the 50-25-25 estimates, as well as the fact that AA worked in different areas for a great many people on a long-term basis. However, there were some drawbacks to these earlier surveys. For instance, samples were either too small or taken from too limited a geographical area. In some cases, respondents were selected rather than picked at random.

 


 

e)       Assessment of Claims of AA Success or Failures and Corroboration

No one can accurately define the number of AA members who stayed sober or returned to drinking beyond AA’s formative years (i.e. 1935 to 1939). Yet the challenge remains to provide a reasonable, figurative estimate of results achieved (past and present). There should also be some measurement of the magnitude of the problem population that AA seeks to assist.

The 2008-2013 National Institute on Alcohol Abuse and Alcoholism (NIAAA) Strategic Plan [20] contains citations on the “Prevalence of Alcohol Problems and Their Consequences.” The plan reports that the US Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) came to similar conclusions in the US and globally, about the toll taken by excessive alcohol use and the population it encompasses:

·      Alcohol was the third leading cause of death in the US in 2003 (an estimated 85,000 deaths).

·      According to the NIAAA National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) [21] the number of American adults who abuse alcohol or are alcohol dependent rose from 13.8 million in 1991-1992 to 17.6 million in 2001-2002 (i.e. 8.5% of the population 18 years and older or about 1 in every 12 adults).

·      A June 10, 2004 National Institutes for Health (NIH) News Release summarized the NESARC survey data as:

1991-1992: alcohol abuse: 5.6 million, alcohol dependence: 8.2 million, total: 13.8 million.

2001-2002: alcohol abuse: 9.7 million, alcohol dependence: 7.9 million, total: 17.6 million.

·      “Alcohol abuse” is defined by the NIAAA as a condition characterized by failure to fulfill major role obligations at work, school, or home, interpersonal social and legal problems, and/or drinking in hazardous situations. It might include a single instance of an adverse alcohol-related consequence. Persons in this category may or may not be alcoholics. Conversely, it would include some measure of those beginning their progression of a downward spiral of alcoholism (problem drinking). It is unfortunate that a more specific proportional distinction cannot be made.

·      The NIAAA defines “alcohol dependence” (also known as “alcoholism”) as a condition characterized by impaired control over drinking, compulsive drinking, preoccupation with drinking, tolerance to alcohol and/or withdrawal symptoms. The estimated 7.9 million in this population is close to the prime candidates that some in AA call "real” or “active” alcoholics or other similar terms.

·      The NIAAA defines “alcohol use disorder” to include the 9.7 million alcohol abuse plus the 7.9 million alcohol dependence for a 2002 total population of 17.6 million.

In 2002, global AA membership was around 2.1 million (1.2 million of that in the US). [22] These membership numbers are likely understated; even so, it is a substantial quantitative indicator of AA success.

·         AA was helping 1 alcoholic for every 7 active alcohol dependent adults. If we consider alcohol abusers to be potential members then it is 1 AA member for every 15 problem drinkers.

·         Corresponding research in the 1991-1992 NIAAA "National Longitudinal Alcohol Epidemiologic Survey” (NLAES) showed similar numbers. There was 1 AA member for every 7 alcohol dependent people and 1 for every 12 either alcohol dependent or alcohol abusers.

As cited earlier in this paper, Bill W reminded the 1960 General Service Conference “... we ought never neglect to soberly reflect that this Society has made only the smallest scratch on the total problem of alcoholism...” No longer just scratching the surface, AA has been helping a major portion of its prospects.


 

Assessment of Claims of AA Success:

A “50% + 25%” claim for successful AA outcomes was first publicly cited in 1940 and numerous times since. It raises a question of what other demographic measurement has remained so rigidly fixed and unchanging over the course of 7+ decades for three categorizations (i.e. 50% immediate success, 25% drink but return and 25% failure).

The basis of the numbers is primarily anecdotal and its repetitive citation has endowed it with an image of absolute precision. Yet, overall, the estimate of AA success may well be reasonable depending on the scope and qualification of the prospect population included in the estimate.

What is all too often missing from discussions of AA success (or failure) outcomes is that out of all the prospects that show up at AA meetings, or are contacted by AA members, only a fraction of them will go beyond that. This population poses a formidable challenge to AA in its worldwide outreach.

Past estimates by AA co-founder Bill W stated that 20% to 40% of the prospects who showed up in AA seriously tried AA. The “50% + 25%” success outcome rate claimed by Bill W applied (and always has applied) exclusively to this segment. Bill W reported this significant qualification in the American Journal of Psychiatry (November 1949), the Foreword to the second edition Big Book (1955) and to the New York City Medical Society on Alcoholism (1958).

Including the 60% to 80% of the prospects who showed up and did not try AA, in any outcome calculation is not only absurd, it is tantamount to measuring the effectiveness of a medical procedure by including those who suffer a medical problem but do not seek medical help. One must at least try a remedy to assess its efficacy - so too with AA’s recovery program.

The subtitle of AA’s Big Book is “The Story of How Many Thousands of Men and Women Have Recovered from Alcoholism.” Quantitatively, over the past 7+ decades it is quite likely that the success of “many thousands” has been passed on to “millions” worldwide.

Assessment of Claims of AA Failure:

The worldwide scale and consequences of alcohol abuse and dependence remain severe. In 2003, nearly 17 million prospects in the US needed help for their alcohol problem. To provide context, one should be aware that “needing help” can open the door to an AA meeting, or a visit by an AA member, however “wanting help” is far more relevant to achieving successful sobriety and recovery.

Since AA’s beginning, about 60% to 80% of those who show up at AA meetings observe but do not really try AA. They engage in investigation but not participation. Many are simply uninterested or unaware that AA can help them. That others are not seeking a solution in AA may well be a function of their intrinsic nature of being averse or unable to admit what they are, recognize how their life is unmanageable and being willing to do something about it.

