A Completer’s Analysis of an Integrated Psychiatric/ Substance Treatment for Adolescents and Young Adults

I first announced the use of Encompass to deliver treatment for Substance Use Disorder in 2013. Four years later, we’re seeing some results.

Study Title: A Completer’s Analysis of an Integrated Psychiatric/ Substance Treatment for Adolescents and Young Adults

Co-authors: Christian Thurstone, MD; Madelyne Hull, MPH; Sean LeNoue, MD; Nicholson Brandt, BA; Paula D Riggs, MD*

Citation: Colorado Journal of Psychiatry & Psychology Volume 2 Number 2 May 2017

Abstract

Background

At least 80% of youth in treatment for substance use disorders have co-occurring psychiatric disorders. However, there are few treatment models that integrate psychiatric and substance treatment for youth. To address the gap, this study evaluates treatment outcomes for adolescents undergoing a clinical implementa on of Encompass, an evidence-based interven on that integrates psychiatric and substance treatment for adolescents and young adults.

Methods

Outcomes from 53 youth (11-20 years) who completed the 16-week outpatient program were collected using the a en on-de cit/hyperac vity (ADHD) disorder symptom check-list, Child Depression Rating Scale-Revised; Child Post-Traumatic Stress Scale; conduct disorder symptom checklist; Multidimensional Anxiety Scale for Children, 2nd edition; Timeline Follow- Back Interview; and urine drug screens. Paired t-tests, or their non-parametric equivalent, were used to evaluate change in psychiatric and substance use disorder severity.

Results

Overall, 54.7% of participants had a month of confirmed, substance abstinence at the end of treatment. Significant reductions in symptoms of ADHD (p<0.0001), anxiety (p=0.01), depression (p<0.0001), post-trauma c stress disorder (p=0.02), and propor on of days used substances (p=0.0004) were observed.

Conclusions

These preliminary findings support the need for multisite, controlled studies with intent-to-treat analyses to assess the efficacy of this type of integrated treatment. Further research is also needed to evaluate whether this intervention is feasible, and sustainable, and achieves outcomes similar to outcomes from controlled research trials.

——-

Background

At least 80% of youth in substance treatment have a co-occurring psychiatric disorder.1-3 The most preva- lent comorbidi es among these youth are conduct disorder (CD; 40%-60%), a en on-de cit hyperac v- ity disorder (ADHD; 30%-50%), major depressive dis- order (MDD; 20%-30%), generalized anxiety disorder (GAD; 20%), and post-trauma c stress disorder (PTSD; 14%). The presence of co-occurring psychiatric disorders, when entering treatment and psychiatric symptoms at the end of treatment or post-treatment, are both associated with worse substance treatment outcomes.

Current best practices call for integrated treatment of psychiatric and substance use disorders. However, estimates show that only about 25% of youth in substance treatment receive such care. Barriers to integrated psychiatric and substance use disorder treatment include separate sources of funding, licensure, and training as well as a paucity of models that integrate these services.

To address this gap, several studies were conducted to treat co-occurring psychiatric and substance use dis- orders in youth. The integrated approach for pro- viding concurrent treatment of co-occurring psychiat- ric and substance disorders included: comprehensive diagnos c and clinical evalua on, mo va onal en- hancement therapy, individual cogni ve behavioral therapy, family sessions, con ngency management, case management, and delity monitoring to ensure adherence to evidence-based prac ces. Urine drug screens were collected at each weekly session. Reliable and validated measures were administered at intake and monthly to track reduc ons in psychiatric and substance symptoms. The findings from the integrated treatment approach employed in these stud- ies led to the development of a manual-standardized 16-week outpa ent treatment program for adoles- cents and young adults with co-occurring psychiatric and substance use disorders called Encompass. We conducted a PubMed literature search including the terms “adolescent,” “substance treatment,” and “co- occurring psychiatric disorder” (using “AND” between search terms) and were unable to find another integrated psychiatric and substance treatment model for this population.

While these components of integrated treatment have been developed and tested in rigorous randomized-controlled studies, they have not been evaluated in non-research clinical se ngs, and have not been evaluated as a single package or model. To address these gaps, this study presents the outcomes of an Encompass implementa on to an urban, safety net hospital. It was hypothesized that patients completing the Encompass treatment program would have significantly reduced (1) frequency of substance use, and (2) severity of specific psychiatric symptoms such as ADHD, anxiety, conduct disorder, depression, and PTSD.

