By John M. Majer, Ph.D., Leonard A. Jason, Ph.D., & Joe Ferrari, Ph.D.
There has been increased interest among social scientists in understanding Twelve Step groups such as Alcoholics Anonymous (AA) and Narcotics Anonymous (NA) in the past two decades. It seems as though things took off in the mid-1980’s shortly after Edward I, Khantzian, MD, posited his “self-medication hypothesis” in a short piece in the American Journal of Psychiatry, claiming that addicted persons turn to their substances of choice to deal with their “painful affect states.” Dr. Khantzian was one of the first advocates for Twelve Step recovery in scientific literature, claiming that AA was the shining prototype of Twelve Step groups.
By the mid-1990’s, more recovery-related articles were getting published in scholarly journals and it seemed as though there were more alcohol- and drug-related journals available in university settings. The days where psychologists and psychiatrists argued through published letters, case studies, and studies-ones that used very small clinical samples aiming to prove which theoretical approach to psychotherapy worked best in treating problem drinking and narcotic addiction-finally gave way to more scholarly reports based on rigorous scientific methods and research designs for understanding Twelve Step involvement among persons in rehabilitative programs for substance use disorders.
Twelve Step Involvement
Operational definitions-such as subjecting participants to various conditions, operations, and procedures in order to obtain observed responses that can be measured and analyzed through statistical analyses-are important elements of scientific methods that permit researchers to examine concepts that cannot be directly measured. “Twelve Step involvement” is a term that has been operationally defined in several ways through the use of established surveys. Investigations on Twelve Step groups such as AA and NA have used terms including “affiliation,” “participation,” “investment,” and “involvement” that seem to tap into the general construct counselors in the field refer to as “Living the program.” For the sake of argument, we simply refer to one’s active participation in various activities practiced and discussed at Twelve Step groups-the kind of active participation that is typically encouraged in Twelve Step-oriented treatment and in Twelve Step groups as “involvement” in this article.
The Alcoholics Anonymou Affiliation Scale (AAAS) and the Anonymous Involvement Scale (AAIS) are two of the most commonly used measures in assessing Twelve Step involvement (Humphreys, Kaskutas, 8: Weisner, 1998; Tonigan, Connors, & Miller, 1998). Scores derived from these measures typically consist of summary scores that include rates of Twelve Step meeting attendance. Although “meeting makers make it,” as the saying goes, there are some mixed results with respect to meeting attendance in the literature. Some researchers have found frequent meeting attendance as being related to good outcomes whereas other researchers have found frequent meeting attendance was related to negative outcomes. The latter seems to make sense when thinking of folks who are court-ordered to Twelve Step groups; some might attend meetings for the sake of placating the courts but have no real interest in getting involved with Twelve-Step-based recovery. However, there is some evidence to suggest that the duration of meeting attendance is perhaps more important than the frequency of meeting attendance. In other words, the number of meetings attended each week may not he as important as the ongoing attendance to Twelve Step meetings.
We argue that involvement in Twelve Step groups like AA and NA might be better understood when examining activities apart from meeting attendance. Meeting attendance, especially in the beginning of one’s recovery, is highly instrumental in getting one involved in personal recovery, but it’s not sufficient because there is more to recovery than meeting attendance. For instance, “old-timers” who have been clean or sober for decades may not necessarily attend the same number of meetings they did early in their recovery, but they continue to attend meetings and chances are they are involved in their recovery by engaging in a number of Twelve Step activities and practices aside from attending meetings.
In addition to meeting attendance, Twelve Step involvement has been assessed with measures that consist of items regarding specific Twelve Step activities. This has led to the calculation of summary scores across a number of activities, including meeting attendance, thereby creating a dimensional approach to assessing Twelve Step involvement. This common practice in research renders a scale score with higher scores indicating greater Twelve Step involvement.
However, some researchers have examined individual Twelve Step activities – having a sponsor, working the Steps, having a home group, being involved in service – apart from meeting attendance as a measure of Twelve Step involvement. Individual Twelve Step activities have been significantly related to outcomes in various studies but findings are not highly consistent across studies. Consistency of findings across research studies is necessary in providing confidence that such findings are reliable, so inconsistent findings regarding specific Twelve Step activities becomes problematic.
Perhaps more importantly, there is a lack of theoretical basis to support the use of these approaches such as averaging and/or summarizing responses across various activities, and examining individual Twelve Step activities, toward measuring involvement in AA and NA groups. Once major problem with a summary score approach to measuring Twelve Step involvement lies in the fact that a summary score approach does inflate and reduce actual involvement in any one Twelve Step activity, within the set of individual activities, covered by these measures. For example, someone who does not have a sponsor, does not work steps and is not involved in Twelve Step service, but considers oneself as a member, claims to have had a spiritual awakening, and attends meetings, could complete a measure like the AAAS or AAIS and receive a scale score of Twelve Step involvement. This score in turn, really reflects an averaged-like rate of involvement across all Twelve Step activities listed in the measure where partial credit is given for being involved in some activities.
Another major problem with using specific items of these commonly used measures of Twelve Step involvement is that some items do not necessarily reflect actual involvement or involvement that is suggested to newcomers. Although most Twelve Step activities listed in the AAS are very similar to those in the AAIS, both instruments include other items that are not directly relevant to new members (e.g. sponsoring others), are not direct behaviors/activities but more of a perception (e.g. considering oneself as being a “member”), or appear to be more of a consequence of an action (e.g. having had a “spiritual awakening” or recovery “birthday”). In addition, there are other Twelve Step activities commonly practiced by members such as the application of the Twelve Steps, practicing spiritual principles, helping the newcomer, and participating in the group conscience of one’s home group, that aren’t covered in these measures.
Nonetheless, we believe that measures such as the AAAS and AAIS have been highly instrumental in raising awareness into the benefits of Twelve Step groups. They have historic significance in Twelve Step research and as a result scientists have learned quite a bit about involvement in AA and NA. Only recently have some researchers taken alternative approaches to scoring these valuable instruments, like examining independent activities such as having a sponsor. However, AA/NA members endeavor to engage in several Twelve Step activities simultaneously – assessing Twelve Step involvement should reflect this practice. That is why we have recently assessed involvement in Twelve Step groups categorically.
“Categorical Twelve Step Involvement” is a term used to indicate simultaneous involvement in several Twelve Step activities. It has been ragued that involvement in various Twelve Step activities should be measured categorically because it is a representative assessment of what members are encouraged to do for their recovery (Majer, Droege & Jason, 2010.) In addition, a categorical approach to assessing Twelve Step involvement controls for artificial reduction or inflation of true scores (for individual activities) created by summary score approaches that are typically used. Recent investigations have demonstrated that categorical Twelve Step involvement in several Twelve Step activities is a significant factor related to increased in important cognitive resources and coping skills for recovery, and a better predictor of abstinence rates compared to summary scores of Twelve Step involvement (Majer Droege, & Jason, 2012; Majer, Jason Ferrari & Miller, 2011).
Abstinence-related outcomes in Twelve Step involvement research investigations have been measured in various ways, including substance use in the past 30 days, substance use in the past 90 days, reductions in substance use, and reductions in alcohol-related problems. In addition, many investigations have measured alcohol and drug use separately. However, these abstinence-related outcomes to not fully capture a major aim of Twelve Step groups; complete and continuous abstinence. Thus it is important to measure continuous abstinence from both alcohol and illicit drugs, not including medications prescribed by a licensed professional, among Twelve Step members while investigating representative involvement in Twelve Step groups. In recent years, we have conducted studies related to Twelve Step involvement by sampling residents living in Oxford Houses.
For over 20 years researches at DePaul University’s Center for Community Research have scientifically evaluated the Oxford House model of care. Oxford Houses are self-run, communal-living settings comprised of recovering alcoholics and addicts (Jason & Ferrari, 2010). The first Oxford House opened in Silver Spring, Maryland in 1975 when several recovering alcoholics who were living in a professionally-run halfway house were faced with being on the streets because their halfway house lost funding and had to close. Collectively, these recovering alcoholics sought help from a few fellow AA members who advised them to take over the house. In short, the first Oxford House was created adapting AA’s Twelve Traditions as a model for its direction. Residents were allowed to stay as long as they wanted to as long as they followed three basic tenets: no drinking and drugging (grounds for immediate eviction), partake in regular business-related House meetings, and contribute to the House in terms of rent and democratically-assigned chores.
Since then, over 1,600 Oxford Houses in the US, Australia, Canada, Ghana and the UK have flourished. One longitudinal investigation found that patients who completed inpatient treatment for substance use disorders who were randomly assigned to living in an Oxford House – versus patients who were randomly assigned to a usual care condition upon discharge – had better outcomes at two years; including twice the amount of abstinence, double the amount of income, and only a third of criminal behavior involvement compared to those who were randomly assigned to usual care (Jason, Olson, Ferrari, & Lo Sasso, 2006). However, this investigation did not control for the effects of Twelve Step involvement.
One could argue that the results from this investigation were related to Twelve Step involvement because Oxford House living is sort of like 24-hour residential AA or NA; nearly every Oxford House resident attends Twelve Step groups. Thus it was important to determine whether Twelve Step involvement was related to abstinence outcomes independent of condition and to raise the bar on abstinence outcomes by measuring continuous abstinence from both alcohol and illicit drug use at two years.
Examining Twelve Step Involvement and Oxford House
We examined categorical Twelve Step involvement and living in an Oxford House in related to continuous abstinence among persons with substance use disorders exiting inpatient treatment over a two-year period. We hypothesized that participants who were categorically involved in Twelve Step activities, compared to participants who were not categorically involved, would be more likely to maintain continuous abstinence at two years. We hypothesized that participants who were randomly assigned to the Oxford House condition would be more likely to maintain continuous abstinence at two years than those who were randomly assigned to a usual care condition. We also hypothesized that categorical Twelve Step involvement would be a better predictor of continuous abstinence outcomes at two years compared to summary scores of Twelve Step involvement.
We recruited 150 persons exiting inpatient treatment centers located in northern Illinois. After participants entered the study, they were interviewed every six months over a two-year period, yielding a total of five assessment waves; baseline and six-, twelve-, eighteen-, and twenty-four month follow-ups. The follow-up rate across the two-year study was comparable for the Oxford House (89%) and usual after-care (86%) conditions.
Categorical involvement in Twelve Step activities was assessed by positive endorsement of four of the following AAAS items at baseline: having a sponsor, reading Twelve Step literature, doing service work, and calling other members for help. Participants who endorsed all four of these items were considered as being categorically involved in Twelve Step activities whereas those who endorsed three or less of these items were not, and thus served as a comparison group.
These four AAAS items were selected because they are recovery-related actions that new members who are early in their recovery are commonly encouraged to take in AA and NA groups, whereas the other three AAAS items – identifying as a member, having had a spiritual awakening, being a sponsor to others – were not. However, we calculated summary scores of the AAAS, which is commonly done in research, to see if summary scores of Twelve Step involvement, and/or our measure of categorical Twelve Step involvement, would be related to continuous abstinence two years later.
Simply put, our investigation sought to examine the independent effects of categorical Twelve Step involvement and condition and Oxford House versus usual care – upon continuous abstinence from alcohol and illicit drug use two years later – while also testing whether categorical involvement or a summary score of Twelve Step involvement was the better predictor of continuous abstinence at two years.
Several findings were particularly revealing:
· Participants who were categorically involved in Twelve Step activities at baseline were 2.78 times more likely to maintain continuous abstinence from both alcohol and illicit drug use at two years.
· Participants who were randomly assigned to the Oxford House condition at baseline were 5.6 times more likely to maintain continuous abstinence at two years.
· Participants with high AAAS scores were only 0.7 times more likely to maintain continuous abstinence from both alcohol and drug use at two years.
