This review provides information about the current evidence-base for screening for atrial fibrillation (AF). The current burden of AF and recommendations related to AF screening are discussed, and randomised-controlled trial evidence for AF detection, clinical outcomes, harms and cost-effectiveness of population-based AF screening are reviewed. Finally, novel methods to refine the population to whom AF screening should be offered, which may improve clinical and cost-effectiveness, are considered.
Introduction
The diagnosis of atrial fibrillation (AF) is made from an electrocardiogram (ECG) showing AF lasting for at least 30 seconds.1 The ECG characteristics of AF are irregularly irregular R–R intervals (where atrioventricular conduction is not impaired), absence of distinct repeating P-waves, and irregular atrial activations.1
AF is the most common sustained arrhythmia in the general population, with an estimated prevalence worldwide of 2% to 4%,2 and this is expected to increase two- to three-fold by 2030.3 In the UK, approximately 1.2 million individuals (1.8% of the populace) have been identified and diagnosed with AF.4,5
AF is associated with an increased risk of stroke, left ventricular systolic dysfunction, cognitive decline, depression, hospitalisation and death.1 Notably, AF increases the risk of stroke by five-fold, and AF-related ischaemic strokes are often severe, recurrent and fatal.6 Oral anticoagulation reduces the risk of AF-related stroke by two-thirds.6 Historically, warfarin was the oral anticoagulant of choice, but this has been superseded by direct oral anticoagulants (DOACs) in routine practice over the last decade.5,7
Recommendations for AF detection in the UK
Presently, the National Institute for Health and Care Excellence (NICE) guidelines recommend pulse palpation and the performance of a 12-lead ECG in patients who experience symptoms of palpitations, shortness of breath, syncope, chest discomfort or transient ischaemic attack/stroke to detect underlying AF.5 Table 1 summarises recommendations regarding AF screening from various task forces.
Table 1. Current atrial fibrillation (AF) screening recommendations
| Issuing group | AF screening recommendations |
| European Society of Cardiology (ESC)1 | Opportunistic screening (≥65 years) and systematic screening offered (≥75 years) |
| National Institute for Health and Care Excellence (NICE)5 | Opportunistic screening in symptomatic patients |
| Heart Foundation and Cardiac Society of Australia and New Zealand (CSANZ)8 | Opportunistic screening offered (≥65 years) |
| United States Preventative Service Task Force (USPSTF)9 | More evidence required to assess benefit of screening |
| Asian Pacific Heart Rhythm Society (APHRS)10 | Opportunistic screening (≥65 years) and systematic screening offered (≥75 years) |
Is there a need for screening for AF?
It is estimated that 30% of people living with AF are undiagnosed, and up to 15% of strokes occur in the context of undiagnosed AF – equating to 30,000 strokes per year in the UK.11 Early detection of AF is a cardiovascular priority in the NHS Long Term Plan,12 and the 2020 European Society of Cardiology (ESC) guidelines for the diagnosis and management of AF recommend opportunistic screening for AF by pulse taking or ECG rhythm strip in patients aged 65 years and over (class I, level of evidence B), and that systematic ECG screening should be considered to detect AF in individuals aged 75 years and over and among those at high risk of stroke (class IIa, level of evidence B).1,13 However, the UK National Screening Committee did not recommend AF screening when they last reviewed the topic in 2019, and at present there is no systematic pathway in the NHS for the early detection or screening of AF.12,14
Evidence for screening for AF
The two strategies used in AF screening to date are systematic and opportunistic screening. In systematic screening, an entire population or a stratum of a population is targeted for screening.15 Opportunistic screening is a strategy in which the participant is offered screening during a healthcare visit not caused by a suspicion of the screened disease.
