Validity of daily self-pulse palpation for atrial fibrillation screening in patients 65 years and older: A cross-sectional study.

BACKGROUND: The European Society of Cardiology guidelines recommend (Class IA) single-time-point screening for atrial fibrillation (AF) using pulse palpation. The role of pulse palpation for AF detection has not been validated against electrocardiogram (ECG) recordings. We aimed to study the validity of AF screening using self-pulse palpation compared with an ECG recording conducted at the same time using a handheld ECG 3 times a day for 2 weeks. METHODS AND
FINDINGS: In this cross-sectional screening study, patients 65 years of age and older attending 4 primary care centers (PCCs) outside Stockholm County were invited to take part in AF screening from July 2017 to December 2018. Patients were included irrespective of their reason for visiting the PCC. Handheld intermittent ECGs 3 times per day were offered to patients without AF for a period of 2 weeks, and patients were instructed in how to take their own pulse at the same time. A total of 1,010 patients (mean age 73 years, 61% female, with an average CHA2DS2-VASc score 2.9) participated in the study, and 27 (2.7%, 95% CI 1.8%-3.9%) new cases of AF were detected. Anticoagulants (ACs) could be initiated in 26 (96%, 95% CI 81%-100%) of these cases. A total of 53,782 simultaneous ECG recordings and pulse measurements were registered. AF was verified in 311 ECG recordings, of which the pulse was palpated as irregular in 77 recordings (25%, 95% CI 20%-30% sensitivity per measurement occasion). Of the 27 AF cases, 15 cases felt an irregular pulse on at least one occasion (56%, 95% CI 35%-75% sensitivity per individual). 187 individuals without AF felt an irregular pulse on at least one occasion. The specificity per measurement occasion and per individual was (98%, 95% CI 98%-98%) and (81%, 95% CI 78%-83%), respectively.
CONCLUSIONS: AF screening using self-pulse palpation 3 times daily for 2 weeks has lower sensitivity compared with simultaneous intermittent ECG. Thus, it may be better to screen for AF using intermittent ECG without stepwise screening using pulse palpation. A limitation of this model could be the reduced availability of handheld ECG recorders in primary care centers.

Introduction

Atrial fibrillation (AF) without anticoagulant (AC) treatment is a risk factor for ischemic stroke [1]. Stroke may be the first presentation of AF [2]. Silent AF detected by continuous ECG monitoring is associated with an elevated stroke risk [3]. A nonrandomized interventional screening study for AF and treatment with ACs suggest a reduction in stroke incidence among a screened population compared with an unscreened population [4]. Thus, detection of AF and initiation of ACs may be important for preventing stroke [1]. AF manifests as paroxysmal attacks and can then progress to persistent or permanent AF [5]. Nonparoxysmal AF could be detected through a single-time–point electrocardiogram (ECG), but it is difficult to detect paroxysmal AF [1]. Thus, repeated daily ECG monitoring is probably more sensitive than single-time–point ECG in detecting AF [6].

According to the European Society of Cardiology guidelines 2016 [1], it is recommended that persons 65 years of age and older are screened for AF. This is based on screening studies [7] using single-time–point ECG with or without pulse palpation. These guidelines even consider systematic screening for AF in persons over 75 years of age based on a screening study [6] that used intermittent ECG monitoring over 2 weeks. A consensus paper [8] for AF screening showed differences in the AF detection rate depending on the screening technique and the characteristics of the screened population. However, very few studies have performed a direct comparison between pulse palpation and ECG recordings. Self-detection of AF through pulse palpation is feasible among the elderly, as shown in an interventional study conducted on an anatomical model [9]. Another study showed a high level of motivation in participants to continue taking their pulse over several weeks [10].

Our aim was to study the validity of repeated self-pulse palpation over 2 weeks compared with ECG monitoring to screen for AF among patients 65 years of age and older who were seeking primary healthcare. Finally, we aimed to evaluate the initiation rate of oral ACs for newly detected AF cases to prevent stroke.

Methods

Design

A cross-sectional screening study was performed in Swedish primary care centers (PCCs). We planned to recruit 3–6 PCCs that had shown an interest in participating. They were located outside Stockholm County. The reason for this geographical limitation was that another large AF screening study was being conducted within the Stockholm area. The study started in June 2017 in 2 PCCs, and then an additional 2 PCCs were recruited to fulfill the sample size within the planned study period of 2 years. A total of 14 PCCs were approached, and 4 PCCs participated in the study. This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (S1 Table). The protocol is described on protocols.io (https://www.protocols.io/view/validity-of-daily-self-pulse-palpation-over-two-we-m2ec8be).

Study population

Patients who, for whatever reasons, were seeking care at a PCC and who were 65 years of age or older were invited by health personnel to participate. Patients who were interested in participating were then directed to a research nurse in the PCC. No incentives were given for participation. Patients with previously known AF or ongoing oral AC treatment were excluded.

Screening procedure

One to two nurses were assigned per center in this study. The nurses had received prior training in AF and pulse palpation. No incentives or accreditation points were given for this training.

