Comparative Efficacy of Family and Group Treatment
for Adolescent Substance Abuse
Douglas C. Smith, PhD,1James A. Hall, PhD,2Julie K. Williams, MSW,1
Hyonggin An, PhD,3Nathan Gotman, MS3*
1Project Iowa SCY, Department of Pediatrics, University of Iowa, Iowa City, Iowa
2Department of Pediatrics, University of Iowa, Iowa City, Iowa
3Department of Biostatistics, University of Iowa, Iowa City, Iowa
Due to the continuing prevalence of adolescent substance
abuse, promising treatment models need to be developed and
evaluated. Thus, the purpose of this study was to evaluate
the efficacy of two promising models, Strengths Oriented
Family Therapy (SOFT) and The Seven Challenges1
(7C). Adolescents who qualified for outpatient treatment
and agreed to participate in our study were randomly
assigned to one of the two treatments and assessed at 3
and 6-months following baseline. Using a two-part, ran-
dom-effects model, we examined the odds of achieving absti-
nence or full symptom remission between treatments and
over time. For those not achieving full abstinence or full pro-
blem remission, we investigated whether frequency of use or
symptom severity were reduced at follow-up. Participants in
both SOFT and 7C demonstrated significant reductions in
substance use and related problems, but treatments did
not differ at 3 and 6 months following baseline. Overall,
treatment services were delivered as planned. Both SOFT
and 7C were efficacious with adolescents who abuse sub-
stances, as participants in both conditions were significantly
more likely to be in symptom remission or abstinent at fol-
low-up interviews versus at baseline. Replication studies are
needed that address this study’s limitations. (Am J Addict
As adolescent substance abuse continues to be a major
societal problem with dire consequences, continued
research is needed on the efficacy of promising drug treat-
ment models.1,2Prior to the 1990’s, only a few studies on
drug treatment existed and these studies, for the most
part, suffered from major design flaws, including poorly
specified treatments, high treatment dropout in control
conditions, poor documentation of service hours, non-
standard and insensitive outcome measures, and high fol-
addressed these design flaws, but additional rigorous ado-
lescent substance abuse outcome studies are needed.4
BACKGROUND AND SIGNIFICANCE
approaches adds to our knowledge about adolescent sub-
stance abuse treatment by demonstrating their short-term
impact. Based on initial results, researchers can decide if
treatments are worthy of additional study and eventual
dissemination.5If treatment models are found to be effi-
cacious, they can be added to the arsenal of evidence-
based treatments available to practitioners. Overall, the
practice field has made it a priority to develop and imple-
ment evidence-based treatments. Additional studies are
also needed on new and unique treatment models that
differ in therapeutic orientation from existing models.
This study addresses these needs by evaluating the
efficacy of two new models for adolescent substance
abuse, Strengths Oriented Family Therapy (SOFT) and
The Seven Challenges1(7C).
Strengths Oriented Family Therapy (SOFT) Overview
Although some research has demonstrated positive
outcomes for existing family therapy models, these mod-
els did not include components such as pre-treatment
motivational sessions with parents, solution-focused lan-
guage, or formal strengths assessments.6,7SOFT com-
solution-focused family therapy, multifamily skills train-
ing groups, and targeted case management to comprise
a distinct family-based intervention.8Since SOFT has
unique qualities, evaluating its efficacy could eventually
Received March 27, 2006; revised May 31, 2006; accepted
August 17, 2006.
?Mr. Gotman is now in Connecticut.
The opinions are those of the authors and do not reflect offi-
cial positions of the government.
Address correspondence to Dr. Hall, Department of Pedia-
trics, 2559 JCP, University of Iowa, Iowa City, IA 52242.
The American Journal on Addictions, 15: 131–136, 2006
Copyright # American Academy of Addiction Psychiatry
ISSN: 1055-0496 print / 1521-0391 online
increase the number of evidence-based, replicable treat-
ments. SOFT is thoroughly described elsewhere.9
Seven Challenges1(7C) Overview
7C treatment combines decision-making exercises, skills
training, interactive journaling, and motivational inter-
viewing concepts to treat adolescent substance abusers.
