Applying Machine Learning Models Derived From Administrative Claims Data to Predict Medication Nonadherence in Patients Self-Administering Biologic Medications for Inflammatory Bowel Disease.

Background: Adherence to self-administered biologic therapies is important to induce remission and prevent adverse clinical outcomes in Inflammatory bowel disease (IBD). This study aimed to use administrative claims data and machine learning methods to predict nonadherence in an academic medical center test population. Methods: A model-training dataset of beneficiaries with IBD and the first unique dispense of a self-administered biologic between June 30, 2016 and June 30, 2019 was extracted from the Commercial Claims and Encounters and Medicare Supplemental Administrative Claims Database. Known correlates of medication nonadherence were identified in the dataset. Nonadherence to biologic therapies was defined as a proportion of days covered ratio <80% at 1 year. A similar dataset was obtained from a tertiary academic medical center's electronic medical record data for use in model testing. A total of 48 machine learning models were trained and assessed utilizing the area under the receiver operating characteristic curve as the primary measure of predictive validity. Results: The training dataset included 6998 beneficiaries (n = 2680 nonadherent, 38.3%) while the testing dataset included 285 patients (n = 134 nonadherent, 47.0%). When applied to test data, the highest performing models had an area under the receiver operating characteristic curve of 0.55, indicating poor predictive performance. The majority of models trained had low sensitivity and high specificity. Conclusions: Administrative claims-trained models were unable to predict biologic medication nonadherence in patients with IBD. Future research may benefit from datasets with enriched demographic and clinical data in training predictive models.

Utilization of long-acting injectable monthly depot buprenorphine for opioid use disorder (OUD) in Kentucky, before and after COVID-19 related buprenorphine access policy changes.

INTRODUCTION: Long-acting injectable buprenorphine (LAI-bup) formulations have advantages over transmucosal buprenorphine (TM-bup), but barriers may limit their utilization. Several policies shifted during the COVID-19 pandemic to promote buprenorphine access. The federal government expanded telemedicine treatment for opioid use disorder and Kentucky (KY) Medicaid lifted prior authorization requirements (PAs) for LAI-bup (i.e., Sublocade®). This retrospective cohort study evaluated changes in LAI-bup access, utilization, and retention before and after these policy changes in KY. METHODS: Individual-level TM-bup and LAI-bup dispensing record data from KY's prescription drug monitoring program examined LAI-bup utilization and retention, without a >30-day gap in coverage, for patients starting a new episode of LAI-bup treatment. Two key time periods were examined: pre-policy changes (Apr 1, 2019 - Dec 31, 2019) and post-policy changes (Apr 1, 2020 - Dec 31, 2020). Data on PA requests among Medicaid managed care organizations and availability of LAI-bup Risk Evaluation and Mitigation Strategy (REMS)-certified pharmacies were also obtained. A multivariable Cox proportional hazard regression model analysis compared pre- versus post-policy period treatment discontinuation. RESULTS: The number of patients initiating LAI-bup increased from 211 to 481 over the two periods. By the end of the post-policy period, 24.3 % of eligible patients were retained on LAI-bup, versus 12.5 % in the pre-policy change period. The adjusted hazard ratio, comparing discontinuation during the post- versus pre-policy change periods, was 0.70 (95 % confidence interval: 0.55-0.89). There were also more REMS-certified pharmacies and providers in the post-policy change period. CONCLUSIONS: LAI-bup access, utilization, and retention increased after several policy changes.

Barriers to Buprenorphine Dispensing by Medicaid-Participating Community Retail Pharmacies.