Given these traits and potential size of the population it should not be surprising that AA successes are made up of a fraction of those who show up at AA meetings or are contacted by AA members. AA faces a daunting challenge trying to help this segment of the problem population because it intrinsically refuses or resists being helped:

The size of this population does not denigrate the efficacy of AA (past or present) or represent AA failure. It illustrates the very difficult nature of AA’s mission to assist prospects who are commonly unwilling and uncooperative in seeking a solution even in the face of alcohol problems having the severest of consequences.

Sometimes labeled as “denial” or “defiance” it is a morbid peculiarity that makes alcoholism so powerfully destructive. It also speaks to the core principle of AA’s First Step, the preeminent action of a prospect admitting to an alcohol problem and being receptive to a well-tested solution (i.e. “We learned that we had to fully concede to our innermost selves that we were alcoholics”). [23]

Some come to AA because of pressure by another person (i.e. family, friend, doctor, nurse, clergy or lawyer) or institution (i.e. job, treatment facility or legal system). However, many who are forced to attend AA get sober anyhow.

Others would prefer a solution that does not involve stopping drinking altogether. Many of them who show up at AA meetings after a time elect to no longer attend meetings or try AA. In addition, the growing number of drug addicts with no drinking history, and the number of involuntary court-ordered, and employer-ordered, attendees at AA meetings further contributes to this population.

·         Some 20% to 40% of the prospect population comes to AA because they want to get sober. They will try what millions before them have successfully tried. They bring with them, or develop some elemental measure of willingness, honesty and open mindedness which AA’s Big Book defines as “the essentials of recovery” and further emphasizes that “... these are indispensible.” [24]

·         Citations of the 50% + 25% success rate by Bill W, et al, were qualified to apply only to this 20% to 40% of the prospect population who “really tried.” Bill W attempted to emphasize this consideration but it is frequently obscured or ignored.

·         The 60% to 80% who do not try are not a failure of AA. It is simple common sense that if zero efforts are put into AA then zero results will come out of it. This is not a function of the efficacy of AA; it is a function of the willingness of a prospect to make a good-faith effort to try AA’s recovery program.

Assessment of Claims of Corroboration and Citations:

Among those who propagate the failure myth of a 5% or less AA recovery rate, assertions are made that AA’s alleged dire (and distorted) recovery rate can be restored based on a certain type of beginner’s meeting or Step choreography or verbal, as opposed to written, inventory or a certain way of praying and meditating or through Scripture, etc., etc.

These claimed restorative acts are a product of wishful imagination and speculation as opposed to reliable and substantiated historical investigation, information and demonstration. Flawed data gathering techniques, and flawed assertions of cause and effect, remain flawed regardless of where they are cited or who constructed them.

Historical analyses should consist of some measure of scholastic study and scrutiny coupled with some minimal attempt at verification or refutation of the accuracy of the data observed.

This is not to imply that anyone is being maliciously deceptive. As an illustration of the complexity of the matter, simply try, in the case of any moderately sized AA group today, to derive a success or failure rate over the last 6 months to one year. AA’s structure, autonomy and emphasis on anonymity make attempts to collect precise outcome data near impossible.

A phenomenon, related to assertions of dismal AA recovery rate outcomes, concerns the growing frequency and propagation of incorrect declarations and citations that “studies have shown …” and claims that “studies” corroborate the assertions made. In far too many cases, the citations merely rise to the level of perfunctory contentions and cite phantom studies. No specific academic papers, or other documents, are identified by topic, author and date.

Many citations merely refer to other spurious citations that also do not reference specific studies but simply claim that they exist when, in fact, they do not. Consequently, the erroneous claims of AA failure take on an appearance of validity due to repetition rather than demonstration.

There are also instances where specific documents are cited that offer no substantiation whatsoever to the failure claims and conclusions asserted.


 

Assessment of Claims of Success in this paper:

Assessment of AA success or failure outcomes is complicated by the fact that there is no consistent record keeping within AA to conclusively verify or refute assertions of positive or negative outcomes. AA’s autonomous and anonymous structure makes it a challenge just to obtain reasonable historical estimates of the number of AA groups and members, much less precise measurements of success or failure outcomes over the 7+ decade history of AA.

However, the AA membership surveys, discussed in this paper have, since 1968, provided a means of obtaining reliable demographic trend statistics provided, as with all sampled populations, the data are randomly sampled, interpreted in proper context, and most importantly, described ethically and accurately.

Sections a) b) and c) of this paper show graphs, tables and analyses of past AA Triennial Membership Survey Data. They illustrate and hopefully clarify how the survey data should actually have been interpreted. The Survey data are also updated with projections through 2007.

In summary, the data reveal that:

·         Of those in their first month of AA meeting attendance, 26% will still be attending AA at the end of that year.

·         Of those in their fourth month of AA meeting attendance (i.e. have stayed beyond 90-days) 56% will still be attending AA at the end of that year.

·         Other Survey results show even better AA retention rates after the first year.

·         The 2004 Survey showed an increase in the length of sobriety over the 2001 Survey (as has every triennial survey since 1983).

·         As of the 2004 Survey, long-term AA sobriety was so prevalent that the "Greater Than Five Years" range of previous surveys was subdivided into: 5-10 Years (14%) , >10 Years (36%), > 5 Years (50%).

·         For growth of AA sobriety ranges, the 1983 Survey showed 25% of AA members sober over 5 years and the 2004 Survey showed 50% of AA members sober over 5 years.

·         For growth of AA sobriety averages, the 1983 Survey found the average AA member sober for 4 years and the 2004 Survey found the average AA member sober for more than 8 years.

The above are not measurements of failure.

 


 

f)        Prescreening In Early AA

To get success rates as high as 50% first time through and 75% eventually (the normally cited figures for good early AA) you have to do either (a) prescreening and/or (b) exclude people from the sample who only came to a relatively few meetings and then dropped out because of lack of real internal motivation.