Participants received written and oral information about the study. All participants provided their consent to participate by signing and submitting a consent form before entering the study. In a 30-minute consultation with a nurse, participants completed a questionnaire (Supplement Case Report Form) about stroke risk stratification according to the CHA2DS2-VASc score. The CHA2DS2-VASc score considers congestive heart failure, hypertension, age, diabetes mellitus, previous stroke/transient ischemic attack, vascular disease, and female sex. Body weight, height, pulse, and blood pressure were measured. The nurse carefully instructed participants in the technique of radial pulse palpation and then checked the participants’ ability to perform self-pulse palpation. The participants were instructed to perform self-pulse palpation at home 3 times a day over a 2-week period, immediately followed each time by a 30-second ECG recording using a Zenicor handheld ECG with an integrated mobile transmitter (https://zenicor.se/; Stockholm, Sweden). We did not specify which time of day the participants should perform self-pulse palpation. However, we preferred them to do it at different times of the day. When handheld ECG findings indicated AF or other suspected pathological findings, the ECG was re-examined by an experienced cardiologist in order to confirm the diagnosis. AF was diagnosed as a 30-second recording with an absolutely irregular rhythm without distinct p-waves [1]. Individuals with unclear or uninterpretable ECG recordings were further investigated using patch ECG monitoring for 5 days. The Zenicor device has a button that participants can use to note when their pulse felt irregular during each ECG recording. The responsible author is an experienced family doctor who interpreted the ECG register remotely each day and phoned the participants reminding them to check their pulse and to check whether they felt any AF symptoms when the ECG showed AF. When AF was verified, the patients were contacted by phone, informed about the detected AF, and asked to consult their family doctor in order to validate the indication for AC treatment in accordance with national guidelines. Thus, the patient’s family doctor was responsible for the initiation of ACs. A follow-up was performed to assess whether the patient had been prescribed AC treatment. Participants with no detected AF returned the handheld ECG devices at the end of the screening with no scheduled follow-up and were then notified by post about their ECG results.

Statistical analyses

Categorical data were summarized by counts and percentages. For all continuous variables, visual inspection of histograms and the Shapiro–Wilk’s test were used to assess the deviation from a normal distribution. This test showed no normally distributed study data. Thus, medians with interquartile ranges were used. Fisher’s exact test was used to analyze categorical variables. Student t test or the Mann–Whitney test were used to compare continuous variables between 2 groups. Odds ratios with 95% CIs were used to test for associations between AF and risk factors. Multivariate logistic regression analyses were conducted to analyze the independent predictors for AF detection. An exponential prediction model was used for AF detection by age because this model was most suitable. For these tests, a two-sided probability value ≤ 0.05 was considered statistically significant. These analyses were performed using Stata statistical software version 10.

The sample size was calculated to show a statistically higher AF detection rate using a 2-week intermittent ECG than an AF detection rate using a single-time–point ECG on inclusion. We assumed a 1.4% AF detection rate using a single-time–point ECG according to a previous meta-analysis [7] for AF screening of patients 65 years of age and older. We expected that total AF detection using a 2-week intermittent ECG would be 3%, depending on a previous study [6] that used this screening method on 76-year–old patients. Using an alpha value of 0.05, power 0.75 to calculate the sample size for the difference in the proportion between 2 dependent groups yielded 955. Thus, we chose a sample size of at least 1,000. We chose 0.75 power because of limited resources.

For pulse palpation for detecting AF, we compared the results of each pulse palpation with the results of a simultaneous ECG recording. We constructed 2 × 2 contingency tables as we had already planned in order to enable calculation of sensitivity, specificity, positive predictive value, and negative predictive value. The planned analyses were not changed during the study, and no data-driven changes of analysis were performed. Calculations with exact Clopper–Pearson CIs were used for sensitivity and specificity, while standard logit CIs were used for the predictive values.

Ethics

This study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Stockholm (DNR 2017/3:3).

Results

Participation

Of the 46,477 individuals who were registered at 4 PCCs in 2018, 9,224 (19.85%) were 65 years of age and older. This proportion was similar to the corresponding proportion of the Swedish population (19.9%) but varied across the 4 centers (Table 1).

Table 1

Participation and AF cases across screening centers.

	PCC1	PCC2	PCC3	PCC4	Total
Individuals registered at a PCC	14,901	10,797	8,344	12,435	46,477
Number of registered individuals ≥65 years (% of all registered individuals)	2,987 (20.4)	2,065 (19.13)	1,253 (15.02)	2,919 (23.47)	9,224 (19.85)
Number of registered individuals ≥65 years who visited the PCC (% of registered individuals in this age group)	2,739 (91.7)	1,773 (85.9)	1,015 (81)	2,885 (98.8)	8,412 (91.2)
Number of registered individuals ≥65 years with known AF (% of all registered individuals in this age group)	216 (7.23)	194 (9.39)	111 (8.86)	105 (3.6)	626 (6.79)
Number of screened individuals (% of visitors in the target age without previously known AF)	702 (27.8)	89 (5.64)	104 (11.5)	115 (4.14)	1,010 (12.97)
Number of new AF cases detected (% of screened individuals)	18 (2.56)	1 (1.12)	4 (3.85)	4 (3.48)	27 (2.67)
Median age of screened patients, years	72.18	70.93	70.86	72.68	71.97
Median age of detected AF cases, years	76.58	72.66	76.18	73.08	76.41
Screening period, months	15	15	10	9	9–15

Abbreviations: AF, atrial fibrillation; PCC, primary care center.