Effect of opportunistic screening on detection rates for AF
Recent randomised-controlled trials (RCTs) have found that opportunistic screening for AF does not increase the detection rates for AF compared with routine care in contemporary practice. The Detecting and Diagnosing Atrial Fibrillation (D2AF) trial was a cluster RCT involving 47 intention-to-screen and 49 usual-care primary care practices in the Netherlands.16 Opportunistic screening consisted of three index tests: pulse palpation, electronic blood pressure measurements with an AF detection algorithm, and ECG with a single-lead hand-held ECG device. The detection of new AF was not significantly different between the intervention arm and the control arm (144/8,874 patients in intervention arm [1.62%] vs. 139/9,102 patients in the control arm [1.53%]; adjusted odds ratio [OR] 1.06, 95% confidence interval [CI] 0.84 to 1.35). In the Screening of Atrial Fibrillation in Older Primary Care Visits (VITAL-AF) trial 16 primary care clinics were randomised 1:1 to AF screening using a handheld single-lead ECG (AliveCor, KardiaMobile) during vital sign assessments or usual care.17 Of 30,715 patients without prevalent AF (n=15,393 screening [91% screened], n=15,322 control), 1.72% of individuals in the screening group had new AF diagnosed at one year versus 1.59% in the control group (risk difference [RD] 0.13%, 95%CI –0.16 to 0.42, p=0.38).
Effect of systematic screening on detection rates for AF
A recent systematic review and meta-analysis has demonstrated that systematic AF screening provides a greater yield of new AF diagnosis than opportunistic screening,18 and table 2 demonstrates a comparison of AF detection rates in the intervention arms of opportunistic and systematic AF screening trials.16,17,19–22
Table 2. Summary of trials using different screening methodologies – opportunistic versus systematic
| Study | Year | Inclusion criteria | Country | Methodology | AF detection rate |
| Opportunistic screening | |||||
| D2AF16 | 2020 | ≥65 years | Netherlands | Pulse palpation and blood pressure assessments followed by intermittent monitoring for 14 days | 1.6% |
| VITAL-AF17 | 2022 | ≥65 years | USA | Intermittent monitoring for 7 days | 1.7% |
| Systematic screening | |||||
| STROKESTOP22 | 2015 | 75–76 years | Sweden | Intermittent monitoring for 14 days | 3.0% |
| REHEARSE-AF19 | 2017 | >65 years with CHA2DS2-VASc ≥2 | USA | Intermittent monitoring over 12 months | 3.8% |
| mSToPS21 | 2018 | ≥65 years + a CHA2DS2-VASc or ≥75 years | USA | Continuous monitoring for 12 days | 4.2% |
| SCREEN AF20 | 2021 | >75 years + hypertension | Canada | Continuous monitoring for 12 days | 5.3% |
| Key: CHA2DS2-VASc = congestive heart failure, hypertension, age >75 (2 points), stroke/transient ischaemic attack/thromboembolism (2 points), vascular disease, age 65–74, sex category | |||||
Clinical outcomes from AF screening
In the STROKESTOP study, participants aged 75 and 76 years, with no exclusions, were randomised 1:1 to an invitation to screening or routine care (invitation to screening n=14,387; control group n=14,381).22 The screening intervention was two weeks of twice-daily intermittent single-lead ECG monitoring with a handheld device for 30 seconds. At a median follow-up of 6.9 years, the rate of composite end point events (ischaemic stroke, haemorrhagic stroke, systemic embolism, bleeding leading to hospitalisation, and all-cause mortality) was significantly lower in the invitation-to-screening group (5.45 events/100 person-years) compared with the control group (5.68 events/100 person-years) with an unadjusted hazard ratio (HR) of 0.96 (95%CI 0.92 to 1.00, p=0.045).
In the LOOP study, individuals without known AF aged 70 to 90 years, with at least one additional stroke risk factor, were randomly assigned 1:3 to implantable loop recorder (ILR) monitoring or routine care (ILR monitoring n=1,501; control group n=4,503).23 After a median follow-up of 5.4 years there was no significant difference in the risk of stroke or systemic embolism between the ILR group (67 of 1,501, 4.5%) compared with the control group (251 of 4,503, 5.6%; HR 0.80, 95%CI 0.61 to 1.05, p=0.11).
