Discontinuation of Buprenorphine Treatment for Opioid Use Disorder During the Coronavirus Disease-2019 Pandemic

Saloner, Brendan; Chang, Hsien Yen; Alexander, G. Caleb; Stein, Bradley D.

doi:10.1097/mlr.0000000000001802

Cited by 4 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Accordingly, identifying patients at risk of premature discontinuation of buprenorphine can help healthcare providers develop patient support systems that improve treatment retention rates (Saloner, 2023). Towards this goal, several studies have sought to identify patient risk factors for buprenorphine discontinuation using either conventional statistical approaches or machine learning procedures (Samples, 2018; Vakkalanka, 2022).…”

Section: Introductionmentioning

confidence: 99%

An Explainable Machine Learning Framework for Predicting the Risk of Buprenorphine Treatment Discontinuation for Opioid Use Disorder

Faysal,

Noor-E-Alam,

Young

et al. 2023

Preprint

View full text Add to dashboard Cite

ObjectivesBuprenorphine is an effective evidence-based medication for opioid use disorder (OUD). Yet premature discontinuation undermines treatment effectiveness, increasing risk of mortality and overdose. We developed and evaluated a machine learning (ML) framework for predicting buprenorphine care discontinuity within 1-year following treatment initiation.MethodsThis retrospective study used United States 2018-2021 MarketScan commercial claims data of insured individuals aged 18-64 who initiated buprenorphine between July 2018 and December 2020 with no buprenorphine prescriptions in the previous six months. We measured buprenorphine prescription discontinuation gaps of ≥30 days within the first year of initiating treatment. We developed predictive models employing logistic regression, decision tree classifier, random forest, XGBoost, Adaboost, and random forest-XGBoost ensemble. We applied recursive feature elimination with cross-validation to reduce dimensionality and identify the most predictive features while maintaining model robustness. We focused on two distinct treatment stages: at the time of treatment initiation and one and three months after treatment initiation. We employed SHapley Additive exPlanations (SHAP) analysis that helped us explain the contributions of different features in predicting buprenorphine discontinuation. We stratified patients into risk subgroups based on their predicted likelihood of treatment discontinuation, dividing them into decile subgroups. Additionally, we used a calibration plot to analyze the reliability of the models.ResultsA total of 30,373 patients initiated buprenorphine and 14.98% (4,551) discontinued treatment. C-statistic varied between 0.56 and 0.76 for the first-stage models including patient-level demographic and clinical variables. Inclusion of proportion of days covered (PDC) measured at one-month and three-month following treatment initiation significantly increased the models’ discriminative power (C-statistics: 0.60 to 0.82). Random forest (C-statistics: 0.76, 0.79 and 0.82 with baseline predictors, one-month PDC and three-month PDC, respectively) outperformed other ML models in discriminative performance in all stages (C-statistics: 0.56 to 0.77). Most influential risk factors of discontinuation included early stage medication adherence, age, and initial days of supply.ConclusionML algorithms demonstrated a good discriminative power in identifying patients at higher risk of buprenorphine care discontinuity. The proposed framework may help healthcare providers optimize treatment strategies and deliver targeted interventions to improve buprenorphine care continuity.

show abstract

Section: Introductionmentioning

confidence: 99%

An Explainable Machine Learning Framework for Predicting the Risk of Buprenorphine Treatment Discontinuation for Opioid Use Disorder

Faysal,

Noor-E-Alam,

Young

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…discontinuation of buprenorphine treatment in the immediate months after the pandemic began. 10 With the increasing deaths due to opioid overdose, as well as the access barriers and additional stress of a global pandemic, innovative solutions are needed to systematically restructure MOUD in the contemporary health care system.…”

mentioning

confidence: 99%

“…Specifically, while there was a leveling of opioid overdose deaths from 2017 to 2019, this was followed by stark increases in deaths in 2020 (68,630 deaths) and 2021 (80,411 deaths), demonstrating a large, unmet need for MOUD. In addition, the COVID-19 pandemic has compounded an already stressed health care system, as demonstrated by the increased discontinuation of buprenorphine treatment in the immediate months after the pandemic began 10 . With the increasing deaths due to opioid overdose, as well as the access barriers and additional stress of a global pandemic, innovative solutions are needed to systematically restructure MOUD in the contemporary health care system.…”