Methods

Procedure. Funding was obtained from private foundations to implement Encompass at a hospital-based, outpatient treatment program for adolescents and young adults with substance use disorders. The clinical program is located at a hospital a liated with a large academic institution. The Encompass team consisted of 4 full-time therapists (2 licensed clinical social workers, 1 licensed professional counselor, and 1 certified addiction counselor) and a child and adolescent psychiatrist. Treatment responses and outcomes as well as therapist adherence to the clinical protocol were regularly assessed. Approval from the Colorado Multiple Institutional Review Board was obtained to use the data collected from Encompass patients for research purposes.

Participants. Participants were 53 consecutive adolescents and young adults who completed Encompass. Participants were considered completers if they completed 17 weeks of treatment. A total of 117 patients enrolled in the program, yielding 53 who completed 17 weeks of treatment. Patients enrolled in the treatment from August, 2013 to June, 2015. Inclusion criteria were: (1) ages 11-24 years, (2) meeting criteria for at least 1 Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) non-nicotine substance use disorder; and (3) completed 17 weeks of Encompass treatment. Completers were chosen for these initial analyses to evaluate clinical changes with a full dose of treatment. Of note, while Encompass is designed to integrate psychiatric and substance treatment, it may be adapted to youth without co-occurring psychiatric disorders.

Measures. Data were obtained from clinical assessments that are systematically tracked as part of the Encompass intervention. The measures were admin- istered by the patient’s therapist or physician. All staff received a 2-day initial training to ensure proper and consistent administration of the measures. Baseline administration of the measures was conducted jointly with the patient’s physician and therapist to ensure diagnostic consensus and to ensure consistency of administering the measures.

The measures included in the intervention were the following:

Kiddie Schedule for Affective Disorders and Schizophrenia Present and Lifetime Version (K-SADS PL). The K-SADS is a reliable and valid semi-structured child and adolescent psychiatric diagnos c interview widely used in research. The instrument has shown to have good reliability for diagnosing ADHD, bipolar disorder, conduct disorder, generalized anxiety disorder, postraumatic stress disorder, oppositional defiant disorder, and substance use disorder in adolescents. For minors, both adolescent and parent/ guardian report were used. This instrument was used to obtain baseline psychiatric and substance use disorder diagnoses using the adolescent and guardian report, if the patient was a minor.

Timeline Followback Interview (TLFB). The TLFB uses anchor points to assess the number of days an indi- vidual used substances in the past 28 days. The TLFB was administered at baseline and weekly during treat- ment. The weekly results were collated into monthly outcomes. The TLFB is clinician-administered and has been shown to be reliable and valid in research and clinical se ngs of youth and young adults.

Child Depression Rating Scale-Revised (CDRS-R).22 The CDRS is a reliable and valid clinician-administered assessment of depression severity in children and adolescents and is widely used in treatment studies of adolescent depression. Raw scores are convertedto t-scores (0-100) with a mean of 50 and a standard devia on of 10 in norma ve samples. The cutoff score for clinically-significant depressive symptoms is 65. The CDRS-R was administered only to youth with a diagnosis of major depressive disorder at baseline and monthly throughout treatment.

Multidimensional Anxiety Scale for Children, 2nd edition (MASC II). The MASC II is a reliable and valid self-report, pen-and-paper ques onnaire assessing anxiety severity in children and adolescents. Raw scores are converted to a t-score with a cuto of 65 for clinically-significant anxiety. The MASC II was only administered to youth with an anxiety disorder at baseline and monthly throughout treatment.

Diagnostic and Statistical Manual of Mental Disorders, 5th edition conduct disorder symptom checklist (DSM- 5). The conduct disorder symptom checklist is a self- report, pen-and-paper ques onnaire created for this interven on that calculates the number of conduct disorder symptoms in the past month. Scores range from 0 (no symptoms) to 14 (maximum symptoms). The conduct disorder checklist was administered to youth with a diagnosis of conduct disorder at baseline and monthly during treatment.

DSM-5 ADHD checklist. The DSM-5 ADHD checklist assesses the severity of ADHD symptoms on a scale of 0 (none) to 54 (most severe). A clinical cuto score is 22. This clinician-administered instrument rates each ADHD symptom on a scale of 0 (none) to 3 (severe) over the last month. For minors, both adolescent and parent reports were used for baseline severity ra ng. This instrument has been widely used in clinical re- search of adolescents and young adults. Consistent with how the instrument was used in these studies, the current study relied on adolescent self-report as the primary outcome for ADHD severity. The ADHD checklist was administered to youth with a diagnosis of ADHD at baseline and monthly during treatment.

The Child PTSD Symptom Scale (CPSS). The CPSS is a 17-item self-report, pen-and-paper ques onnaire used to assess the severity of post-trauma c stress disorder (PTSD) symptoms. Scores range from 0 (no symptoms) to 51 (maximum symptoms) with a stan- dard clinical cuto of 15. The CPSS was only admin- istered to youth with a diagnosis of PTSD at baseline and monthly throughout treatment.

Urine drug screen. A commercially available, point-of-care urine drug screen was used weekly during treatment to evaluate substance use. The urine drug screen evaluated for alcohol, amphetamine, benzodi- azepines, cocaine, marijuana, and opioids.

Intervention. The Encompass intervention is a manual- standardized treatment for adolescents and young adults with co-occurring psychiatric and substance use disorders. Patients usually came to the clinic weekly for the Encompass interven on. What follows is a description of Encompass and how it was implemented in this study.
Comprehensive and ongoing assessment. Each patient was evaluated jointly with a master’s-level therapist and board-certified child/adolescent and adult psychiatrist. Baseline DSM-5 diagnoses were obtained using the K-SADS PL, adapted for use with the updated DSM-5. Frequency of substance use was obtained using the TLFB, and severity of ADHD, anxiety, conduct disorder depression, and PTSD were obtained using the ADHD checklist, MASC II, conduct disorder check- ist, CDRS-R, and CPSS respectively.
Motivational enhancement therapy and cognitive behavioral therapy (CBT). Encompass uses motivational interviewing throughout treatment to mo vate youth for posi ve change. CBT has been shown to effectively treat multiple psychiatric and substance use disorders in youth.

All sessions are individual, not group, in order to tailor treatment to the specific co-occurring psychiatric disorders with which a patient was diagnosed.

Contingency management. Youth have point-of-care urine drug screens with each session. The results are used with the sh-bowl technique to reward clean urine drug screens with an escala ng, posi ve rein- forcement schedule. Patients also get a drawing for showing up and 1 to 2 chances to draw for com- ple ng 1 or 2 pro-social activities that were collaboratively established in the prior therapy session. The fish bowl contains chits that are labeled “good job,” “small prize,” “medium prize,” and “jumbo prize.” Small prizes were of approximately $1 to $5 in value; medium prizes were approximately $10 to $20 in value; and the jumbo prize was about $100 in value. Patients received an automatic medium prize for the first time they obtained 2 consecutive clean urine drug screens.

Family treatment. Families were encouraged to a end up to 3 sessions to work on goal-settng, communication, and problem-solving. These sessions could occur at any time after the fourth session. Families could choose not to participate in family sessions.

Medication-assisted treatment. Medications were prescribed as indicated for psychiatric and substance use disorders.

Ongoing assessment. Substance use was formally assessed weekly with urine drug screens and monthly with the TLFB. Psychiatric symptoms were formally assessed monthly with the ADHD checklist, CDRS, conduct disorder checklist, CPSS, and MASC II.
Fidelity monitoring. The clinical team a ended a 2-day didactic session prior to beginning treatment. The didactic was led by the Encompass team, and included the rateonale for the Encompass interventions as well as information and skill building about the assessments and intervention. The team had weekly phone consultations with the Encompass physician, lead therapist, and operations manager to discuss cases for the first year of the implementation of the treatment. The frequency of phone consultation decreased to every other week in year 2. Encompass staff also conducted monthly site visits in the first year. The Encompass therapists a ended weekly group supervision and weekly individual supervision with the Encompass therapy supervisor in year 1. The supervision was decreased to every other week group supervision in year 2. Therapy sessions were audio-taped and randomly scored by the Encompass therapy supervisor for fidelity using an Individual or Family Session Rating Scale. These scales rated therapist adherence on different domains common to CBT such as agenda setting, use of role plays, and collaborative agreement on at-home practice. The scales also assess for common practices of motivational interviewing such as use of empathy and reflective statements as well as respect for patient autonomy. Adherence was rated on a scale of 1 (minimal adherence) to 5 (maximum adherence), with a 3 being a passing score. One session per month was evaluated for adherence to the treatment. Prior research shows that ongoing fidelity monitoring is crucial for therapist adherence to manual-standardized treatments. These therapist rating scales were adapted by the Encompass team from previous psychotherapy research studies.

