Inpatient vs. Outpatient Treatment Outcomes for Adolescent Substance Abuse

Literature Search Methodology

Sources for this review were gathered using a number of keyword searches and combinations of terms, such as "adolescent," "substance abuse," "treatment," "drug," "outcome," "drug abuse," "juvenile drug abusers," "juvenile drug use," "juvenile drug treatment," "adolescent drug treatment," and "adolescent drug use." These keyword searches were conducted in online databases including PsychInfo, PubMed, NCBI, JSTOR, the National Institute of Mental Health, the National Criminal Justice Reference Service, and Google Scholar.

These searches yielded a number of studies that appeared useful for this analysis. The five studies chosen were selected on the basis of common keywords and their relevance to this review's central focus: whether adolescent males with substance use disorder who receive inpatient treatment, rather than outpatient treatment, have a greater chance of maintaining sobriety.

Article Descriptions and Technique

Article 1: Branson, Clemmey, and Mukherjee (2013)

This study used a sample of 48 adolescents, roughly half Latino and half African-American, male and female. Treatment was outpatient only, making the study useful primarily as background on outpatient therapies and their success rates. The focus of the study was on reducing attrition by incorporating text message reminders of therapy sessions sent to patients. The study found that text messages reduced attrition rates, and only 4% of the sample did not receive text messages throughout the duration of the trial due to phone complications.

The study groups were comparable in that all participants were minorities receiving outpatient treatment, keeping the focus tight. Post-treatment, the completion rate improved: pretest attrition rates were in the 40th percentile, while post-test attrition rates improved considerably, with completion rising to approximately the 60th percentile. The demographic of the study was largely homogenous, consisting of a minority population (African-American and Latino). Attrition within the trial sample itself was negligible. The Hawthorne effect was not taken into consideration, as there was no mention of it in the study.

The reliability and validity of the measures used in the study are substantial. There is no significant temporal validity concern beyond the single trial, which spanned 7 months. The criterion and construct validity are sound: the outcome confirms that text messaging does reduce attrition, and construct validity demonstrates that appropriate inferences can be made about the usefulness of text message reminders in helping youths complete outpatient treatment.

The external validity of this study may be greater than it initially appears, as it suggests that one key to improving outpatient treatment success rates is to leverage the tools familiar to younger generations — specifically mobile phones — to remind them of scheduled appointments. A text message is a simple way to convey a sense of care that the recipient can perceive even through such a brief medium. This finding may relate more broadly to adolescent male substance abusers by offering a practical "way in" to the user's life and providing the ongoing support that can help facilitate a "way out" of substance abuse.

Article 2: Esposito-Smythers, Spirito, Kahler, Hunt, and Monti (2011)

This study examined cognitive behavioral therapy (CBT) protocol as a treatment for adolescents with co-occurring substance abuse disorder and suicidality. The sample was randomized, and the study was conducted as a clinical trial. Treatment consisted of integrated outpatient CBT. Results showed that outpatient treatment was successful in reducing suicide attempts, inpatient stays, emergency room visits, and arrests. Adolescents also demonstrated reduced suicidal ideation as well as reductions in alcohol and marijuana use.

Groups were comparable in age and ethnicity, with the overwhelming majority (approximately 80%) identifying as Caucasian. There was no baseline heterogeneity, as the sample was largely homogenous, treatment protocols were similar across participants, and patients shared the same substance dependencies. Pre-treatment, patients showed higher suicidality; post-treatment, this risk was reduced. The demographic composition of the group did not appear to be a primary concern of the researchers, and there was no indication of attrition or consideration given to the Hawthorne effect.

The temporal validity of the study is of some use given the time frame employed and the nature of the study's impact, which centers on CBT outpatient treatment — an applicable therapy for adolescents. Inferences regarding construct validity appear accurate, and the criterion validity is consistent, as variables align with outcomes and CBT appears to be a model with genuine applicability to this subject population.

