Assessment of Sex Differences in the Initial Symptom Burden, Applied Treatment Strategy, and Quality of Life in Japanese Patients With Atrial Fibrillation

Key Points

Question  Are there differences in symptoms, provided care, and quality of life between female and male patients with atrial fibrillation in Japan?

Findings  This Japanese registry-based cohort study included 1534 new patients referred for the management of atrial fibrillation. Overall, 29.9% (n = 458) were women; although women experienced more atrial fibrillation–related symptoms and had worse quality of life at baseline, they were less likely to receive rhythm control treatment compared with men, and the gender gap on quality of life tended to grow.

Meaning  Management of atrial fibrillation in women appears to be both less aggressive and less effective than in men.

Importance  The clinical characteristics and outcomes of women and men with atrial fibrillation (AF) are reported to be different. However, whether sex-related differences extend to patients’ symptom burden and perceived quality of life (QOL) or the management pattern of AF has been rarely studied, particularly in Asian countries.

Objective  To assess the differences in symptoms, treatment, and QOL between Japanese female and male patients with AF.

Design, Setting, and Participants  Retrospective cohort study using data from the multicenter outpatient registry Keio Interhospital Cardiovascular Studies–Atrial Fibrillation (KiCS-AF), which collects information regarding health status and the treatment of patients with newly diagnosed or referred AF. One-year follow-up data were available for 1534 patients at 11 referral centers in the Tokyo, Japan, area who were enrolled between September 2012 and December 2015. All data available up to the 1-year follow-up examination through July 31, 2017, were included.

Main Outcomes and Measures  Sex, symptoms, AF treatment, and QOL as determined by Atrial Fibrillation Effect on Quality of Life (AFEQT) questionnaires at their initial visit and 1 year later.

Results  Of 1534 patients, 1076 (70.1%) were men. Compared with men, women were more likely to be older (median age, 73 years [interquartile range {IQR}, 65-78 years] vs 65 years [IQR, 57-73 years], P < .001) and have higher median brain-type natriuretic peptide levels (102.8 pg/mL [IQR, 47.3-235.5 pg/mL] vs 74.1 pg/mL [IQR, 28.5-150.5 pg/mL], P < .001). Women also had lower median AFEQT overall summary scores than men (75 [IQR, 61-85] vs 80 [IQR, 69-90]; P < .001) but similar treatment satisfaction at baseline. During follow-up, women were less likely to be treated with a rhythm control strategy (48.1% [n = 214] vs 58.0% [n = 621], P < .001), including catheter ablation of AF (adjusted hazard ratio, 0.77 [95% CI, 0.62-0.95]; P = .02). At 1-year follow-up, women and men had improved QOL scores, regardless of their baseline characteristics (eg, age or brain-type natriuretic peptide levels) or treatment strategies, yet the sex gap persisted and grew (adjusted change in AFEQT overall summary score during 12 months, 5.89 [95% CI, 2.24-9.54] in women vs 8.94 [95% CI, 5.59-12.30] in men; P = .02).

Conclusions and Relevance  In contemporary Japanese clinical practice among unselected patients with AF, women were initially seen with greater QOL impairment, and the sex gap grew 1 year after presentation. The present study underscores the need for focused efforts to better understand and close this observed sex gap over the initial year of treatment for patients with AF.

Atrial fibrillation (AF) can significantly impair quality of life (QOL).¹ Various management strategies (eg, catheter ablation) for AF are available to improve patients’ symptom burden and QOL.^2,3 It is also widely recognized that there are sex-related differences in background characteristics and clinical outcomes for patients with AF.^4,5 For instance, women with AF have a higher risk for stroke and mortality than men.^4,5 However, whether sex-related differences extend to the symptom burden, the management pattern of AF, or patients’ perceived QOL has been rarely studied, particularly in Asian countries. Furthermore, prior studies have been limited by a narrow description of QOL based on visual analog scales and lack information on patients’ concerns or satisfaction after their initial management.⁶ Given the rapid increase in the burden of AF in Asian populations,⁷ a deeper understanding of their symptom burden, QOL, and satisfaction with treatment is needed. This is of particular importance in Japan, where an aging society, which is more prone to AF, is becoming a public health concern.^8,9