The unique activities of this therapeutic approach include
journaling and receiving feedback from clinicians on jour-
nal entries, careful therapist role definition, and use of
peer-developed readings. This innovative approach shares
components with other interventions such as the use of
active listening and non-confrontational challenging, deci-
sional balance activities, amplification of client ambiva-
lence about change, and the use of the stages of change.
Comprehensive descriptions of 7C exist elsewhere.10,11
Clients in both treatments were expected to exhibit sig-
nificant reductions in substance use and related problems.
Based on previous literature, we did not expect that one
condition would produce significantly better reductions
in substance use frequency or related problems.4,12
All procedures were approved by the University of
Iowa’s Institutional Review Board before any data were
collected. The study was conducted in a catchment area
that included both metropolitan and rural counties.
Treatment services in this study were provided in colla-
boration with the Mid-Eastern Council on Chemical
Abuse (MECCA), a private not-for-profit agency.
Initially, adolescent participants were interviewed with
the Global Appraisal of Individual Needs (GAIN-I) while
the parent completed collateral measures (i.e., GAIN-CI).
Then, a few days later, the youth and parent returned to
discuss the results of the assessment.13At this feedback
session, a therapist provided clinical recommendations
and informed adolescents referred to outpatient treatment
that they were eligible for our treatment study. A research
assistant then met with eligible teens and their parents to
invite them to participate. Active consents were obtained
from both parent and teen. If the youth and parent chose
not to participate, the family was referred directly to stan-
dard drug treatment at MECCA. Participants were then
oriented to study procedures, signed consent documents,
provided information used for follow up location, com-
pleted additional intake measures, and were randomized
to condition. The youth returned for assessments at 3,
6, 9, and 12-months following baseline while parents were
assessed only at 6 and 12 months following baseline.
Enrollment into the randomized study began in June
2003 and ended in August 2005.
These data are a subset of a larger longitudinal study
comparing SOFT to 7C. Adolescents could be included
in this study if they were: a) 12–18 years old, b) living in
the catchment area or could commute, c) conversant in
English, d) referred to adolescent outpatient treatment,
and e) covered by either public or private insurance. For
the current study, we selected cases that had completed
both baseline and six-month assessments.
Out of 210 eligible adolescents, 103 (i.e., 49%) con-
sented to participate. Five cases were excluded from ana-
lysis because it was determined after randomization that
they did not meet study inclusion criteria. Of these, four
had insurance difficulties, and one was randomized to
SOFT but chose to enroll in 7C instead. We retained
92% and 92% of participants for 3 and 6-month fol-
low-ups, respectively. The final analysis sample included
Demographic and Clinical Characteristics (N ¼ 98)
The mean age of participants was 15.8 years. The
sample was 29% female, 24% minority, 39% lived in a
single-parent household, and 71% reported current juve-
nile justice system involvement. For the 90 days prior to
intake, youth reported risky behaviors including, (a)
using substances while driving, operating machinery, or
playing sports (13%), (b) using substances at work,
school, or while babysitting (31%), and (c) having unpro-
tected sex (30%). Using data on past year symptoms, we
diagnosed 1% with alcohol dependence, 12% with mari-
juana dependence, 12% with some AOD dependence,
22% with alcohol abuse, 37% with marijuana abuse,
and 38% with some AOD abuse. Using lifetime data,
we diagnosed 90% with substance abuse and 47% with
substance dependence. Also using lifetime data, 68% of
participants reported being physically, sexually, or emo-
tionally abused, with 42% reporting past year abuse
experiences. Over 80% of participants reported three or
more past year substance-related, psychological, beha-
vioral or legal problems.
Strengths-Oriented Family Treatment Procedures
The major activities of SOFT treatment included sev-
eral sessions and tasks. Initially, the youth and parent
participated in a pre-treatment motivational feedback ses-
sion (SORT). After completing an Immediate Concerns
Checklist, the therapist interviewed the youth and parent
using a formal Strengths and Resources Assessment.