Potential harms from AF screening
The main side effect from direct anticoagulation is excessive bleeding, which can increase hospitalisation episodes, morbidity and mortality. However, in STROKESTOP there was no statistical difference in hospitalisation from major bleeding between intervention and control groups.22
Cost-effectiveness of AF screening
Ischaemic and haemorrhagic strokes are associated with increased hospitalisation and reduced quality of life. Reviews of systematic and opportunistic screening for AF detection indicate that both were more cost-effective than routine practice for those aged 65 years and older.24 The first health-economic study using actual long-term clinical follow-up data from the STROKESTOP study, showed that systematic screening for AF was associated with both lower costs and gained quality-adjusted life-years (QALYs).25 Analysis based on the Markov cost model concluded that for every 1,000 patients undergoing screening over a three-year period, 77 life-years and 65 QALYs were gained, as well as £1.7 million lower incremental cost, as compared with current AF management.25
Targeting the population for AF screening
Low detection rates in AF screening trials will limit both the clinical and cost-effectiveness of AF screening. In the Apple Heart Study anyone in the US aged ≥22 years without known AF who owned an Apple Watch could enrol in the study.26 Of 419,297 participants enrolled, with a mean age of 41 years, only 153 had new AF diagnosed, equating to a yield of 0.0003%. The mSTOPS, REHEARSE-AF, STROKESTOP and SCREEN-AF studies demonstrated AF detection rates of 3% to 5.3% by targeting screening to older patients who had vascular risk factors.19–22 In STROKESTOP II, participants were stratified into high and low risk based on N-terminal pro-B-type natriuretic peptide (NT-proBNP) level (high risk >125 ng/L; low risk <125 ng/L).27 Patients in the high-risk group were allocated intermittent ECG monitoring for two weeks, resulting in the identification of 164 cases with AF, representing a yield of 4.4%. In the UK, Screening for Atrial Fibrillation with ECG to reduce Stroke (SAFER) is an ongoing RCT that aims to understand targeted screening further by inviting patients aged above 70 years and no previous anticoagulant to participate in intermittent digital monitoring for the purposes of new AF detection.28
Can a prediction algorithm assist in targeting screening for AF?
Multi-variable prediction algorithms can classify individuals into risk categories for incident AF based on their medical data, and could play an important role in identifying patients who are at high risk for developing AF. In the UK, 98% of the population are registered in primary care with an electronic health record (EHR), and so this data source could be used to target AF screening in the general population at scale.29 Several prediction models have been developed and/or validated in European community-based EHRs for AF prediction, but they have a number of shortfalls (table 3).30–35
Table 3. AF prediction models
| Algorithm | Study aim | EHR cohort (country) | Age eligibility, years | Discrimination, c-statistic | Follow-up, years |
| Models originally derived for another purpose but tested for prediction of incident AF | |||||
| CHA2DS2-VASc30 | EV | Nivel-PCD (NL) | ≥40 | 0.669 | 5 |
| Machine-learning models | |||||
| CPRD31,32 | D | CPRD (UK) | ≥30 | 0.827 | 11 |
| EV | Discover (UK) | ≥30 | 0.870 | 8 | |
| Regression models derived in EHRs | |||||
| C2HEST33 | EV | DCRS, DNPR, DPR (DK) | 65 | 0.588 | 5 |
| 70 | 0.594 | ||||
| 75 | 0.593 | ||||
| InGef34 | D | InGef (G) | ≥45 | 0.829 | 1 |
| Regression models derived in a prospective cohort design | |||||
| CHARGE-AF31 | EV | CPRD (UK) | ≥30 | 0.725 | 11 |