The multicenter Keio Interhospital Cardiovascular Studies–Atrial Fibrillation (KiCS-AF) registry collects information regarding health status and the treatment of patients with newly diagnosed or referred AF in Japan and provides a unique opportunity to address current knowledge gaps regarding sex and AF. Therefore, we investigated whether symptoms, practice patterns of physicians, and QOL differed between Japanese female and male patients with AF. Understanding sex-related differences in different regions could encourage new efforts for better management of AF to minimize health status imbalances between women and men.

We obtained data from the KiCS-AF multicenter registry for the period September 2012 to December 2015. The KiCS-AF is a multicenter registry-based retrospective cohort study designed to collect clinical variables and outcomes data from consecutive patients with AF who were newly diagnosed at or referred to an outpatient clinic at each of 11 participating hospitals within the Kanto area of Japan (Saitama, Tochigi, Chiba, Kanagawa, and Tokyo Prefecture). Most were large tertiary care referral centers (8 had >500 beds). Dedicated clinical research coordinators were assigned to each hospital, and approximately 150 variables were collected for each patient. Staff members at participating hospitals were instructed to record and register data from consecutive hospital visits for patients with AF using an internet-based data collection system. Data quality assurance was achieved through education and training of the clinical research coordinators, automatic data validation, and reporting of data completeness. To ensure consecutive case enrollment, the senior study coordinator (I.U.) and investigator (S.K.) performed on-site auditing to ensure proper registration of each eligible patient.

The institutional review board at each hospital approved the study protocol, and all participants provided written informed consent. Almost all patients agreed to participate. For example, the refusal rate was 2.9% at the core center (Keio University Hospital, Tokyo, Japan). The KiCS-AF steering committee was responsible for overall study guidance, including the study protocol, data collection forms, data analysis, and interpretation of the results. Before launching the KiCS-AF registry, information regarding the objectives of the study and its social significance was provided for clinical trial registration with the University Hospital Medical Information Network (identifier UMIN000022229). This network is recognized by the International Committee of Medical Journal Editors as an ‘‘acceptable registry’’ according to a statement issued in September 2004. This study followed the Standards for Quality Improvement Reporting Excellence (SQUIRE) reporting guidelines.

Patients completed a detailed questionnaire about their perception of their QOL and treatment at baseline and at 1 year using the internationally validated Atrial Fibrillation Effect on Quality of Life (AFEQT) questionnaire.¹⁰ The development and validation of the AFEQT have been previously described.¹¹ It is a 20-item questionnaire that quantifies 4 domains of AF-related QOL, including symptoms, daily activities, treatment concern, and treatment satisfaction by using 7-point Likert-type response scales. An AFEQT overall summary score can be calculated from the first 3 domains, which range from 0 (worst health status) to 100 (best possible health status [no impairment]). Recent analysis has suggested that a 5-point change in AFEQT overall summary score is observed among patients who demonstrate change by 1 European Heart Rhythm Association (EHRA) functional status class, which is a clinically important difference (D. N. Holmes, MS; written communication; January 19, 2019). A culturally and linguistically translated version of the AFEQT for Japan was used.

Patients’ initially reported problems included palpitations, dyspnea on exertion, difficulties with activities, dizziness, fatigue, chest pain, and syncope. These were elicited by a physician at an initial medical interview in an outpatient clinic before access to patients’ AFEQT questionnaires and immediately after enrollment.