Even from the beginning, they wanted to make sure that they were working with someone who wanted to get sober, to quit drinking for good. In the Big Book (page 182 in the 4th edition), the man who became “Alcoholics Anonymous Number Three” describes when AA co-founders Bill W and Dr Bob first visited him in the hospital:

They said to me, “Do you want to quit drinking? It’s none of our business about your drinking. We’re not up here trying to take any of your rights or privileges away from you, but we have a program whereby we think we can stay sober. Part of that program is that we take it to someone else who needs it and wants it. Now, if you don’t want it, we’ll not take up your time, and we’ll be going and looking for someone else.”

The use of prescreening in early AA was common to both verify the prospects were (typically chronic or “low bottom”) alcoholics and have them experience a “surrender” prior to introducing them (or having prescreening members “sponsor” them) to the groups.

In early Akron, AA co-founder Dr Bob did very vigorous prescreening before he would begin working with prospects. With a good eye, one can always increase success rate by prescreening in this fashion.

After Dr Bob's death, Sister Ignatia continued to have high statistics at St Thomas Hospital because she made the AA members do her prescreening for her. She insisted that an Akron AA member in good standing had to "sponsor" the newcomer. In addition to making sure he had AA visitors, this meant agreeing to pay the newcomer's hospital expenses if he dropped out and did not pay it himself. You can be certain that this made the prescreening rigorous indeed. Sister Ignatia normally allowed people only one chance to go through hospital treatment. On rare occasions, she would let a patient come back for a second try, but that patient would be completely isolated from the other incoming alcoholics, so as not to tear down morale. In addition, no one at all got a third chance. That also increases the statistics on "successes." Similar practices appear to have been followed in New York City and Cleveland.

When an alcoholism treatment center allows people to come back for treatment several times in a row, it greatly pulls down the apparent success rate. References to this can be found in Mary Darrah's biography “Sister Ignatia.” Sgt. Bill S also notes it in his book “On the Military Firing Line in the Alcoholism Treatment Program” discussed later in this section. Sgt. Bill recounts the year he spent visiting Sister Ignatia at St Thomas Hospital and seeing how she did it.

When Cleveland members separated from the Oxford Group (and Akron, OH meetings), they adopted a very rigorous prescreening procedure for prospects. Today, some AA members would consider it to be in direct opposition to the spirit and letter of AA’s Tradition Three. Others might interpret it as a rather rigorous set of beginners meetings separate from regular meetings. Akron and Cleveland members worked very hard with these prospects. It should also be noted that the early AA membership population primarily consisted of alcoholics that were considered hopeless and beyond any help.

In “Dr Bob and the Good Oldtimers” one of the Cleveland successes states:

After Clarence talked to me at my home, others would come over and talk to me. [25] They wouldn’t let you in a meeting just by one guy talking to you, as they do now. They felt you should know something about what you were going to hear and the purpose of the program. Then Clarence made me go to the home of one of the newer members every night for 3 months,[26] and they had 9 or 10 people talking to me. Then I had to read the Big Book before I went to my first meeting. As a result, I think I had a better understanding of what they were trying to do.”

The following are additional excerpts [27] from Dr Bob and the Good Oldtimers” that condition the context of a claimed “93%” success rate for the Cleveland, OH groups. It would likely be better qualified as Cleveland achieving a 93% success rate among prescreened prospects that were already demonstrated to be successful.

Meetings in Cleveland evolved somewhat differently from those in Akron. “We opened with an audible prayer,” said Clarence S. “The speaker, who was chosen four weeks in advance, spoke for 45 minutes, and we closed with the Lord’s Prayer. Then, we would reopen for informal comments, questions, and so forth. The total meeting might go on anywhere from one and a half to two hours. No smoking was allowed in the first part of the meeting, only in the informal part.” “That’s the trouble,” Clarence said. “They take it so casually today, I think a little discipline is necessary. I think AA was more effective in those days. Records in Cleveland show that 93 percent of those who came to us never had a drink again. When I discovered that people had slips in AA, it really shook me up. Today it’s all watered down so much. Anyone can wander in now.”

Cleveland’s rigorous prescreening of prospects is also described in “Dr Bob and the Good Oldtimers.” It does not appear to be a practice that was done for any negative or harsh reasons, but for practical reasons. A possible intent behind the prescreening of a very difficult and challenging population of chronic alcoholics was that the early membership base could serve as a dependable and living example that the AA recovery program worked.

The active or even recently active alcoholic was definitely not welcome at early meetings in Cleveland. In September 1940, Clarence wrote Bill that “several groups do not permit a rummy to attend unless he has been hospitalized or talked to by ten men.” Clarence noted that they then had a “definite setup” with three hospitals and two sanitariums, and that there were ten to 15 hospitalized at all times. By January 1941, requirements had eased up - slightly. Clarence wrote that “most groups” required either hospitalization, being talked to by a least five members, or being passed by a committee before a new person could attend meetings.

In Youngstown, it was usual for two couples to visit the prospective member before he attended his first meeting. The husband would tell the man about AA, and the woman would talk to the wife. “That way, they would know what it was all about when they finally got to AA” said Norman V. “Various groups have various distinctions” Clarence wrote. “But the general idea is to try and prepare a fellow and give him a pretty good understanding of the aims and principles of AA before he comes to meetings. This eliminates much of the nuisance of entertaining boys under the influence at our meetings.”

