Intervention Review

You have free access to this content

Effectiveness of systematic screening for the detection of atrial fibrillation

  1. Patrick S Moran1,*,
  2. Martin J Flattery1,
  3. Conor Teljeur1,
  4. Mairin Ryan1,
  5. Susan M Smith2

Editorial Group: Cochrane Heart Group

Published Online: 30 APR 2013

Assessed as up-to-date: 22 AUG 2012

DOI: 10.1002/14651858.CD009586.pub2


How to Cite

Moran PS, Flattery MJ, Teljeur C, Ryan M, Smith SM. Effectiveness of systematic screening for the detection of atrial fibrillation. Cochrane Database of Systematic Reviews 2013, Issue 4. Art. No.: CD009586. DOI: 10.1002/14651858.CD009586.pub2.

Author Information

  1. 1

    Health Information and Quality Authority, Health Technology Assessment, Dublin, Dublin, Ireland

  2. 2

    Royal College of Surgeons, Department of General Practice, Dublin, Ireland

*Patrick S Moran, Health Technology Assessment, Health Information and Quality Authority, George's Court, George's Lane, Smithfield, Dublin, Dublin, D7, Ireland. patrick.s.moran@gmail.com.

Publication History

  1. Publication Status: New
  2. Published Online: 30 APR 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Hobbs 2005

MethodsMulti-centre cluster-randomised controlled trial involving 50 (computerised) primary care centres across the West Midlands, UK, over a 12 month period. Randomisation was stratified by levels of deprivation (Townsend quartiles) and practice size. A subsidiary trial was embedded in the intervention arm, comparing two different screening strategies. Overall time period was from October 2001 to February 2003.


ParticipantsMale and female patients over 65 years of age attending general practices in the UK.
Age range was 65 - 98 years, average age of 73.5 years.
A random sample of 10,000 patients from the intervention group were allocated randomly to either systematic or opportunistic screening. Randomisation was stratified according to whether or not AF had been previously diagnosed in order to have an equal prevalence of known AF on both arms.
A random sample of 5000 was selected from the control population. After sampling, lists were returned to practices to remove those who had died, moved or were terminally ill. These were replaced from a back-up list which had been randomised at the same time as the initial list.

Final number of participants in control arm = 4963 from 25 general practices.

Final number of participants in intervention arms = 4933 for opportunistic screening and 4933 for systematic screening from 25 general practices.

Baseline AF prevalence in the control population higher than in the intervention populations (7.9% versus 6.9%)


InterventionsTraining:

Staff at primary care centres in the intervention arms were given training on the importance of AF detection, available treatment options, and were encouraged to consider opportunistic screening of patients. Staff at control centres were given no training. Practice nurses received ECG training prior to starting ECG clinics.

Systematic screening:

All patients in the systematic screening arm were sent an invitation to attend a screening clinic along with an information sheet. Non-responders were sent a reminder.

Opportunistic screening:

Patients in the opportunistic screening arm had their records flagged to encourage staff to undertake pulse recordings during routine consultation. Patients who had an irregular pulse were given an information sheet and invited to attend a screening clinic.

Screening clinics:

Screening clinics were run by practice nurses, who took patient histories, checked radial pulse rate and whether it was regular or irregular, and recorded a 12-lead ECG. The patient was then asked to complete a questionnaire on the acceptability of the intervention. All 12-lead ECGs were sent to two cardiologists for reporting. If there was disagreement over the diagnosis a third cardiologist decided. Patients were informed of the results within two weeks.


OutcomesPrimary outcomes:

New cases of atrial fibrillation detected within the 12 month study period

Incremental cost per case detected

Secondary outcomes:

Cost-effectiveness of screening in the UK

Community prevalence and incidence of AF

Acceptability of AF screening and patient uptake


FundingThis research was funded by the NHS research and development health technology assessment programme (No 96/22/11)


NotesIntention-to-treat analysis was performed, patients who already had a diagnosis of AF excluded from the calculation of newly detected cases.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskProbably done for control and intervention groups "After stratification for practice size and deprivation (based on Townsend score), we used MINITAB to select randomly two equal size groups from those practices within a particular stratum. We used a simulated value from a Bernoulli distribution, comprising two values equally likely to occur, to determine which group became the intervention arm (the other being the control arm)"

Also probably done for embedded trial within the intervention arm: "We used SPSS to allocate patients randomly from this list to either systematic or opportunistic screening to create two equal size groups of patients within each stratum so that each strategy (systematic or opportunistic screening) had an equal chance of detecting known, unknown, and suspected atrial fibrillation (n=4933). Which group then became the systematic arm (the other being opportunistic) was again decided by using a simulated value from a Bernoulli distribution, comprising two values equally likely to occur."

