Key result
Machine learning risk prediction fails to increase atrial fibrillation detection over routine primary care.
Why the study?
The study was conducted to evaluate whether a machine learning risk-prediction algorithm combined with diagnostic testing could effectively identify undiagnosed AF in primary care in England.
Does a machine learning risk-prediction algorithm combined with diagnostic testing improve the identification of undiagnosed atrial fibrillation in adults aged ≥30 years?
RCT (n=23,745)
Open-label
Simple randomization in a 1:1 ratio
Yes
Does a machine learning risk-prediction algorithm combined with diagnostic testing improve the identification of undiagnosed atrial fibrillation in adults aged ≥30 years?
While an AI risk-prediction algorithm did not significantly increase overall AF diagnoses in the intention-to-treat high-risk population, it doubled the diagnosis rate among those who completed diagnostic testing.
Answers are grounded in this paper and cite their sources.
ML-guided AF screening does not increase diagnoses in high-risk primary care; challenges targeted diagnostic strategies and requires refined implementation.
Aims The aim of the PULsE-AI trial was to assess the effectiveness of a machine learning risk-prediction algorithm in conjunction with diagnostic testing for identifying undiagnosed atrial fibrillation (AF) in primary care in England. Methods and results Eligible participants (aged ≥30 years without AF diagnosis; n = 23 745) from six general practices in England were randomized into intervention and control arms. Intervention arm participants, identified by the algorithm as high risk of undiagnosed AF (n = 944), were invited for diagnostic testing (n = 256 consented); those who did not accept the invitation, and all control arm participants, were managed routinely. The primary endpoint was the proportion of AF, atrial flutter, and fast atrial tachycardia diagnoses during the trial (June 2019–February 2021) in high-risk participants. Atrial fibrillation and related arrhythmias were diagnosed in 5.63% and 4.93% of high-risk participants in intervention and control arms, respectively {odds ratio (OR) [95% confidence interval (CI)]: 1.15 (0.77–1.73), P = 0.486}. Among intervention arm participants who underwent diagnostic testing (28.1%), 9.41% received AF and related arrhythmia diagnoses [vs. 4.93% (control); OR (95% CI): 2.24 (1.31–3.73), P = 0.003]. Conclusion The AF risk-prediction algorithm accurately identified high-risk participants in both arms. While the proportions of AF and related arrhythmia diagnoses were not significantly different between high-risk arms, intervention arm participants who underwent diagnostic testing were twice as likely to receive arrhythmia diagnoses compared with routine care. The algorithm could be a valuable tool to select primary care groups at high risk of undiagnosed AF who may benefit from diagnostic testing.
No takes yet. Share an insight, caveat, or question.
Hill et al. (2022) conducted an RCT in Undiagnosed atrial fibrillation (n=23,745). Machine learning risk-prediction algorithm and diagnostic testing (12-lead ECG ± KardiaMobile) vs. Routine care was evaluated on Proportion of diagnoses of atrial fibrillation, atrial flutter, and fast atrial tachycardia during the trial in high-risk participants (OR 1.15, 95% CI 0.77-1.73, p=0.486). A machine learning risk-prediction algorithm coupled with diagnostic testing did not increase the diagnosis of atrial fibrillation and related arrhythmias among high-risk primary care patients compared to routine care.
Synapse has enriched 10 related papers. Consider them for comparative context: