Title: From iPod to iGod Are 12step Groups Hip Enough for Adolescents
1From iPod to iGod? Are 12-step Groups Hip
Enough for Adolescents?
- John F. Kelly, Ph.D.
- Associate Professor in Psychiatry
- Harvard Medical School
- Associate Director
- Massachusetts General Hospital
- Center for Addiction Medicine
2Outline
- Developmentally-related barriers to participation
in 12-step mutual-help groups among adolescents - Review evidence regarding 12-step participation
and its influence on relapse - 8-year course of recovery for young people in
relation to AA/NA - The implications of the available evidence for
clinical practice guidelines
3Acknowledgments
- UCSD/VA San Diego Healthcare System
- Sandra A. Brown, Ph.D.
- Mark G. Myers, Ph.D.
- Brown University
- Christopher Kahler, Ph.D.
- Ana Abrantes, Ph.D.
- NIAAA (R01AA007033-09 R01 5AA015526-03)
4Treatment and Continuing Care
- Alcohol/drug disorders viewed as chronic -
frequently require ongoing management/monitoring,
re-intervention for lasting remission (McLellan,
2002 Dennis et al, 2005 White et al, 2006) - To aid recovery efforts informal continuing care
resources (e.g., AA and NA) are often utilized
have been shown to be cost-effective adjuncts to
formal treatment (Bond et al., 2003 Gossop et
al, 2007 Humphreys, 2004 Humphreys and Moos,
2001 2007 Kaskutas et al., 2002 Kelly et al.,
2006 Kissin et al., 2003 Magura et al., 2005
McKay, 2001 Tonigan et al., 2002 Weiss et al.,
2005 White, 2009). - Thus, newer interventions incorporate standard
referral (Longabaugh et al., 2005 Pettinati et
al., 2005) and addiction and mental health
organizations advocate in adult practice
guidelines (e.g., APA, 2006 IOM, 1998 NIDA,
2000 VHA, 2001)
5Limited Youth-specific Knowledge
- Comparatively little known about AA/NA as youth
continuing care resources - Yet widespread use of 12-step treatment
approaches and referrals to AA/NA by youth
providers (Brown, 2004 Drug Strategies, 2003
Kelly et al., 2008 Knudsen et al, 2008).
6AA/NA as continuing care resource for adolescents
- More than three quarters of treatment programs
espouse a 12-step orientation (Roman et al, 2004)
yet it remains one of the least evaluated (Miller
Willbourne, 2002). - 66 or more of youth facilities incorporate
12-step elements to some degree (Drug Strategies,
2003) only 9 base treatment solely on 12-step
model but 47 require AA/NA participation during
treatment, and 85 of programs link adolescents
to AA/NA groups (Knudsen et al, 2008). - From 12-step treatment perspective, ongoing
community AA/NA participation is seen as key
mediator of successful, long-term remission of
the disorder (McElrath, 1997) - Other treatment orientations may see community
AA/NA groups as a useful continuing care resource
irrespective of theoretical orientation and most
pts appear to be referred (Humphreys, 1999
Laudet White, 2005 Kelly et al, 2008)
7Clinical Staff Perceptions, Expectations, and
Referrals to AA/NA
Kelly, Yeterian Myers, 2008, Alcoholism
Treatment Quarterly.
8In theory 12-step groups promising treatment
adjunct
- Lots of meetings, in most communities, at
high-risk times (evenings/ weekends) when
professional support often not available - Access to a supportive abstinent social network
- Free of charge, pts can attend as intensively for
as long as desired - Provide an adaptive community based
recovery-oriented system, responsive to
undulating relapse risk (Kelly Yeterian, in
press)
9In actuality, youth face a number of
developmentally-related Barriers
- AA Membership 11.3 ltage 30 and 2.3 lt21 (AA,
2008) - Clinical differences
- Addiction severity (substance use
profile/withdrawal/consequences) - Problem recognition/motivation for abstinence
- Greater psychological distress
- Barriers to identification with pertinent
life-stage issues - marriage, employment, NIH funding problems
- 12-step Specific barriers
- Disinterest or opposition to spiritual/religious
content - May signify poor fit with 12-step fellowships
emphases on abstinence and spiritual growth
10Youth-Specific AA/NA Knowledge Limited
Existing evidence suggests participants better
outcomes at least in the short-term
11Youth-Specific AA/NA Knowledge Further steps
needed(2)
- Given small number of studies, we still we know
very little regarding -
- Which youth attend and at what rate?
