UNITED by Technology: A New Medication Safety Program at the VA

Paul Heidenreich, MD

Like many other tools, technology can be used for good or ill, to enlarge gaps between people or to bridge them. But for Paul Heidenreich, MD, professor of cardiovascular medicine (and, by courtesy, of health research and policy at the Palo Alto Veterans Affairs Health Care System), technology can be used to create a "community of practice."

Heidenreich serves as vice chair for clinical, quality, and analytics in the Department of Medicine and is currently heading the MedSafe project, sponsored by the VA’s Quality Enhancement Research Initiative, which seeks to improve medication safety.

This aspect of medicine needs improvement. In 2011, 12 percent of veterans were prescribed a potentially inappropriate new medication with an incidence of six percent per year. Heidenreich explains that this happens for various reasons: Patients may be inappropriately prescribed a high-risk medication or be on a high-risk medication without appropriate lab monitoring. “Our systems are not such that we can catch that or realize it happened every time,” he says.

Interventions to improve safety

The VA has initiated programs to improve medication safety. The interventions come in various forms, from educating physicians about risks, to writing draft medication or lab test orders for physicians to sign, to reaching out to patients to remind them about needed tests or medications. While a variety of these interventions have been implemented, they lacked a way to measure their effectiveness. This is where Heidenreich’s project comes in.

“One of our goals was to look at all the interventions the VAs in different states were using, see which sites were most effective and had the best safety records, and then note what were they doing to manage things,” he states.

To that end, his group designed MedSafe, which is government-funded and set to run for five years with access to all patient records within the VA system. The VA serves more than 8.9 million veterans at 168 VA Medical Centers and 1,053 outpatient clinics each year. Data about all patients is tracked and can be fed back to the various hospitals and clinics.

The project consists of three subprojects. While Heidenreich’s group studies the effectiveness of various interventions, another group is putting the interventions (suggestions, orders, etc.) into the electronic dashboard so physicians and hospital staff can immediately access the information. A third subproject, headed by Mary Goldstein, MD, a professor of medicine at VA Palo Alto Health Care System, focuses on developing “clinical decision support (CDS)” integrated with the dashboard to guide providers through the process of implementing the interventions.

“In addition to updating the knowledge bases to newer evidence and guidelines, we are linking the CDS to a clinical dashboard,” Goldstein states. “For example, if a patient with diabetes is out of range for glucose control, our CDS system will generate recommendations for the primary care team.”  

The project is too new to have conclusive data, but Heidenreich expects the “more active, targeted, interruptive interventions” to be the most effective. On a past project they “found that physicians were for the most part very willing to receive a draft order for a diagnostic test,” and he believes that the same will hold true for this CDS project, which plans to provide recommendations for medications and lab tests.

improving patient SAFETY is also important to operations people

The VA, like many governmental institutions spread across states, is both a local and a national organization. This can sometimes cause friction, but Heidenreich sees his project as potentially both a centralized and a localized effort. “I think in the long run there’s no reason why it couldn’t be centralized,” he says. “It’s not clear that physicians need to see a recognizable name before they’re going to look at the recommendations in the dashboard. The VA system is still fairly decentralized in terms of medical records and care, so our goal would be to see which things are the most effective and then go back to all 100-plus facilities and encourage them to adopt those interventions.”

He’s optimistic about the adoption. “We don’t do these projects just as isolated researchers,” he explains. “We do them in partnership with the operations people. The nice thing is that improving patient safety is also important to operations people, and since everything we’re doing is improving care, we’re all in sync.” 

Implementing the interventions

This mutual interest can be drawn on in the next stage, as the project yields results that need to be implemented. Heidenreich’s team has ideas for this as well. In the past, he explains, they used what they called “a community of practice.”

In one case, they invited the lead pharmacists of all VA facilities to get together and then “we would present data or, even better, we’d have different facilities present things that they’d done. We would then show effects and which things seemed to work well. We were able to link people and also provide them with information like, ‘This is how they did it, this is how you can do it, this was the cost to implement it.’ To get them all talking to each other is one of the ways we’ll be implementing MedSafe.”

Goldstein agrees that the project “holds a lot of potential. In working with newer technology, such as dashboards with CDS, it can be helpful for groups to talk with each other to share ideas of what works best. We know of some clinical groups who are using the dashboard to share information within their teams, and we hope that they will be able to take this a step further by using the recommendations from the CDS. We plan to talk with health professionals from multiple teams to learn about what works for them, and we hope later in this project that the teams will share best practices with each other.”

Technology that unites

Goldstein is a believer in the power of this technology to unite: “I think what drives the community of practice is the shared goal of providing best care for patients. I see the technology as something that, if designed and introduced to the clinical setting in a way that is helpful to the health professionals working there, can be part of an overall approach to providing best care. In my view it’s never about the technology per se, but about the technology making it easier for the health professionals, ideally freeing up time from rote work so that they can spend more time interacting with patients— doing the things that humans do well, attending to relationships, emotion, patient goals—and less time with the computer.”

It’s a sentiment echoed by Heidenreich. The efforts, he says, “give a sense of community to those people, especially some who are at smaller facilities. I think it helps them feel engaged in a larger effort.”

The MedSafe project ultimately seeks to do just that: use technology as a tool to create stronger bonds among far-flung hospitals and clinics. This information sharing creates a broad community of practice and practices, funneling research, technology, and real-world knowledge into something that ultimately benefits the individual at the heart of all of this: the patient.