| FIND-AF35 | EV | CPRD (UK) | ≥30 | 0.824 | |
| Key: AF = atrial fibrillation; CHADS2 = congestive heart failure, hypertension, age >75, diabetes mellitus, prior stroke or transient ischaemic attack (2 points); CHA2DS2-VASc = congestive heart failure, hypertension, age >75 (2 points), stroke/transient ischaemic attack/thromboembolism (2 points), vascular disease, age 65–74, sex category; CHARGE-AF = Cohorts for Heart and Aging Research in Genomic Epidemiology; C2HEST = coronary artery disease/chronic obstructive pulmonary disease (1 point each), hypertension, elderly (age ≥75, 2 points), systolic heart failure, thyroid disease (hyperthyroidism); ClalitHS = Clalit Health Services; CPRD = Clinical Practice Research Datalink; D = derivation; DCRS = Danish civil registration system; DK = Denmark; DNPR = Danish national patient register; DPR = Danish prescription register; EHR = electronic health record; EV = external validation; FIND-AF = Future Innovations in Novel Detection – Atrial Fibrillation; G = Germany; InGef = Institute for Applied Health Research Berlin; Nivel-PCD = Netherlands Institute for Health Services Research Primary Care Database; NL = Netherlands; UK = United Kingdom | |||||
First, many algorithms show only moderate discriminative performance, that is, the ability to distinguish between individuals who will and will not experience the condition. Second, algorithm prediction horizons are often five or 10 years, making it difficult to judge the merits of investigating individuals in the short term. Third, many algorithms often require variables frequently missing from routinely collected data, such as height, weight and blood pressure – thereby, restricting the population to which they can be applied.
The Future Innovation in Novel Detections of Atrial Fibrillation (FIND-AF) model predicts AF occurrence within the next six months for individuals, is accurate (c-statistic 0.824), and is scalable throughout UK primary care EHRs because it does not require observations or laboratory measures.35 The effectiveness of risk-based AF screening using FIND-AF is being tested in a British Heart Foundation (BHF) funded study in primary care (BHF Bristol Myers Squibb Cardiovascular Catalyst Award – CC/22/250026). Risk prediction calculation using FIND-AF will incorporate baseline demographics, such as age, sex and ethnicity, as well as past medical history, including hypertension, diabetes mellitus, vascular disease, left ventricle systolic dysfunction, stroke, chronic obstructive pulmonary disease (COPD), hyperthyroidism and valvular heart disease.
Conclusion
AF is common, yet often undetected, leading to potentially avoidable adverse clinical events. Systematic screening for AF is feasible and increases rates of AF detection and initiation of oral anticoagulation compared with routine care. For AF screening to be effective, it may require further refinement of the target population.
Key messages
- Atrial fibrillation prevalence is rising with increasing morbidity outcomes
- Use of different detection modalities for atrial fibrillation is gaining more interest
- Introduction of prediction algorithms can be considered to bridge the gap between yield detection and cost-effectiveness of a future screening programme for atrial fibrillation
Conflicts of interest
CPG reports personal fees from AstraZeneca, Amgen, Bayer, Boehringer-Ingelheim, Daiichi Sankyo, Vifor, Pharma, Menarini, Wondr Medical, Raisio Group and Oxford University Press. He has received educational and research grants from BMS, Abbott Inc., the British Heart Foundation, National Institute for Health Research, Horizon 2020, and from the European Society of Cardiology, outside the submitted work. AW and RN: none declared.
Funding
None.