· When we ran out analytic model using both measures of Twelve Step involvement, categorical involvement, and AAAS scores, in addition to condition as predictor variables, only categorical Twelve Step involvement and the Oxford House condition were significant predictors of continuous abstinence at two years. AAAS summary scores were not significant predictors
Participants who were categorically involved with Twelve Step activities, compared to those who were not, were more likely to maintain continuous abstinence at two years, consistent with Twelve Step ideology. Most treatment providers would agree that it is better to be involved in a number of Twelve Step activities for sustaining ongoing recovery and abstinence than it is to be partially involved. Our findings are consistent with the “basics” that are discussed at AA and NA meetings and affirm the Twelve Step cliché, “Half measures availed us nothing.” Although a continuous measure of Twelve Step involvement, or the AAAS summary score, was related to abstinence outcomes, the likelihood of complete abstinence due to AAAS scores was not as robust as it as among participants who were categorically involved in Twelve Step activities. This suggested that a categorical approach to assessing Twelve Step involvement might be a better indication of actual involvement in Twelve Step groups, consistent with previous investigations (Majer, Droege & Jason, 2012; Majer Jason, Ferrari & Miller, 2011). Our findings add empirical support to Twelve Step theory in this respect.
In addition, findings in the present study extend previous investigations in that we examined continuous abstinence from both alcohol and illicit drugs (together, not separately) throughout the course of two years instead of the past 30 or 90 days at key assessment intervals. Our use of corroborated self-reports through collateral informants provided some assurance of complete and continuous abstinence throughout the course of our investigation. We encourage researchers to examine continuous abstinence from both alcohol and illicit drugs in future investigations because this best reflects the aim of Twelve Step groups like AA and NA.
Furthermore, participants who were randomly assigned to the Oxford House condition were also more likely to remain completely abstinent throughout the course of the investigation. Our randomized clinical trial allowed us to examine more closely the effects of categorical Twelve Step involvement upon continuous abstinence, independent of condition and vice versa as other investigations on categorical Twelve Step involvement were limited to Oxford House residents, did not include comparison groups, and did not examine continuous abstinence from both alcohol and illicit drugs over time (Majer, Droege, & Jason, 2012; Majer, Jason, Ferrari & Miller, 2011).
Limitations, Recommendations and Conclusion
Summary scores from the AAAS in the present study included participants’ rates of Twelve Step meeting attendance – a Twelve Step activity that might not necessarily be directly related to positive outcomes. Although meeting attendance rates I the present longitudinal study were not controlled for throughout our two-year investigation, because we collected these data only at baseline, it is possible that meeting attendance rates might have accounted for the weak relationship between AAAS scores and continuous abstinence at two-year follow-up. However, a recent longitudinal investigation that included multiple measures of Twelve Step involvement demonstrated that summary scores of Twelve Step involvement did not significantly predict abstinence outcomes when controlling for meeting attendance (Majer, Jason, Ferrari, & Miller, 2011). We believe that more investigations regarding Twelve Step meeting attendance rates and duration of attendance are needed to better understand this important activity in relate to categorical Twelve Step involvement, and to determine how and why members change their rates of meeting attendance over the course of long-term recovery in maintaining their complete and continuous abstinence.
Findings in the present study suggest that categorical Twelve Step involvement and the Oxford House model are mutually exclusive factors related to recovery. Even though the vast majority of Oxford House residents are involved in Twelve Step groups, findings in the present study are consistent with previous research that found Oxford House living provides additional benefits to recovering persons apart form their Twelve Step involvement (Majer, Jason & Olson, 2004). Nonetheless, the use of clinical samples in the present study might be a limitation, and future investigations with community-based samples would help us further understand the effects of categorical Twelve Step involvement.
In addition, findings in the present study suggest that categorical involvement in a set of Twelve Step activities that are encouraged at AA and NA meetings, and Oxford Housing living, is likely to empower persons with substance use disorders in maintaining their abstinence. Clinicians should encourage clients’ active and concurrent involvement in a number of Twelve Step activities early in their recovery such as the ones used I the present study, and consider referrals to self-run, communal-living settings like Oxford Houses, The Oxford House network of substance abuse recovery homes as been endorsed by SAMSHA as being effective.
Categorical twelve-step involvement and continuous abstinence at two years.
Journal of Substance Abuse Treatment, 44, 46-51.