Allocation concealment (selection bias)Unclear riskThe authors state that "there was no deliberate concealment of allocation to the trial arms...the trial statistician determined allocation, which was implemented by the trial coordinator". However the clusters (GP practices) were identified and recruited before randomisation was conducted, so allocation was concealed from the people providing permission for the cluster to be included in the trial. Similarly, patients in the intervention arm were identified and randomly allocated into two groups before it was known to anyone involved in the trial which group would be allocated to which treatment (opportunistic or systematic), since this was decided at the end of the randomisation process using a simulated value from a Bernoulli distribution, comprising two values equally likely to occur. However since there was no deliberate attempt to conceal allocation it is unclear to what extent a risk of selection bias might have arisen from practices in the intervention arm knowing they were in the intervention arm and not the control arm prior to the recruitment of participants.

Blinding of participants and personnel (performance bias)
All outcomes
High riskIt was not possible to blind patients, who were notified by letter that they were being offered the opportunity to participate in an AF screening clinic or were encouraged to have their pulse recorded during routine consultation. Neither were primary care physicians and healthcare staff blinded, since the intervention arm received training where they were informed of the importance of detecting AF and its treatment. Practice nurses at screening clinics who took the patients' medical history, pulse and ECG were probably not blinded to whether the patient came from the systematic or opportunistic arm. Blinding is not feasible in a situation where well informed patients who need to decide whether of not they want to avail of screening are a key component of the systematic screening intervention. However since inability to blind a study is not equal to a blinded study, it is classified as high risk. Screening clinics were used to test patients from each group according to the same protocol and with the aid of a 12-lead ECG machine (Biolog).

Blinding of outcome assessment (detection bias)
All outcomes
Low riskBlinding was performed where possible; cardiologists who interpreted the 12-lead ECG reading in order to make a diagnosis of AF were blinded as to the allocation of the patient from whom the ECG was taken.

Incomplete outcome data (attrition bias)
All outcomes
Low riskAfter random sampling to identify participants from the cluster-randomised primary care centres, general practices were contacted to exclude people who had died, moved away or were terminally ill. These exclusions were randomly filled from a reserve list of 10% of the practice patients which was randomised at the same time as the original list. Immediately prior to sending screening invitations or flagging notes, the general practices were again contacted to exclude people who had since died, moved or were terminally ill and these exclusions were not replaced, with the numbers in each arm reported. The primary outcome was calculated taking the original figure using an ITT approach. Patients within each group who already had a diagnosis of AF were excluded from the calculation of the primary outcome (new cases detected). This necessitated a review of patient record to identify those with a pre-existing diagnosis. Records for some people in each of the groups were missing and are reported for each group individually. Both patients with AF and those with missing notes were excluded from the calculation of the rate of new cases detected.

Selective reporting (reporting bias)Low riskAll outcomes specified in the trial protocol were reported

Other biasUnclear riskThere is the potential for recruitment bias and contamination in the study. Recruitment bias could have been introduced at the stage where general practitioners were asked to exclude unsuitable patients from the opportunistic and systematic screening arms within the intervention group. Advice was given to exclude those who had died, moved away or were terminally ill from both groups. People who were excluded at this stage were replaced from a back-up list of patients that had been randomised at the same time as the groups. No data are provided about how many from each group were replaced at this stage, nor is the breakdown of the reasons for their exclusion given. Immediately prior to the intervention GPs were again asked to exclude any patients who had died, moved away or were terminally ill from both groups. Data concerning exclusions at this stage are reported and there was a considerable difference in the numbers excluded between the two arms; 500 were excluded from the systematic screening arm (10% of the total) and 195 (4% of the total). However the individual reasons for exclusion from the systematic screening arm are also reported and only a small minority of these (9 people, 0.2% of total) were deemed unsuitable, as opposed to having died or moved away (491people, 9.9% of total). An ITT analysis was performed that included patients that were removed from the intervention group at this stage in the calculation of the primary outcome.

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

ACTRN12612000406808Ineligible study design - no control group. This is an ongoing non-randomised registered trial where community pharmacists will screen members of the general public for atrial fibrillation using a combination of a manual pulse check and a handheld single-lead ECG (using the AliveCor Heart Monitor for iPhone). This will be a once off screening of approximately 5-10 minutes duration. Following screening, the pharmacist will contact the participant's GP via letter, stating the provisional diagnosis. A cardiologist will review all of the single-lead ECG recordings to ensure the pharmacists interpretation is correct. The GP will be further contacted by the research team if the diagnosis is other than reported by the pharmacist. The screening trial will be conducted over a 6-month period.

Baxter 1998Ineligible study design - this was a pilot study of self screening for AF in an older population (age range 55-75 years). No controls were used and irregular pulse readings were not confirmed using ECG. Communication with corresponding author indicated that the study had not been continued further following this publication.

Claes 2012Ineligible study design - no control group. This report describes a study where "patients over 40 years were invited through different channels (TV, radio, journals, web site, posters, leaflets) for a free screening in 69 hospitals allocated over Belgium during one week. After filling in a question on their personal history of AF, they had to fill in a questionnaire about their CHAD2-score. Afterwards a one channel ECG was taken using a versatile Heart Scan Device (Omron HCG-801-E©) by a trained nurse or a physician. If the ECG was positive for AF the patient was referred to their physician for follow-up." No control group receiving routine care was included or no time series data were recorded to examine the effect of the intervention compared with no screening. 10,758 people over 40 years participated, resulting in 167 new diagnoses of AF. When calculated on the basis of those who responded to the media campaign the detection rate for new cases of AF is approximately 1.56%. It is not possible to calculate the rate of detection based on the total number of people who received an invitation.

DeRuijter 2008Ineligible study design - this was a prospective cohort study to evaluate whether routinely performed ECGs in older people from the general population have added value for cardiovascular risk management beyond the information that is already available from their medical records.

Ho 2004Ineligible study design - no controls. In this study "500 subjects were drawn by two-stage random sampling from 5,002 subjects aged 70 years and over living at home. Subjects were screened for atrial fibrillation and left ventricular systolic dysfunction using electrocardiography and echocardiography." This was a prevalence study with no data on the effect of screening compared with routine care.

Hoefman 2005Ineligible population - participants in this study were consecutive patients presenting with unexplained symptoms suggestive of arrhythmia.

Johnson 2010Ineligible study design - no controls. This conference abstract describes a study that employed a strategy of random screening in a public venue (an inner city public market) to determine stroke risk. However no controls were used and results were not compared to multiple time points pre and post-intervention. No diagnoses of AF were made.

Maeda 2004Ineligible study design - this was an economic evaluation which modelled the clinical outcomes and costs associated with a screening programme in Japan.

Marek 2011Ineligible study design - this a retrospective cohort study of large-scale electrocardiographic screening of young adults.

Morgan 2002Ineligible study design - no controls. This was a randomised trial comparing two different screening strategies. Patients were randomised either to nurse led screening or to prompted opportunistic case finding. Irregular pulses found during opportunistic screening did not need to be confirmed by ECG. The study was carried out over a 6-month period. Uptake in the systematic screening arm was 73%, compared to 29% (for pulse palpation alone) in the opportunistic arm. The detection rate of new cases of AF in the systematic arm was 0.8%, compared with 0.4% in the opportunistic arm.

Munschauer 1999Ineligible study design - this study was designed to determine whether individuals taken from the general community could be taught to find and classify the pulse of another as very irregular, implying AF, or regular, implying normal sinus rhythm (NSR). No data on the effectiveness of a screening programme compared to routine practice were reported.

Somerville 2000Ineligible study design - this study compared different methods of identifying cases in general practice using patients over 65 recruited from a general practice. 56% of invitees accepted an invitation for testing (86/154) but no data were reported on the rate of detection of new cases. The study was not designed to investigate the effect of screening compared with routine practice.

Sudlow 1998Ineligible study design - no controls. This study compares three methods of diagnosing AF in a sample of 1235 over 65's invited from 9 general practices in the UK. The three methods of screening used were 1) checking for a digoxin prescription 2) pulse palpation and 3) limb lead ECG. Response rate was 74% (916/1235). No data on rate of new diagnoses were reported.

Wheeldon 1998Ineligible study design - all patients over 65 years of age from a single primary care practice with 4 GPs were invited to attend for a 12-lead ECG to detect AF. An uptake rate of 85% was achieved (1207/1422). The overall detection rate of new cases of AF was approximately 0.4%.

Wright 2007Ineligible study design - no controls. This study randomised primary care centres to either implementing AF or TIA guidelines. The type of AF testing that was associated with the AF guidelines was unclear but effects on the rate of diagnosis of new cases of AF was reported. However there was no control arm receiving routine care.