mentioning

confidence: 99%

(RE-)AIMing for Rapid Uptake

Heavey,

Beehler,

Funderburk

2023

Medical Care

View full text Add to dashboard Cite

Objective: Medication for opioid use disorder (MOUD) is an effective, evidence-based treatment, but significant gaps in implementation remain. We evaluate one novel approach to address this gap: a Hub and Spoke model to increase buprenorphine access and management. Methods: This outcome evaluation was guided by the Reach, Effectiveness, Adoption, Implementation, and Maintenance framework using secondary data analysis of clinical and administrative data to characterize program outcomes for program Reach, Effectiveness, Adoption, and Maintenance. Implementation was assessed through a chart review of provider progress notes and through key informant interviews with program staff to understand why this site was able to introduce a novel approach to MOUD. Results: Nearly half of patients with opioid use disorder (45.48%, n=156) were reached by the program over 2 years. Of those, 91.67% had 1 or more program visits after an initial intake appointment, and 78.85% had a buprenorphine prescription. Patients in the program were 2.44 times more likely to have a buprenorphine prescription than those in comparator site that did not have a Hub and Spoke program (95% CI: 1.77–3.37; P<0.001). There was significantly greater program reach in year 1 than year 2, suggesting rapid initial uptake followed by modest program growth. Key informant interviews illustrated several themes regrading program implementation, including the importance of process champions, the beneficial impact of MOUD for patients, and addressing facility performance metrics. A supportive organizational culture and a receptive climate were also key factors for implementation. Conclusions: This program led to rapid improvement in MOUD uptake across the facility. Future efforts should focus on improving program maintenance, including supporting the exchange of patients from the hub to appropriate spokes.

show abstract

Associations Between State Policies Facilitating Telehealth and Buprenorphine Episode Initiation and Duration Early in the COVID Pandemic

Stein,

Saloner,

Sheng

et al. 2024

J GEN INTERN MED

View full text Add to dashboard Cite

Importance State policies facilitating telehealth implemented early in COVID may support buprenorphine treatment of opioid use disorder. However, little empirical information is available about those policies’ effects. Objective Examine association between state policies that may facilitate telehealth use and buprenorphine treatment. Design, Setting, Participants Retrospective cohort study using 2019–2020 national pharmacy data on dispensed buprenorphine prescriptions. Exposures State policies implemented after March 3, 2020, public health emergency declaration requiring private insurers’ telehealth reimbursement to be commensurate with in-person service reimbursement, authorizing Medicaid reimbursement for audio-only telehealth, allowing physicians to provide cross-state telehealth services, and allowing psychologists to provide cross-state telehealth services. Main Outcomes and Measures (a) Duration of treatment episodes started between March 1 and March 13 in both 2019 and 2020, and (b) daily numbers of new buprenorphine treatment episodes from March 13 through December 31 in each year. Key Results We found little change in the number of new buprenorphine treatment episodes started in 2020 compared to 2019 and an increase in treatment duration of 10.3 days (95%CI 8.3 to 12.2 days) for episodes started in March 2020 before the public health emergency declaration compared to the comparable 2019 period. States implementing a telehealth parity policy in 2020 had 7.3% (95%CI − 13.3% to − 0.4%) fewer new buprenorphine treatment episodes. States joining the psychologist interstate compact in 2020 after the public health emergency declaration had treatment episodes 7.97 days longer (95%CI 0.78 to 15.16) than other states. None of the other policies examined was associated with changes in new treatment episodes or treatment duration. Conclusions and Relevance Policies undertaken during the pandemic we examined were associated with few significant changes in buprenorphine treatment initiation and duration. Findings suggest realizing the benefits of telehealth and other policy changes for buprenorphine may require more extensive implementation and infrastructure support.

show abstract

Discontinuation of Buprenorphine Treatment for Opioid Use Disorder During the Coronavirus Disease-2019 Pandemic

Cited by 4 publications

References 24 publications

An Explainable Machine Learning Framework for Predicting the Risk of Buprenorphine Treatment Discontinuation for Opioid Use Disorder

An Explainable Machine Learning Framework for Predicting the Risk of Buprenorphine Treatment Discontinuation for Opioid Use Disorder

(RE-)AIMing for Rapid Uptake

Associations Between State Policies Facilitating Telehealth and Buprenorphine Episode Initiation and Duration Early in the COVID Pandemic

Contact Info

Product

Resources

About