Skip to main content

New Best-Practice Alert Helps Clinicians Identify Patients at Risk of Preeclampsia

The EPIC feature has eliminated racial disparities among patients

Low-dose aspirin is proven to reduce the risk of preeclampsia for pregnant women at high risk of the condition, including those over 40 years old and Black women. But during routine prenatal visits, clinicians may miss certain preeclampsia risk factors, which prevents some patients from receiving a simple treatment for a common, life-threatening condition. To address this issue, Cedars-Sinai has created a new alert in patients’ EPIC electronic health record.

“It’s difficult for any clinician to remember the risk factors off the top of their head,” said Cedars-Sinai maternal-fetal medicine specialist Melissa Wong, MD, who created the alert. “We have 15 or 16 risk factors scattered throughout the entire prenatal record, some common, some uncommon or not elicited, like family history of preeclampsia. You can look them up, but it’s not something that you’re going to routinely do.”

Wong was inspired to create the best-practice alert, which was unveiled in March 2022, to improve health disparities after she uncovered data that showed that Cedars-Sinai clinicians had recommended aspirin to Black patients less frequently than to white patients.

“Our numbers from 2016 to 2017 suggested we recommended aspirin about 3% of the time when it was indicated for Black patients compared to about 16% of the time for white patients, so over a fivefold racial disparity,” Wong said. “This disparity probably stems from the fact that we as clinicians forget that being Black is one among the list of moderate risk factors for preeclampsia, and so often an indication for aspirin.”

If data listed in the discrete fields of a patient’s EPIC health record shows that they have one high preeclampsia risk factor or two moderate risk factors, an alert pops up asking clinicians if the patient is taking low-dose aspirin. This reminds physicians to prescribe the medication if appropriate.

In a retrospective cohort study to determine the effectiveness of the EPIC alert, Wong performed a manual review of 677 patient medical charts from before the alert was implemented. Preliminary findings show that clinicians recommended aspirin to 23.6% of at-risk patients. When EPIC checked discrete fields for risk factors of these patients, aspirin was recommended 58.1% of the time. When natural-language processing helped check the patient’s entire health record, the detection rate rose to 73.3%; however, EPIC cannot accommodate AI learning at this time.

The EPIC alert has improved aspirin administration to all at-risk patients, including Black women.

“We’ve eliminated that racial disparity,” Wong said. “Among those for whom the alert triggers, we now treat them equitably.”

As with all AI tools, EPIC’s ability to identify at-risk patients is limited by the available data.

“The goal of this clinical decision support tool is to take some of that mental burden off of clinicians, but it ultimately relies on clinical documentation in discrete fields of the health record,” Wong said. “If you haven’t logged a patient’s weight, for example, then it won’t be able to trigger an alert on the patient’s BMI.”

Wong advises physicians at other health systems to follow Cedars-Sinai’s lead. To increase the likelihood of discovering preeclampsia risk factors at every prenatal appointment, she recommends adding a new dot phrase to the EHR template.

“You could include, in your own template, ‘aspirin candidate?’ as a prompt each time,” she said. “Forty percent of our patients meet criteria for preeclampsia risk. If you’re going to be a ‘yes’ almost one in two times, that’s worth prompting yourself every time.”

Wong also has developed an online risk calculator to help patients learn their own preeclampsia risk. When the calculator goes live, Wong hopes to provide clinicians with a QR code they can place in waiting rooms for patients.

“A patient-facing risk calculator will have details that will never be elicited in the electronic health record, like a family history of preeclampsia or sensitive things like low socioeconomic status,” Wong said. “There’s potential to put power in the patients’ hands.”