The prospect had to demonstrate the ability to abstain from alcohol for up to 3 months prior to being allowed to attend a meeting. This type of prescreening restriction excluded those who could not stay sober from being factored into Cleveland’s reported “93%” success rate. A similar reference can be found in the book “Slaying the Dragon.” [28]

As AA completed its developmental separation from the Oxford Group and moved toward publication of the Big Book, other significant but less observable milestones occurred. Rules evolved (rules that were later relaxed) governing when a potential member, known variably as a "prospect," "baby," "pigeon," "fish," or "suspect," could first attend a meeting. Several Cleveland groups, for example, would not allow any prospective member to attend a meeting until he had either been detoxified in a hospital or talked to by ten members. A Denver Group would not allow prospects to attend meetings until they had taken the Steps. (41)

Endnote (41): P., Wally (1995) "But, For the Grace of God...How Intergroups & Central Offices Carried the Message of Alcoholics Anonymous in the 1940s" Wheeling, WV: The Bishop of Books.

The Cleveland area groups essentially “cherry picked” prospects who had already previously achieved recovery and demonstrated the ability to stay sober. They effectively went through a probationary or introductory period before being allowed to attend Cleveland meetings.

Under the circumstances, portraying the Cleveland practices as achieving a “93%” success rate is tantamount to claiming a 93% success rate among prescreened prospects that were already demonstrated to be 100% successful for up to 3 months before they attended their first meeting.

Without an indication of the number of prescreened “unsuccessful” alcoholics who were not allowed to attend Cleveland meetings, the reputed “93%” success rate is so dominantly biased to a demonstrated favorable subset of the prospect population as to be a dubious claim at best.

Bill W had solid reasons for being selective in which prospects to approach, especially in getting a new group going. It was not to “get the numbers up.” He saw it as essential to the success of Alcoholics Anonymous. In a 1939 letter [29] to Earl T, a member who was trying to start AA in Chicago, Bill advised:

It is usually a big job, in fact a hell of a job, to get a group functioning in a new locality, but once you have eight or ten really on the ball, things go faster and much easier. Our experience shows that we cannot in the beginning walk into public hospitals or snatch lushers off the street willy-nilly and have much but a headache.

It is very easy in this way to attract a big fellowship of panhandlers and mentally defective people. Surely, they are all as important in God’s sight as any of the rest of us. They have just had a tougher break, and we are finding that later on, when a group gets size and power, quite a number of such individuals can be assimilated, and those who can’t, or won’t, fall away quickly; but if you get too many of them in the beginning, you are likely to find that your home becomes a drinking club, a hospital, a bank, or a nursery.

A June 1947 “Survey Midmonthly” magazine article (cited previously) titled “Problem Drinkers” noted the use of prescreening with the early prison and probation population:

AA chapters have been established in more than thirty penal and correctional institutions. Meeting once a week in groups of thirty or forty, under the supervision of prison authorities and chapter members, inmates hear talks by visiting AA members, and are encouraged to ask questions, make suggestions, and join in the discussion. New members are carefully screened to make sure of their sincere desire to stop drinking (emphasis added). In AA's work with alcoholic probationers, the prisoner is placed on probation to an AA "sponsor” (emphasis added) who is responsible for his supervision. The sponsor introduces him to a local AA clubhouse, and goes with him to regular meetings. The new member is made to feel that he "belongs" to the group and is "one of the family.”

An April 1949 issue of “Survey” (cited previously) reported on an alternative to AA whose founder claimed an impressive success rate and used prescreening. In part, it states:

Edward J McGoldrick Jr, director of New York City's Bureau of Alcoholic Therapy, established within the Department of Welfare in 1943, is an individualist among alcoholic therapists for he also holds out against the theory that alcoholism is a disease. ...

The method which he uses at Bridge House, the bureau's convalescent home for selected male alcoholics (emphasis added) is based on theories of will and thought control ... Bridge House, with only twenty beds, serves about 350 alcoholics a year, both on a resident and a non-resident basis. Its record of success, using Mr. McGoldrick's measurement of one year of complete sobriety, is 66 percent - a good record but one not affecting some 200,000 alcoholics in New York City who do not reach Bridge House, nor any of the city's alcoholic women. It is, however, a project being watched throughout the country.

Sgt. Bill S, in his book “On the Military Firing Line in the Alcoholism Treatment Program” explains how he obtained a similar success rate in the strongly AA-linked alcoholism treatment program that he set up at Lackland Air Force Base in San Antonio in the early 1950s. 50% got sober and stayed sober the first time around, and additional people, after being bumped from the program, eventually realized how they had missed the boat, came back to AA meetings on their own, and got sober.

Sgt. Bill S (who got sober on Long Island in 1948) also prescreened people before he would begin working with them. There had to be evidence of strong motivation (one of Dr Bob's key criteria) and Sgt. Bill also excluded people who had serious mental problems. His criterion for the latter was that he refused to admit Air Force personnel into the program whose mental problems were so serious that they would have had to be released from the service anyway.

Nancy O's book “With a Lot of Help from Our Friends” has chapters on the successful treatment program that Dr Joe Zuska set up for the Navy in the late 1960s. This program was a team effort, in which an AA member, a retired Navy Commander named Dick J, was given major input at all times, and in which Dr Joe (a nonalcoholic himself) learned to listen to Dick J and take him seriously.

Submarine Bill, a retired submarine commander went through the Navy program that Dr Zuska originally devised. They still did very careful prescreening and booted people out of not only the treatment center, but also the Navy, if they refused to work the program. Therefore, they too had very impressive official success statistics.

Although early AA members were directly involved in the creation of many treatment programs and facilities such as “rehab” centers and “detox” centers, these now operate in a spirit of "cooperation without affiliation." They are completely separate from AA. Their continued importance can be seen in the fact that approximately one third of Alcoholics Anonymous members today list a "treatment facility" as one of the two most important factors in their coming to AA in the first place. [30]

Can AA still obtain similar success rates today?

Submarine Bill, and another AA member, have done a study of an excellent small AA meeting in Osceola, Indiana, which follows good old-time AA practices and procedures. Over the past twelve years plus, 90% of newcomers who attended every Tuesday evening without fail (or close to that) for a full year stayed sober for that whole year. In addition, 90% of those who stayed sober that year, even if they later moved and started going to other AA meetings, are still sober today. That is around an 80% success rate, measured that way.