References
1. Hindricks G, Potpara T, Dagres N et al. 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;42:373–498. https://doi.org/10.1093/eurheartj/ehaa612
2. Benjamin EJ, Muntner P, Alonso A et al. Heart disease and stroke statistics – 2019 update: a report from the American Heart Association. Circulation 2019;139:e56–e528. https://doi.org/10.1161/CIR.0000000000000659
3. Chugh SS, Havmoeller R, Narayanan K et al. Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 study. Circulation 2014;129:837–47. https://doi.org/10.1161/CIRCULATIONAHA.113.005119
4. Wu J, Nadarajah R, Nakao YM et al. Temporal trends and patterns in atrial fibrillation incidence: a population-based study of 3.4 million individuals. Lancet Reg Health Eur 2022;17:100386. https://doi.org/10.1016/j.lanepe.2022.100386
5. National Institute for Health and Care Excellence. Atrial fibrillation: diagnosis and management. London: NICE, 2021. Available from: https://www.nice.org.uk/guidance/ng196
6. Ruff CT, Giugliano RP, Braunwald E et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet 2014;383:955–62. https://doi.org/10.1016/S0140-6736(13)62343-0
7. Cowan JC, Wu J, Hall M, Orlowski A, West RM, Gale CP. A 10 year study of hospitalized atrial fibrillation-related stroke in England and its association with uptake of oral anticoagulation. Eur Heart J 2018;39:2975–83. https://doi.org/10.1093/eurheartj/ehy411
8. NHFA CSANZ Atrial Fibrillation Guideline Working Group, Brieger D, Amerena J et al. National Heart Foundation of Australia and the Cardiac Society of Australia and New Zealand: Australian clinical guidelines for the diagnosis and management of atrial fibrillation 2018. Heart Lung Circ 2018;27:1209–66. https://doi.org/10.1016/j.hlc.2018.06.1043
9. US Preventive Services Task Force, Davidson KW, Barry MJ et al. Screening for atrial fibrillation: US Preventive Services Task Force recommendation statement. JAMA 2022;327:360–7. https://doi.org/10.1001/jama.2021.23732
10. Chan NY, Orchard J, Agbayani MJ et al. 2021 Asia Pacific Heart Rhythm Society (APHRS) practice guidance on atrial fibrillation screening. J Arrhythm 2022;38:31–49. https://doi.org/10.1002/joa3.12669
11. Gladstone DJ, Sharma M, Spence JD; Embrace Steering Committee and Investigators. Cryptogenic stroke and atrial fibrillation. N Engl J Med 2014;371:1260. https://doi.org/10.1056/NEJMc1409495
12. National Health Service. NHS Long Term Plan – cardiovascular disease. London: NHS, 2019. Available from: https://www.longtermplan.nhs.uk/online-version/chapter-3-further-progress-on-care-quality-and-outcomes/better-care-for-major-health-conditions/cardiovascular-disease/
13. Schnabel RB, Marinelli EA, Arbelo E et al. Early diagnosis and better rhythm management to improve outcomes in patients with atrial fibrillation: the 8th AFNET/EHRA consensus conference. Europace 2023;25:6–27. https://doi.org/10.1093/europace/euac062
14. UK National Screening Committee. Adult screening programme: atrial fibrillation. Available at: https://view-health-screening-recommendations.service.gov.uk/atrial-fibrillation/
15. National Health Service. When you’ll be invited for breast screening and who should go. Available at: https://www.nhs.uk/conditions/breast-screening-mammogram/when-youll-be-invited-and-who-should-go/
16. Uittenbogaart SB, Verbiest-van Gurp N, Erkens PM et al. Detecting and Diagnosing Atrial Fibrillation (D2AF): study protocol for a cluster randomised controlled trial. Trials 2015;16:478. https://doi.org/10.1186/s13063-015-1006-5
17. Lubitz SA, Atlas SJ, Ashburner JM et al. Screening for atrial fibrillation in older adults at primary care visits: VITAL-AF randomized controlled trial. Circulation 2022;145:946–54. https://doi.org/10.1161/CIRCULATIONAHA.121.057014
18. Whitfield R, Ascencao R, da Silva GL, Almeida AG, Pinto FJ, Caldeira D. Screening strategies for atrial fibrillation in the elderly population: a systematic review and network meta-analysis. Clin Res Cardiol 2023;112:705–15. https://doi.org/10.1007/s00392-022-02117-9
