Five decades of research have illuminated the role of nonvalvular atrial fibrillation (AF) in the pathogenesis of stroke, heart failure, dementia, and premature death. Given the often-asymptomatic nature of the arrhythmia and the clear benefit of premorbid interventions including anticoagulation for stroke prevention, it makes intuitive sense that screening for asymptomatic AF would affect outcomes. However, the evidence to support screening remains elusive and is complicated by questions as basic as who to screen, how best to screen, how much AF is actionable, and if we can do more good than harm by finding and treating AF early.

It is in this context that the US Preventive Services Task Force (USPSTF) reports that “the current evidence is insufficient to assess the balance of benefits and harms of screening for AF” (I statement).1 The evidence report and systematic review have also been updated.2 This conclusion is not a recommendation against screening for AF but indicates that the evidence is insufficient to recommend either for or against screening. It mirrors the 2018 statement on the same topic, but several recent advances bolster both sides of the argument.

The USPSTF statement does not currently support electrocardiogram (ECG) screening, but opportunistic screening via pulse palpation at the time of physical examination is recommended by multiple professional organizations in Europe, Canada, and Australia and by the World Heart Federation for those older than 65 years followed by an ECG if pulse irregularity is detected.3-5 By contrast, the USPSTF considers opportunistic screening to be a facilitated case finding rather than actual screening and is considered to be part of routine or usual care.

However, studies have not demonstrated a superiority of ECG compared with a thorough physical examination in the context of clinical trials although in usual clinical practice, the examination is infrequently performed,6 is likely to detect more persistent AF, and can have high false-positive rates owing to ectopy and a high false-negative rate in older individuals owing to intrinsic or iatrogenic slowing of ventricular rates. Two randomized clinical trials of single-time opportunistic single-lead ECG screening in primary care did not find an increased rate of AF detection in the ECG screening practices compared with control practices, with the exception of those 85 years or older in the Screening for Atrial Fibrillation Among Older Patients in Primary Care Clinics (VITAL-AF) study.7,8 Still, ECG remains the criterion standard for diagnosing AF, although it too can be misinterpreted and may trigger additional and often unnecessary testing owing to other abnormal findings even if the rhythm is normal.

Regardless of how the diagnosis is made, for AF screening to be fully endorsed, it must first be demonstrated that screen-detected AF carries the same prognosis and responds similarly to interventions as clinically detected AF. Data from recent retrospective studies support both these requirements. Compared with symptomatic patients with AF in the Global Anticoagulant Registry in the Field-Atrial Fibrillation (GARFIELD-AF) study,9 those who were asymptomatic at the time of diagnosis—a situation analogous to single-time screening—showed similar rates of all-cause mortality, hemorrhagic stroke, systemic embolism, and major bleeding when treated with anticoagulation.

An analysis of patients with incident AF in the UK Primary Care Clinical Practice Research Datalink also showed similar results with asymptomatic, incidentally detected AF diagnosed in the outpatient setting carrying the same risk of ischemic stroke and death compared with those diagnosed with symptomatic AF or AF diagnosed in-hospital.10 Notably, stroke rates in asymptomatic and symptomatic AF presentations were reduced similarly by anticoagulants in both cohort studies. Further evidence comes from a prospective screening study in Hong Kong using a handheld single-lead ECG that found similar stroke risk in screen-detected vs clinically detected AF, along with a comparable reduction in strokes for screen-detected AF treated with anticoagulation.11

It is important to recognize that the cohort studies that showed equivalence of incidentally detected or screen-detected and clinically detected AF relied on screening performed at a single assessment. However, the diagnostic yield of AF screening increases with the number, duration, and temporal dispersion of screening sessions. As a result, a screening approach incorporating repeated measures over time or continuous monitoring is more likely to detect paroxysmal AF or intermittent persistent AF that still may benefit from treatment.

Such strategies were assessed in 2 recent large-scale randomized clinical trials. The STROKESTOP trial12 randomized 28 768 individuals aged 75 or 76 years in 2 counties in Sweden to be invited to screen for AF or to a control group who were not contacted. The screening group used a handheld, single-lead ECG twice daily for 2 weeks. There was a small but statistically significant reduction in the risk for the combined end point of ischemic or hemorrhagic stroke, systemic embolism, bleeding requiring hospitalization, and all-cause mortality.