- Whether youth benefit over the short and
longer-term as they transition into young
adulthood - What levels of utilization are associated with
benefits
12What we do know lacks rigor
- In most studies 12-step participation and
outcomes examined concurrently obfuscating causal
attributions related to temporal precedence
(Nock, 2007) - In many studies relevant time-invariant confounds
related to better prognosis (e.g., baseline
substance involvement) not controlled - No existing youth studies have taken into account
influential time-varying factors (e.g.,
professional txs, levels of substance use in
preceding follow-up periods) - Thus, a major aim of current investigation was to
use rigorous longitudinal methods (GEE), lagged
approach, controlling for static and time-varying
covariates
13Which youth attend? Predictors of AA/NA
Attendance (1)
- Based on theory and prior evidence we examined a
number of predictor variables - Demographic (e.g., gender)
- Psychiatric (e.g. conduct disorder)
- Severity of substance involvement (e.g.
dependence sxs, lifetime intoxication,
withdrawal) - Psychological (ability to moderate alcohol/drug
use, abstinence self-efficacy) - Prior-treatment
- Religious (affiliation, practices)
14Do youth benefit in the short and
longer-term?Influence of Early Ongoing AA/NA
- Critical juncture - inpatients discharged back to
community first 6m requires demanding
adjustments - remains highest risk relapse period
(Brown, 1993 Brown et al., 1989 Godley et al.,
2002 Hunt et al., 1971 Vik et al., 1992). - Findings from adult studies suggest also that pts
who achieve immediate, rather than delayed, AA/NA
may experience better long-term outcomes (Moos
and Moos, 2004). - Given pervasive use of 12-step treatment (Drug
Strategies, 2003) and pressure for evidence-based
practice (IOM, 1998), we aimed to provide
knowledge regarding extent treated youth meet the
key proximal outcome and whether its associated
with better long-term outcome (Finney, 1995
Suchman, 1967).
15What levels of utilization are associated with
benefits?
- Prior studies support notion linear relationship
between AA/NA and outcome - Stress 90 in 90? Or would 1 or 2 meetings a
week help my patient? What would be optimal?
Little empirical basis for recommendations - Hence, further aim here to analyze relationship
between levels of AA/NA participation and
abstinence using non-parametric regression (i.e.,
locally-weighted scatterplot smoothing LOWESS
Cleveland Devlin, 1988 Cleveland, 1993,
chapter 3)
16Study Aims
- Describe rates and predictors of participation in
AA/NA over 8-years following inpatient treatment - Examine the relationship between early, and
ongoing AA/NA participation, and substance use
outcome using lagged, longitudinal models
(Generalized Estimating Equations GEE)
controlling for static and time-varying
covariates - Explore dose-response thresholds of AA/NA
attendance and outcome (LOWESS)
17Methods 1
- Participants (N166 40 female M age 16 75
Caucasian) consecutive admits to adolescent
12-step oriented, inpatient, SUD programs in San
Diego (M stay 4 wks) - Followed at 6m and 1, 2, 4, 6, and 8yrs
(follow-up rates gt 84) - Neither demographic nor tx/clinical vars found
associated with follow-up (psgt.27).
18Methods 2
- Prior outpatient 47
- Prior inpatient 8
- School Attendance
- 73.8 was enrolled in public or private school
- 19.2 reported quitting school
- 6.2 were expelled or suspended.
- Alcohol -most commonly used drug.
- Reported drug of choice was
- meth/amphetamines (53)
- cannabis (32)
- hallucinogens (7)
- alcohol (4)
- cocaine (3)
- opiates (lt1)
19Results Rates of Attendance
20AA/NA Predictors Results
21Relationship to Outcome Zero-Order Spearman rs
AA/NA x PDA Across 8 Years
Correlation is significant at the 0.01 level
(2-tailed). Correlation is significant at the
0.05 level (2-tailed).
