Comparing Two Training Methods for Opioid Wizard

The prevalence of opioid use disorder (OUD) and opioid-related deaths has risen dramatically in recent years. Effective treatments, including medications for opioid use disorder (MOUDs; e.g., buprenorphine-naloxone and methadone) are under-utilized. There are few evidence-based interventions for changing attitudes toward Opioid Use Disorder (OUD) in the general public and especially among healthcare clinicians. This study proposed an innovative intervention to change attitudes of Primary Care Clinicians (PCCs) toward persons with OUD. Study participants were stratified into one of two online learning courses: the intervention training was compared with an attention-control training.

Primary care offers an ideal setting in which to treat OUD; however, few clinicians are waivered to prescribe buprenorphine and of those who are waivered, less than one-third do prescribe. One potential barrier to increasing access to MOUDs are primary care clinician (PCC) attitudes towards people with OUD. This study used a randomized controlled trial design to examine a novel intervention to change attitudes towards people with OUD among PCCs. PCCs in clinics randomized to the intervention in COMPUTE 2.0 (PCCs in 15 of the 30 clinics randomized in the parent study) were randomized 1:1 to the intervention or comparison training, stratified by clinic and waiver status. Training was conducted via MyLearning, an online learning software. All PCCs were asked to complete a brief training (25-35 minutes for both trainings) on the shared decision-making tool (SDM) called the Opioid Wizard, which alerts PCCs to screen, diagnose, and treat people with OUD. PCCs in the intervention arm heard patient narratives designed to shift attitudes about patients with OUD. PCCs in the comparison training had training on using the SDM tool. The PCCs were asked to complete a survey of attitudes and intentions to get waivered to prescribe buprenorphine immediately following the online training. Use of the CDS was monitored in both groups for 6 months.

This study used a randomized controlled trial (RCT) design embedded in a larger, multisite trial of a clinical decision support (CDS) tool to help PCCs identify, diagnose, and treat patients with OUD. PCCs in clinics randomized to Opioid Wizard were randomized 1:1 to the intervention or comparison training and stratified by clinic.

Methods for concealment of allocation. PCCs were blind to their randomized MyLearning training assignment. The study statistician generated the randomization assignment, and a study team member submitted that list to HealthPartners (HP) MyLearning staff. Study team members collecting outcome data from surveys, Opioid Wizard, and Epic (including use rates, buprenorphine prescriptions) also were blinded to treatment assignment.

This arm received a case-based training on how to use the Opioid Wizard tool, including patient narratives and videos and person-first language.

This intervention consisted of two different interactive online trainings delivered through an online learning platform at HealthPartners.

Attitudes toward people with OUD were measured using the Difference, Disdain, and Blame scales, which measure attitudes toward people with mental illness and substance use disorders (SUDs). Difference, Disdain, and Blame scales measured stigma toward people with OUD, with three questions in each domain. Items are scored on a 9-point agreement scale. The range is 1-9; some items are reverse-scored. Items are averaged to obtain a score, with higher scores reflecting more stigma (range = 1-9). Evidence suggests that the scales demonstrate good internal consistency and are positively associated.

Non-waivered PCCs rated one question on their intention to get waivered to prescribe buprenorphine ("How likely are you to get waivered to prescribe buprenorphine in the next year?") on a five-point Likert-type scale ranging from 1 (I definitely will not) to 5 (I definitely will). Higher scores correspond to greater intention to get waivered.

Non-waivered PCCs rated one question on their intentions to prescribe buprenorphine if a waiver were no longer required ("If your patient with OUD requested buprenorphine in the next year and a waiver were no longer required, would you prescribe buprenorphine?") on a five-point Likert-type scale ranging from 1 (I definitely would not) to 5 (I definitely would). Higher numbers indicate greater likelihood to prescribe buprenorphine.

PCC Opioid Wizard use was defined as clicking within the tool, such as screening for OUD, making a diagnosis, providing a referral, prescribing a medication, printing patient education materials, or prescribing naloxone. This variable was dichotomized as whether the PCC ever clicked in the tool (1=yes; 0=no) in the 6 months following the training.

Inclusion Criteria: Primary Care Clinician: Had to be a family physician, general internist, adult-care non-obstetric nurse practitioner, or a physician assistant Had to practice at a study-eligible primary care clinic already involved in the main study intervention arm Exclusion Criteria: PCCs were ineligible to participate if they had fewer than 5 eligible patient encounters between Sep 2020 and Feb 2021.

As part of the HEAL initiative, the main study results, as described in the initial publication, will be shared on the HEAL Data Platform.

You will not need to log in in order to: browse the study metadata on the Discovery Page or read the pre-made tutorial notebooks in the "Example Analysis" tab. You will need to log in and obtain authorization (access) in order to: register your own study; access studies with controlled data perform analyses in workspaces; and download data files and file manifests. run interactive tutorial notebooks in workspaces