A second study, the LOOP trial,13 randomized 6004 volunteers (mean age, 74.7 years), with 1 or more additional stroke risk factors in a 1:3 ratio to receive an insertable cardiac loop recorder capable of detecting AF episodes more than 2 minutes in duration or usual care. There was no significant difference in the primary end point of stroke or systemic embolism or the secondary end points of cardiovascular death or all-cause mortality.

Together, these studies make an underwhelming argument for AF screening, but important caveats are worth noting. In the STROKESTOP trial, nearly half the patients invited to screening declined to participate yet were appropriately included in the intention-to-treat analysis. These patients were older, had more comorbidities, and had less favorable sociodemographic factors compared with those who did participate. Indeed, for the secondary end point of ischemic stroke, individuals randomized to the screening arm who actually underwent screening experienced a 24% reduction in the risk of stroke compared with controls and 34% reduction compared with nonparticipants.

A planned sensitivity analysis of the LOOP trial excluding those in the intervention arm who prematurely terminated monitoring showed a significant 30% reduction in stroke or systemic embolization. It is also relevant to consider that 42% of the qualifying AF episodes in the LOOP trial were less than 5.5 hours long, a duration below the threshold associated with increased stroke in prior studies using continuous monitoring. These findings emphasize the need for screening techniques that are widely accessible and easy to perform and underscore that participants in screening studies generally have a better prognosis than nonparticipants, a major limitation in this type of study design.

While the focus of most of the work in this area to date has been on evaluating the effect of early detection on stroke, screen-detected AF has other implications. For patients with no or few other stroke risk factors that would provide an indication for anticoagulation, screen-detected AF may serve as a wake-up call for the evaluation and treatment for upstream diseases such as sleep apnea or the adoption of lifestyle modifications that may reduce AF progression including weight loss, exercise, and alcohol reduction or abstinence. Even a sinus rhythm 12-lead ECG may prove useful in screening. For example, developments in artificial intelligence can now be used to identify patients at particularly high risk of developing AF in the future, providing an opportunity to distinguish a cohort that may benefit from closer surveillance.14

The advent of direct-to-consumer, smartphone-based devices capable of assessing pulse regularity using photoplethysmography or recording a single-lead ECG has changed the paradigm of how and where screening can be performed. The historical focus of the AF screening debate has been on the utility of screening initiated by health care professionals using pulse palpation and/or ECG recordings. When compared with routine care, repeated twice weekly at-home ECG recordings performed over a year using a direct-to-consumer single-lead ECG device had a higher rate of AF detection.15 Studies of wearable devices such as smartwatches that combine photoplethysmography for frequent pulse sampling and the ability to confirm an AF diagnosis with a single-lead ECG are ongoing.16

These advances will likely highlight the trade-off of continuous monitoring strategies that provide increased detection of lower-burden AF representing a lower risk of stroke that may not benefit from anticoagulation. To be effective, the movement toward consumer-based screening must first show that such an approach improves outcomes. It must also deal with the paradox that those at highest risk of AF and AF-related stroke may be the least likely to own these technologies unless supported by the health care system. In addition, appropriate care pathways for confirming the diagnosis and, if necessary, initiating appropriate treatment in individuals with positive findings will need to be established. It will also be critical to ensure that device costs and variable technological literacy do not create barriers to making screening accessible to all those at risk.

Interest in AF screening has been fueled not only by a greater understanding of the long-term effect of the arrhythmia but also by major advances in stroke prevention and AF monitoring. More information will come from a prospective individual-patient meta-analysis of randomized clinical trials17 as well as large-scale ongoing trials including the SAFER and SAFER-AUS trials.18 As the USPSTF statement implies, we have yet to convincingly prove that screening for the most common sustained arrhythmia reduces stroke, prolongs life, or improves the quality of life. However, with exponential growth in AF risk factors putting millions of individuals at risk, the stakes are too high to not continue to try.

Corresponding Author: Rod Passman, MD, MSCE, Northwestern University, 676 N St Claire St, Ste 600, Chicago, IL 60611 (r-passman@northwestern.edu).