22Relationship to Outcome Generalized Estimation
Equations (GEE) Models (1) AA/NA in relation to
Outcome
- Preliminary Analyses Static control variables
(sig. on 2 or more follow-ups) - Examined 5 sets of variables
- Demographic (age, gender, ethnicity, religion)
- Substance-related (pre-treatment frequency of
use, lifetime alcohol/drug intoxication, DSM-IV
alcohol/drug dependence sxs, withdrawal sxs,
beliefs about ability to use alcohol/drugs in
moderation) - Psychological (conduct disorder diagnosis,
self-efficacy) - Treatment (prior outpatient or inpatient)
- Family (family history of alcohol/drug
dependence)
23Generalized Estimation Equations (GEE) Models
(2) AA/NA in relation to Outcome
- Preliminary Analyses
- Two sig. results for (static) control variables
- Gender predicted outcome at 4-, 6-, and 8-years
(girls/young women having more abstinent days
than boys (M r -.30, pslt.001). - Greater perceived ability to use alcohol/drugs in
moderation consistent predictor of less
abstinence at every follow-up time point through
8 years (M r -.21 range -.18 to -.22, ps
.008-.03). - Thus, in addition to baseline use (past 90 days),
gender, perceptions of ability to moderate use
included - Time-varying covariates
- Substance use (prior period)
- Aftercare meetings (6m)
- Any outpatient/inpatient SUD treatment (all
periods)
24Lagged GEE Model of Youth Treatment Outcome over
8 Years in Relation to AA/NA Attendance in the
first 6 months post-treatment
1. Square root transformed N 150
Kelly, Brown, Abrantes, Kahler, Myers (2008)
Alcoholism Clinical Experimental Research, 32, 8
1468-1478.
25Lagged GEE Model of Youth Treatment Outcome in
relation to AA/NA attendance over 8 Years
1 Sq root transformed 2 Time varying covariate
26Levels of participation in relation to
outcomeDose-Response Thresholds for Attendance
(LOWESS)
- LOWESS Locally weighted fitting and smoothing
scatterplots Cleveland, 1979 1991) - Instead of forcing the data to fit some a priori
function of x on y, this method that lets the
data speak for themselves (thus, no single
parameter estimate of relationship) - Data is modeled locally by WLS polynomial
regression- weighted more for local points
(robust to outliers)
27(No Transcript)
28(No Transcript)
29Results Summary
- AA/NA attendance early post-treatment was common
and relatively intensive, but declined steadily
over time - Objective and subjective indices related to
addiction severity predicted youth AA/NA
attendance following treatment - Despite declining attendance rates over the 8
years, early post-treatment attendance predicted
long-term salutary outcomes - Effects associated with ongoing attendance
remained significant and consistent - Exploratory LOWESS analyses suggested that even
relatively small amounts of AA/NA associated with
benefits, with a strong linear association
evident up through 3 meetings per week
30Discussion (1)AA/NA Participation over 8 years
- Participation very high early post residential
tx, but quickly declines in proportion
attending/freq AA/NA - Clinical staff survey (n114 5 programs Kelly
et al, 2008) suggest desire is for long-term
AA/NA participation (especially if treated in
12-step oriented program) and staff refer most
patients - Suggests youth may encounter other developmental
barriers to AA/NA participation (intrinsic,
age-barriers)
31Relation between Age Composition of Attended
Meetings and Percent Days Abstinent
Kelly, Myers Brown, (2005) Journal of Child and
Adolescent Substance Abuse
32Predictors of AA/NA Attendance
- Similar to other studies youth in the current
study were more likely to participate if they had
- greater hx of alcohol/drug involvement
- thought they had a problem with alcohol
- considered themselves unable to control their
substance use (possible screening question) - Results consistent with HBM (Rosenstock, 1974),
construct of perceived severity (Finney Moos,
1995) - Religious involvement did not predict attendance
and girls/young women as likely as boys/young men
to attend
33Outcome in Relation to Early AA/NA Attendance
- Effects from early attendance (0-6m) suggests
what is accomplished therapeutically early
post-treatment has diminishing, but longer-term,
recovery ramifications (Moos and Moos, 2004) - Specific mediators of effects remain to be
clarified (e.g., avoidance of high risk social
environments or cognitive-emotional shifts)
34How might AA/NA reduce relapse risk?
