Abstract
OBJECTIVE: Clinical trials for substance use disorder (SUD) often assess outcomes by aggregating substance use data into weekly proportions. However, daily substance use patterns may reveal how motivation changes during treatment. Neurocognitive principles of recency and expectancy indicate the salience of treatment mechanisms would increase on days proximal to therapy sessions. This study tested whether substance use decreased on days near treatment sessions. METHOD: Bayesian multilevel models were used in a secondary analysis of randomized clinical trial data comparing cognitive behavioral therapy (CBT), computerized CBT, and treatment as usual during outpatient SUD treatment (n = 94; 76% males; M(age) = 38; 46% African American, 38% White, 6% multiracial, 10% other; 17% Hispanic). The number of days before/after a therapy session was used to predict daily substance use assessed by weekly self-reports. RESULTS: The models suggested that substance use increased as more days passed after a therapy session (b = 5.23) and then decelerated before the next therapy session (b² = -8.20). The evidence indicated that substance use was less likely on therapy days and the days after therapy. Primary drug type, SUD severity, and treatment condition moderated these findings. CONCLUSIONS: Substance use patterns during treatment were consistent with recency and expectancy neurocognitive principles. Substance use decreased on days closer to a therapy session, suggesting that treatment mechanisms were more salient on those days. More frequent therapeutic contact and targeted timing of treatment delivery may enhance treatment efficacy. (PsycInfo Database Record (c) 2023 APA, all rights reserved).