 
Characteristics of ongoing studies [ordered by study ID]
NCT01291953

Trial name or titleEffectiveness of Early Detection of Atrial Fibrillation (FAMDAP)

MethodsMulti-centre cluster-randomised controlled trial. Primary care centre professionals will be randomised to either the intervention group or a control group involving routine practice.

ParticipantsMen and women over 65 years of age who are attending a primary care centre for any reason. Patients with a prior diagnosis of AF will be excluded. Anticipated enrolment of 12,870 participants.

InterventionsOpportunistic screening of people aged 65 year or more presenting at primary care services. Opportunistic screening will involve pulse taking and requesting an ECG if an irregular pulse is found.

OutcomesPrimary outcome is the numbers of new diagnoses of AF using opportunistic screening versus routine practice.

Starting dateJanuary 2011

Contact informationPrincipal Investigator: Luis Angel Pérula de Torres, Andalusian Health Service, langel.perula.sspa@juntadeandalucia.es

NotesTrial ongoing, protocol due to be submitted for publication. Estimated completion in late 2012.

ClinicalTrials.gov identifier: NCT01593553

NCT01593553

Trial name or titleSystematic ECG Screening for Atrial Fibrillation Among 75 Year Old Subjects in the Region of Stockholm and Halland, Sweden

MethodsRandomised controlled trial

ParticipantsMen and women 75-76 years of age living in the region of Stockholm or Halland. Anticipated enrolment of 6500 participants.

InterventionsECG screening for atrial fibrillation with intermittent ECG recording for 14 days. Introduction of anticoagulants in the case of atrial fibrillation.

OutcomesReduced incidence of stroke among 75 year old subjects.

Starting dateMarch 2012

Contact informationAnna Hollander, RN LicMedSci +46-8-51778214 anna.hollander@karolinska.se

NotesTrial ongoing, estimated completion March 2019

ClinicalTrials.gov identifier: NCT01291953

 
Comparison 1. Detection of new cases of atrial fibrillation versus routine practice

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Systematic Screening versus Routine Practice19075Odds Ratio (M-H, Fixed, 95% CI)1.57 [1.08, 2.26]

 2 Opportunistic Screening versus Routine Practice19088Odds Ratio (M-H, Fixed, 95% CI)1.58 [1.10, 2.29]

 3 Gender Subgroups (Systematic)19075Odds Ratio (M-H, Fixed, 95% CI)1.56 [1.08, 2.26]

    3.1 Men
13838Odds Ratio (M-H, Fixed, 95% CI)2.68 [1.51, 4.76]

    3.2 Women
15237Odds Ratio (M-H, Fixed, 95% CI)0.98 [0.59, 1.62]

 4 Age Subgroups (Systematic)19075Odds Ratio (M-H, Fixed, 95% CI)1.58 [1.09, 2.29]

    4.1 Aged 65-74 years
15034Odds Ratio (M-H, Fixed, 95% CI)1.62 [0.90, 2.91]

    4.2 Aged >74 years
14041Odds Ratio (M-H, Fixed, 95% CI)1.56 [0.97, 2.50]

 5 Gender Subgroups (Opportunistic)19088Odds Ratio (M-H, Fixed, 95% CI)1.58 [1.10, 2.29]

    5.1 Men
13821Odds Ratio (M-H, Fixed, 95% CI)2.33 [1.29, 4.19]

    5.2 Women
15267Odds Ratio (M-H, Fixed, 95% CI)1.20 [0.74, 1.93]

 6 Age Subgroups (Opportunistic)19088Odds Ratio (M-H, Fixed, 95% CI)1.61 [1.12, 2.33]

    6.1 Aged 65-74 years
15100Odds Ratio (M-H, Fixed, 95% CI)1.63 [0.91, 2.92]

    6.2 Aged > 74 years
13988Odds Ratio (M-H, Fixed, 95% CI)1.60 [1.00, 2.57]

 
Comparison 2. Detection of new cases of atrial fibrillation versus other screening

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Systematic versus Opportunistic Screening19137Odds Ratio (M-H, Fixed, 95% CI)0.99 [0.72, 1.37]

 2 Gender Subgroups19137Odds Ratio (M-H, Fixed, 95% CI)0.99 [0.71, 1.36]

    2.1 Men
13899Odds Ratio (M-H, Fixed, 95% CI)1.15 [0.74, 1.79]

    2.2 Women
15238Odds Ratio (M-H, Fixed, 95% CI)0.82 [0.50, 1.33]

 3 Age Subgroups19137Odds Ratio (M-H, Fixed, 95% CI)0.98 [0.71, 1.36]

    3.1 Aged 65-74 years
15190Odds Ratio (M-H, Fixed, 95% CI)0.99 [0.60, 1.64]

    3.2 Aged > 74 years
13947Odds Ratio (M-H, Fixed, 95% CI)0.97 [0.64, 1.48]

 
Summary of findings for the main comparison. Screening versus routine practice for the detection of atrial fibrillation