35Conceptualization of How AA May Reduce Relapse
Risk
Decreased
Coping
Self-efficacy
probability
Response
of
relapse
High Risk Situation
AA
Increased
Initial use
No coping
Decreased
Abstinence
probability
of
response
self
-
efficacy
violation
of relapse
substance
effect
Dissonance
Positive
Outcome
conflict and
Expectancies
self
-
for initial use
attribution
of the
(guilt
substance
loss control)
Source Marlatt Gordon, 1985
36plt.001 plt.01 plt .05 plt.10 N99
Kelly, Myers, Brown (2000) Psychology of
Addictive Behaviors
37plt.001 plt.01 plt .05 plt.10 N99
Kelly, Myers, Brown (2000) Psychology of
Addictive Behaviors
38Ongoing AA/NA Participation in Relation to Outcome
- Although attendance declines substantially,
consistent effects of AA/NA over time, after
accounting for formal care utilization and other
confounds - for every meeting attended there was a subsequent
gain of about 2 days of abstinence, independent
of other factors also associated with good
outcome - Findings suggest AA/NA may be beneficial for some
youth in their recovery in the short- and
long-term, particularly those more severely
affected and who believe it unlikely they can
moderate substance use - Ongoing recovery-specific support from AA/NA may
help reduce relapse risk, in part, by providing a
social context where role-models are available
and friendships can develop that provide
alternative socially rewarding activities (Brown,
2001 Moos, 2007 Longabaugh et al., 1998).
39Substance Use and Problem Onset and Offset
NSDUH Age Groups
100
Severity Category
90
No Alcohol or Drug Use
80
70
Light Alcohol Use Only
60
Any Infrequent Drug Use
50
40
Regular AOD Use
30
Abuse
20
10
Dependence
0
65
12-13
14-15
16-17
18-20
21-29
30-34
35-49
50-64
NSDUH and Dennis Scott under review
40Prevalence of DSM-IV Alcohol Dependence across
the Lifespan
Source Grant, Dawson et al, 2004
41Serious Psychological Distress (NSDUH, 2007)
42Dose-Response
- 3 or more AA/NA meetings per week, on average,
appears associated with complete abstinence early
post-treatment - However, even 1 or 2 meetings per week was
associated with sharp increases in abstinence - Suggests even though no 12-step-specific actions
completed general group-therapeutic processes may
nevertheless be helpful (e.g. Kelly, Myers
Rodolico, 2008).
43What do you like best about AA/NA meetings? (6m
follow-up n53)
Source Kelly, Myers Rodolico (2008) Journal of
Substance Abuse
44Limitations
- Although rigorous test of AA/NA participation
over time cause-effect conclusions limited - Aggregated data over long time periods (up to 2
years at a time) do not lend themselves to
interpreting the more complex relationship of
attendance topography in relation to outcome - Our measure of AA/NA participation was frequency
only, did not measure degree of involvement in
AA/NA or age composition of attended meetings - Although all met for AUD, many identified main
substance as meth/amphetamine (53) - Inpatient, 12-step oriented, treatment (Drug
Strategies, 2003)
45Conclusions
- AA/NA attendance is likely to be high initially,
when patients are treated in 12-step-oriented
residential programs encouraged/facilitated to
attend - Even small amounts of participation appear
related to improvement strong linear
relationship up through 3 meetings per week - Similar to adults, more severe youth attend, and
youth who continue attending appear to experience
better outcomes into young adulthood (M range
16-24yrs) - Social contexts facilitated by AA/NA may help
buffer youth relapse during stage of life when
rates of use/disorders peak in the general
population and where support for recovery may be
minimal
46Provisional Clinical Guidelines
- Substance dependent youth should be encouraged to
try NA or AA meetings - Locate and direct youth to meetings where some
other youth may be present - Prepare youth for what to expect
- Emphasize potential benefits reported by other
youth that go (e.g., feeling connected, less
alone that somebody cares) - Emphasize social component