Smartwatch monitoring increased new-onset AF detection in elderly patients compared to standard care

This randomized controlled trial was conducted in secondary care centers in the Netherlands. The study population consisted of 437 patients aged 65 years or older with elevated stroke risk, specifically defined as a CHADS-VASc score of 2 or higher for men and 3 or higher for women. The primary objective was to evaluate the detection of new-onset atrial fibrillation in this high-risk cohort.

The intervention group received 6-month (180-day) monitoring using a smartwatch equipped with photoplethysmography and single-lead electrocardiogram (ECG) functions. The comparator group received standard care. The primary outcome was defined as new onset AF, which required a confirmed episode lasting 30 seconds or longer detected on single-lead ECG or standard ECG methods.

Primary outcome results demonstrated a marked difference between the two groups. In the intervention group, 21 patients (9.6%) experienced new-onset AF. In the control group, 5 patients (2.3%) experienced new-onset AF. The risk difference was 7.3 percentage points, with a hazard ratio of 4.40. The statistical significance was established with a p-value of 0.001, and the 95% confidence interval for the risk difference ranged from 2.9 to 11.7 percentage points.

No secondary outcomes were reported in the study data. Safety and tolerability findings were not reported, including adverse events, serious adverse events, discontinuations, or general tolerability metrics. Consequently, no specific adverse event rates or discontinuation numbers are available for this analysis.

The direction of the effect indicated an increased risk of new-onset AF in the intervention group compared to the control group. This finding must be interpreted with caution regarding causality. The elevated detection rate in the intervention group likely reflects the superior sensitivity of the smartwatch monitoring in identifying previously undiagnosed arrhythmias in a population with high baseline risk, rather than the device causing the condition.

Key methodological limitations and potential biases were not reported in the provided data. The study design is a randomized controlled trial, which generally minimizes selection bias, but the lack of reported limitations prevents a full assessment of potential confounding factors or measurement biases inherent in self-worn devices versus clinical monitoring.

Clinical implications suggest that clinicians should recognize that sensitive monitoring tools may unmask new-onset AF in elderly patients with elevated stroke risk. This has significant implications for stroke risk stratification and anticoagulation decisions. However, because safety and adverse event data were not reported, the full risk-benefit profile of this monitoring strategy remains incomplete.

Several questions remain unanswered. The lack of reported safety data prevents a comprehensive assessment of patient tolerance and potential harms associated with prolonged smartwatch use. Furthermore, the absence of reported limitations limits the ability to generalize these findings to other settings or populations. The study does not establish a causal relationship between the monitoring device and the occurrence of AF, but rather highlights the high yield of detection in this specific demographic.

In conclusion, while the study provides robust evidence of increased AF detection in patients using smartwatch monitoring, the absence of safety reporting and limitations data necessitates a conservative interpretation. Clinicians should consider these findings when evaluating the utility of consumer-grade monitoring devices for AF screening, keeping in mind that the increased detection rate reflects diagnostic sensitivity rather than a pathological effect of the device.