Using EHRs to Predict Stroke Risk? New Study Shows Promise

Each year, according to data from the American Stroke Association, almost 800,000 people across the United States experience a stroke. What’s more, strokes are responsible for approximately one out of every 20 annual deaths, killing almost 130,000 people. They also account for about $33 billion in annual health care spending, per the Centers for Disease Control and Prevention (CDC). All of this, of course, serves to make stroke prevention a key public health priority. And as Fierce Healthcare first reported, researchers may have identified one way to predict the risk of stroke by using data found in electronic health records (EHRs) to create a “scoring system.” The results of the retrospective cohort study, published in the journal Cardiology, could be promising for providers treating post-stroke patients. “The scoring system we developed is simple to use and the results could help physicians tailor treatment to individual patients,” study co-author Albee Ling explained.

According to the study’s authors, who are based at Stanford University, the University of California, San Francisco, and the Santa Clara Valley Medical Center, patients who suffer but survive one stroke are often at risk for a second one—which could be deadly. This makes identification of risk factors in patients crucial. As a press release accompanying the study from the Stanford School of Medicine explains, “one important risk factor for that perilous second stroke is an irregular heart beat called atrial fibrillation.” In most cases, providers monitor post-stroke patients for atrial fibrillation while they’re hospitalized; the American Heart Association recommends, too, that patients undergo 30 days of monitoring within six months of the stroke occurring. But as study co-author Calvin Kwong explained in the Stanford press release, this rarely actually happens: “Once they go home — after about a week — clinicians aren’t usually too vigilant about monitoring them for atrial fibrillation,” he said of stroke patients, citing cost as one explanation for this. What’s more, not all patients require such monitoring.

With that in mind, and with access to a database of more than 9,500 stroke patients’ health information, the researchers sought to develop a way to more easily spot atrial fibrillation in post-stroke patients, eventually arriving at an algorithm that examined EHRs for certain key words associated with atrial fibrillation. As outlined in the press release, “By ranking the clinical attributes of patients whose medical records indicated they went on to be diagnosed with atrial fibrillation, the team was able to assemble a set of seven risk factors that, when combined, predicted which stroke patients were the most likely to develop the condition and should be monitored after hospitalization.” While the study’s authors note that further research is needed to validate their findings, they are optimistic about the system they created. As Fierce Healthcare and Stanford both note, the study’s results are part of an ongoing effort at the university to prioritize precision medicine. Earlier this year, we highlighted research from another Stanford faculty member focusing on wearable devices. Euan Ashley and his team examined how well wearable devices track such measures as heart rate, believing that the devices could be potentially useful for improving cardiac health outcomes.

Click here to read the article from Fierce Healthcare on the EHR algorithm study.

Click here for the article abstract from Cardiology (full article requires payment). 

Click here for the press release from Stanford University on the EHR/stroke study.

Share

Leave a Reply