The study has external validity in that it can serve as a basis for further research — something the researchers themselves advocate in their concluding remarks, noting that they cannot say with certainty that integrated CBT is the primary reason for successful outcomes, or whether other outpatient treatments would be equally effective. For this particular study, integrated CBT did show positive results, but more research must be conducted to confirm these findings.

Article 3: Friedman, Terras, and Ali (1998)

This study drew an outpatient sample that was approximately 70% male, collected randomly from cities in four states (Ohio, Texas, Pennsylvania, and Oregon). The inpatient sample was approximately half male and half female. Participants were assessed using the Adolescent Drug Abuse Diagnosis (ADAD) instrument both pre- and post-test. The study was conducted to test prognostic variables and predictors of outcome. Results showed that fewer than half of the variables examined were effective in predicting outcomes in either outpatient or inpatient cases of adolescent substance users seeking treatment. The variables were drawn from 157 historical cases of adolescent inpatient substance users and 296 historical cases of adolescent outpatients seeking treatment.

The groups were heterogenous, and there were signs of baseline heterogeneity: several variables produced different results across cases, including age, gender, attribution, and type of therapy. Demographics were not treated as significant given the random sampling approach. The study showed no awareness of attrition or consideration of the Hawthorne effect.

The outcome criterion validity was defined as a reduction in substance use. Of the variables examined, 12 were helpful in predicting inpatient outcomes, and 15 in predicting outpatient outcomes. Two variables were equally poor predictors in both inpatient and outpatient settings: school performance and failure to recognize substance use as harmful. Prognostic variables were also not the same for adolescents as they have been historically for adults, suggesting some external validity for future research. For example, increased substance use in follow-up sessions predicted increased use at admission in adults — but in adolescents, this pattern held only for outpatients.

The study's external validity lies in its creation of a framework for predictive variables that may be useful to the field. For instance, the assumption that schooling and positive or negative associations with substance abuse would serve as outcome indicators proved misleading, with no meaningful correlation found. This is a strength of the study. One weakness in external validity, however, is a methodological one: treatment phases were not homogenous, and the scope of the sample and study focus was too broad to produce substantial findings. More research is needed to corroborate these results, but the study serves as a useful starting point for understanding predictors of treatment outcome.

Article 4: Waldron, Slesnick, and Brody (2001)

This study used a randomized sample enrolled in a clinical trial designed to assess different forms of therapy: CBT, family therapy, combined individual and family therapy, and group intervention. The sample consisted of 114 adolescent substance users. The study focused primarily on outpatient therapy and provided no reference to inpatient treatment, although it acknowledged that "treatment outcomes may differ for youths in inpatient" settings.

The sample was heterogenous, comprising Anglo Americans and Hispanic-Americans, with the majority being Anglo. As a result, the study may have limited external validity when it comes to applying findings to un-acculturated Hispanics, African-Americans, Native Americans, or other minority groups. The groups were comparable in terms of primary substance use (marijuana), but further research is needed to identify effective treatment modalities, given that several modalities were employed simultaneously in this study. The Hawthorne effect was not considered, as there was no mention of it in the study.

The reliability of this study is hampered by its inclusion of multiple modalities, and its validity is constrained to a narrow window in terms of substance type and sample composition. Additionally, the number of treatment sessions per case was not equal across participants, which limits construct validity and temporal validity. That said, the inferences drawn appear appropriate and well-associated, so the study does retain some construct validity.

The primary limitation of this study's external validity is that it examines only outpatient treatment within a sample that is predominantly Anglo and English-speaking Hispanic-American, which limits its generalizability to other demographic groups. Its chief strength is its focus on a single substance, which tightens the analytic frame. However, the inclusion of multiple treatment modalities reduces the precision of any generalizations that might otherwise be drawn.