Based on these strengths and concerns, the therapist
and family developed a Solution Plan which identified
the immediate and long-term goals for the family. Youth
and parents attended several multifamily group sessions in
which they received family communication skills training.
Finally, therapists provided targeted case management to
address ecological factors relevant to treatment (e.g.,
132Efficacy of Family and Group TreatmentSupplement 1 2006
school, friends, etc.). Families also developed a relapse
prevention plan together. Because SOFT is a solution-
focused model, great emphasis is placed on adolescents
and parents making concrete action plans for reducing
substance use and improving family communication.
During SOFT treatment, adolescents and their parents
attended ten weekly multifamily groups and five conjoint
family therapy sessions. Clinicians met with adolescents
individually in limited situations, which is commonly
allowed in family treatments to promote engagement.
Multifamily group sessions lasted about two hours while
conjoint family sessions were approximately two hours in
duration. We planned that SOFT participants would
receive approximately 30 hours of treatment. Typically
all SOFT sessions were provided within the first three
months following baseline.
Seven Challenges Procedures
Youth met with therapists for individual sessions and
with other teens in peer groups. During 7C treatment,
participants completed journaling exercises that examined
barriers to being honest about their use, what they liked
about drugs and alcohol, the impact of drugs and alcohol
on their lives, their and others’ responsibility for their
problems (i.e., including victimization), goals and dreams
in life, decisions about use, and steps needed to follow
through on those decisions. The therapists wrote com-
ments in the margins of the clients’ journals empathizing,
clarifying, or asking the client to expand on themes.
Therapists also discussed journal content in individual
sessions, which had a heavy emphasis on building rapport
and maintaining a non-confrontational stance. During
Seven Challenges therapy groups, adolescents completed
a structured check-in activity, read a reading on one of
the Seven Challenges followed by a brief lecture and
discussion, and spent the remainder of the group doing
Seven Challenges Dosage
Youth attended 10 weekly group sessions and 5
bi-weekly individual sessions with their counselors occur-
ring within the 3 months following baseline. As group
sessions averaged about two hours and individual ses-
sions lasted one hour, we planned that 7C1participants
would receive about 25 hours of treatment.
Treatment Dosage Variations
Minor variations in service hours existed for both con-
ditions, which depended on whether the adolescent was in
need of intensive outpatient or regular outpatient, or lived
in rural or urban locations. For extremely rural adoles-
cents (8%), we could not provide group treatment as they
could not always travel to our treatment location and the
pool of available clients varied. Thus, we modified
treatment models so that group content was presented
in individual sessions.
Staff Training and Supervision
Both SOFT and 7C therapists were trained initially by
the principal developers of each model. As part of these
initial trainings, we reviewed the treatment manuals with
the therapists, and provided intensive supervision of
initial SOFT and Seven Challenges cases by tape reviews
and weekly supervision. Manuals were referred to
regularly throughout the study. Additionally, therapists
recorded intervention dosage using a web-based manage-
ment information system. We conducted spot checks on
Three master’s level therapists provided SOFT—one
male and two females. The male therapist had six years
of experience, but the other two therapists had no adoles-
cent substance abuse treatment experience. Four thera-
pists provided the Seven Challenges intervention—two
with master’s degrees and two with bachelor’s degrees,
one male and three females, and an average of 2 years
experience treating adolescent substance abusers. Thera-
pists were originally nested within condition, but for a
brief period SOFT therapists were crossed over to provide
7C after receiving appropriate training. We monitored the
7C cases of these therapists closely to prevent intervention
Global Appraisal of Individual Needs (GAIN)
We administered the GAIN, a widely used, reliable
and valid, semi-structured interview, with each adolescent
at baseline.14–16Trained interviewers were certified after
2–3 tapes were satisfactorily reviewed. Approximately
5% of all audio taped assessments were reviewed. For this
study, two GAIN scales were used, the Substance Fre-
quency Scale (SFS) and the Substance Problem Scale
(SPS). The Substance Frequency Scale (SFS) is a reliable
(a ¼ .87) and valid scale measuring the average percent
of days reported of any AOD use, days of heavy AOD
use, days of problems from AOD use, and days of
alcohol, marijuana, crack=cocaine and heroin use. The
Substance Problem Scale (SPS) is a reliable (a ¼ .88;
test-retest r ¼ .73) and valid scale measuring the presence
of substance-related problems (i.e., DSM-IV criteria).