Screening versus routine practice for the detection of atrial fibrillation

Patient or population: patients with the detection of atrial fibrillation
Settings:
Intervention: screening
Comparison: routine practice

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Routine practiceScreening

Systematic Screening versus Routine Practice
Number of new diagnoses
Follow up: 12 months
Study populationOR 1.56
(1.08 to 2.24)
9075
(1 study)
⊕⊕⊕⊝
moderate1

10 per 100016 per 1000
(11 to 23)

Moderate

10 per 100016 per 1000
(11 to 22)

Opportunistic Screening versus Routine Practice
Number of new diagnoses
Follow up: 12 months
Study populationOR 1.57
(1.1 to 2.26)
9088
(1 study)
⊕⊕⊕⊝
moderate1

10 per 100016 per 1000
(11 to 23)

Moderate

10 per 100016 per 1000
(11 to 22)

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Given the nature of the intervention it was not possible to blind the participants in this study. No deliberate attempt to conceal allocation was made but failure to do this is not judged to introduce a risk of selective enrolment.
 
Summary of findings 2. Systematic screening compared to opportunistic screening for the detection of atrial fibrillation

Systematic screening compared to opportunistic screening for the detection of atrial fibrillation

Patient or population: patients with the detection of atrial fibrillation
Settings:
Intervention: systematic screening
Comparison: opportunistic screening

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Opportunistic screeningSystematic screening

Systematic versus Opportunistic Screening
Number of new diagnoses
Follow up: 12 months
Study populationOR 0.99
(0.72 to 1.36)
9137
(1 study)
⊕⊕⊕⊝
moderate1

16 per 100016 per 1000
(12 to 22)

Moderate

16 per 100016 per 1000
(12 to 22)

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Given the nature of the intervention it was not possible to blind the participants in this study. No deliberate attempt to conceal allocation was made but failure to do this is not judged to introduce a risk of selective enrolment.
 
Table 1. Number of new AF cases detected through screening versus routine practice

GenderAge GroupTotal




MenWomen65 - 7475+

Systematic Screening44/195830/260430/256244/200074/4562

Opportunistic Screening38/194137/263431/262844/194775/4575

Routine Practice16/188031/263318/247229/204147/4513

 Data taken from Hobbs 2005 (reported in Fitzmaurice 2007)
 
Table 2. Uptake of screening

GroupSystematic ScreeningOpportunistic Screening

All53%46%

Men57%49%

Women50%41%

Aged 65 - 7461%49%

Aged 75+ age43%42%

 Rates of uptake of screening based on data reported in Hobbs 2005. Rate of uptake of opportunistic screening is based on those who consented to have their pulse taken AND undergo an ECG if an irregular pulse was found.
 
Table 3. Prevalence data (by gender, age group)

 MenWomen 

Total



 Group65 - 7475 - 84≥ 8565 - 7475 - 84≥ 85

Baseline Prevalence 

Control74/1216 (6.1)84/703 (11.9)25/156 (16.0)44/1378 (3.2)106/1050 (10.1)56/420 (13.3)389/4923 (7.9)

Opportunistic70/1304 (5.4)63/650 (9.7)24/148 (16.2)48/1448 (3.3)91/1005 (9.1)44/375 (11.7)340/4930 (6.9)

Systematic69/1318 (5.2)67/647 (10.4)15/154 (9.7)68/1391 (4.9)70/1022 (6.8)50/396 (12.6)339/4928 (6.9)

12 month prevalence

Control81/1213 (6.7)91/699 (13.0)27/151 (17.9)55/1377 (4.0)122/1044 (11.7)60/418 (14.4)436/4902 (8.9)

Opportunistic90/1303 (6.9)77/647 (11.9)28/148 (18.9)59/1443 (4.1)109/1001 (10.9)52/373 (13.9)415/4915 (8.4)

Systematic90/1312 (6.9)82/643 (12.8)23/154 (14.9)77/1387 (5.6)88/1012 (8.7)53/398 (13.5)413/4906 (8.4)

 Data taken from Hobbs 2005 (reported in Fitzmaurice 2007), Figures are number (percentages)