Screening periods varied from 9 to 15 months in the 4 participating centers. The screening ended at all centers in December 2018 because the prespecified sample size had been included.

Screening was halted during the summer vacation (June and July) at all centers. Furthermore, one of the first 2 centers stopped screening for 3 months because the nurse in charge was on sick leave.

About 90% of registered individuals in the target population visited their PCC during the screening period (Table 2). The prevalence of known AF was 6.79% (95% CI 6.29%–7.32%). Between 4%–28% of all visitors without known AF were screened, with only some of these visitors being invited to be screened. This invitation to the screening depended on the capacity of the research nurse, who carried out her clinical work at the same time. No real estimate of agreement to screening was made, although the recruiting health personal considered the agreement rate to be around 80%–90%. The median age of the screened individuals was 72.1 years (IQR 68.7, 75.9), of which 61.6% were female.

Table 2

Distribution of target population.

Total number of registered individuals ≥65 years in 4 PCCs	9,224
Number of registered individuals ≥65 years who did not visit the PCCs (% of all registered individuals ≥65 years)	812 (8.8)
Number of registered individuals ≥65 years with known AF (% of all registered individuals ≥65 years)	626 (6.79)
Number of registered individuals ≥65 years without known AF who visited the PCCs (% of all registered individuals ≥65 years)	7,786 (84.41)
Number of registered individuals ≥65 years without known AF who visited the PCCs but did not participate in the screening (% of registered individuals ≥65 years without known AF who visited the PCCs)	6,776 (87.03)
Number of screened individuals (% of registered individuals ≥65 years without known AF who visited the PCCs)	1,010 (12.97)

Abbreviations: AF, atrial fibrillation; PCC, primary care center.

AF detection

A total of 1,010 individuals were screened, and 27 (2.7%, 95% CI 1.8%–3.9%) new cases of AF were detected. Only 2 new cases showed persistent AF, while the other cases were paroxysmal. The 2 cases of persistent AF and 3 other cases of paroxysmal AF were detected by the first ECG recording at inclusion. Thus, the rate ratio for AF detection by intermittent ECG compared with single ECG measurement was 5.4 (95% CI 2.3–12.6). Of the 27 new AF cases, 16 (59%) were asymptomatic. Forty-two individuals had nonconclusive ECG recordings, mainly showing a frequent atrial ectopic beat, and were further investigated using BioTelemetry ePatch continuous ECG monitoring (https://www.gobio.com/) for 5 days. AF could be verified in 4 of these individuals.

AC treatment was initiated by the patient’s family doctor in 26 of 27 new AF cases (96%, 95% CI 81%–100%). Non-vitamin-K–antagonist oral ACs were initiated in 25 cases, and one other case initially received oral ACs but then changed to low-molecular–weight heparins because of bleeding as a side effect. Only one patient was not treated, as the doctor in charge considered that the patient had a relatively low thromboembolic risk (CHA2DS2-VASc score = 1).

Table 3 shows the characteristics of newly detected AF cases with a higher age, predominantly male (70.4%) and with more prevalent heart failure. Age and male gender were independent predictors for detection of new AF cases, with an odds ratio (95% CI) of 1.14 (1.07–1.21) and 4.46 (1.9–10.43), respectively. Fig 1 shows the age prediction for detection of AF using an exponential prediction model.

Table 3

Characteristics of newly detected AF cases compared with those without AF.

	AF—27 Patients	No AF—983 Participants	P-Value
Age mean (SD)	76.8 (7)	72.7 (5)	0.0001
Age median (IQR)	76.4 (72,82)	72 (69,76)	0.0016
Female gender number (%)	8 (29.6)	614 (62.5)	0.0005
Heart failure number (%)	2 (7.4)	13 (1.3)	0.0099
Hypertension number (%)	15 (55.6)	558 (56.8)	0.9004
Diabetes mellitus number (%)	3 (11.1)	175 (17.8)	0.3680
Stroke/TIA number (%)	3 (11.1)	77 (7.8)	0.5338
Myocardial infarction/ peripheral artery disease number (%)	3 (11.1)	74 (7.5)	0.4888
CHA2DS2-VASc score median (IQR)	3 (2,4)	3 (2, 4)	0.9912
Systolic BP median (IQR)	142 (132, 166)	140 (129, 152)	0.1235
Diastolic BP median (IQR)	83 (77, 92)	83 (76, 90)	0.6001
BMI female median (IQR)	28 (25, 31)	25 (23, 29)	0.1939
BMI male median (IQR)	25 (24, 29)	26 (24, 28)	0.5783

Abbreviations: AF, atrial fibrillation; BMI, body mass index; BP, blood pressure; IQR, interquartile range; SD, standard deviation; TIA, transient ischemic attack.

