Study uses Real-World Data to describe important patterns in the treatment of opioid use disorder

Holmusk, a leading global data science and digital health company building the largest Real-World Evidence platform for behavioral health, today announced study results that could help identify which patients with opioid use disorder are at the highest risk for ending treatment early.

The findings were published in collaboration with researchers from JPS Health Network in Neuropsychiatric Disease and Treatment. The authors of the manuscript, entitled “Factors that Affect Patient Attrition in Buprenorphine Treatment for Opioid Use Disorder,” developed a predictive model for treatment dropout. The model predicts with high accuracy (90.5% precision and 75.2% recall) which patients are at risk for treatment dropout within three months of initiating treatment.

Published in collaboration with researchers from JPS Health Network, the manuscript assessed factors associated with treatment retention for buprenorphine among patients with opioid use disorder (OUD). Leveraging NeuroBlu, Holmusk’s Real-World Evidence platform for behavioral health, the authors developed a predictive model for treatment dropout within three months of initiation with high accuracy (90.5% precision and 75.2% recall). 

Buprenorphine is typically prescribed as part of medication-assisted treatment (MAT) to help people reduce or quit use of opiates, and use of BUP has been found to improve clinical outcomes including harm reduction, treatment retention, and decreased risk of illicit drugs. However, premature treatment discontinuation remains a major obstacle in improving outcomes across OUD patient populations.

“Buprenorphine is an effective treatment for OUD, but it is well known that patients have problems staying in treatment,” commented Miguel Rentería, PhD (Principal DataScientist, Holmusk), an author on this study. “To our knowledge, this is the largest observational study using Real-World Data to elucidate factors associated with use of buprenorphine for OUD. This study, which focuses on a critical public health issue, could be used to inform new strategies to improve outcomes for patients."

This study found that a high proportion of OUD patients who start buprenorphine treatment drop out of treatment early, especially those who start treatment while being hospitalized; for this group, only 16% remained in treatment after three months. Several factors were associated with higher treatment retention, including initiating treatment as an outpatient, having higher functioning and less severe condition at the start of treatment, initiating at a higher buprenorphine dose, and being on fewer prescriptions. This study leveraged clinical data from NeuroBlu, the industry-leading behavioral health clinical dataset with over 20 years of data on 560,000+ patients and 20+ million encounters.

“There is a dearth of evidence regarding the factors that contribute to buprenorphine treatment retention and dropout,”shared Dustin DeMoss, DO (Department of Psychiatry, JPS Health Network). “Predicting ahead of time which patients will do better or worse in treatment can help allocate resources and optimize outcomes to help address the ongoing opioid crisis."

Full text available here: https://www.dovepress.com/articles.php?article_id=70097

Citation information: Ker S, Hsu J, Balani A, Mukherjee SS, Rush AJ, KhanM, Eichehabi S, Huffhines S, DeMoss D, Rentería ME, Sarkar J. (2021). Factors That Affect Patient Attrition in Buprenorphine Treatment for Opioid Use Disorder: A Retrospective Real-World Study Using Electronic Health Records. Neuropsychiatric Disease and Treatment. 17:3229-3244. https://doi.org/10.2147/NDT.S331442