Reducing High-Risk Geriatric Polypharmacy via EHR Nudges

Project: Research project

Project Details

Description

Overall goal: Devise and test EHR-mediated behavioral nudges to reduce rates of inappropriate polypharmacy among elderly patients in the primary care setting

Specific Aims
R21/Planning phase (project years 1-2)
1. Develop and pilot one or more behavioral nudges to improve prescriber decision making in clinical partner EHRs. The exact set of nudges will be determined in close partnership with the clinical partners.

2. Operationalize measures of polypharmacy in the EHRs and other datasets of the clinical partners. These datasets will be supplemented by versions of the measures calculated from Medicare claims data, to capture any prescription fills and other medical utilization not directly observed by the clinical partners.

R33/Implementation phase (project years 3-5)
1. To refine the nudges within the Northwestern EHR (as needed) and implement the nudges and polypharmacy measures within the Sutter Health Epic EHR.

2. To evaluate the effects of each EHR-based nudge on the measures of polypharmacy, via randomized controlled trial. This will be a cluster randomization in which primary care clinics are randomized to no intervention, one intervention, or more than one intervention (using a factorial design, if multiple interventions are developed). Prior to randomization, clinicians will be invited to participate, and those who consent will be enrolled. The trial intervention(s) will run for 18 months, followed by 12 months of subsequent observation of effects on measures of polypharmacy, to assess persistence of effects.

3. To assess qualitatively the acceptability of the EHR-based nudges to clinicians and to describe the effects of these nudges on clinician workflow.
StatusFinished
Effective start/end date9/30/178/31/19

Funding

  • RAND Corporation (9920180010//R21AG057396)
  • National Institute on Aging (9920180010//R21AG057396)

Fingerprint Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.