Data regarding patients’ demographics, clinical comorbidities, symptoms, prior and current drug therapy (including oral anticoagulants [OACs] and antiplatelet agents), electrocardiograms, echocardiograms, and blood sample test results were abstracted from the patients’ medical records. Additional patient background information included the type of visit (ie, diagnosis at health screening or referral from an emergency department). We defined patients diagnosed during a routine medical visit, regardless of whether they were asymptomatic or symptomatic, as being diagnosed by health screening. Patient data on variables for calculating risk scores for stroke included age 75 years or older, hypertension, diabetes, heart failure, and stroke or transient ischemic attack for the CHADS₂ score and age 75 years or older, hypertension, diabetes, cardiac failure or dysfunction, and stroke (including female sex, age 65-74 years, and vascular disease) for the CHA₂DS₂-VASc score. Oral anticoagulants consisted of warfarin sodium and direct OACs (ie, dabigatran etexilate, rivaroxaban, apixaban, and edoxaban tosylate), and antiplatelet agents comprised aspirin, clopidogrel bisulfate, and prasugrel. Blood samples were analyzed for complete blood cell count, liver function, kidney function, coagulation, and serum brain-type natriuretic peptide (BNP) level.

Yearly follow-up examinations were performed for all patients by mail, phone interviews, and medical record reviews. Patients completed the AFEQT questionnaire during clinic visits or by mailed questionnaires. Trained study personnel subsequently transcribed the completed AFEQT questionnaires and updated the status of comorbidities, medication use, catheter ablation of AF, and intercurrent adverse events (all-cause mortality, stroke, and bleeding). All data available up to the 1-year follow-up examination through July 31, 2017, were included. At that time, 1769 consecutive outpatients with AF were registered in our registry, and 1-year follow-up data were available for 1534 (86.7%). During 1-year follow-up, 27 patients (1.7%) died, and these patients were excluded from analyses.

Complete baseline and 1-year follow-up data were stratified and analyzed overall by sex. We compared baseline characteristics between women and men, including patients’ demographics, symptoms, AF history, prior and current medical therapies, electrocardiograms, echocardiograms (left ventricular ejection fraction and left atrial diameter), and blood sample test results. The AFEQT scores at baseline and 1-year follow-up were calculated and compared between women and men. To more completely define patient and treatment factors associated with QOL,¹² we performed prespecified subgroup analyses based on baseline age (≥75 and <75 years), baseline BNP level (≥200 and <200 pg/mL), type of visit (eg, diagnosis at health screening), and use of catheter ablation of AF within 1 year of registration (to convert BNP level to nanograms per liter, multiply by 1.0).

Continuous variables are presented as medians and interquartile ranges (IQRs); categorical variables are presented as numbers and percentages. Group differences were evaluated using the χ² test for categorical variables and the Wilcoxon rank sum test for continuous variables. We defined change in overall AFEQT score within 1 year as overall AFEQT score at 1 year minus overall AFEQT score at baseline. A positive change represents improved QOL, and a negative change represents worsening QOL. In addition, a sensitivity analysis was performed to exclude patients with high QOL at baseline (eg, patients with overall AFEQT score ≥90) because these patients might not be eligible to experience improvement, rendering the interpretation of changes in overall AFEQT score difficult.

To evaluate the association between female sex and change in QOL, the mean changes in overall AFEQT score within 1 year were compared between women and men using analysis of covariance adjusted for baseline AFEQT score. In addition, a general linear mixed model was constructed in an effort to adjust the AFEQT scores for observed differences between women and men, which included age, body mass index, smoking, dyslipidemia, heart failure, obstructive sleep apnea, chronic obstructive pulmonary disease, renal function (estimated glomerular filtration rate as a contentious variable), diagnosis at health screening, referral from an emergency department, type of AF, and catheter ablation of AF within the 1-year study period. Furthermore, to investigate whether female sex was independently associated with a meaningful improvement in QOL, defined as a 5-point increase in AFEQT overall summary score, we constructed a logistic regression model to adjust for the aforementioned clinically relevant variables. Moreover, we constructed Cox proportional hazards regression models to estimate the adjusted hazards ratios (HRs) and 95% CIs to determine whether female sex was independently associated with the time to implementation of catheter ablation of AF after adjusting for the aforementioned clinically relevant variables, except for treatment with catheter ablation. In addition, participating hospitals were included as a random effect to account for clustering of patients by site in these models.