The two people who made the survey phrased the criteria in that fashion to exclude first of all  those who only attended a relatively few meetings and then dropped out because of lack of any real internal motivation, although there was also an additional small number of people who stopping going to meetings later on during the first year.

One challenge is to present the AA recovery message in a way that overcomes the natural “denial” and resistance of many alcoholic prospects. It helps if they stay around long enough that they eventually “get it.” The success example cited above demonstrates what can be accomplished when prospects “sincerely try” AA’s program of recovery.

As Sgt. Bill explains in his book, there were both good AA groups and poor AA groups even in the old days, and even in the greater Akron area. He gives an account of a group near Akron in the mid-1940's which did an abysmally poor job of helping newcomers, and explains in detail why it failed to help nearly as many people as it could have.

There is a moral dilemma involved in doing prescreening or excluding certain groups from AA meetings.

·         If standards are set too high, members will refuse to work with some alcoholic prospects who could have made it, and condemn them to a miserable fate, perhaps even condemn them to the death penalty.

·         If standards are set too low, the possibility exists to tear down morale (as Sister Ignatia emphasized) and that too will mean that alcoholics who should have made it will end up failing and going to their doom.

 


 

g)      AA Membership Composition and Growth Trend

In October 1937, AA’s co-founders, Bill W and Dr Bob met in Akron, OH. According to “AA Comes of Age” they “compared notes” on the membership that existed at the time. “40 alcoholics were sober (more than 20 over a year). All had previously been diagnosed as hopeless.” This meeting led, among other things, to a proposal to develop a book of experience to help those who were distant from the Akron and New York City areas. The book of experience became AA’s “Big Book.” [31] In his later book, “The Twelve Steps and Twelve Traditions” (12&12), Bill W wrote: [32]

"Alcoholics Anonymous" published when our membership was small, dealt with low-bottom cases only. Many less desperate alcoholics tried AA, but did not succeed because they could not make the admission of hopelessness.

It is a tremendous satisfaction to record that in the following years this changed. Alcoholics who still had their health, their families, their jobs, and even two cars in the garage, began to recognize their alcoholism. As this trend grew, they were joined by young people who were scarcely more than potential alcoholics. They were spared that last ten or fifteen years of literal hell the rest of us had gone through …”

The tables that follow show an “AA Membership Growth Trend” and “Estimated Counts of AA Groups and Members.” From 1954 on, the figures were derived from final General Service Conference reports. Data for years prior to 1954 are from a May 1953 AA Grapevine article titled “How Many AAs?” which contained the following excerpt:

If the first million is the hardest, Alcoholics Anonymous has done one-eighth of the job … worldwide membership in our Society has passed the eighth-of-a-million mark, according to the new 1953 AA World Group Directory: 128,361 strong, to be exact.

Exact membership statistics in AA, however, are not and never will be available, (emphasis added) since protection of individual members’ anonymity guarantees that general membership rolls will not be kept.

The latest count of total members in 58 countries around the globe is based on reports from 5,243 groups and 127 “loners.” And, as all observant group members know, the active membership of a group may vary at any given moment, sometimes according to the interpretation of “when is a member?” [sic] In addition, many groups fail to report new membership figures from year to year.

Another variable, which follows an unpredictable pattern, is the precise method of “counting noses” used by the various groups. Some suggest that those who attend and take part regularly in meetings declare themselves members (or not, as the case may be) of that group; others may count everyone who has been dry and shows up fairly regularly for at least three months; still others base the count on a rough division of the average financial collection; some count only those who come out in bad weather, never turn down a Twelfth Step call, and are always willing with the broom or dish-pan. There may even be some, it has been unauthoritatively suggested, that count the cigarette butts after a meeting …

Since General Service Headquarters never issues instructions, rules and regulations or anything else that might suggest an AA “government” the compilers of AA’s annual census rely, by and large, on the faithful reporting by each group secretary of what appears to be the more or less stable core of membership in the group. In New York, a staff of four toiled for three months, processing the thousands of information cards, and the countless letters received from the groups who preferred supplying the desired information in that form, to produce the new Directory.

Continuing to today, the autonomous and anonymous nature of AA makes the derivation of a precise census a very difficult undertaking that is inherently inaccurate.


 

The data in tables 4 and 5 below are figurative and should not be construed literally. The numbers are likely understated. Group membership counts include only those groups asking the AAWS/GSO to be listed (thousands do not). Groups may or may not report membership estimates or update previously submitted estimates over time. Members can be counted in multiple group estimates and the composition of the numbers has changed at various times from “reported” to “estimated.” The data must be interpreted carefully, skeptically and in proper context.

AA is in about 150 countries (with over 50 GSOs overseas). The AAWS/GSO obtains data from other GSOs and groups requesting to be listed in their records. Where current data are lacking, earlier year’s figures are used. Estimates of membership of non-reporting groups are arrived at by taking an average of reporting groups. From the beginning, the numbers are at best, “fuzzy” and need to be interpreted prudently to avoid drawing erroneous conclusions. The data are not an accurate measure of a specific year’s increase or decrease. However, decade trends are indicative (but not exact) of AA groups reaching more places and more AA members achieving recovery.

