Does AA Work? What the Evidence Shows

Key takeaway

The best single answer comes from a 2020 Cochrane review that pooled 27 studies and 10,565 people. It found high-certainty evidence that structured, clinician-delivered programs which get people into Alcoholics Anonymous lead to higher rates of continuous abstinence over months and years than other active treatments such as cognitive behavioral therapy.

Last updated: Sun Oct 18 2026 00:00:00 GMT+0000 (Coordinated Universal Time)

If you or someone you love is drinking too much, the question "does AA even work?" is not academic. You are deciding where to spend your evenings and your hope. You deserve a straight answer, not a slogan. So here is what the research actually shows — where the evidence is strong, and where people on both sides of the argument tend to overreach.

Start with the biggest review we have

The reference point is a Cochrane review published in 2020 by researchers including John Kelly of Harvard and Keith Humphreys of Stanford. Cochrane reviews are systematic: they gather the studies, weigh their quality, and report what holds up. This one pooled 27 studies covering 10,565 people.

It separates two things people often blur together. First, Alcoholics Anonymous itself — the free, peer-led mutual-help fellowship. Second, what researchers call Twelve-Step Facilitation, a clinician-delivered approach that introduces people to AA and helps them get involved. The review's clearest findings are about the second kind: structured, manualized programs that increase a person's AA participation.

What it found

The headline result, in Cochrane's own summary, is that clinically delivered and manualized programs designed to increase AA participation "can lead to higher rates of continuous abstinence over months and years" when compared with other active treatments such as cognitive behavioral therapy.

The estimate behind that: around 42% of people in the AA-based groups were completely abstinent one year later, compared with about 35% of those receiving other treatments, including CBT. The review rated this evidence as high certainty — meaning the authors judged it reliable.

There is a detail in there worth catching. The benefit showed up largely because the programs kept people involved in AA after the formal treatment ended. In other words, the meetings were doing real work, not just the clinical sessions.

What the review does not say

On some other measures, the difference fades. AA-based approaches performed about as well as the comparison treatments at reducing how heavily people drank, cutting alcohol-related consequences like health and family problems, and lowering addiction severity. "About as well" is not the same as "better."

The review is also about alcohol, and mostly about adults with alcohol use disorder. It is not a study that proves AA treats opioid or methamphetamine use, and anyone who tells you otherwise is borrowing certainty that isn't there.

And no honest reading of a group average tells you what will happen to you. A 42% figure is built from many studies with many kinds of people. Some of them were not ready to stop. Some had co-occurring conditions that treatment did not reach. The number describes the group; it does not predict your story.

What it does not compare

One common misreading is that this was "AA versus nothing." It mostly wasn't. The main comparisons were AA and Twelve-Step Facilitation against other active treatments — CBT, motivational enhancement, and different styles of facilitation — and sometimes against each other. When the review looked at structured versus unstructured facilitation, the structured kind did better. So the practical takeaway is not "treatment or AA." It is that programs which connect people to AA, and keep them connected, tend to do well.

What this means if you are the one suffering

Here is the human version of all that research. AA is free, it exists in nearly every town, you can walk in and say nothing, and the strongest evidence we have says that getting involved in it is linked to staying sober longer. It is not the only path, and it is not a guarantee. But it is not a gamble either — it is one of the few things in this field that has been studied this hard and come out looking good.

If the words "higher power" or the God language put you off, that is worth naming. There are other peer programs, and NIAAA's Alcohol Treatment Navigator walks you through evidence-based options step by step. You are allowed to pick the door you can actually walk through.

What you can do tonight

You can find a meeting tomorrow — search "AA meetings near me" plus your city, or look for a local intergroup. You can attend one meeting and just listen. You do not have to introduce yourself, and you do not have to decide anything about your life to sit in a chair for an hour.

If drinking has you frightened, and you want to talk through what real treatment would look like, call or text (800) 653-9376 and we will walk through the options with you. SAMHSA's National Helpline is free and confidential at 1-800-662-HELP (4357). If withdrawal is making you shaky or sick, or you have been drinking heavily and want to stop, talk to a doctor first — alcohol withdrawal can be dangerous and should be medically supervised. If you are in crisis, call or text 988. In an emergency, call 911.

Additional Resources

Sources cited on this page:

Common Questions

Does AA actually work for alcohol problems?

The largest and most rigorous review we have says yes, with an important detail attached. The 2020 Cochrane review found high-certainty evidence that clinically delivered, manualized programs designed to get people into Alcoholics Anonymous produce higher rates of continuous abstinence over months and years than other active treatments such as cognitive behavioral therapy.

What did the review actually measure, and who took part?

It pooled 27 studies with 10,565 people, mostly adults with alcohol use disorder. It looked at whether AA involvement and Twelve-Step Facilitation — a clinician-delivered approach that gets people into AA — changed abstinence, how much people drank, alcohol-related consequences, and addiction severity.

I keep hearing 42% versus 35%. What does that mean?

It is the review's estimate that about 42% of people in an AA-based group stayed completely abstinent one year later, compared with about 35% who received other treatments such as cognitive behavioral therapy. It is a group average from the pooled studies, not a promise about any one person.

Does the review mean AA beats therapy?

Not exactly. The clearest advantage was for continuous abstinence, and it came mainly from helping people stay involved in AA after the structured program ended. On other measures — how heavily people drank, alcohol-related consequences, addiction severity — AA-based approaches performed about as well as the comparison treatments.

Does this same evidence apply to drugs like opioids or methamphetamine?

No. The review was about alcohol use disorder. It is the strongest evidence for AA and alcohol. For other substances, the picture is different, and this page is not the place to stretch those findings further than they go.