Importance: Controlled substances have regulatory requirements under the US Federal Controlled Substance Act that must be met before pharmacies can stock and dispense them. However, emerging evidence suggests there are pharmacy-level barriers in access to buprenorphine for treatment for opioid use disorder even among pharmacies that dispense other opioids. Objective: To estimate the proportion of Medicaid-participating community retail pharmacies that dispense buprenorphine, out of Medicaid-participating community retail pharmacies that dispense other opioids and assess if the proportion dispensing buprenorphine varies by Medicaid patient volume or rural-urban location. Design, Setting, and Participants: This serial cross-sectional study included Medicaid pharmacy claims (2016-2019) data from 6 states (Kentucky, Maine, North Carolina, Pennsylvania, Virginia, West Virginia) participating in the Medicaid Outcomes Distributed Research Network (MODRN). Community retail pharmacies serving Medicaid-enrolled patients were included, mail-order pharmacies were excluded. Analyses were conducted from September 2022 to August 2023. Main Outcomes and Measures: The proportion of pharmacies dispensing buprenorphine approved for opioid use disorder among pharmacies dispensing an opioid analgesic or buprenorphine prescription to at least 1 Medicaid enrollee in each state. Pharmacies were categorized by median Medicaid patient volume (by state and year) and rurality (urban vs rural location according to zip code). Results: In 2016, 72.0% (95% CI, 70.9%-73.0%) of the 7038 pharmacies that dispensed opioids also dispensed buprenorphine to Medicaid enrollees, increasing to 80.4% (95% CI, 79.5%-81.3%) of 7437 pharmacies in 2019. States varied in the percent of pharmacies dispensing buprenorphine in Medicaid (range, 73.8%-96.4%), with significant differences between several states found in 2019 (χ2 P < .05), when states were most similar in the percent of pharmacies dispensing buprenorphine. A lower percent of pharmacies with Medicaid patient volume below the median dispensed buprenorphine (69.1% vs 91.7% in 2019), compared with pharmacies with above-median patient volume (χ2 P < .001). Conclusions and Relevance: In this serial cross-sectional study of Medicaid-participating pharmacies, buprenorphine was not accessible in up to 20% of community retail pharmacies, presenting pharmacy-level barriers to patients with Medicaid seeking buprenorphine treatment. That some pharmacies dispensed opioid analgesics but not buprenorphine suggests that factors other than compliance with the Controlled Substance Act influence pharmacy dispensing decisions.

Effect of the Communities that HEAL intervention on receipt of behavioral therapies for opioid use disorder: A cluster randomized wait-list controlled trial.

BACKGROUND: The U.S. opioid overdose crisis persists. Outpatient behavioral health services (BHS) are essential components of a comprehensive response to opioid use disorder and overdose fatalities. The Helping to End Addiction Long-Term® (HEALing) Communities Study developed the Communities That HEAL (CTH) intervention to reduce opioid overdose deaths in 67 communities in Kentucky, Ohio, New York, and Massachusetts through the implementation of evidence-based practices (EBPs), including BHS. This paper compares the rate of individuals receiving outpatient BHS in Wave 1 intervention communities (n = 34) to waitlisted Wave 2 communities (n = 33). METHODS: Medicaid data included individuals ≥18 years of age receiving any of five BHS categories: intensive outpatient, outpatient, case management, peer support, and case management or peer support. Negative binomial regression models estimated the rate of receiving each BHS for Wave 1 and Wave 2. Effect modification analyses evaluated changes in the effect of the CTH intervention between Wave 1 and Wave 2 by research site, rurality, age, sex, and race/ethnicity. RESULTS: No significant differences were detected between intervention and waitlisted communities in the rate of individuals receiving any of the five BHS categories. None of the interaction effects used to test the effect modification were significant. CONCLUSIONS: Several factors should be considered when interpreting results-no significant intervention effects were observed through Medicaid claims data, the best available data source but limited in terms of capturing individuals reached by the intervention. Also, the 12-month evaluation window may have been too brief to see improved outcomes considering the time required to stand-up BHS. TRIAL REGISTRATION: Clinical Trials.gov http://www. CLINICALTRIALS: gov: Identifier: NCT04111939.

Neurological diagnoses in hospitalized COVID-19 patients associated with adverse outcomes: A multinational cohort study.

