The Role of Trajectories of Strength Changes in Predicting SUD Treatment Completion

Background: In the United States, an estimated 40.7 million people aged 12 or older struggle with illicit drug use (nearly 1 in 12) (SAMHSA, 2022). To address this challenge, proactive treatment for substance use disorder (SUD) and successful program completion are crucial. Studies consistently show that completing treatment programs leads to positive outcomes such as longer abstinence periods, reduced relapse and overdose risks, and lower recidivism rates (Brorson et al., 2013; Gottlieb et al., 2022). Unfortunately, high dropout rates remain a persistent issue. This study aims to bridge this knowledge gap by examining the trajectories of strength changes, characteristics, and behavioral health needs associated with successful SUD treatment completion among young adults with SUD.

Methods. This study investigated how the development of strengths (i.e., family strength, social connectedness, optimism, talents/interests, education, volunteering, job history, spiritual/religious beliefs, community involvement, natural supports, and resiliency) in young adults (aged 18-25) predict substance use disorder (SUD) treatment outcomes over 24 months. We analyzed statewide, publicly funded behavioral health program data, including demographics, strengths in Adult Needs and Strengths Assessments (ANSA), and national outcome measures. The focus was on participants enrolled in programs during the state fiscal year 2021 whose program participation ended within the study timeframe. From a total sample of 3,699 participants, the findings revealed that 92% (n=2,909) engaged in substance use disorder (SUD) treatment within the 24-month follow-up period.

Analysis. Descriptive statistics were initially used to assess differences between the dropout and SUD treatment completion groups. Subsequently, the Chi-square Automatic Interaction Detection (CHAID) approach was implemented to develop treatment completion algorithms. This approach considered the trajectories of strength changes (i.e., improved, remained negative, or worsened), behavioral health needs improvement, and individual-level variables. Variables such as gender, race/ethnicity, living location, living arrangement, primary age of 1^st substance use, legal system involvement, type of substances used, and ANSA twelve-strength trajectories were all taken into account. All analyses were conducted using IBM SPSS Modeler version 18.4.

Results. Findings revealed that improved behavioral health (indicated by a reduced number of behavioral health needs) emerged as the strongest predictor of successful SUD treatment completion (χ² _{(4, 2296)} = 289.16, p = 0.00). This was followed by strength-related factors showing development or positive change, including social connectedness, family strength, job history, and optimism. Other factors, such as legal system involvement, residency (rural vs. urban), and the presence of opioid use disorder, were also considered significant predictors. The CHAID algorithm achieved an area under the ROC curve (AUC) value of 0.80.

Conclusion and Implications: This study identified well-defined variables that can be used to build a predictive model for young adults based on their development or change in strength-related factors. This model has the potential to guide us in identifying the specific needs and support areas that contribute to successful SUD treatment outcomes in young adults. Ultimately, it could serve as a foundation for further in-depth analyses of behavioral health services, such as exploring effective programs that can divert young adults with SUD toward positive treatment outcomes.