A silent killer is now easier to identify through the use of electronic health records.

Hypertension, known as the silent killer because of its asymptomatic ability to cause heart attacks and strokes, has not always been diagnosed easily or accurately by health care providers.

But a group of researchers have successfully used an approach to identify patients with high blood pressure that uses EHRs to bypass some of the systemic obstacles to diagnosing the condition. The results were published in the July/August issue of Annals of Family Medicine.

Hypertension is a huge health problem. The Centers for Disease Control and Prevention estimates that 31 percent of American adults have high blood pressure, raising their risk for heart- and kidney-related conditions. At the same time, about 20 percent of people who have high blood pressure are not aware they have it, says Northwestern Medicine Evanston family physician Michael Rakotz, M.D., the principal investigator for the study.

There are many temporary variables that can lead to an inaccurate elevated blood pressure reading, such as a recently smoked cigarette or chugged cup of coffee. A physician sometimes chalks up what in his mind is a one-off high reading to one of those variables. The physician then might let that high reading pass, possibly not even telling the patient, Rakotz says. Other factors contribute to that tendency — known as clinical inertia — such as a lack of time to get an accurate reading, he says.

So Rakotz and the other investigators designed a simple way to skip the human elements that can prevent a proper hypertension diagnosis. They set up computer screens based on blood pressure readings recorded in EHRs that offer strong indicators about whether a patient is a candidate for hypertension.

“It’s as though you’ve got a computer warning system watching your back,” Rakotz says.

The screens were applied to roughly 92,000 patients previously undiagnosed with hypertension or related conditions, producing a final list of 1,432 patients who potentially had undiagnosed hypertension.

Those patients’ names were given to physicians in the groups for follow-up, along with some education as to why blood pressure readings in the office are commonly wrong.

“I wasn’t real popular because every doctor on average had 14 patients and I was telling them [that those patients] may have hypertension,” Rakotz says. “They looked at it as [my saying], ‘You may have missed 14 patients with hypertension,’ “ he says. “But they understood when I explained to them that this was not a phenomenon unique to them.”

After disqualifying some patients for various reasons, and follow-up checks with the rest, 361 of the patients were identified as having hypertension, 290 had a related but not full hypertension diagnosis, 89 had normal blood pressure and 293 remained unclassified and at risk.

Rakotz says the approach was included as a prototype site with an American Medical Association initiative to reduce high blood pressure.

“This is the best use of an EHR, to help doctors practice better medicine.”