MIDAS Cluster Randomized Controlled Trial of Implementation Strategies to Optimize Use of Medications in VA Clinical Settings

Background The adoption and sustainment of evidence-based practices (EBPs) is a challenge within many healthcare systems, especially in settings that have already strived but failed to achieve longer-term goals. The Veterans Affairs (VA) Maintaining Implementation through Dynamic Adaptations (MIDAS) Quality Enhancement Research Initiative (QUERI) program was funded as a series of trials to test multi-component implementation strategies to sustain optimal use of three EBPs: 1) a deprescribing approach intended to reduce potentially inappropriate polypharmacy; 2) appropriate dosing and drug selection of direct-acting anticoagulant medications (DOACs); and 3) use of cognitive behavioral therapy as first-line treatment for insomnia before pharmacologic treatment. We describe the design and methods for a harmonized series of cluster-randomized control trials comparing two implementation strategies. Methods For each trial, we will recruit 8-12 clinics (24-36 total). All will have access to a clinical dashboard that flags patients who may benefit from the target EBP at that clinic and provider. For each trial, clinics will be randomized to one of two implementation strategies to improve use of the EBPs: 1) individual-level academic detailing (AD); or 2) AD plus the team-based Learn. Engage. Act. Process. (LEAP) quality improvement (QI) learning program. The primary outcomes will be operationalized across the three trials as a patient-level dichotomous response (yes/no) indicating patients with potentially inappropriate medications (PIMs) among those who may benefit from the EBP. This outcome will be computed using month-by-month administrative data. Primary comparison between the two implementation strategies will be analyzed using generalized estimating equations (GEE) with clinic-level monthly percent of PIMs as the dependent variable. Primary comparative endpoint will be at 13-18 months post-baseline. Each trial will also be analyzed independently. Discussion MIDAS QUERI trials will focus on fostering sustained use of EBPs that previously had targeted but incomplete implementation. Our implementation approaches are designed to engage frontline clinicians in a dynamic optimization process that integrates use of actionable clinical dashboard data and making incremental changes, designed to be feasible within busy clinical settings.