We examined both simple counts of session types, as
well as actual time spent on activities to yield measures
of total non-session minutes (i.e., case management), total
session minutes, and total treatment minutes (i.e., case
management plus counseling). In Table 1 we converted
minutes to hours for ease of presentation.
Smith et al.Supplement 1 2006133
We investigated the validity of self-reported past
month marijuana use using the rate of false negatives
(i.e., positive urine test, negative self report) as the criter-
ion for a subset of assessments (n ¼ 245). We found that
15 participants out of 154 (9.7%) did not report using
marijuana but tested positive for THC.
A mixed-effects, two-part model was used that first
estimated the logistic probability that a client achieved
abstinence (i.e., SFS) or had full symptom remission
(i.e., SPS). The model then predicted differences in the
severity of non-zero outcomes using a log-normal
regression. We used the SAS macro ‘‘MIXCORR,’’ for
the mixed-effects two-part model.17,18In all analyses,
age, gender, race, days in controlled environments,
weekly use in the home, and age of onset were included
as covariates. We controlled for the last two variables
because they were not distributed evenly across condi-
tions at baseline.
Data on number of sessions, total service hours, and
percentages of clients completing treatment (i.e., achiev-
ing 75% of expected dose) are presented in Table 1. Over-
all, service patterns for both groups reflected adherence to
Changes over Time and Between Groups
Many clients reported abstinence (SFS) or full symp-
tom remission (SPS), respectively. The percentages of
abstinent participants in 7C and SOFT were 8% and
3% at baseline, 34% and 27% at 3 months, and 39%
and 31% at 6 months, respectively. The percentages of
symptom free adolescents in 7C and SOFT were 33%
and 26% at baseline, 50% and 60% at month 3, and
61% and 60% at month 6, respectively. As shown in
Table 2, clients in both treatments were significantly more
likely to be abstainers or in full remission at the three and
six month interviews than at baseline. For 7C partici-
pants, the odds of continued use (SFS) at month 3 and
month 6 were reduced by 93% and 95% respectively,
and the odds of continued substance-related problems
(SPS) at month 3 and month 6 were reduced by 52%
and 69%, respectively. Similarly, for SOFT participants,
the odds of continued use (SFS) at month 3 and month
6 were reduced by 94% and 95% respectively, and the
odds of continued substance-related problems (SPS) at
month 3 and month 6 were reduced by 76% and 76%,
respectively. No significant differences were
between the two conditions on SFS or SPS, indicating
that neither treatment was superior at reducing the odds
of continued use or continued symptoms.
When considering those that had not achieved full
abstinence or full symptom remission, participants in
both conditions significantly reduced their frequency of
use at three months. Reductions in the severity of sub-
stance problems were also found for SOFT participants,
TABLE 1.Treatment dosage by condition
(n ¼ 58)
(n ¼ 98) (n ¼ 40)
Mean Sessions (SD)
Single Family Counseling?
Multifamily Group Counseling?
Group Counseling-Teen Only?
Group Counseling-Parents Only
Follow-Up Counseling Session
Mean Hours (SD)
Treatment Completion (%)y
?Statistically significant, p < 0.01.
yBased on 0 minutes for none (i.e. received no treatment), percentage of clients receiving below 75% of expected dose (partial), and percentage of
participants receiving equal to or above 75% of expected dose (Completed) calculated at 1125 minutes for 7 Challenges, and 1350 for SOFT.
134 Efficacy of Family and Group TreatmentSupplement 1 2006
but not 7C participants, at both follow up points. Neither
condition was found to be superior to the other in redu-
cing frequency of use or substance problem severity at
the follow up assessments.
