Author
BENDER Kimberley; SPRINGER David W; KIM Johnny S.
Title
Treatment effectiveness with dually diagnosed adolescents: a systematic review.
Journal citation/publication details
Brief Treatment and Crisis Intervention, 6(3), August 2006, pp.177-205.
Summary
A statistical synthesis of the findings of six randomised controlled trials (RCTs) of seven treatment interventions for dually diagnosed adolescents shows that several are effective in producing improvements in the three domains of substance abuse and internalising or externalising mental health problems. Family behaviour therapy and individual cognitive problem-solving therapy stand out as showing large effect sizes across all three domains. Preliminary treatment guidelines based on the results of the studies are suggested, but further research is needed.
Context
Adolescents with comorbid substance abuse and mental health disorders are ‘a very challenging population to treat’, and among substance abusers it is estimated (in US studies) that 50-90% also have a psychiatric illness. Few interventions have been developed or tested specifically with this group, and this review focuses on the limited amount of evidence currently available.
Methods
What sources were used?
The following databases were searched: ERIC (Education Resources Information Center); PsycINFO; Medline; Social Services Abstracts; Social Work Abstracts; and ‘the Campbell Collaboration and Cochrane databases’. The first of these is presumably C2-SPECTR (Social, Psychological, Educational and Criminological Trials Register), while the second could be the Cochrane Library as a whole or an element of it, such as the Cochrane Central Register of Controlled Trials (CENTRAL).
What search terms/strategies were used?
The terms used ‘included’: adolescent, youth, teen, juvenile, substance abuse, drug abuse, treatment outcome, intervention, efficacy, mental health, co-occurring, dual diagnosis, comorbid.
What criteria were used to decide on which studies to include?
Eligible studies were RCTs of treatments for dually diagnosed disorders among adolescents aged 12-18, published in an English language peer reviewed journal in the previous ten years. Prevention studies were excluded.
Who decided on their relevance and quality?
The initial results of searches, and subsequent relevance and quality assessment, are not reported. Methodological quality issues are, however, referred to in the text.
How many studies were included and where were they from?
Six studies examining seven interventions met the inclusion criteria and are summarised in Table 2, which gives details of data collection, outcomes for substance abuse and internalising/externalising mental health problems, sample size, gender, race/ethnicity, age range, diagnosis, attrition, and treatment delivery mode. Geographical settings are not reported, but all or most of the studies appear to be from the USA.
How were the study findings combined?
The synthesis is statistical. Effect sizes were calculated for three categories of outcome measure (externalising problems, internalising problems and substance abuse problems), and for each treatment modality. The procedures used in the calculations are fully reported.
Findings of the review
In addition to the RCTs, the searches also identified ten non-randomised pre-test/post test outcome studies, five with follow-up from three to 24 months. These are summarised in Table 1 and show that ‘treatment is a complex task, often resulting in mixed outcomes’. The treatment of substance abuse disorders, including the maintenance of lasting improvement, is particularly difficult and made more so by the comorbid disorder, especially if it is externalising. Although positive outcomes for some symptoms are possible, ‘reducing them to clinically meaningful levels is difficult.’
The six RCTs covered the following interventions:
Multisystemic therapy (MST)
MST is a family- and community-based treatment, based on the notion that behaviour is determined by multiple forces. A trial conducted by its developer, Scott Henggeler, compared MST to ‘usual community services’ (often minimal) in treating substance abusing juvenile offenders, of whom over 70% were dually diagnosed. MST was found to reduce alcohol, marijuana and other drug use, as well as the number of days spent in out-of-home placement. However, improvement was not maintained at six month follow-up, and reductions in criminal activity were less than in other MST studies. Henggeler et al suggest that this was probably ‘due to difficulty in transporting MST from its developers into practice.’