Few studies examining the patient outcomes of concurrent neurological manifestations during acute COVID-19 leveraged multinational cohorts of adults and children or distinguished between central and peripheral nervous system (CNS vs. PNS) involvement. Using a federated multinational network in which local clinicians and informatics experts curated the electronic health records data, we evaluated the risk of prolonged hospitalization and mortality in hospitalized COVID-19 patients from 21 healthcare systems across 7 countries. For adults, we used a federated learning approach whereby we ran Cox proportional hazard models locally at each healthcare system and performed a meta-analysis on the aggregated results to estimate the overall risk of adverse outcomes across our geographically diverse populations. For children, we reported descriptive statistics separately due to their low frequency of neurological involvement and poor outcomes. Among the 106,229 hospitalized COVID-19 patients (104,031 patients ≥18 years; 2,198 patients <18 years, January 2020-October 2021), 15,101 (14%) had at least one CNS diagnosis, while 2,788 (3%) had at least one PNS diagnosis. After controlling for demographics and pre-existing conditions, adults with CNS involvement had longer hospital stay (11 versus 6 days) and greater risk of (Hazard Ratio = 1.78) and faster time to death (12 versus 24 days) than patients with no neurological condition (NNC) during acute COVID-19 hospitalization. Adults with PNS involvement also had longer hospital stay but lower risk of mortality than the NNC group. Although children had a low frequency of neurological involvement during COVID-19 hospitalization, a substantially higher proportion of children with CNS involvement died compared to those with NNC (6% vs 1%). Overall, patients with concurrent CNS manifestation during acute COVID-19 hospitalization faced greater risks for adverse clinical outcomes than patients without any neurological diagnosis. Our global informatics framework using a federated approach (versus a centralized data collection approach) has utility for clinical discovery beyond COVID-19.

Impact of the Communities That HEAL Intervention on Buprenorphine-Waivered Practitioners and Buprenorphine Prescribing: A Prespecified Secondary Analysis of the HCS Randomized Clinical Trial.

Importance: Buprenorphine significantly reduces opioid-related overdose mortality. From 2002 to 2022, the Drug Addiction Treatment Act of 2000 (DATA 2000) required qualified practitioners to receive a waiver from the Drug Enforcement Agency to prescribe buprenorphine for treatment of opioid use disorder. During this period, waiver uptake among practitioners was modest; subsequent changes need to be examined. Objective: To determine whether the Communities That HEAL (CTH) intervention increased the rate of practitioners with DATA 2000 waivers and buprenorphine prescribing. Design, Setting, and Participants: This prespecified secondary analysis of the HEALing Communities Study, a multisite, 2-arm, parallel, community-level, cluster randomized, open, wait-list-controlled comparison clinical trial was designed to assess the effectiveness of the CTH intervention and was conducted between January 1, 2020, to December 31, 2023, in 67 communities in Kentucky, Massachusetts, New York, and Ohio, accounting for approximately 8.2 million adults. The participants in this trial were communities consisting of counties (n = 48) and municipalities (n = 19). Trial arm randomization was conducted using a covariate constrained randomization procedure stratified by state. Each state was balanced by community characteristics including urban/rural classification, fatal opioid overdose rate, and community population. Thirty-four communities were randomized to the intervention and 33 to wait-list control arms. Data analysis was conducted between March 20 and September 29, 2023, with a focus on the comparison period from July 1, 2021, to June 30, 2022. Intervention: Waiver trainings and other educational trainings were offered or supported by the HEALing Communities Study research sites in each state to help build practitioner capacity. Main Outcomes and Measures: The rate of practitioners with a DATA 2000 waiver (overall, and stratified by 30-, 100-, and 275-patient limits) per 100 000 adult residents aged 18 years or older during July 1, 2021, to June 30, 2022, were compared between the intervention and wait-list control communities. The rate of buprenorphine prescribing among those waivered practitioners was also compared between the intervention and wait-list control communities. Intention-to-treat and per-protocol analyses were performed. Results: A total of 8 166 963 individuals aged 18 years or older were residents of the 67 communities studied. There was no evidence of an effect of the CTH intervention on the adjusted rate of practitioners with a DATA 2000 waiver (adjusted relative rate [ARR], 1.04; 95% CI, 0.94-1.14) or the adjusted rate of practitioners with a DATA 2000 waiver who actively prescribed buprenorphine (ARR, 0.97; 95% CI, 0.86-1.10). Conclusions and Relevance: In this randomized clinical trial, the CTH intervention was not associated with increases in the rate of practitioners with a DATA 2000 waiver or buprenorphine prescribing among those waivered practitioners. Supporting practitioners to prescribe buprenorphine remains a critical yet challenging step in the continuum of care to treat opioid use disorder. Trial Registration: ClinicalTrials.gov Identifier: NCT04111939.