Peer Pressure May Push Physicians to Do Better at Prescribing Statins
Tools that enable doctors to see which patients need statins, and know how well they match up to their colleagues, could make a difference.
The best motivator for encouraging physicians to prescribe necessary statins could be a bit of healthy competition, a new study suggests.
In the clinical trial, called Pragmatic Evaluation of Statin Active Choice to Reach Improved Outcomes Based on Evidence (PRESCRIBE), researchers found that providers who were reminded how their statin prescription rates compared with those of their peers were more likely to prescribe the medication to other eligible patients.
Despite their proven role in primary and secondary prevention of cardiovascular events, statins are only prescribed in roughly half of patients who qualify. “What we wanted to do was design an experiment to see if we could nudge clinicians—primary care physicians—to prescribe a statin,” said Mitesh Patel, MD (University of Pennsylvania, Philadelphia), lead author of the study.
To do so, they created a “dashboard” that listed all of a physician’s patients who met national guidelines for a statin but did not yet have a prescription. They also sent some physicians emails on how their average statin prescription rates compared with those of their peers.
For their study published on July 27, 2018, in JAMA Network Open, the researchers randomly selected 96 physicians from 32 different practices within the University of Pennsylvania Health System (UPHS) for the study. They then evaluated whether these clinicians’ patients met the 2013 American College of Cardiology/American Heart Association (ACC/AHA) guidelines for statin therapy: any form of atherosclerotic cardiovascular disease, an LDL cholesterol level of 190 mg/dL, and, for patients between 40 and 75 years old, either diabetes with an LDL cholesterol level between 70 and 189 mg/dL or an estimated 10-year risk score of 7.5% or greater. Patients were included if they met these guidelines but, according to their UPHS electronic health record (EHR), had never been prescribed a statin.
The cohort ultimately included 4,774 patients with a mean age of 62. Approximately two-thirds of study participants were white and 27.6% were black.
Physicians in the study were split up into three groups. One was a control group providing usual care. Two of the groups received an email link with access to a secure automated online dashboard listing of all of their patients who met the 2013 ACC/AHA guidelines for statin therapy but had no record of a prescription.
In the intervention groups, doctors could either select a generic prescription for all patients on the list (20 mg of atorvastatin once daily) or go through the list individually, selecting specific statin prescriptions or choosing not to prescribe the medication. These prescriptions were then sent to a research coordinator, who would input them into the EHR for a provider to sign.
“By offering multiple options to prescribe a guideline-indicated statin and requiring a reason to say no, clinicians may be nudged toward prescribing a statin,” the authors say.
On top of this process, one of those two intervention groups received added incentive: being informed of how well their statin patterns met national guidelines compared with their peers, with top performers getting extra praise.
Extra Motivation Only Goes So Far
Over a 2-month period in 2017, the dashboard plus peer-comparison group had a 5.8% higher prescription rate compared with the control group (P = 0.008). Physicians who didn’t know how well they stacked up against their peers didn’t do any better than usual care.
While fewer physicians in the dashboard plus peer-comparison group visited the dashboard (12 out of 32), eight physicians submitted statin prescriptions through the tool and later signed them in the electronic medical record. In this group, 8% of patients received a statin prescription by the end of the trial.
In the dashboard-only group, half of 32 physicians accessed the interface. Of the physicians who did so, only four ordered prescriptions through the tool and just two later signed these prescriptions in the EHR. By the end of the 2-month trial, 6.7% of patients in that group got a statin.
In the control group, by comparison, just 2.6% of patients got a statin over the course of the study.
This study hits on a question that health systems around the country are grappling with, Patel told TCTMD. While these types of population health tools are readily available for physicians for a variety of conditions, “the challenge is actually getting physicians to use these dashboards,” he said. This study shows that access to the tools may not be enough on its own.
“I think it was the combination of the active-choice dashboard, which enabled them to make decisions, [plus] peer-comparison feedback that really nudged clinicians to use the dashboard more,” he added.
EHR Integration Necessary
Patel said the difference between groups is meaningful but that there is room for improvement.
“Despite the differences in statin prescription states in our study, physician engagement with the dashboard was low,” the researchers say. This in part could be attributed to the study design. For example, the emails and reminders sent to physicians came from the clinical trial team, rather than more familiar sources like practice managers, they write. Additionally, this dashboard was not embedded in the EHR, which could explain why the impact of the intervention was “modest,” the authors say.
The PRESCRIBE team is currently designing another trial that will be integrated into the UPHS electronic health record. It will also nudge patients to discuss prescription options with their primary care physicians, Patel said.
Despite the stumbling blocks, these types of interventions are a step in the right direction, the researchers say, referencing the overall low cost of PRESCRIBE’s intervention. “It didn't take a lot to build the dashboard to send an email or to do the peer comparisons, but [these actions] can have really dramatic impact in how people behave,” Patel said. “We really think that this shows evidence that if you can combine these types of dashboards with these insights in behavioral economics—the subtle low-cost nudges—that it really might help to improve how health systems address these types of gaps in care.”
Patel MS, Kurtzman GW, Kannan S, et al. Effect of an automated patient dashboard using active choice and peer comparison performance feedback to physicians on statin prescribing: the PRESCRIBE cluster randomized clinical trial. JAMA Network Open. 2018;1(3):e180818.
- Patel reports receiving personal fees from Catalyst Health LLC, Healthmine Services, and Life.io outside of the submitted work.