These results support the efficacy of SOFT and 7C in
clinical practice to reduce drug use frequency and related
problems. Many participants in both conditions achieved
full abstinence or symptom remission, and even for those
that didn’t there were still significant reductions in fre-
quency of use at three months. Neither SOFT nor 7C
demonstrated superior results when directly compared.
Nevertheless, initial empirical support was found for
Strengths and Limitations of Current Study
Although the sample size for this study (n ¼ 98) did
not allow adequate power to include other variables of
interest, rigorous design features and strengths should
be noted. First, treatment manuals and supervision proce-
dures ensured treatment integrity. Second, dosage for
treatment interventions was documented to address pre-
vious critiques. Third, standardized measures with quality
assurance procedures (i.e., certified data collectors, tape
reviews with feedback, support from coordinating center)
were used to ensure standardized data collection. Fourth,
follow-up attrition rates were low (i.e., less than 20%),
which is very important because drop-out from follow-
up assessments introduces bias into findings.19Fifth, this
study controls for an important confound of longitudinal
substance abuse outcomes, time spent by participants in a
controlled environment during the follow-up period.20
Sixth, the sample included adolescents from a wide range
of ages. Taken together, this trial was methodologically
Several limitations should also be noted. First, treat-
ment manuals were finalized during the initial months
of the study. Thus, although didactic trainings, supervi-
sion, and the production of manuals were all components
of this study, the developmental nature of this study may
have impacted results. Second, case flow issues dictated
that we modified the intervention format for rural clients.
Third, we used open groups, which meant that partici-
pants in this study attended groups with non-study parti-
cipants. Finer grain analyses of how nesting adolescent
participants within these groups affected outcomes were
not possible due to a lack of data on non-study partici-
pants. Fourth, this study presents the 6 month results of
an on-going study following adolescents through 12
months. Although findings demonstrate that SOFT and
7C reduce drug use frequency and substance-related pro-
blems at 6 months, little is known about the long-term
intervention effects. Fifth, because we excluded partici-
pants without insurance, results may not generalize to
underserved populations. Finally, it was difficult to ade-
quately address therapist effects in this analysis model,
TABLE 2.Between group differences and changes over time for substance use frequency and substance-related problems
Part I: Logistic occurrence of abstinence (SFS) or symptom remission (SPS)y
Baseline to 3 months: 7C
Baseline to 3 months: SOFT
Baseline to 6 months: 7C
Baseline to 6 months: SOFT
SOFT vs. 7C: 3 months
SOFT vs. 7C: 6 months
Part II: Lognormal severity of frequency of use or abuse=dependence symptomsz
Baseline to 3 months: 7C
Baseline to 3 months: SOFT
Baseline to 6 months: 7C
Baseline to 6 months: SOFT
SOFT vs. 7C: 3 months0.07
SOFT vs. 7C: 6 months
yLogistic occurrence of zero or non-zero values (i.e., abstinence or full symptom remission) on the Substance Frequency Scale (SFS) and Substance
Problem Scale (SPS). Using the formula 1-OR (i.e., Odds Ratio) we calculated the reduction in odds of continued use or continued symptoms at follow-
zPrediction of lognormal values after zeros (i.e., participants in remission or participants reporting abstinence) were removed in the first part of the
two part model. Odds ratios (OR) are not applicable for this portion of the model. Instead the coefficient (Est.; b) is presented. Negative estimates indi-
cate reduced severity at follow-up. Standard errors (S.E.) of the estimates are also presented.
?Statistically significant at p < .05.
Smith et al.Supplement 1 2006135
due to the low number of clients seen by therapists that Download full-text
were crossed over into 7C. Future studies should address
Participants receiving both the Strengths Oriented
Family Therapy (SOFT) and The Seven Challenges1
treatments showed significant reductions in substance
use frequency and substance-related problems. Future
research should replicate these promising findings on
the efficacy of SOFT and 7C while addressing this study’s
This study was funded by cooperative agreement TI13354
from SAMHSA’s Center for Substance Abuse Treatment,
Rockville, MD (Dr. Hall, Principal Investigator).