Fig 1

AF detection rate by age.

AF, atrial fibrillation.

Validity of pulse palpation

A total of 53,782 simultaneous ECG recordings and pulse measurements were registered for the 1,010 screened individuals, corresponding to a median of 51 recordings/individuals. Of the 27 detected AF cases, AF was verified in 311 ECG recordings (Table 4), but the pulse was palpated as irregular in only 77 of these recordings, yielding a 25% (95% CI 20%–30%) sensitivity and a 98% (95% CI 98%–98%) specificity per measurement occasion (Table 5).

Table 4

Contingency table comparing pulse results with ECG results.

Irregular versus regular pulse	AF—27 Individuals	No AF—983 Individuals
Irregular pulse—202 individuals	15	187
Regular pulse—808 individuals	12	796
Pulse measurements	AF—311 Measurements	No AF—53,471 Measurements
Irregular pulse—1,046 measurements	77	969
Regular pulse—52,736 measurements	234	52,502
Irregular versus regular pulse	AF—5 Individuals	No AF—1,005 Individuals
Irregular pulse—25 individuals	4	21
Regular pulse—985 individuals	1	984

Abbreviations: AF, atrial fibrillation; ECG, electrocardiogram.

Table 5

Validity results of pulse palpation in detecting AF.

Variable	Pulse Palpation by Nurse on Inclusion versus Single ECG Measurement		At Least One Irregular Pulse by Individual Self-Pulse Palpation over 2 Weeks versus at Least One ECG Recording with AF		Individual Self-Pulse Palpation versus Simultaneous ECG
	Estimate	95% CI	Estimate	95% CI	Estimate	95% CI
Sensitivity, %	80	28.36–99.49	55.56	35.33–74.52	24.74	20.06–29.94
Specificity, %	97.91	96.82–98.7	80.98	78.38–83.39	98.19	98.07–98.3
Positive predictive value, %	16	9.39–25.94	7.43	5.29–10.32	7.36	6.09–8.88
Negative predictive value, %	99.9	99.42–99.98	98.51	97.75–99.02	99.56	99.53–99.58

Abbreviations: AF, atrial fibrillation; ECG, electrocardiogram.

Of these 27 AF cases, 15 cases felt an irregular pulse on at least one occasion, resulting in 56% (95% CI 35%–75%) sensitivity per individual (Table 4). Of individuals without AF, 187 felt an irregular pulse on at least one occasion, with 81% (95% CI 78%–83%) specificity, and the positive predictive value was 7%.

Pulse palpation by nurse on inclusion was irregular in 25 cases, and 5 new AF cases were detected on inclusion, in which the pulse was irregular in 4 of the cases (Table 4), yielding 80% (95% CI 28%–99%) sensitivity, 98% (95% CI 97%–99%) specificity, and 16% (95% CI 5%–36%) positive predictive value (Table 4).

Discussion

This is the first AF screening study to compare self-pulse palpation with an ECG recording. We found that self-pulse palpation had a low sensitivity and high specificity for AF detection compared with intermittent ECG. Almost all newly detected AF cases were treated with ACs.

Participation appears to depend more on the stability of health personnel and their motivation rather than the motivation of the target population, as only a minority of the care-seeking patients were invited to the screening, although the majority of the target population visited their PCC during the screening period.

Compared with the Swedish general population in the target age (http://www.statistikdatabasen.scb.se/pxweb/en/ssd/), the screened individuals were relatively younger and comprised more females (Fig 2). Thus, we would probably have detected more AF cases if the participants had corresponded more to the Swedish age and gender distribution. The prescreening prevalence of AF was higher than the AF prevalence observed in other studies [4], probably indicating better routine care with repeated ECG measurements in our studied population.

Fig 2

Age and gender distribution of the study participants compared with the Swedish population.

Our detection rate of AF cases was comparable with a 3% detection rate in a previous screening study [6] that used the same ECG technique among 75- to 76- year–old individuals, although our detection rate was lower compared to a similar study in primary care showing a 5.5% AF detection rate with a median age of 72 years. However, our participants had lower morbidities and were predominantly female. In our study, 59% of the detected AF cases were asymptomatic, compared with 75% in the previous screening study [11].

In our study, almost all of the newly detected AF cases were paroxysmal AF. This rate is higher than the paroxysmal AF rates in previous screening studies of 62.5% and 75%, respectively [11, 12]. This may indicate that the majority of nonparoxysmal AF had already been detected via regular healthcare at our screening centers. Thus, in our study, the rate ratio for AF detection by intermittent ECG compared with single ECG measurement was 5.4. Previous screening studies [6, 11] showed similar rate ratios of 4.8 and 5.3, respectively. This indicates the need for prolonged screening in order to detect more new AF cases.