There were missing data for less than 2% of all candidate variables, except for estimated glomerular filtration rate (5.3%). To account for missing data, single mean imputation was used. A statistical software program (SPSS, version 24.0; IBM Corp) was used for all analyses. All reported P values were 2 sided, with P < .05 considered statistically significant. In subgroup analyses, to correct for the inflation of type I error, all P values with a threshold of P < .005 were considered statistically significant via Bonferroni correction (ie, there were 10 hypothesis tests within each of the relevant subgroups [.05 / 10 = .005]).

There were 458 women (29.9%) and 1076 men (70.1%) in our analytic cohort (Table 1 and Figure 1). Compared with men, women were older (median age, 73 years [IQR, 65-78 years] vs 65 years [IQR, 57-73 years], P < .001) and had higher median BNP levels (102.8 pg/mL [IQR, 47.3-235.5 pg/mL] vs 74.1 pg/mL [IQR, 28.5-150.5 pg/mL], P < .001). Women also had lower median AFEQT overall summary scores than men (75 [IQR, 61-85] vs 80 [IQR, 69-90], P < .001) but had similar treatment satisfaction at baseline (Table 2). In addition, women more often had dyslipidemia (171 [38.7%] vs 332 [31.1%], P = .005), heart failure (74 [16.6%] vs 129 [12.1%], P < .001), and chronic kidney disease (estimated glomerular filtration rate <60 mL/min/1.73 m²) (209 [49.2%] vs 399 [39.7%], P < .001) (Table 1). Women were less likely to have AF diagnosed during health screening than men (79 [17.9%] vs 324 [30.4%], P < .001), even among symptomatic patients (39 [12.9%] vs 111 [19.5%], P = .01).

At the initial visit, women were more likely to have been referred from an emergency department (50 [11.3%] vs 69 [6.5%], P = .001) (Figure 1) and to have experienced palpitations (222 [50.3%] vs 409 [38.4%], P < .001), dyspnea (87 [19.7%] vs 153 [14.4%], P = .009), and exercise intolerance (19 [4.3%] vs 15 [1.4%], P = .001) than men (Figure 2). Of note, 31.4% of women had never experienced symptoms compared with 46.8% of men (P < .001). There were no significant differences in distributions of AF types between women and men, yet women had shorter median AF durations (108 days [IQR, 39-402 days] vs 190 days [IQR, 58-1100 days], P < .001) (Table 1).

At baseline, women had lower median AFEQT overall summary scores than men (75 [IQR, 61-85] vs 80 [IQR, 69-90], P < .001) (Table 2 and Figure 3). Consistently, women had lower median AFEQT scores for each domain, including symptoms, daily activities, and treatment concern, than men but not treatment satisfaction. Regardless of advanced age (≥75 years), baseline BNP levels (≥200 and <200 pg/mL), or type of visit (ie, diagnosis at health screening or not), these differences between women and men persisted (Table 2 and eTable 1 in the Supplement).

At baseline, women were more likely to be treated with diuretics (116 [26.2%] vs 185 [17.4%], P < .001) but were less likely to undergo treatment involving a rhythm control strategy (211 [47.8%] vs 619 [58.1%], P < .001) (Table 1). There were no significant differences by sex in the use of OACs for patients with a CHA₂DS₂-VASc score greater than 2 (266 [88.0%] vs 288 [90.8%], P = .88).

Overall median AFEQT score improved in both women and men. However, women continued to have lower median QOL scores than men (82 [IQR, 70-92] vs 90 [IQR, 80-96], P < .001) 1 year after presentation (Table 2). These sex differences persisted when the analysis was restricted to prespecified subgroups or after the exclusion of patients with preserved QOL at baseline (eg, patients with overall AFEQT score ≥90) (Table 2, Figure 3, and eTable 1 and eTable 2 in the Supplement). After adjusting for baseline AFEQT score, the mean change in overall mean AFEQT score within 1 year was 4.97 (95% CI, 3.74-6.20) in women and 10.20 (95% CI, 9.50-11.05) in men. Further adjustment for all baseline characteristics that differed between women and men and their treatments consistently showed that improvement in overall AFEQT was less in women than in men (5.89 [95% CI, 2.24-9.54] vs 8.94 [95% CI, 5.59-12.30], P = .02) (eTable 3 in the Supplement). Furthermore, women were less likely to achieve a meaningful improvement in QOL (ie, defined as a 5-point increase in AFEQT overall summary score) within the 1-year study period than men (adjusted odds ratio, 0.71 [95% CI, 0.53-0.96]; P = .03) (eTable 4 in the Supplement).

During follow-up, women were less likely to be treated with a rhythm control strategy (48.1% [n = 214] vs 58.0% [n = 621]; P < .001), and 28.8% (n = 127) of women and 43.3% (n = 462) of men underwent catheter ablation of AF (P < .001). After adjusting for differences between women and men, women were less likely to undergo catheter ablation of AF within the year after presentation than men (adjusted HR, 0.77 [95% CI, 0.62-0.95], P = .02) (eTable 5 in the Supplement). In terms of the clinical effectiveness of catheter ablation of AF, follow-up electrocardiogram data were available for 59.5% (n = 897) of patients. In this sample, overall percentage of successful catheter ablation of AF, defined as maintenance of sinus rhythm when followed up, regardless of antiarrhythmic drug therapy, was 89.1% (n = 286), and there was no difference between women (92.6% [n = 75]) and men (87.9% [n = 211]) (P = .24).

Regarding the AFEQT scores of those who underwent catheter ablation of AF, women continued to have lower median QOL scores than men (85 [IQR, 75-95] vs 94 [IQR, 86-98], P < .001) (Table 2). The median AFEQT scores were consistent for patients who successfully underwent catheter ablation of AF, defined as maintenance of sinus rhythm when followed up, regardless of antiarrhythmic drug therapy (85 [IQR, 77-95] for women vs 95 [IQR, 87-99] for men, P < .001).

While sex-based disparities in care and outcomes have been reported in many areas of cardiovascular disease, less study has been directed toward AF, particularly in non-Western countries. In a large, 11-center Japanese registry of newly recognized AF, we report the following findings: (1) although distributions of AF types were similar between women and men, women were initially seen with different demographic and comorbidity profiles and were more likely to be symptomatic, with worse QOL; (2) women were less likely to be treated with a rhythm control strategy, including catheter ablation of AF, than men; and (3) at 1 year, women and men had improved QOL scores, but the sex gap not only persisted but grew. Collectively, these data suggest less aggressive care in women, with worse QOL outcomes over the first year of treatment. Novel strategies are needed to minimize the observed health status disparities between women and men and to optimize their care.

To our knowledge, this is the first description of sex-associated differences in patient-reported outcomes among patients with AF in Japan. The incidence of AF increases rapidly with advancing age,¹³ and an estimated 700 000 people in Japan have AF, which is projected to increase by 2050 to more than 1 million,⁹ despite a predicted decrease in the total population. Compared with previous reports from a Western registry, the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF),¹⁴ patients enrolled in the KiCS-AF were younger, more likely to be male, and less likely to have cardiac comorbidities; consequently, they had lower CHA₂DS₂-VASc scores. However, regardless of these differences, our findings regarding baseline comorbidities and the greater symptom burden in women with AF are congruent with the results of previous studies from Western countries.^14-16 Prior studies^17,18 have also demonstrated that women with AF were less likely to receive catheter ablation of AF than men. These imbalances persisted even after adjusting for age and burden of comorbidities; therefore, our results are consistent with those of previous reports.