Interactional group therapy (IT)
IT focuses on the importance of interpersonal relationships, with the aim of developing insight and self-esteem, and improving self-care. Two studies compared IT with cognitive behaviour therapy (CBT) for 32 adolescents aged 13-18 who had left a partially hospital-based treatment programme, the second study reporting on follow-up 15 months after treatment ended. CBT was more effective in reducing substance use than IT at three months post-treatment, but both groups showed a significant reduction at the 15 month point. These findings confirm research with adults showing the maintenance of treatment gains independent of therapy type, but are limited in usefulness because of the lack of a no-treatment control group, high attrition rates and the lack of objective measurement of substance use at follow-up.
Family behaviour therapy (FBT)
FBT views physiological dependence and peer pressure as reinforcers of continued substance use, and targets multiple factors believed to influence substance use and anti-social behaviour. One study compared FBT with Individual Cognitive Problem Solving, which focuses on developing self-control and problem-solving abilities. This study is described as rigorous, though with a rather small sample size (56 individuals aged 12-17), and shows no differences between FBT and ICPS in reducing alcohol and illicit drug use, and improving conduct and mood. In both areas, there were significant improvements from pre-test to post-test, and these were maintained at follow-up. These findings confirm earlier research on the effectiveness of FBT in reducing behavioural problems, and providesthe first evidence of the effectiveness of ICPS in reducing drug use.
Cognitive behaviour therapy (CBT)
One study compared CBT with psychoeducational therapy (PET) among 88 predominantly dually diagnosed youths aged 13-18 who were receiving outpatient treatment. CBT was associated with lower substance abuse relapse rates than PET at three months post-treatment, but this superiority was not maintained by the time of follow-up at nine months when both treatments had similar positive effects on relapse rates. Like the IT study (conducted by the same research team), this one shows the same pattern of early differential effects but similar effects in the longer term.
Ecologically based family therapy (EBFT)
EBFT is targeted at runaway adolescents with numerous problems and is based on the notion of family preservation. One study of 124 youths in a runaway shelter compared EFBT with usual care. It showed that the EFBT group had a greater overall reduction in substance use but that both groups showed significant and equal improvement in psychological functioning, family functioning and HIV knowledge. ‘Overall’, reductions in high risk behaviour were maintained at follow-up. The findings confirm those of earlier studies supporting the family treatment of substance abusing adolescents, but ‘the retention rates and engagement…are uncharacteristically high’ for a study involving runaway youth. The study also lacks measures of treatment fidelity.
Seeking Safety Therapy (SS)
SS is a manualised psychotherapy designed to treat co-occurring substance use disorder and post-traumatic stress disorder through the development of coping skills. A study of 33 adolescent girls comparing SS with treatment as usual found that the SS group had significantly better outcomes in areas such as substance use and knowledge/understanding of their conditions. However, only some gains were maintained by follow-up at three months, although attrition meant ‘the power to detect significant relationships’ at this stage was limited.
Authors' conclusions
The methodological variability of the studies means that the analysis of findings ‘was a complex and difficult task’. However, overall ‘the effect sizes…are impressive’.
The authors conclude that, in respect of mental health outcomes, externalising effect sizes were large for the MST, FBT and ICPS groups, and had increased further in the MST and ICPS groups at follow-up. Internalising effect sizes were large for the IT, CBT, FBT and ICPS groups, and in all four cases (but especially IT and CBT) they had risen at follow-up. For substance abuse outcomes, effect sizes were large for the FBT, ICPS, PET and CBT groups, while newer and less established treatments such as EFBT and SS also had moderate effect sizes.
Taking all outcomes into account, FBT and ICPS were the only two approaches to produce large effect sizes in all three of the externalising, internalising and substance abuse domains, and these were sustained over time.
Implications for policy or practice
The authors found ‘few clear treatment guidelines for effective treatment of dually diagnosed adolescents’ when searching the literature, and Table 6 presents some ‘tentative’ advice based on the findings of their review. However, this is intended only as ‘a general gauge of how to tailor treatment’ for this group, and the limitations of the evidence base are discussed. These include the small number of studies, and methodological deficiencies in several of them, including the widespread use of self-report measures. Further research ‘is clearly warranted’ and some general guidelines for future studies are offered.