Anticoagulation could be initiated by primary care physicians in almost all newly detected AF cases. This high initiation rate was similar to a previous screening study [11] in primary care.

In our study, age and male gender were independent predictors for the detection of new AF cases. This confirms previous evidence [8] that age and male gender is a strong predictor for AF.

Our analysis that compared self-pulse palpation with simultaneous ECG measurements for AF detection showed low sensitivity. Thus, pulse might be of limited use to AF screening. However, individual analysis based on repeated pulse palpation increases sensitivity when an individual detects an irregular pulse at least once over 2 weeks. Screening tests should be sensitive. A sensitivity of 56% may be not high enough to motivate stepwise screening with self-pulse palpation followed by intermittent ECG recording when the pulse is irregular. Our results showed one AF case detected in 13 individuals with at least one irregular pulse, compared with detecting one AF case in 37 individuals irrespective of pulse palpation.

Our analysis, which was based on a trained nurse checking the pulse as a single measure in the PCC, showed a higher sensitivity to AF detection on inclusion. A meta-analysis [13] of such pulse palpation showed a higher sensitivity of 92% (95% CI 85%–96%). However, it is difficult to detect most paroxysmal AF cases through such single-time measurements. In our study, we detected only 19% of all new AF cases by an initial single ECG recording. Repeated pulse palpation (even with low sensitivity) detected more AF cases than single-time–point pulse palpation (15 versus 4 cases, as shown in Table 6). Repeated self-pulse palpation at home is inexpensive and easy and requires no special equipment.

Table 6

AF detection by single-time–point measurement pulse palpation and ECG on inclusion versus repeated pulse palpation and ECG measurement over 2 weeks.

	27 AF Cases Detected by Intermittent ECG (5.4 Times than Those Detected on Inclusion)	5 AF Cases Detected at First ECG on Inclusion
Irregular pulse	15* (56%)	4** (80%)
Regular pulse	12*	1**

*Self-pulse palpation at home.

**Pulse palpation by nurse.

Abbreviations: AF, atrial fibrillation; ECG, electrocardiogram.

A meta-analysis [13] showed the highest sensitivity of single pulse measurement using a smartphone to detect AF of 97% (95% CI 95%–99%), and a recent study showed comparable sensitivity [14, 15]. Again, it is difficult to detect most paroxysmal AF using this single-time–point pulse palpation. Moreover, this screening requires the availability of a smartphone. A handheld smartphone [14, 15] single-lead ECG with an interpreting program can be used for AF screening without the need for pulse evaluation. Thus, if such a smartphone ECG were available, repeated measurements could be a sensitive screening method for AF without pulse checking.

This study has a few limitations. Firstly, PCCs were not randomly selected for recruitment. This could affect the reproducibility of our results. Secondly, a minority of patients who visited PCCs were invited to the screening, and there was no real estimate of participation among the invited patients. The study nurses recruited relatively younger and probably healthier patients. This could cause selection bias. However, the recruitment of patients with higher morbidity would result in a higher AF detection rate. Finally, ectopic heart beats could be felt as an irregular pulse, and this could reduce the specificity of pulse palpation for AF. However, specificity was high in our study.

Conclusion

AF screening using self-pulse palpation 3 times per day for 2 weeks has lower sensitivity compared with simultaneous intermittent ECG. Using such an ECG is more effective in detecting AF than a single-time–point ECG. Thus, it may be better to screen for AF using intermittent ECG without stepwise screening using pulse palpation. In the future, there is a need to conduct a randomized control screening study for stroke prevention using intermittent ECG.

STROBE Checklist.

STROBE, Strengthening the Reporting of Observational Studies in Epidemiology.

(DOCX)

Click here for additional data file.

Data reporting.

(XLSX)

Click here for additional data file.

Case reporting form.

(DOCX)

Click here for additional data file.

5 Nov 2019

Dear Dr. Ghazal,

Thank you very much for submitting your manuscript "Validity of daily self-pulse palpation over two weeks for screening for atrial fibrillation among patients 65 years of age and older seeking primary care: A cross-sectional study" (PMEDICINE-D-19-03337) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Nov 26 2019 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Title: suggest shortening to “Validity of daily self-pulse palpation for Atrial Fibrillation screening among patients 65 years of age and older: A cross-sectional study”

Abstract: first sentence perhaps should start with “XXX guidelines recommend single time-point screening for atrial fibrillation”. Presumably these are ESC guidelines?

Abstract methods and findings- please provide a brief description of which primary care centres participated. Perhaps something like “Primary care centres in stockholm”?. Please also provide brief demographics of the participants. Last sentence of this section should include a limitation of your study design

Abstract and author summary- the writing needs to reflect that the study compared the validity of self palpation and handheld ECG recordings. Since you use the word “simultaneously”, it is not clear that these two are being compared.

Introduction= please introduce AF and AC on first view in Line 78

Line 79 please rephrase to “may be important to prevent stroke” as there are other risk factors for ischaemic stroke?