Published Online: January 25, 2022. doi:10.1001/jamacardio.2021.5873

Conflict of Interest Disclosures: Dr Passman reported personal fees from Janssen and personal fees from Medtronic outside the submitted work. Dr Freedman reported grants and personal fees from Bristol Myers Squibb/Pfizer to their institution for investigator-initiated studies; personal fees and travel support from Daiichi Sankyo and OMRON; and nonfinancial support from Alivecor outside the submitted work.

1.US Preventive Services Task Force.  Screening for atrial fibrillation: US Preventive Services Task Force recommendation statement.   JAMA. 2021;327(4):1-8. doi:10.1001/jama.2021.23732Google Scholar2.Kahwati  LC, Asher  GN, Kadro  Z,  et al.  Screening for atrial fibrillation: updated evidence report and systematic review for the US Preventive Services Task Force.   JAMA. 2021;327(4):1-16. doi:10.1001/jama.2021.21811Google Scholar3.Hindricks  G, Potpara  T, Dagres  N,  et al; ESC Scientific Document Group.  2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): the Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC.   Eur Heart J. 2021;42(5):373-498. doi:10.1093/eurheartj/ehaa612PubMedGoogle ScholarCrossref 4.Brieger  D, Amerena  J, Attia  JR,  et al.  National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Australian clinical guidelines for the diagnosis and management of atrial fibrillation 2018.   Med J Aust. 2018;209(8):356-362. doi:10.5694/mja18.00646PubMedGoogle ScholarCrossref 7.Uittenbogaart  SB, Verbiest-van Gurp  N, Lucassen  WAM,  et al.  Opportunistic screening versus usual care for detection of atrial fibrillation in primary care: cluster randomised controlled trial.   BMJ. 2020;370:m3208. doi:10.1136/bmj.m3208PubMedGoogle Scholar8.

Lubitz  S, Atlas  S, Ashburner  J,  et al VITAL AF screening for atrial fibrillation among older patients in primary care clinics (VITAL-AF). Presented at: American Heart Association Scientific Sessions; November 16, 2020; virtual.

9.Gibbs  H, Freedman  B, Rosenqvist  M,  et al; GARFIELD-AF Investigators.  Clinical outcomes in asymptomatic and symptomatic atrial fibrillation presentations in GARFIELD-AF: implications for AF screening.   Am J Med. 2021;134(7):893-901.e11. doi:10.1016/j.amjmed.2021.01.017PubMedGoogle ScholarCrossref 10.Wallenhorst  C, Martinez  C, Freedman  B.  Risk of ischemic stroke in asymptomatic atrial fibrillation incidentally detected in primary care compared with other clinical presentations.   Thromb Haemost. Published online June 30, 2021. doi:10.1055/a-1541-3885PubMedGoogle Scholar11.Sun  W, Freedman  B, Martinez  C, Wallenhorst  C, Yan  BP.  Atrial fibrillation detected by single time-point handheld electrocardiogram screening and the risk of ischemic stroke.   Thromb Haemost. Published online August 16, 2021. doi:10.1055/a-1588-8867PubMedGoogle Scholar12.Svennberg  E, Friberg  L, Frykman  V, Al-Khalili  F, Engdahl  J, Rosenqvist  M.  Clinical outcomes in systematic screening for atrial fibrillation (STROKESTOP): a multicentre, parallel group, unmasked, randomised controlled trial.   Lancet. 2021;398(10310):1498-1506. doi:10.1016/S0140-6736(21)01637-8PubMedGoogle ScholarCrossref 13.Svendsen  JH, Diederichsen  SZ, Højberg  S,  et al.  Implantable loop recorder detection of atrial fibrillation to prevent stroke (The LOOP Study): a randomised controlled trial.   Lancet. 2021;398(10310):1507-1516. doi:10.1016/S0140-6736(21)01698-6PubMedGoogle ScholarCrossref 14.Attia  ZI, Noseworthy  PA, Lopez-Jimenez  F,  et al.  An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction.   Lancet. 2019;394(10201):861-867. doi:10.1016/S0140-6736(19)31721-0PubMedGoogle ScholarCrossref 16.HEARTLINE - A Heart Health Study Using Digital Technology to Investigate if Early AF Diagnosis Reduces the Risk of Thromboembolic Events Like Stroke IN the Real-world Environment. ClinicalTrials.gov. Identifier: NCT04276441. Posted February 9, 2020. Accessed December 30, 2021. https://clinicaltrials.gov/ct2/show/NCT04276441