It has also been inferred that the greater QOL influence of AF for women may be attributed to increased sensitivity to disease and disease manifestation symptoms, differences in illness perceptions, or lower thresholds for reporting illness burden.¹⁹ Furthermore, sex differences have been associated with depression,^15,20 which is known to be more common among women than men and has been shown to strongly influence QOL measures of other cardiovascular conditions.²¹ In the present study, women had shorter AF durations than men; nonetheless, distributions of AF types were similar between women and men. In addition, these observations suggest that there may be important variables, beyond AF burden, that influence patients’ QOL, and it remains unclear what accounts for the large sex differences among patients with AF. Future studies ascertaining psychological and physical effects and factors underlying these differences will inform AF practice to further optimize the sex disparities in AF.

In our study, women with AF had more symptoms and were initially seen earlier than men, which suggests less aggressive screening in women compared with men. Health screening processes vary widely between countries and health care systems. Typical Japanese workers are eligible to undergo annual health screenings that include a 12-lead electrocardiogram examination (nonpermanent employees are not eligible to receive health screenings). The effectiveness of health screening programs for women may be limited. However, there are substantial numbers of female patients with asymptomatic AF; given that women with AF are known to be at higher risk for stroke and death than men,⁵ further investigations regarding lower rates of diagnosis in women are warranted. Because of the high proportion of older women due to an aging society, appropriate identification of women who may benefit from health screening is an important opportunity to further optimize the outcomes of women.

For a thorough understanding of our results, several limitations should be acknowledged. First, nonrandomized observational research involves inherent limitations, although it is the best way to describe the current treatment patterns and outcomes of care. There is likely to be unmeasured confounding, such as depression, frailty, or economic status, that may explain some of the observed differences in QOL. Second, our sample sizes were small, particularly for the analyses of clinical outcomes, and the study was likely underpowered to describe sex-based differences in these events. Third, the applicability of the adjusted change in AFEQT overall summary score during the first year of treatment (5.89 in women vs 8.94 in men), while statistically significant, is less than the 5-point difference associated with a change in EHRA class (D. N. Holmes, MS; written communication; January 19, 2019). Nevertheless, an EHRA class is a modestly large change in health status, and further work is needed to better clarify the clinical importance of these differences. However, the interval estimates for the differences in the AFEQT score at 1 year were robust in adjusted analysis. Furthermore, our findings regarding the greater symptom burden in women are in line with previous results.¹⁴ Fourth, not all patients with AF in Japan participated in the KiCS-AF registry, and the sampling bias and generalizability of the study results to Japan are a potential concern, although we strictly included those seen at the participating centers with new-onset AF. Despite this potential limitation, our study involved one of the most representative Japanese databases of patients with AF, and our results arguably comprise the most complete assessment of current practice patterns and QOL outcomes in Japan.

Within a Japanese network of hospitals, women with AF were initially seen with greater QOL impairment than men, and the sex gap persisted 1 year after their treatment. Understanding sex-specific factors independently associated with worse QOL can be a foundation for tailoring treatment.

Accepted for Publication: January 28, 2019.

Published: March 29, 2019. doi:10.1001/jamanetworkopen.2019.1145

Open Access: This article is published under the JN-OA license and is free to read on the day of publication.

Corresponding Author: Shun Kohsaka, MD, Department of Cardiology, Keio University School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan (sk@keio.jp).

Author Contributions: Dr Kohsaka had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ikemura, Kohsaka, Ueda, Katsumata, Tanimoto, Momiyama, Akaishi, Mitamura, Fukuda, Takatsuki.

Acquisition, analysis, or interpretation of data: Ikemura, Kohsaka, Kimura, Katsumata, Nishiyama, Aizawa, Tanimoto, Akaishi, Spertus, Takatsuki.

Drafting of the manuscript: Ikemura, Kohsaka, Ueda, Katsumata, Nishiyama, Aizawa, Tanimoto, Akaishi.

Critical revision of the manuscript for important intellectual content: Ikemura, Kimura, Tanimoto, Momiyama, Akaishi, Mitamura, Fukuda, Spertus, Takatsuki.

Statistical analysis: Ikemura.

