Dr. Marc Scott joined lead author Ignacio Bórquez Infante and colleagues at NYU's Center for Opioid Epidemiology & Policy Team, as well as at Columbia University and Brandeis University, in a study that examined dispensing practices among opioid treatment programs in the US using sequence analysis.
The article, "State sequence analysis of daily methadone dispensing trajectories among individuals at United States opioid treatment programs before and following COVID-19 onset" was published in the journal Addiction and discusses how COVID-19-related regulatory, patient (e.g., demographic, substance use, among others) and clinic characteristics shape treatment regimes (e.g., methadone take‐home doses). State sequence analysis (SSA) often involves unsupervised machine learning to organize nominal longitudinal repeated measures in a manner that reveals common structures. This analysis employed an SSA technique known as discrepancy analysis, which is analogous to analysis of variance (ANOVA) in this domain. SSA presents a vital opportunity to address essential healthcare-related questions, such as the trajectories of utilization or pharmacoepidemiology, particularly with the increasing availability of real-world data, including electronic health records, administrative, and claims databases. The study concluded that variability in dispensing trajectories was primarily driven by clinic site practices rather than individual patient characteristics or treatment responses, both before and after the COVID-19 regulatory changes.