Table 4: AA Membership Growth Trend

Time

Span

Total

Members

Amt

Changed

%

Changed

Notes

1935

5

5

 

Founding of Akron #1 and NY [33]

1935 to 1939

100

95

n/a[34]

At publication of Big Book

1939 to 1949

75,625

75,125

n/a%

Growth mostly after WW II

1949 to 1959

151,606

75,981

100%

Annual growth leveled off in 1952

1959 to 1969

297,077

145,471

96%

 

1969 to 1979

867,411

570,334

192%

Overseas numbers revised

1979 to 1989

1,793,236

925,825

107%

 

1989 to 1993

2,062,011

268,775

15%

Counting system changed

1993 to 1999

1,989,740

-72,271

-4%

Counting system changed

1999 to 2007

2,044,655

54,915

3%

 

In 1979, overseas numbers were revised to be double that of 1978. In 1993 and 1994, a major revision occurred in the US/Canada GSO’s counting methods and records system. The number of groups reported no longer included those described as “meetings” which chose not to be considered “groups.” Those “meetings” (typically special interest such as “alcohol and pill” and “family” meetings) are included in prior year’s data and inflate the numbers. The 1993 and 1994 revisions are often erroneously cited as a steep drop in AA membership from prior years when, in fact, it simply reflects procedural changes in the GSO administrative counting methods and criteria used.

For a number of years, GSO also offered “estimates” of what they believed the actual counts to be. The “estimates” were substantially higher than the data reported (sometimes triple). It should also be noted that the baseline for each year is not consistent for the month range contained. Data for the years 1960 through 1982 contain counts reported up to April 1 of the designated years. Data for the years 1983 through contemporary data contain counts reported from January 1 through December 31 of the designated year. Data in contemporary Conference reports apply to the year prior to the Conference year (i.e. the 2006 Conference report contains count data for calendar year 2005).

AA’s overall membership estimates, if observed as a broad indicator, signal that that AA is doing something right and has been doing so for quite some time. Over the course of 7+ decades, estimated AA membership has grown from 2 members and 2 groups in the United States in 1935, to an international Fellowship of approximately 2,000,000 members and over 100,000 groups today.

That is not a measurement of doing something wrong.


 

 

Table 5: Estimated Counts of AA Groups and Members

Yr

Group Counts

Member Counts

Sources and Notes

Decade Growth

% Chg

Total

US

Can

O'seas

Hosp

Prisons

% Chg

Total

US

Can

O'seas

Hosp

Prisons

% Chg

Chg

1935

 

2

 

 

 

 

 

 

5

 

 

 

 

 

1935-1953 membership counts are from the May 1953 Grapevine article

"How Many AAs?"

 

 

1936

0%

2

 

 

 

 

 

200%

15

 

 

 

 

 

1935
to
1939

1937

0%

2

 

 

 

 

 

167%

40

 

 

 

 

 

1938

0%

2

 

 

 

 

 

63%

65

 

 

 

 

 

 

1939

50%

3

 

 

 

 

 

54%

100

 

 

 

 

 

 

n/a

97

 

1940

 

 

 

 

 

 

 

400%

500

 

 

 

 

 

Growth mostly in Cleveland

 

 

1941

 

200

 

 

 

 

 

300%

2,000

 

 

 

 

 

Jack Alexander article in Sat Eve Post

 

 

1942

 

 

 

 

 

 

 

200%

6,000

 

 

 

 

 

 

 

 

1943

 

 

 

 

 

 

 

33%

8,000

 

 

 

 

 

 

 

 

1944

 

325

 

 

 

 

 

25%

10,000

 

 

 

 

 

 

 

 

1945

72%

560

 

 

 

 

 

40%

14,000

 

 

 

 

 

WW II demobilization

 

 

1946

79%

1,000

 

 

 

 

 

107%

29,000

 

 

 

 

 

 

1939
to
1949

1947

65%

1,650

 

 

 

 

 

38%

40,000

 

 

 

 

 

 

1948

21%

2,000

 

 

 

 

 

50%

60,000

 

 

 

 

 

 

1949

50%

3,000

 

 

 

 

 

26%

75,625

 

 

 

 

 

 

n/a

75,525

 

1950

17%

3,500

3,500

 

 

 

 

28%

96,475

 

 

 

 

 

 

 

 

1951

27%

4,436

3,606

399

216

82

133

24%

120,000

 

 

 

 

 

 

 

 

1952

11%

4,925

3,938

 

713

98

176

-4%

114,724

 

 

 

 

 

 

 

 

1953

20%

5,905

4,663

504

354

151

233

12%

128,361

 

 

 

 

 

 

 

 

1954

0%

5,927

4,484

559

441

184

259

-2%

125,856

89,780

8,671

10,766

4,511

12,128

Conference report data used from 1954 on

 

 

1955

5%

6,249

4,610

574

564

211

290

8%

135,905

93,951

9,545

14,397

4,956

13,056

GSO estimated members were around

triple of that reported

 

 

1956

8%

6,779

4,893

624

710

256

296

3%

139,798

95,661

10,235

15,129

5,936

12,837

1949
to
1959

1957

0%

6,793

4,755

680

743

290

325

1%

141,795

93,746

10,548

17,501

6,500

13,500

Overseas = GSO guess

1958

14%

7,765

5,443

775

915

285

347

3%

145,830

98,858

11,181

14,471

7,077

14,243

 

1959

6%

8,211

5,719

801

1,033

302

356

4%

151,606

101,724

11,175

16,651

6,509

15,547

 

100%

75,981

1960

5%

8,615

5,875

833

1,112

360

435

7%

161,549

108,004

11,524

17,973

6,675

17,373

GSO estimated members > 300,000

 

 

1961

8%

9,305

6,208

876

1,274

445

502

9%

176,474

109,923

12,554

22,768

10,778

20,451

 

 

 

1962

8%

10,070

6,600

946

1,484

476

564

7%

189,702

117,953

13,093

25,642

10,602

22,412

 

 

 

1963

9%

10,956

7,033

1,075

1,690

514

644

10%

209,434

122,483

15,410

32,371

11,417

27,753

GSO estimated members > 350,000

 

 

1964

7%

11,761

7,543

1,146

1,854

570

648

4%

217,368

128,498

15,928

37,178

10,764

25,000

 

 

 

1965

6%

12,444

7,821

1,241

2,136

569

677

6%

231,477

133,786

17,180

43,580

10,774

26,157

 

 

 