Line 84-85 please rephrase to “it is recommended to screen persons 65 years of age and older for AF” for clarity

Please rephrase “head to head” to “direct comparison” or something similar

Introduction- requires further background on atrial fibrillation and the risks of undetected AF. Also provide further detail on self-palpation and whether this is a feasible method to detect AF. Please provide further detail on the 65+ age group and self palpation, along with the various modalities used to detect, diagnose AF.

Methods- Lines 103-104 can be omitted and please provide details of which primary care centres were included. Please also clarify how many PCCs were approached and how many consented to participate for greater transparency

Methods- please provide a citation to the health questionnaire referred to in Line 125 or provide a copy of this questionnaire as supplementary info. Please also provide a brief summary of what health scores from this questionnaire might mean ( scale of 1 to 5 ? or lower scores are worse than higher?)

Methods- please provide a citation to Zenicor handheld ECG if available and mention which anticoagulants were offered to patients

Line 173- please clarify the time of the summer vacation during which no screening took place

Line 197- please mention again 26 of 27 (?) new AF cases?

Fig 1- please remove pie chart and replace with table containing exact numbers and not approximations

Line 200-201- please provide further detail on why one case was not given treatment and what a score of one means

Fig 2- its not clear how the prediction of AF detection was calculation- please provide this detail in the methods and results sections as necessary

Please present and organize the Discussion as follows: a short, clear summary of the article's findings; what the study adds to existing research and where and why the results may differ from previous research; strengths and limitations of the study; implications and next steps for research, clinical practice, and/or public policy; one-paragraph conclusion. Please remove all results from the discussion and include these in the results section only.

Please remove assertions of primacy in Line 235-237

Figures 3 is hard to understand, please simplify in line with comments from Ref 1

Figure 4 should be modified into a table

Lines 256-260 require revision for clarity and grammar

Discussion-please discuss in greater detail the potential of remote ECG monitoring in at risk or general populations for AF screening using smartphone apps or other smart devices.

Discussion- you mention ectopic beats here but there is no context provided previously that the presence of these ectopics can skew the self palpation. Please provide brief summary in the introduction

Conclusion and abstract did this study involve stepwise screening? If so, this needs to be made clearer throughout the text

Data availability statement- you say the data are provided in SI files but we are unable to locate these. Please provide the data underlying tables, figures and charts as per PLOS Data policy

Please provide p values along with confidence intervals as applicable

Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis—including those made in response to peer review comments—should be identified as such in the Methods section of the paper, with rationale.

Comments from the reviewers:

Reviewer #1: I confine my remarks to statistical aspects of this paper. The basic approach is fine, but I have some issues to resolve before I can recommend publication.

Line 39 - Should "patients" here be "recordings"?

Line 150 The usual power is 0.80. Why did the authors choose 0.75? (It isn't necessarily wrong to do so, but it needs justification).

Lines 153-155 How were the CIs calculated (there are several methods).

Line 182 Give the interquartile range.

Fig 1 - pie graphs are not good. Here, a simple table would be fine. Also, why were two of the N's approximate?

Fig 3 - this is hard to read. If exact age is available, then it would be better to use two overlaid density plots - one for men and one for women - with lines for screened and general population for each. If age is only available in categories, then two mosaic plots could be presented, or these could be combined into one three-way mosaic plot.

Fig. 4 Stacked bar charts are not good and 3-D makes them worse. I'm not sure what to suggest here. Maybe just a table. Maybe a mosaic plot. Maybe a bar plot in 2D with the bars adjacent to each other.

Peter Flom

Reviewer #2: Ghazal et al have report a cross-sectional screening study in patients >65 years of age visiting primary care centers and taking part in AF screening. Screening was performed with intermittent ECG recordings three times per day for a period of 2 weeks and simultaneous pulse palpation. A total of 1010 patients participated in the study and 27 (2.7%) new cases of AF were detected in 311 ECG recordings, of which the pulse was palpated as irregular in 77 patients (25%). 187 individuals without AF felt an irregular pulse on least one occasion. The specificity per measurement occasion and per individual was (98%). They conclude that AF screening using self-pulse palpation three times daily for two weeks has lower sensitivity compared with simultaneous intermittent ECG. Thus, it may be better to screen for AF using intermittent ECG without stepwise screening using pulse palpation.

The study idea is important, since silent AF is a common problem and stroke is too often its 1st manifestation (PLOS One 2016;11:e0168010). The execution of the study is adequate and the sample size is sufficient for the present study questions. The report is generally well-written.

The weakness of this study is the unclear selection process of study patients weakening the generalizability of the results.

The sensitivity of pulse palpation was lower than expected showing that this method is not suitable for every elderly subject. In an earlier study, elderly (> 75 years of age) subjects the accuracy of pulse palpation after a training given by study nurse was: sinus rhythm 97%, extra beats 74.3%, slow AF 81.8% and fast AF 91.9% when assessed with the help of anatomic human arm model programmed with various rhythms ( Scand J Prim Health Care 2017;35:93-98). The main problem with pulse palpation is the primary motivation and ability (Parkinson, severe arthrosis in fingers etc) and later the motivation to continue the habit of regular palpation.

