Measurement Error-Corrected Estimation of Clinically Significant Change Trajectories for Interventions Targeting Comorbid PTSD and Substance Use Disorders in OEF/OIF Veterans
Abstract: In randomized control trials (RCTs), a focus on average differences between treatment arms often limits our understanding of whether individuals show clinically significant improvement or deterioration. The present study examined differences in individual-level clinical significance trajectories between Concurrent Treatment of PTSD and Substance Use Disorders Using Prolonged Exposure (COPE) and Relapse Prevention (RP). Eighty-one treatment-seeking veterans with a comorbid PTSD/SUD diagnosis were randomized to COPE or RP; data from an additional n = 48 patients who did not meet criteria for both disorders was used to establish a normative threshold. A newly developed, modernized approach to the Jacobson and Truax (1991) clinically significant change framework, using (a) moderated nonlinear factor analysis (MNLFA) scale scoring and (b) measurement error-corrected multilevel modeling (MEC-MLM) was used; this approach was compared to other approaches using conventional total scores and/or assuming no measurement error. Using a conventional approach to estimating the Reliable Change Index (RCI) yielded no differences between COPE and RP in the percentage of patients achieving statistically significant improvement (SSI; 88.9% for both groups). However, under MNLFA/MEC-MLM, higher percentages of patients receiving COPE (75.0%) achieved SSI compared to RP (40.7%). Findings suggest that, even though COPE and RP appear to reduce the same number of PTSD symptoms, MNLFA scoring of outcome measures gives greater weight to interventions that target and reduce “hallmark” PTSD symptoms.
Abstract: Novel and automated means of opioid use and relapse risk detection are needed. Unstructured electronic medical record data, including written progress notes, can be mined for clinically relevant information, including the presence of substance use and relapse-critical markers of risk and recovery from opioid use disorder (OUD). In this study, we used natural language processing (NLP) to automate the extraction of opioid relapses, and the timing of these occurrences, from veteran patients' electronic medical record. We then demonstrated the utility of our NLP tool via analysis of pre-/post-COVID-19 opioid relapse trends among veterans with OUD. For this demonstration, we analyzed data from 107,606 veterans OUD enrolled in Veterans Health Administration, comparing a pandemic-exposed cohort (n = 53,803; January 2019-March 2021) to a matched prepandemic cohort (n = 53,803; October 2017-December 2019). The recall of our NLP tool was 75% and our precision was 94%, demonstrating moderate sensitivity and excellent specificity. Using the NLP tool, we found that the odds of opioid relapse postpandemic onset were proportionally higher compared to prepandemic trends, despite patients having fewer mental health encounters from which to derive instances of relapse postpandemic onset. In this research application of the tool, and as hypothesized, we found that opioid relapse risk was elevated postpandemic. The application of NLP Methods: to identify and monitor relapse risk holds promise for future surveillance, risk prevention, and clinical outcome research.