In light of the recent Robin Williams tragedy I think it is high time that we asked the question of whether or not 12 step programs can induce depression and suicidality. Let us remember that 12 step programs are not the creation of mental health professionals, but were invented by a failed stockbroker, Bill Wilson, who suffered lifelong depression which was not alleviated by the 12 step program which he invented.
Anyone who has attended 12 step meetings for any period of time has heard numerous stories of member suicides. It was only a few months after I started attending AA that I heard my first funeral announcement; one of the old timers who had been around for decades and who had founded and led many local AA meetings had committed suicide in his garage. He was not using drugs or alcohol. This is one of many things which raised doubts about 12 step programs in my mind; none of my drinking friends had ever committed suicide. This leads us to the question: do the 12 steps themselves induce depression and lead people to commit suicide?
As far back as the 1970s Aaron Beck defined the cognitive triad (link is external): people who are depressed believe that the self is powerless, the world is unfair, and the future is unchanging. This activates schema of self-deprecation and self-blame which cause and maintain the state of depression. But powerlessness, self-blame, and self deprecation are all key elements of 12 step programs.
In television shows like “Dexter” we hear about confronting the demons of one’s past as being “stepwork”; but this is all fiction dreamed up in the mind of some TV writer. In actuality 12 step programs forbid anger, forbid confronting the past, and emphasize the fact that one is to blame for all bad things which have happened to one in the past. Anger is never permitted; the AA Big Book Says, “If we were to live we had to be free of anger (p. 66).” In the step 4 inventory people are expect to list their resentments, list the names of people, institutions or principles with whom they are angry, and ask themselves the following questions: Why am I angry? Where was I to blame? What is the exact nature of my wrong (4th step workshop with Dallas B (link is external)).
Even if someone was raped as a child, AA asks the victims to blame themselves for being raped and to make amends to their rapist. There is no exorcism of demons of the past through confrontation. This schema of self-blame is a highly depressogenic aspect of 12 step programs.
The other highly depressogenic aspect of 12 step programs is found in step one where one is forced to declare that one is powerless; this is an example of learned helplessness (link is external) and Martin. Seligman pointed out long ago that learned helplessness induces depression.
What Does the Research Say?
Epidemiological studies show that the comorbidity substance dependence and depression is very high, for example, the National Comorbidity Study (NCS; Kessler et al. 1996) showed that, among those diagnosed with a (DSM-III-R) substance use disorder, there was a 12-month prevalence rate of almost 23% for major depression. Other studies have shown similar high rates of comorbidity between substance dependence and depression. However, in spite of this high rate of comorbidity, there is a dearth of research on the interaction of 12 step programs and depression. As search of PubMed turns up only three articles about the effects of 12 step programs on individuals with comorbid substance dependence and depression: Kelly et al. 2003 (link is external) which looks at the effects of 12 step affiliation on both substance use and depression, Worley et al. 2012 (link is external) which also looks at the effects of 12 step affiliation and attendance on both substance use and depression, and Worley et al. 2013 (link is external) which looks at the effects of 12 step affiliation and attendance on substance use only. We will discuss the first two of these studies in detail below.
The Kelly 2003 study (link is external) followed substance dependent male veterans who had completed inpatient substance abuse treatment (21 to 28 days) and reported on outcomes at the one and two year mark. This study compared two groups: one group with substance dependence only, and one group with comorbid substance dependence and major depressive disorder. In this study 12 step affiliation was measured using the following five variables: meeting attendance, incorporation of steps into life, reading 12 step literature, sponsor interaction, and number of AA friends. The Kelly study found that both groups did equally well in terms of abstinence and reduced substance use in spite of the fact that the comorbid group showed significantly less 12 step affiliation than the dependence-only group. Moreover, in the dependence-only group improvements in substance abuse were positively correlated with 12 step affiliation; in the comorbid group there was no correlation between improvements in substance use and 12 step affiliation. Finally, there was no correlation between 12 step affiliation and improvements in depression in the comorbid group. The authors of the study note that the comorbid group had two and a half times more mental health visits than did the dependence-only group and hypothesize that these mental health visits may have been a factor in the improvements in substance use in this group.
The Worley 2012 (link is external) study looked at the effects of 12 step attendance and 12 step affiliation on depression as measured by the Hamilton Depression Scale. Twelve step affiliation was defined in terms of the following four factors: having a sponsor, doing service, reading literature, having a spiritual awakening. Twelve step meeting attendance was measured separately from 12 step affiliation. All subjects in this study had comorbid substance dependence and major depressive disorder The subjects were divided into two groups, one of which received Twelve Step Facilitation Therapy (TSF group) and one which received Cognitive Behavioral Therapy (CBT group). The treatment was outpatient and the duration of the treatment was 6 months. At the end of this 6 month treatment period the TSF group scored significantly higher than the CBT group on the variables of 12 step meeting attendance and improvement in depressive symptoms. Twelve step meeting attendance predicted lowered depression independently of 12 step affiliation. Twelve step affiliation did not independently predict lowered depression. Although the TSF group had significantly fewer drinking days than the CBT group at the end of treatment, they had significantly more drinking days than the CBT group at the 18 month follow up period (Worley 2013 (link is external))
Although at first glance the Worley 2012 study seems to suggest that 12 step programs lead to improvement in depression, there are a number of confounds in this study which make this conclusion questionable:
There was no control of social interaction. Increased social interaction has been shown in itself to have a salutary effect on depression (see Cruwys et al. 2014 (link is external) for example). It is quite possible that attendance at a church, mosque, yoga group, book club or SMART (link is external) or HAMS (link is external) group would have had a far more salutary effect than 12 step attendance. This question remains to be investigated.
The TSF therapy given the subjects in this study was based on the Project MATCH TSF Manual (link is external) which is a radical reinterpretation of the 12 steps which frequently blatantly contradicts AA conference approved literature by emphasizing concepts taken from harm reduction and Marlatt’s Relapse Prevention such as valuing every abstinence day whether consecutive or not. Powerlessness is reframed as having a problem and people are not shamed for slips and put back to day one. This radical reframing may have a protective effect against the depressogenic effects of the 12 steps as presented in meetings.
There was no long-term follow-up of depression scores; since drinking scores were far higher at the 18 month follow-up for the TSF group compared to the CBT group, the same might also have been true of depression scores had they been measured.
One would like to see the effect of standard 12 step treatment on depression scores as compared to TSF. The Kelly 2003 study apparently used standard 12 step treatment rather than TSF and found it had no effect on depression scores.
One would like to know if the improvement in depression scores was due to the effects of the meeting itself or of the socializing which normally takes place after the meeting ends.
One would like to see if a comparison group using SMART or other non 12 step meetings had better outcomes.
One must conclude that the jury is still out on the question of whether 12 stepping leads to depression and suicide and that far more and better research is needed. Although there have been a couple of studies on the 12 steps in relation to depression there have been none at all on suicide and the 12 steps. It is my own personal conclusion is that people who benefit from 12 step groups do so in spite of the steps and because of the increased social interaction. Anything which increases social interaction in a mentally healthy fashion would probably have far superior outcomes than AA meetings and result in far fewer suicides as well.