Article 5: Williams, Chang, and Addiction Centre Adolescent Research Group (2000)

This study is a comprehensive and comparative review of studies on treatment provided to adolescents with substance use disorder. The sampling probability is systematic in nature, producing a tightly focused set of treatment-related outcomes. The study found that most available research is methodologically weak. Those studies with stronger methodology consistently indicate that the majority of treated adolescents show decreased substance use and a concurrent reduction in life problems in the year immediately following the end of treatment.

Across studies, the average rate of sustained non-substance use over a 6-month period was 38% (range: 30–55%), dropping slightly to 32% over a 12-month period (range: 14–47%). Variables most consistently associated with positive treatment outcomes include: completion of treatment, a lower level of substance use prior to treatment (heavier use being harder to treat), and the support structure of friends, family, and peers in keeping the patient away from substances. Overall, the reviewed studies suggest that any treatment is better than no treatment, but the evidence does not support a sufficient comparison between treatment types to determine which is superior. One exception is that outpatient family therapy appears to be the most effective form of outpatient treatment, because it incorporates both the patient and the support system — both of which are crucial to positive outcomes. The internal validity of this systematic review is strong, as it maintains consistency in its analysis of variables across studies.

Methodology across the included studies is inconsistent: time frames are not standardized, with some evaluations occurring at the end of treatment and others being conducted up to 6 years post-treatment. Typical evaluations occur at 6-month and 12-month post-treatment intervals. The definition of "success" varies from study to study, and methods of reporting success are often weak — completion of a treatment phase alone is frequently treated as a marker of success. The reliability of this systematic review lies in its breadth: there is temporal validity in its examination of 48 studies spanning multiple years; reliability validity in its function as a systematic review; criterion validity in its well-defined evaluative criteria; and construct validity in its comprehensive qualitative analysis.

The strength of this study's external validity is its capacity to generate generalizable inferences about the state of the literature on adolescent substance abuse treatment: greater consistency and stronger methodology are urgently needed, particularly in how studies record and report outcomes, so that success rates can be measured and understood more accurately.

Conclusion

These studies were chosen using keyword searches in various online databases and then selected for relevance to the topic of inpatient versus outpatient outcomes in the treatment of adolescent substance abusers. Not all of the studies addressed this comparison directly — most examined outpatient outcomes only — but those studies were still useful in illuminating methods, approaches, and criteria used to gauge results.

The most robust studies demonstrated that the majority of literature in this field suffers from weak methodologies that do not utilize consistent time frames, treatment phases, or modalities. They also suffer from a lack of precision in defining terms such as "success." With roughly half of studies in this field defining success as abstinence and the other half defining it as a reduction in use, the field is fundamentally divided on what even constitutes a successful outcome in the treatment of adolescent substance abusers. It is therefore essential, going forward, to establish and maintain a precise, well-reasoned definition of success — one that is adequate to the overall purpose of the research and that can meaningfully inform treatment providers working in this field.

References

Branson, C., Clemmey, P., & Mukherjee, P. (2013). Text message reminders to improve outpatient therapy attendance among adolescents: A pilot study. Psychological Services, 10(3), 298–303.

Esposito-Smythers, C., Spirito, A., Kahler, C., Hunt, J., & Monti, P. (2011). Treatment of co-occurring substance abuse and suicidality among adolescents: A randomized trial. Journal of Consulting and Clinical Psychology, 79(6), 728–739.

Friedman, A., Terras, A., & Ali, A. (1998). Differences in characteristics of adolescent drug abuse clients that predict to improvement: For inpatient treatment versus outpatient treatment. Journal of Child and Adolescent Substance Abuse, 7(3), 97–119.

Waldron, H., Slesnick, N., & Brody, J. (2001). Treatment outcomes for adolescent substance abuse at 4- and 7-month assessments. Journal of Consulting and Clinical Psychology, 69(5), 802–813.

Williams, R., Chang, S., & Addiction Centre Adolescent Research Group. (2000). A comprehensive and comparative review of adolescent substance abuse treatment outcome. Clinical Psychology: Science and Practice, 7(2), 138–166.