1. Kandel DB, Davies M, Karus D, Yamaguchi K. The consequences
in young adulthood of adolescent drug involvement: An overview.
Archives of General Psychiatry. 1986;43:746–754.
2. Dennis ML, Dawud-Noursi S, Muck RD, McDermeit M. The need
for developing and evaluating adolescent treatment models. In:
Stevens S, Morral A, eds. Adolescent Substance Abuse Treatment
in the United States: Exemplary Models from a National Evaluation
Study. New York: The Haworth Press; 2003:3–34.
3. Godley SH, Jones N, Funk R, Ives M, Passetti LL. Comparing out-
comes of best-practice and research-based outpatient treatment pro-
tocols for adolescents. Journal of Psychoactive Drugs. 2004;36:35–48.
4. Dennis ML, Godley SH, Diamond G, et al. The Cannabis Youth
Treatment (CYT) Study: Main findings from two randomized trials.
Journal of Substance Abuse Treatment. 2004;27:197–213.
5. Rounsaville BJ, Carroll KM, Onken LS. A stage model of beha-
vioral therapies research: Getting started and moving on from stage
I. Clinical Psychology: Science and Practice. 2001;8:133–142.
6. Waldron HB. Adolescent substance abuse and family therapy out-
come: A review of randomized trials. In: Ollendick TH, Prinz RJ,
eds. Advances in Clinical Child Psychology. New York: Plenum
7. Williams RJ, Chang SY. A comprehensive and comparative review
of adolescent substance abuse treatment outcome. Clinical Psychol-
ogy-Science & Practice. 2000;7:138–166.
8. Berg IK, Miller SD. Working with the Problem Drinker: A Solution-
Focused Approach. New York: W.W. Norton; 1992.
9. Smith DC, Hall JA. Strengths Oriented Family Therapy for Adoles-
cent Substance Abuse, Social Work Under review.
10. Schwebel R. The Seven Challenges: Challenging Ourselves to make
wise Decisions about Alcohol and other Drugs. Tucson, AZ: Viva
11. Schwebel R. The Seven Challenges Manual. Tucson, AZ: Viva Press;
12. Luborsky L, Singer B, Luborsky L. Comparative studies of psy-
chotherapies: Is it true that ‘‘everyone has won and all must have
prizes’’? Archives of General Psychiatry. 1975;32:995–1008.
13. Smith DC, Hall JA. Strengths-Oriented Referral for Teens (SORT):
Giving balanced feedback to teens and families. Health & Social
Work. in press.
14. Dennis ML, Titus JC, White MK, Unsicker JI, Hodgkins D. Global
Appraisal of Individual Needs (GAIN): Administration guide for the
GAIN and related measures. Bloomington, IL: Chestnut Health Sys-
15. Dennis ML, Funk R, Godley SH, Godley MD, Waldron H. Cross-
validation of the alcohol and cannabis use measures in the Global
Appraisal of Individual Needs (GAIN) and Timeline Followback
(TLFB; Form 90) among adolescents in substance abuse treatment.
Addiction. 2004;99(Suppl 2):120–128.
16. Buchan BJ, Dennis ML, Tims FM, Diamond GS. Cannabis use:
Consistency and validity of self-report, on-site urine testing and
laboratory testing. Addiction. 2002;97(Suppl 1):98–108.
17. Tooze JA, Grunwald GK, Jones RH. Analysis of repeated measures
data with clumping at zero. Statistical Methods in Medical Research.
18. SAS Institute. SAS=STAT 9.1 User’s guide. Cary, NC: SAS Publish-
19. Stinchfield RD, Niforopulos L, Feder SH. Follow-up bias in adoles-
cent substance abuse treatment outcome research. Journal of Studies
on Alcohol. 1994;55:285–289.
20. Godley SH, Dennis ML, Godley MD, Funk RR. Thirty-month
relapse trajectory cluster groups among adolescents discharged from
out-patient treatment. Addiction. 2004;99(Suppl 2):129–139.
136 Efficacy of Family and Group TreatmentSupplement 1 2006