Some of the critics claim that modern A.A. has a dismally low success rate, only 5% or less.

The statistics cannot be supported either from U.S. government research or from A.A.’s own careful records.

THE CORRECT FIGURE IS A CLOSE TO 36% SUCCESS RATE, when measured in the same way that the government would evaluate success in treating diabetes, heart disease, or cancer.

The 95% dropout rate [claimed by some of A.A.'s modern critics] is inaccurate and based on flawed statistics.

Rather than 5% of A.A. members remaining at the end of one year a more accurate estimate is that 36% remain attending A.A. at the end of one year and 32% are still attending at the end of 20 years.

The 1989 AA General Service Office internal memo “Comments on A.A.’s Triennial Surveys,” Appendix C: “The First Year” has been widely misquoted as a measure of retention or dropouts. It is not a measure of retention or dropout, it is a simple frequency distribution of individuals by months since first came to AA and is limited to individuals with 12 or fewer months.

The limitations of the A.A. membership surveys are well described in the GSO internal memo on page 4. A major limitation is the significant under counting of the A.A. membership resulting from the many groups that were not surveyed.

The “A.A. Fact File” that I received from GSO shows the estimated A. A. membership in the U.S. as 1,168,990 members. This is a significant undercount of members.

Using a national representative survey of the U.S., in the NIAAA 1991-1992 National Longitudinal Alcohol Epidemiologic Survey (NLAES), I found that in 1991-1992, 2.4 million individuals reported attending an A.A. meeting during the last year.

I have posted a brief summary of my findings on A.A. membership on my website Alcohol Reports as a Data Brief, dated July 29, 2007.
by Loran Archer

An update;

National Longitudinal Alcohol Epidemiologic Survey (NLAES) Findings on Alcoholics Anonymous Membership

Estimated Alcoholics Anonymous Membership 1991-1992
  • New members during past year – 0.9 million
  • On-going members – 1.5 million
  • Total membership – 2.4 million
Continuation Rate in Alcoholics Anonymous

In 1991-1992 4.8 million respondents reported ever attending an Alcoholics Anonymous (AA) meeting, for reasons related to their drinking, prior to the last 12 months.

  • 31% reported continued AA attendance during the last 12 months

Rate of continued AA attendance was associated with years since first AA meeting

  • 1-4 years since first AA meeting – 36% remained
  • 5-9 years since first AA meeting – 30% remained
  • 10-19 years since first AA meeting – 29% remained
  • 20 years or more since first AA meeting – 32% remained
Comparison of Past Year Drinking Status –

Dropouts and Continuing AA Members


  • Abstinent 33%
  • Low risk drinking 14%
  • High risk drinking 53%


  • Abstinent 62%
  • Low risk drinking 9%
  • High risk drinking 29%
Note the advantage of continuing AA attendance.

low risk drinking = never exceed 4 drinks per day (male) or 3 drinks per day (female)

high risk drinking = exceeds 4 drinks per day (male) or 3 drinks per day (female)

Data Source:
NIAAA 1991-1992 National Longitudinal Alcohol Epidemiologic Survey (NLAES).


Related Reading:

Everything I Never Wanted to Be: A Memoir of Alcoholism and Addiction, Faith and Family, Hope and Humor
Under the Influence: A Guide to the Myths and Realities of Alcoholism
The Big Book   of Alcoholics Anonymous
Seven Days Sober: A Guide to Discovering What You Really Think About Your Drinking