EHR analysis identifies patients at risk for HIV and drug candidates

Two large-scale studies have demonstrated the efficacy of leveraging algorithms to analyze electronic health records so doctors can identify patients at risk for HIV who may benefit from antiretroviral drugs.

As part of the pre-exposure prophylaxis (PrEP) prevention strategy, healthy people take one or more antiretroviral drugs to reduce their risk of getting HIV.

However, the problem is that the approach remains greatly underutilized because physicians underprescribe PrEP, according to the National Institutes of Health, which supported the two studies published last week in The Lancet HIV.

In the first study, researchers utilized machine learning algorithms to predict incident HIV infections with 180 potential predictors of HIV risk, using EHR data from more than 1 million patients at Atrius Health, an ambulatory group practice in Massachusetts.

“Automated algorithms can efficiently identify patients at increased risk for HIV acquisition,” concluded the first study. “Integrating these models into EHRs to alert providers about patients who might benefit from PrEP could improve prescribing and prevent new HIV infections.”

Also See: App helps HIV patients report medication, substance use

Krakower-Doug-CROP.jpg

“The incorporation of automatic screening algorithms into EHRs could help busy clinicians identify and assess patients who may benefit from PrEP more efficiently and empower them to prescribe PrEP more frequently,” says study author Douglas Krakower, MD, of Beth Israel Deaconess Medical Center and Harvard Medical School.

According to Krakower, the first study “suggests nearly 40 percent of new HIV cases could potentially have been averted had clinicians received alerts to discuss and offer PrEP to their patients with the highest 2 percent of risk scores.”

In the second study, researchers developed a machine learning algorithm to predict who would become infected with HIV during a 3-year period using the EHRs of 3.7 million Kaiser Permanente patients.

“Our model was able to identify nearly half of the incident HIV cases among males by flagging only 2 percent of the general patient population,” says Julia Marcus of Harvard Medical School and Harvard Pilgrim Health Care Institute, who led the second study. “Embedding our algorithm into the Kaiser Permanente EHR could prompt providers to discuss PrEP with patients who are most likely to benefit.”

“It is critical that we identify our patients at risk of HIV acquisition,” says senior author Jonathan Volk, MD, an infectious disease physician who treats patients with HIV at Kaiser Permanente San Francisco Medical Center. “We used our electronic medical record to develop a tool that could be implemented in a busy clinical practice to help providers identify patients who may benefit most from PrEP.”

For reprint and licensing requests for this article, click here.