19. Halcox JPJ, Wareham K. Response by Halcox and Wareham to letter regarding article, “Assessment of Remote Heart Rhythm Sampling Using the AliveCor Heart Monitor to Screen for Atrial Fibrillation: The REHEARSE-AF Study”. Circulation 2018;137:2193–4. https://doi.org/10.1161/CIRCULATIONAHA.118.033773
20. Gladstone DJ, Wachter R, Schmalstieg-Bahr K et al. Screening for atrial fibrillation in the older population: a randomized clinical trial. JAMA Cardiol 2021;6:558–67. https://doi.org/10.1001/jamacardio.2021.0038
21. Steinhubl SR, Waalen J, Edwards AM et al. Effect of a home-based wearable continuous ECG monitoring patch on detection of undiagnosed atrial fibrillation: the mSToPS randomized clinical trial. JAMA 2018;320:146–55. https://doi.org/10.1001/jama.2018.8102
22. 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;398:1498–506. https://doi.org/10.1016/S0140-6736(21)01637-8
23. Svendsen JH, Diederichsen SZ, Hojberg S et al. Implantable loop recorder detection of atrial fibrillation to prevent stroke (the LOOP study): a randomised controlled trial. Lancet 2021;398:1507–16. https://doi.org/10.1016/S0140-6736(21)01698-6
24. Moran PS, Teljeur C, Ryan M, Smith SM. Systematic screening for the detection of atrial fibrillation. Cochrane Database Syst Rev 2016;2016(6):CD009586. https://doi.org/10.1002/14651858.CD009586.pub3
25. Lyth J, Svennberg E, Bernfort L et al. Cost-effectiveness of population screening for atrial fibrillation: the STROKESTOP study. Eur Heart J 2023;44:196–204. https://doi.org/10.1093/eurheartj/ehac547
26. Perez MV, Mahaffey KW, Hedlin H et al. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med 2019;381:1909–17. https://doi.org/10.1056/NEJMoa1901183
27. Engdahl J, Svennberg E, Friberg L et al. Stepwise mass screening for atrial fibrillation using N-terminal pro B-type natriuretic peptide: the STROKESTOP II study design. Europace 2017;19:297–302. https://doi.org/10.1093/europace/euw319
28. Williams K, Modi RN, Dymond A et al. Cluster randomised controlled trial of screening for atrial fibrillation in people aged 70 years and over to reduce stroke: protocol for the pilot study for the SAFER trial. BMJ Open 2022;12:e065066. https://doi.org/10.1136/bmjopen-2022-065066
29. Nadarajah R, Wu J, Hogg D et al. Prediction of short-term atrial fibrillation risk using primary care electronic health records. Heart 2023;109:1072–9. https://doi.org/10.1136/heartjnl-2022-322076
30. Himmelreich JCL, Veelers L, Lucassen WAM et al. Prediction models for atrial fibrillation applicable in the community: a systematic review and meta-analysis. Europace 2020;22:684–94. https://doi.org/10.1093/europace/euaa005
31. Hill NR, Ayoubkhani D, McEwan P et al. Predicting atrial fibrillation in primary care using machine learning. PLoS One 2019;14:e0224582. https://doi.org/10.1371/journal.pone.0224582
32. Sekelj S, Sandler B, Johnston E et al. Detecting undiagnosed atrial fibrillation in UK primary care: validation of a machine learning prediction algorithm in a retrospective cohort study. Eur J Prev Cardiol 2021;28:598–605. https://doi.org/10.1177/2047487320942338
33. Lip GYH, Skjoth F, Nielsen PB, Larsen TB. Evaluation of the C(2)HEST risk score as a possible opportunistic screening tool for incident atrial fibrillation in a healthy population (from a nationwide Danish cohort study). Am J Cardiol 2020;125:48–54. https://doi.org/10.1016/j.amjcard.2019.09.034
34. Schnabel RB, Witt H, Walker J et al. Machine learning-based identification of risk-factor signatures for undiagnosed atrial fibrillation in primary prevention and post-stroke in clinical practice. Eur Heart J Qual Care Clin Outcomes 2022;9:16–23. https://doi.org/10.1093/ehjqcco/qcac013
35. Nadarajah R, Wu J, Frangi AF, Hogg D, Cowan C, Gale C. Predicting patient-level new-onset atrial fibrillation from population-based nationwide electronic health records: protocol of FIND-AF for developing a precision medicine prediction model using artificial intelligence. BMJ Open 2021;11:e052887. https://doi.org/10.1136/bmjopen-2021-052887