1966

7%

13,279

8,177

1,282

2,522

561

737

9%

251,615

140,379

18,630

48,417

14,384

29,805

 

1959
to
1969

1967

7%

14,154

8,484

1,411

2,784

645

830

4%

262,562

142,566

18,757

55,062

16,307

29,870

GSO estimated members > 400,000

1968

4%

14,747

8,595

1,556

3,057

648

891

8%

283,329

148,574

21,676

60,756

18,548

33,775

GSO estimated members > 425,000

1969

6%

15,624

9,047

1,590

3,350

742

895

5%

297,077

156,974

22,706

63,366

20,160

33,871

GSO estimated members > 450,000

96%

145,471

 

1970

5%

16,459

9,541

1,667

3,559

767

925

5%

311,450

167,313

26,008

67,044

18,604

32,481

GSO estimated members > 500,000

 

 

1971

8%

17,776

10,342

1,815

3,921

783

915

6%

329,429

181,419

29,073

71,737

16,900

30,300

GSO estimated members > 575,000

 

 

1972

17%

20,829

12,137

2,031

4,761

914

986

20%

394,742

211,686

32,740

86,344

28,995

34,977

 

 

 

1973

8%

22,467

12,869

2,122

5,428

1,030

1,018

7%

420,754

225,911

35,091

102,073

22,915

34,764

GSO estimated members > 725,000

 

 

1974

11%

25,030

14,448

2,427

6,088

1,046

1,021

19%

502,283

287,699

42,922

107,643

28,057

35,962

GSO estimated members > 800,000

 

 

1975

6%

26,456

15,308

2,511

6,153

1,351

1,133

6%

533,209

292,646

43,903

126,991

32,711

36,958

GSO estimated members > 1,000,000

 

 

1976

11%

29,352

16,557

2,683

7,597

1,285

1,230

8%

574,018

320,913

47,843

135,147

30,792

39,323

 

1969
to
1979

1977

8%

31,587

18,382

2,950

7,597

1,400

1,258

7%

612,483

352,807

50,783

135,144

33,544

40,205

Overseas 1977 data used

1978

5%

33,241

18,926

3,157

8,737

1,285

1,136

2%

627,021

356,383

53,601

149,943

30,788

36,306

 

1979

20%

39,964

20,359

3,349

14,016

1,156

1,084

38%

867,411

387,875

56,672

363,572

23,956

35,336

Overseas figures being revised

192%

570,334

 

1980

5%

42,105

22,169

3,624

14,016

1,182

1,114

5%

907,067

414,434

61,531

363,360

24,822

42,920

 

 

 

1981

14%

47,797

24,293

3,781

17,650

935

1,138

3%

937,212

455,505

64,244

355,000

24,310

38,153

Much overseas data estimated

 

 

1982

12%

53,576

26,608

3,948

20,669

1,030

1,321

14%

1,064,784

518,790

66,344

412,949

25,750

40,951

GSO stopped quoting estimates

 

 

1983

9%

58,576

29,827

4,197

22,156

1,052

1,344

12%

1,191,408

585,823

69,931

467,419

25,899

42,336

Base changed to January 1

 

 

1984

7%

62,860

31,754

4,286

24,221

1,047

1,552

13%

1,351,282

630,679

71,632

576,236

26,175

46,560

 

 

 

1985

7%

67,019

33,840

4,445

27,054

 

1,680

7%

1,445,502

676,234

74,277

644,591

 

50,400

Hospital (TF) category dropped by GSO

 

 

1986

9%

73,192

36,002

4,540

30,868

 

1,782

8%

1,555,796

727,145

76,377

698,814

 

53,460

 

1979
to
1989

1987

4%

76,184

38,276

4,654

31,387

 

1,867

4%

1,616,770

775,040

78,057

707,663

 

56,010

 

1988

12%

85,270

40,693

4,749

38,060

 

1,768

7%

1,734,195

835,489

81,293

762,605

 

54,808

 

1989

3%

87,696

43,107

4,866

38,060

 

1,663

3%

1,793,236

896,033

82,949

762,701

 

51,553

 

107%

925,825

1990

7%

93,914

46,450

5,046

40,755

 

1,663

14%

2,047,252

1,012,623

87,532

894,563

 

52,534

 

1989
to
1993

1991

3%

96,458

48,747

5,173

40,755

 

1,783

4%

2,119,744

1,079,719

90,735

894,508

 

54,782

 

1992

-8%

89,215

50,325

5,275

31,700

 

1,915

-3%

2,048,519

1,135,395

94,986

770,437

 

47,701

Overseas figures revised

1993

1%

90,155

50,541

5,287

32,190

 

2,137

1%

2,062,011

1,134,500

96,500

778,829

 

52,182

GSO records system revision underway

15%

268,775

1994

-1%

89,239

49,443

5,133

32,578

 

2,085

-13%

1,790,169

1,127,471

95,546

516,169

 

50,983

GSO records revision continues

 

 

1995

7%

95,166

50,671

5,259

37,082

 

2,154

7%

1,921,936

1,153,795

97,397

614,611

 

56,133

Only 18 out of 40 foreign GSOs

sent in updated figures

 

 

1996

2%

96,997

50,681

5,275

38,765

 

2,276

2%

1,959,513

1,158,850

98,720

642,769

 

59,174

1993
to
1999

1997

1%

97,568

50,997

5,277

38,895

 

2,399

0%

1,967,121

1,166,079

102,499

636,414

 

62,129

 

1998

1%

98,710

51,183

5,257

39,804

 

2,466

1%

1,988,777

1,166,927

101,786

657,062

 

63,002

 

1999

0%

99,024

51,151

5,132

40,222

 

2,519

0%

1,989,740

1,161,436

97,054

666,527

 

64,723

 

-4%

-72,271

 

2000

2%

100,766

51,735

5,104

41,423

 

