medwireNews: Screening rates for primary aldosteronism (PA) among at-risk patients with hypertension can be improved by implementing a best-practice advisory (BPA) in electronic health records (EHRs) that prompts physicians to order the relevant tests, study findings indicate.
“The study results suggest that noninterruptive BPAs are potentially promising PA screening-assistance tools, particularly among primary care physicians,” write Adina Turcu (University of Michigan, Ann Arbor, USA) and co-authors in JAMA Internal Medicine.
“Combined with artificial intelligence algorithms that optimize the detection yield, refined BPAs may contribute to personalized hypertension care,” they add.
Turcu et al explain that although guidelines recommend PA screening for at-risk patients, fewer than 2% to 4% of screening candidates undergo testing.
To address this, they initiated a prospective quality improvement study in which they developed an EHR algorithm to identify PA screening candidates, that is, adults with hypertension and at least one of the following PA risk factors: use of four or more current antihypertensive medications; hypokalemia; age younger than 35 years; adrenal nodule(s); or obstructive sleep apnea. Individuals who had plasma aldosterone and renin measurements within the past 5 years or an active diagnosis of PA were excluded.
When a patient met the algorithm criteria, a noninterruptive BPA alert was displayed on their EHR. The BPA included a PA screening smart order set (including plasma aldosterone, renin, and basic metabolic panel) and a link to guidance on interpreting the test results. Clinicians had the option to use, ignore, or decline the BPA.
During the 15-month study, 57,334 adults with hypertension visited an outpatient clinic that provided primary care or secondary hypertension expertise at the University of Michigan. Among them, the BPA identified 14,603 candidates (mean age 66 years, 50% women) for PA screening.
Of these, 48.1% had treatment-resistant hypertension, 43.5% had hypokalemia, 10.5% were younger than 35 years, and 3.1% had adrenal nodule(s).
The researchers report that 14.0% of the PA screening candidates received orders for testing.
“While this rate remains suboptimal, it represents a promising increment compared with retrospective reports of unassisted PA screening across indications from our institution and other academic or populational studies,” Turcu et al remark.
Most (70.5%) patients given an order completed the recommended screening within the system, and 17.4% had positive screening results. Screening orders were most likely to be placed by internists (40.0%) and family medicine physicians (28.1%).
After adjustment for potential confounders, Black race and Asian race were associated with a significantly increased likelihood of PA screening, versus White race, at odds ratios (ORs) of 1.55 and 1.38, respectively. People with obesity (OR=1.64) or dyslipidemia (OR=1.34) were also significantly more likely to receive a screening order than those without, whereas people with chronic kidney disease (OR=0.81), coronary artery disease (OR=0.73), atrial fibrillation (OR=0.58), or heart failure (OR=0.50) were significantly less likely to receive the order than those without these conditions.
Turcu and colleagues say that, to their knowledge, they developed “the first initiative of enhancing PA screening via a noninterruptive BPA that targeted high-risk patient groups, and the results suggested a substantial increment compared with typical screening rates.”
They add that the EHR-delivered BPA was “particularly effective among primary care physicians,” which “suggests an opportunity to implement innovative clinical decision supporting systems that may meaningfully improve PA awareness and enhance testing in high-risk patient populations.”
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2024 Springer Healthcare Ltd, part of the Springer Nature Group