Obtained funding: Kohsaka.

Administrative, technical, or material support: Ikemura, Kohsaka, Kimura, Ueda, Katsumata, Nishiyama, Tanimoto, Fukuda.

Supervision: Kohsaka, Katsumata, Tanimoto, Momiyama, Akaishi, Mitamura, Fukuda, Takatsuki.

Conflict of Interest Disclosures: Dr Kohsaka reported receiving an unrestricted research grant for the Department of Cardiology at Keio University School of Medicine from Bayer Pharmaceutical and Daiichi Sankyo; reported receiving grants from Bayer Yakuhin Ltd and Daiichi Sankyo; and reported receiving personal fees from Bristol-Myers Squibb. Dr Kimura reported receiving grants from Bayer Yakuhin Ltd. Dr Spertus reported receiving personal fees from Novartis, AstraZeneca, Janssen, Bayer, Boehringer Ingelheim, Regeneron, Corvia, and United Healthcare; reported receiving grants from Bayer and Abbott Vascular; and reported owning equity in Health Outcomes Sciences. Dr Takatsuki reported receiving grants and personal fees from Bayer and reported receiving personal fees from Daiichi Sankyo and Bristol-Myers Squibb. No other disclosures were reported.

Funding/Support: This study was funded by Grant-in-Aids for Scientific Research from the Japan Society for the Promotion of Science (grants 16H05215, 16KK0186) and by an unrestricted research grant from Bayer Yakuhin Ltd.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Site investigators were Yukihiko Momiyama, Munehisa Sakamoto, Jun Fuse, Kojiro Tanimoto, Yoko Tanimoto, Yukinori Ikegami, and Kohei Inagawa (National Hospital Organization Tokyo Medical Center); Iwao Nakamura, Jyunji Suzuki, Tomohiro Matsuhashi, and Hiroshi Shiga (Hino Municipal Hospital); Seiji Takatsuki, Yoshiyasu Aizawa, Nobuhiro Nishiyama, Takahiko Nishiyama, Yoshinori Katsumata, Shin Kashimura, Akira Kunitomi, Kazuaki Nakajima, and Taishi Fujisawa (Keio University School of Medicine); Masahiro Suzuki, Takaharu Katayama, Keisuke Matsumura, Tomohiko Ono, Hanako Tokuda, Ryutaro Yamaguchi, and Hiroaki Tanaka (National Hospital Organization Saitama National Hospital); Shigetaka Noma, Takashi Yagi, Kenichiro Shimoji, Koji Ueno, and Satoshi Mogi (Saiseikai Utsunomiya Hospital); Takashi Koyama, Shiro Ishikawa, Hideaki Kanki, Takashi Akima, Masahito Munakata, and Kazutaka Miyamoto (Saitama City Hospital); Hideo Mitamura, Kazunori Moritani, Masaru Shibata, and Toshimi Kageyama (Tachikawa Hospital); Takahiro Oki, Akiyasu Baba, Yoshinori Mano, and Hiroaki Sukegawa (Tokyo Dental College Ichikawa General Hospital); Kouji Negishi, Takahiro Koura, Daisuke Shinmura, Kotaro Fukumoto, and Hiroyuki Yamakawa (Yokohama Municipal Citizen’s Hospital); Keiichi Nagami, Kazuhiro Oyamada, Kotaro Naitou, and Keijiro Chiba (Keiyu Hospital); Megumi Shimada (Tokai University Oiso Hospital); and Makoto Akaishi (Tokai University Tokyo Hospital). Clinical coordinators were Aki Kato, Ikumi Koishi, Miho Matsuoka, Takako Nozaki, Hiroaki Nagayama, Chieko Tamura, Reiko Tamura, Junko Susa, Miho Umemura, and Itsuka Saito. We are grateful to all study coordinators, investigators, and patients who participated in the Keio Interhospital Cardiovascular Studies–Atrial Fibrillation (KiCS-AF) registry. None of the additional contributors received compensation.