Yao X, Rushlow DR, Inselman JW, et al. - In the setting of routine primary care, early diagnosis of low ejection fraction (EF) can be achieved by using an artificial intelligence (AI) algorithm based on electrocardiograms (ECGs).

• This is a pragmatic clinical trial to determine if an ECG-based, AI-powered clinical decision support tool allows early diagnosis of low EF.

• A total of 120 primary care teams from 45 clinics or hospitals were included, and cluster-randomized to either the intervention arm (access to AI outcomes; 181 clinicians) or the control arm (usual care; 177 clinicians).

• A total of 22,641 adults (n = 11,573 intervention; n = 11,068 control) without prior heart failure underwent ECG examinations as part of routine care.

• Increased diagnosis of low EF in the overall cohort was achieved with the intervention [1.6% in the control arm vs 2.1% in the intervention arm, odds ratio (OR) 1.32].

• Also, the intervention increased the diagnosis of low EF in those with a high chance of low EF (that is, positive AI-ECG, 6% of the overall cohort) (14.5% in the control arm vs 19.5% in the intervention arm, OR 1.43).

• Similar echocardiogram utilization between the two arms was evident in the overall cohort; for those with positive AI-ECGs, more echocardiograms were obtained in the intervention vs the control arm (38.1% control vs 49.6% intervention).