EHR 'Nudge' Prompts Ineffective for Increasing Statin Prescribing Rates

Mitesh Patel, MD

Often described as burdensome, new research from the University of Pennsylvania School of Medicine whether a slight alteration to a clinician’s electronic health record (EHR) could encourage them to prescribe statins to patients with a clear indication for use.

The study, which was designed as a 3-arm randomized clinical trial, examined the effect of adding active and passive prompts to a cardiologists’ EHR to improved guideline-directed statin prescribing in a cohort of more than 11,000 patients of 82 cardiologists from 16 medical centers.

"Active choice prompts are used commonly in electronic health records, but they often are not rigorously tested head-to-head against other approaches," said lead investigator, Mitesh Patel, MD, the director of Penn Medicine's Nudge Unit, in a statement. "By systematically testing these interventions we can build upon the approaches that do work and turn off the ones that don't."

To examine how passive choice or active choice decision support interventions might impact guideline-directed prescribing, investigators from the University of Penn designed a 3-arm, cluster randomized clinical trial including cardiologists from practices within the University of Penn Health System in Pennsylvania and New Jersey. Cardiologists included in the study were randomized to usual care or 1 of 2 interventions in the EHR—a passive choice using a non-interruptive alert or an active choice using an interrupted alert requiring action before the clinician can proceed.

The trial included a 6-month preintervention period lasting from March 24, 2018 through September 23, 2018 and a 6-month intervention period lasting from September 24, 2018 through March 23, 2019. All patients included in the study were candidates for statin prescribing by either the 2013 American College of Cardiology/American Heart Association (AHA/ACC) guidelines or the 2015 National Lipid Associations recommendations.

Investigators chose the change in percentage not patients prescribed statin therapy at an optimal dose as the primary outcome measure of the study. The secondary outcome measure chosen by investigators was the change in percentage of eligible patients prescribe any dose of a statin.

Of the 11,693 patients included in the study, 58% were men, 66% were white, 24% were black, and the cohort had a mean age of 63.8 (9.1) years. Additionally, the mean 10-year atherosclerotic cardiovascular disease (ASCVD) risk score was 15.4 (10.0) and 68% of patients had a clinical diagnosis of ASCVD.

At baseline, prescribing rates at the optimal dose were 40.3% in the control arm, 39.1% in the passive choice arm and 41.2% in the active choice arm. Upon analysis, investigators found no significant difference in change in statin prescribing rates at optimal dose from control for passive choice (adjusted difference in percentage points, 0.2; 95% CI, -2.9 to 2.8; P=.86) or active choice (adjusted difference in percentage points, 2.4; 95% CI, -0.6 to 5.0; P=.08).

When assessing a subset of patients with clinical diagnosis of ASCVD, results suggested the active choice intervention was associated with a significant increase in statin prescribing at optimal dose when compared to the control arm (adjusted difference in percentage points, 3.8; 95% CI, 1.0-6.4; P=.008). Investigators noted no other subset analyses produced significant results and there were no significant changes in statin prescribing at any dose for either intervention.

In an invited commentary published in JAMA Cardiology, Thomas Maddox, MD, MSc, reflects on the struggles associated with EHRs and how, if not designed properly, this can result in negative outcomes for the patients or clinicians.

“One of the major lessons from this study is the importance of user-centered design. Also known as human-centered design, this concept involves obtaining deep understanding of those who use CDS and integrating their needs, perspective, and environment into the CDS design.This concept has been successfully deployed in a wide variety of industries over the years but has only recently achieved traction in health care,” wrote Maddox.

This study, “Effect of Passive Choice and Active Choice Interventions in the Electronic Health Record to Cardiologists on Statin Prescribing: A Cluster Randomized Clinical Trial,” was published in JAMA Cardiology.