Analyses. Data were analyzed with SAS EnterpriseGuide 5.1.35 Depending on tests of normality, con nu- ous data are presented as means with standard deviation or as medians with inter-quar le range, or IQR; categorical variables are presented as counts (%). A difference between the propor on of days completers used at least 1 non-tobacco substance in the month prior to program initiation and the month before program comple on was assessed using a Wilcoxon signed-rank test. Completers reporting no non-tobacco substance use for the month prior to entering the program were excluded from this assessment, leaving a sample size of 40 for the pre-post comparison of propor on of days used substances. Paired t-tests were used to determine change in pre- and post- program anxiety, ADHD, and MDD severity. All tests were two-tailed and used a p-value of 0.05 to detect statistical significance.

Results

Based on the mean age and frequency of baseline characteristics, the typical profile for an Encompass program completer is a 16-year-old male with at least 1 psychiatric diagnosis, most likely major depressive disorder, and who uses cannabis (Table 1). On average, program completers participated in 92% of the 16 sessions offered in the program. All 4 therapists achieved passing scores on their adherence rating scales.
Over half (54.7%) of all completers had at least 1 month of abstinence by the end of treatment as measured by both urine drug screen and self-report. The proportion of days used substances also showed a significant pre-post decrease. The average number of negative urine drug screens during treatment is 7.9 (SD=6.6). Participants diagnosed at intake with ADHD, anxiety, depression, and/or PTSD showed significant improvement in severity scoring upon comple on of the program. For the 3 youth with conduct disorder, the average number of conduct disorder symptoms in the past month decreased from 5.3 to 3.3, but no formal statistical analysis was conducted for this variable because of the small sample size.

Discussion

Those who completed Encompass treatment had a significant decrease in ADHD, anxiety, depression, and PTSD severity. They also had a significant decline invthe proportion of days used at least 1 non-tobacco substance. Fifty-five percent of completers had 1 month of confirmed substance abstinence at the end of treatment.
Previous research studies of individual CBT and the prescription of psychotropic medications show pre- post reductions in ADHD, conduct disorder, depression, and substance use during adolescent substance treatment. However, these evaluations were conducted in research settings, and the potential impact of these treatments in primarily clinical settings are lacking. This study is the first to present such outcomes. These ndings are also significant because Encompass is the only adolescent substance treat- ment model, of which we are aware, that is specifically designed to integrate psychiatric and substance use disorder treatment for adolescents.
The main limitation of this study is that it is not a controlled trial and includes only those who completed treatment. Therefore, these current analyses do not demonstrate a causal link between treatment and reduction in clinical symptoms. Furthermore, to limit the amount of potentially-identifying information in communications between the clinic and Encompass staff, the data collected did not include information related to participant race, ethnicity, or socio-economic status. Therefore, it is not possible to evaluate the relationship between various demographic variables and treatment outcome. Finally, the results of this single-site implementation may not generalize to other sites.
Given these limitations, there are several future directions for this line of research. First, it is difficult to compare these completer’s outcomes to published results, which are typically intent-to-treat. A recent study of adolescents with cannabis use disorder (12- 18 years, n=153), who underwent 14 weeks of MI/CBT and contingency management, showed that 53% of the intent-to-treat sample had at least 4 weeks of abstinence during treatment. Therefore, further intent-to-treat analyses are needed for this Encompass data set to see how the real-world clinical outcomes compare to outcomes from a controlled research trial. Second, other analyses, such as comparisons between completers and non-completers, are also needed to evaluate predictors of treatment response. Third, data from multiple sites will inform whether or not these findings generalize to other clinical settings.

Finally, given that only 1 in 10 adolescents with a substance use disorder receives substance treatment, research is needed to promote increased access to effective, integrated psychiatric and substance treatment. Such research involves developing ways to disseminate and sustain models such as Encompass. It also may involve adapting and evaluating this care to non-traditional settings such as schools, or adapting it to be delivered via telemedicine.

Acknowledgements

Funding/Support
This study was funded by: Caring for Colorado, Colo- rado Health Founda on, the Colorado Trust, Daniels Fund, and Denver Health Founda on.

Additional Contributions
The authors wish to acknowledge the following people who made this project possible: Lisa Gawenus, Robert House, Amanda Ingram, Rose Marie Medina, Brad Metzler, Brian Milton, Julia Timmerman, Audrey Vincent, Philippe Weintraub.