PROMPT Trials, Electronic Alerts, and Enhancing CV Outcomes

The Pragmatic Trial of Messaging to Providers (PROMPT) Trials are a series of randomized studies that focus on implementing electronic medical-based alerts to enhance patients outcomes and overall care. At present, these trials are grouped into 4 divisions: lipids, heart failure (HF) inpatient, HF outpatient, and magnetic resonance angiography (MRA). In sum, the PROMPT programs trigger alerts to help guide evidence-based, FDA-recommended interventions for patients with cardiovascular disease or complications.

In this interview, Nihar Desai, MD, MPH, cardiovascular medicine specialist, Yale School of Medicine, speaks to the influence of these alerts and the lessons providers have learned while implementing them.

This transcript has been lightly edited for clarity; captions were auto-generated.

Transcript

In the PROMPT Trials you led at Yale, how did electronic alerts impact patient outcomes?

Great question. We did a series of studies under the broad umbrella called PROMPT looking at the impact of decision support tools on clinical prescribing patterns, and then, ultimately, looking both at LDL [low-density lipoprotein] and lipids, and also for heart failure, for patients with systolic dysfunction.

In both of those studies, we saw that a timely, real-time, individualized alert that was at the point of clinical interaction was associated with a significant increase in the use of evidence-based therapies, both for LDL cholesterol lowering and for patients with systolic heart failure.

How did health care providers respond to the electronic alerts, and were there any major barriers to implementation?

I would say the biggest thing that we learned from the series of PROMPT studies was to have users, the end users, involved in the design of the actual alert. They told us what they wanted in the alert. And our promise to them was, “You wouldn't see this alert until we built what you wanted us to build.” And what they told us was, “Hey, make it real-time, make it timely, make it individualized, and make it so it actually pulls information that I would otherwise have to navigate through different areas of the medical record for—and so make my job a little bit easier.”

When we did that, we saw that it certainly improved prescribing behavior and use of evidence-based therapies, but the feedback that we got from our providers was also quite positive. I think that gave us a lot of enthusiasm for not just building additional decision support tools, but to really build decision support tools that take the end user into account, our colleagues, our clinicians, that are delivering care. If we build what they want, I think then it will help us have the kind of impact that we all want.