Did the study nurses evaluate the ability of subjects to measure their pulse after the training session?

What were the predictors for missing AF with pulse palpation?

Did the subjects really palpate the pulse and did they report heart rate during each recording?

What was the reason for misdiagnosis of AF?

Reviewer #3: This nicely executed study looks at an important issue, which to my knowledge has not previously been assessed. It aims to determine if self-performed pulse palpation is as accurate at detecting unknown AF as a hand-held ECG, over a period of 2-weeks. This is an important issue to clarify, as the economics of hand-held devices for AF detection are much higher than just educating people to take their own pulse. Overall, this is a nicely presented paper. I believe the methodology is appropriate and the study is powered sufficiently to answer the research question.

There are a few minor things in the paper that I think need to be clarified, and a few places where the English grammar may need a little editing.

1. Line 46-47 - I ma not quite sure that the last sentence of the abstract conclusion means - it reads as if you are suggesting NOT to do step-wise screening, but I do not think that is the case

2. Line 60 - "who searched care" - should this be "who sought care"

3. Line 78 - The initial use of the acronyms AF and AC are not defined on their first use in the body of the manuscript. Additionally, it is more common for OAC to be used rather than AC

4. Line 102 states you recruited in Stockholm county and then line 104 states you screened OUTSIDE Stockholm county

5. Lines 118- 130 - More information is required in the methods section under the screening protocol to describe: how the participants recorded the pulse palpation results; how the ECG and pulse palpation results were assessed to determine AF or no AF; and what other tests were done; how symptoms were determined and recorded; what follow up occurred if AF was identified and how the patient was advised they had AF; who determined the treatment; and how OAC was prescribed

6. Line 147 - there is an updated meta-analysis of screening published by the same authors which may be better to quote instead of reference 4. DOI: 10.1371/journal.pmed.1002903

7. Line 166-167 - Can you explain this a little better - I am not quite sure what this sentence means and it may be a source of bias in the selection of practices that could be worth noting in the limitations

8. I think a further point for the discussion is that the pulse palpation performed by patients themselves is significantly less sensitive and specific than pulse palpation performed by health professionals. There are some previously older studies that have also identified this

Any attachments provided with reviews can be seen via the following link:

[LINK]

23 Nov 2019

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Dec 2019

Dear Dr. Ghazal,

Thank you very much for submitting your manuscript "Validity of daily self-pulse palpation for atrial fibrillation screening in patients 65 years and older: A cross-sectional study" (PMEDICINE-D-19-03337R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

We note that the responses to reviewers have not been included within the main text of the manuscript but explained only within the rebuttal letter. Please incorporate the responses to reviewers within the manuscript and ensure that the STROBE checklist has been adhered to as a number of reporting details in the methods appear to be missing. Please see comments from Reviewer 3. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Dec 12 2019 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Comments from the reviewers:

Reviewer #1: The authors have addressed my concerns and I now recommend publication

Peter Flom

Reviewer #2: I have no further comments. The authors have responded adequately to my previous comments.

Reviewer #3: The authors have done a good job in revising the manuscript, but I note that many comments were not addressed within the manuscript itself - they were just addressed in the response to reviewers. I would suggest that the authors go back through the original reviewer comments and check that each point raised by the reviewers has been answered within the manuscript, unless there is a good reason for why it should not be altered, in which case the response should indicate that nothing was altered because xxx reason.

For example, Reviewer 1 requested that an explanation be provided for the choice of 0.75 statistical power rather than 0.8. This point should be explained in the manuscript, not just to the reviewer.

The methods still require some further clarification - mostly with information that has been brought up in the original reviewer comments - eg:

* how did the nurses identify the eligible patients - were they patients that they were seeing for a clini8cal consultation or did the doctors refer their patients to them

* were the practices offered any incentives for participation, or given anything for the training - did the nurses get accreditation points etc

* did the nurse test/check the patients ability to pulse palpate after training

* when were people asked to do the recordings at home - and how often

* Were they asked to do the pulse and ECG at the same time

* How did they record their pulse palpation result (did they have a diary to write it in?)

* How was the pulse palpation and ECG recordings compared

* If they had symptoms with the pulse palpation how did they record these

* Who reviewed the ECGs and was this done remotely or with the patient

* Did the patient come back into the centre at the end of the period of time for a final review and to return the device

* The methods should indicate that the family doctor was responsible for prescription of the AC - this is mentioned later in the manuscript, but not in the methods

Limitations

* I don't understand what is meant by the first limitation. It is vague and needs rewording.

* The authors mention in their comments that the different study nurses at each site can introduce a selection bias, but this is not mentioned in the limitations section

The authors state in their response that this was not step-wise screening, but the conclusion of the abstract refers to stepwise screening - this is confusing. Why is step-wise screening introduced as a concept here?