2,504

9%

2,159,700

1,162,112

98,816

833,100

 

65,672

 

 

 

2001

-1%

100,131

51,245

4,965

41,390

 

2,531

3%

2,214,978

1,160,651

97,124

891,167

 

66,036

 

 

 

2002

4%

103,768

51,537

4,903

44,762

 

2,566

-6%

2,092,460

1,169,204

96,100

760,214

 

66,942

 

 

 

2003

1%

104,589

52,735

4,884

44,425

 

2,545

-1%

2,066,851

1,187,373

96,446

716,523

 

66,509

 

 

 

2004

1%

105,298

52,651

4,872

45,209

 

2,566

1%

2,082,980

1,190,860

95,984

729,173

 

66,963

 

1999
to
2007

2005

1%

106,227

52,050

6,214

45,436

 

2,527

-6%

1,947,662

1,068,761

110,449

702,609

 

65,843

Canada numbers “blip” looks unlikely.

2006

8%

114,561

53,665

4,874

53,590

 

2,432

2%

1,988,968

1,213,269

95,443

616,899

 

63,357

 

2007

-1%

113,168

53,595

4,745

52,469

 

2,359

3%

2,044,655

1,220,993

93,559

669,057

 

60,912

 

3%

54,915

 

Notes:

Ø  “% Chg” trend data equals the percentage of growth or decline of the current year compared to the prior year.

Ø  The baseline year is the year prior to the decade (e.g. 1939 is the baseline for growth measurement from 1940 to 1949).

Ø  In 1979, overseas numbers were revised to be double that of 1978.

Ø  Data for 1960 through 1982 contain counts reported up to April 1 of the designated years. 1983 through 2006 data contain counts reported from January 1 through December 31 of the designated year. Data in contemporary Conference reports apply to the year prior to the Conference year (e.g. 2006 Conference report data is for calendar year 2005).

Ø  In 1993 and 1994, a major revision occurred in the GSO counting methods and records system. The number of groups reported no longer included those described as “meetings” which chose not to be considered “groups.” Those “meetings” (typically special interest “alcohol and pill” and “family” meetings) are included in prior year data and inflate the numbers. These years are often erroneously viewed as a drop in AA membership.


 

[1] The General Service Office (GSO) of the United States and Canada located in New York City (also noted as AAWS/GSO in this paper).

[2] The term “Big Book” is used for AA’s basic recovery text in lieu of its title “Alcoholics Anonymous” (both terms © AAWS, Inc).

[3] 1990 report titled “Comments on AA’s Triennial Surveys” on five membership surveys from 1977-1989.

[4] “Pass It On” page 226 © AAWS, Inc.

[5] Roman numeral page xx in the 3rd and 4th edition Big Book

[6] Bill W’s commentary in the introductions to the personal stories, 2nd edition Big Book unnumbered pages 167-169.

[7] July 1938 letter from Bill W to Dr Richard C Cabot, GSO Archives ref: 1938-81, Alcoholic Foundation, R 28, Bx 59.

[8] The minutes are erroneously dated April 11, 1938, GSO Archives ref: 1938-19, Alcoholic Foundation, R 10, Bx 22.

[9] March 1941 Saturday Evening Post; also see AA publication (P-12) “Jack Alexander Article About AA” © AAWS, Inc.

[10] “Alcoholics Anonymous Comes of Age” pages 35, 192, 310 and “Pass It On” page 266 both © AAWS, Inc.

[11] See also “Alcoholics Anonymous Comes of Age” pages 309-310 © AAWS, Inc.

[12] “Medicine Looks at Alcoholics Anonymous” May 1944 Bill W talk to the Medical Society of the State of NY © AAWS, Inc.

[13] Later reprints excluded the paragraphs stating estimated success rates and membership size.

[14] April 1949, “Survey” article titled “Hope for the Alcoholic” by Kathryn Close.

[15] Second edition Big Book, AA Publishing, Inc. 1955. Introduction to Part III “They Nearly Lost All” unnumbered page 167.

[16] Second edition Big Book, AA Publishing, Inc. 1955. Introduction to Part I “Pioneers of AA” unnumbered page 169.

[17] Bill W’s April 1958 talk is preserved in AA pamphlet (P-6) “Three Talks to Medical Societies” © AAWS, Inc.

[18] “AA Today” is a souvenir book for the 1960 International Convention and AA’s 25th anniversary © AA Grapevine, Inc.

[19] The membership survey was taken in June and July 1968. A population of 11,355 AA members at 466 AA meetings participated.

[20] For NIAAA Strategic Plan details re:: http://pubs.niaaa.nih.gov/publications/StrategicPlan/NIAAASTRATEGICPLAN.htm

[21] For NESARC details re: http://niaaa.census.gov/

[22] See the AA membership count tables at the end of this paper.

[23] The Big Book Chapter 3 “More About Alcoholism” page 30.

[24] The Big Book Appendix II “Spiritual Experience” page 568, 4th edition.

[25] “Dr Bob and the Good Oldtimers” page 169 © AAWS, Inc.

[26] Today, this might be called a “90-in-90” or doing “ninety meetings in ninety days.”

[27] “Dr Bob and the Good Oldtimers” pages 261 and 263 © AAWS, Inc.

[28] “Slaying the Dragon” page 133 © William L White.

[29] “Pass it On” page 225.

[30] The “2004 AA Membership Survey” © AAWS, Inc shows that 31% credit a treatment facility and another 8% credit a counseling agency.

[31] “AA Comes of Age” pages 144-146 © AAWS, inc.

[32] “Twelve Steps and Twelve Traditions” Step One essay, page 23 © AAWS, Inc.

[33] AA’s first group, Akron #1, began July 4, 1935 - the date that Bill D (AA #3) was discharged from Akron City Hospital.

[34] n/a is shown because the % numbers are so large and out of the ordinary in terms of context.


Return to the West Baltimore Group Main Page