Any attachments provided with reviews can be seen via the following link:

[LINK]

10 Dec 2019

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

29 Jan 2020

Dear Dr. Ghazal,

Thank you very much for re-submitting your manuscript "Validity of daily self-pulse palpation for atrial fibrillation screening in patients 65 years and older: A cross-sectional study" (PMEDICINE-D-19-03337R2) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Feb 05 2020 11:59PM.

Sincerely,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

Abstract background: “screening” should be “screening”

Abstract- last sentence of the methods and findings section should be a limitation of your study design.

Conclusion- the last sentence is perhaps overreaching and I would suggest that it is removed

Author summary- could you replace “paroxysmal” with a word that might be more accessible to a non specialist audience?

Please provide a space between text and reference brackets. No additional spaces between the references and full stop are required.

Design section Line 113 should be revised to “ A cross-sectional study was carried out in Swedish primary care centers (PCCs) to screen patients for AF”

Line 126- “whatever reasons” is not appropriate for a research article. Please rephrase this sentence to “Patients seeking care at a PCC above the age of 65 or older were invited by health personnel to participate in our study”. At Line 136, please mention who explained the study objectives to patients and received informed consent.

Line 182-183, in line with comments from Ref 3, please cite the more recent meta-analysis in addition to the currently cited meta analysis (ref 7) noting that the prevalence is the same in both studies. You may say “The AF detection rate of 1.4% was confirmed by a more recent meta-analysis [ref] which was published after we had completed the planning of this study” or similar.

Line 234- should say “inconclusive” ECG recordings

Line 272-275 could you please clarify in the text whether the irregular pulse noted at inclusion meant that the patients would be excluded from the study at that time, AF verified and AC started? I suspect this information ought to be in the methods too. If the patients were not excluded, please do mention that they carried on self palpating and checking ECG measurements

Discussion- please avoid assertions of primacy and replace with “To our knowledge, self-pulse palpation has not been compared with ECG recordings”

Line 282-285 appears to be somewhat unnecessary as participation in any form of clinical research does rely on the health personnel recruiting the subjects. I would suggest removing this paragraph.

Line 289-290 please explain what a Swedish age and gender distribution is, perhaps supported by a reference

The discussion section requires some revision to structure as it jumps from study findings to how these compare with previous studies. Please present and organize the Discussion as follows: a short, clear summary of the article's findings; what the study adds to existing research and where and why the results may differ from previous research; strengths and limitations of the study; implications and next steps for research, clinical practice, and/or public policy; one-paragraph conclusion.

Line 322- I don’t believe stepwise screening has been mentioned anywhere else in accordance with reviewer comments so we recommend this mention is also removed.

Table 6 and any related comments ought to be presented in the results section

Line 349- it is perhaps the generalisability also that is somewhat sacrificed by not randomly selecting various primary care centres in a city or country. It is unclear how you have speculated younger and healthier patients were invited by nurses? Please remove this if there is no reason to believe this might have been the case

You have mentioned ectopic beats in Line 354, 355 but do not mention in the methods if the nurses commented on this while explaining self palpation to patients. Please revise as needed

When resubmitting, please only resubmit the revised version of the manuscript with track changes and a clean version of this. Please do not submit any previous versions of the manuscript.

Comments from Reviewers:

Any attachments provided with reviews can be seen via the following link:

[LINK]

21 Feb 2020

Dear Dr. Ghazal,

On behalf of my colleagues and the academic editor, Dr. Kazem Rahimi, I am delighted to inform you that your manuscript entitled "Validity of daily self-pulse palpation for atrial fibrillation screening in patients 65 years and older: A cross-sectional study" (PMEDICINE-D-19-03337R3) has been accepted for publication in PLOS Medicine.

PRODUCTION PROCESS

Before publication you will see the copyedited word document (in around 1-2 weeks from now) and a PDF galley proof shortly after that. The copyeditor will be in touch shortly before sending you the copyedited Word document. We will make some revisions at the copyediting stage to conform to our general style, and for clarification. When you receive this version you should check and revise it very carefully, including figures, tables, references, and supporting information, because corrections at the next stage (proofs) will be strictly limited to (1) errors in author names or affiliations, (2) errors of scientific fact that would cause misunderstandings to readers, and (3) printer's (introduced) errors.

If you are likely to be away when either this document or the proof is sent, please ensure we have contact information of a second person, as we will need you to respond quickly at each point.

PRESS

A selection of our articles each week are press released by the journal. You will be contacted nearer the time if we are press releasing your article in order to approve the content and check the contact information for journalists is correct. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact.

PROFILE INFORMATION

Now that your manuscript has been accepted, please log into EM and update your profile. Go to https://www.editorialmanager.com/pmedicine, log in, and click on the "Update My Information" link at the top of the page. Please update your user information to ensure an efficient production and billing process.

Thank you again for submitting the manuscript to PLOS Medicine. We look forward to publishing it.

Best wishes,

Adya Misra, PhD

Senior Editor

PLOS Medicine

plosmedicine.org