Abstract
Background
Comprehensive monitoring of cardiovascular disease (CVD) is essential in rapidly aging societies such as Japan. The Japanese Circulation Society (JCS) launched the Japanese Registry Of All cardiac and vascular Diseases-Diagnosis Procedure Combination (JROAD-DPC) registry, linking annual JROAD questionnaires with nationwide DPC administrative claims to enable patient-level analyses of hospitalized CVD care. This Protocol Paper presents a comprehensive overview of the registry.
Methods and Results
Using anonymized data (April 2012–March 2023), we described temporal trends in patient demographics, principal CVD diagnoses, major interventions, disease-specific severity, and hospital characteristics. From FY2012–FY2022, participating facilities increased from 610 to 860, with registered patients more than doubling. Median age rose from 73.0 to 75.0 years; patients aged ≥90 years nearly quadrupled. The proportion of angina pectoris admissions declined (26.8% to 11.7%), while absolute numbers remained stable. Atrial fibrillation/flutter admissions rose in both proportion (4.1% to 5.9%) and absolute number. Heart failure admissions increased steadily, with its proportion showing a U-shaped trend. Catheter ablations for atrial fibrillation/flutter increased over fivefold, exceeding 64,000, while percutaneous coronary interventions for acute myocardial infarction surpassed 46,000.
Conclusions
JROAD-DPC now captures over 1.5 million annual CVD hospitalizations, providing a nationwide, large-scale longitudinal view of cardiovascular care in Japan. Its scale and validated coding enable robust analyses of trends and outcomes, supporting national CVD policy evaluation and improvement.
Key Words: Administrative claims data, Cardiovascular disease, JROAD-DPC Registry, Real-world data
The rapid aging of Japanese society has highlighted the critical need for comprehensive management of cardiovascular diseases (CVD).1 In response to this challenge, the Japanese Circulation Society (JCS) and the Japan Stroke Society jointly announced their “The Five-Year Plan for Overcoming Cardiovascular Disease” in December 2016.2 This strategic initiative aimed to extend healthy life expectancy by addressing 3 major disease categories: stroke, heart failure (HF), and vascular diseases (including acute myocardial infarction [AMI], acute aortic dissection, aortic aneurysm rupture, and peripheral arterial disease). One of the plan’s primary objectives was to promote nationwide disease registration systems for these conditions.2
The JCS established the Japanese Registry Of All cardiac and vascular Diseases (JROAD) in 2004 to evaluate clinical activities in healthcare facilities with dedicated CVD beds.3 This registry system was designed to provide appropriate feedback to participating institutions for quality improvement in patient care.4 All participating affiliated training hospitals have contributed to the collection of nationwide primary data through annual surveys, including information about resources (hospitals, beds, and CVD specialists), workload (number of hospitalized patients), and outcomes (CVD deaths and autopsy rates).3 However, JROAD’s effectiveness was limited by its inability to capture individual patient-level data, which prevented more robust and reliable investigations of nationwide clinical outcomes. To address this limitation, the JCS, in collaboration with the National Cerebral and Cardiovascular Center (Suita, Osaka, Japan), initiated the development of the JROAD-Diagnosis Procedure Combination (JROAD-DPC) registry in 2014.1 This enhanced registry system integrates data from Japan’s Diagnosis Procedure Combination (DPC) / Per-Diem Payment System (PDPS), enabling more detailed analysis of clinical information at the individual patient level.5,6
This Protocol Paper (PP) presents a comprehensive overview of the JROAD-DPC registry, describing its key features and summarizing the research findings that have emerged from its utilization.
Methods
Data Source and Study Population
The JROAD-DPC registry integrates data from the JROAD survey, conducted by the JCS, with patient-level administrative claims data from Japan’s DPC/PDPS. The JROAD-DPC registry (Figure 1) was constructed through a 2-stage facility selection process. Initially, DPC-eligible facilities were identified from the JROAD survey, conducted by the JCS. Subsequently, these eligible facilities were invited to participate in providing DPC data. Facilities that consented to data provision contributed their information annually for each fiscal year (FY: defined in Japan as April 1 to March 31 of the following calendar year). From each participating facility, anonymized DPC data in the specified survey format were processed using a DPC data collection tool to create the dataset for submission. This prepared data was then provided to the JROAD office either online or via physical media. Patient data from the JROAD-DPC registry collected between April 1, 2012, and March 31, 2023, were used for this study. Inclusion in the study cohort required that a patient’s record met specific criteria related to CVD. Specifically, patients were included if their DPC Form 1 indicated ≥1 diagnosis corresponding to CVD codes in any of the following fields: main diagnosis, admission diagnosis, or the diagnosis that accounted for the most resource consumption. Alternatively, patients were included if their record documented medical procedures or interventions related to CVD.
Figure 1.
Overview of the JROAD-DPC registry system. Hospital survey and DPC data from cardiovascular divisions nationwide flow to the National Cerebral and Cardiovascular Center Data Center for integration. The Japanese Circulation Society IT/Registry Committee uses this integrated registry to generate annual reports, research proposals, and quality indicator summaries for participating hospitals. IT, Information Technology; JROAD-DPC, Japanese registry of all cardiac and vascular diseases-diagnosis procedure combination.
Covariates
Data elements collected for the current study encompassed patient demographics (sex and age at admission), admission diagnoses, comorbid conditions present at the time of admission, and complications that developed after admission. These diagnoses, comorbidities, and complications were accompanied by their corresponding International Classification of Diseases, 10th Revision (ICD-10) codes. Further information extracted included details of surgical/ interventional procedures and medical treatments, their respective dates of implementation, specifics on drugs and medical materials utilized, the total length of hospital stay, patient discharge outcomes (e.g., discharged to home, transferred to another acute or long-term care facility, or in-hospital death), and comprehensive healthcare cost information. Various clinical severity scores, such as the Canadian Cardiovascular Society (CCS) functional classification for angina, Killip classification for MI, and activities of daily living scores, were also available depending on the patient’s condition. The CCS functional classification grades angina severity based on the level of physical activity that precipitates symptoms, ranging from Class I (angina only during strenuous or prolonged activity) to Class IV (inability to perform any physical activity without discomfort).7 The JROAD-DPC registry provides rich administrative and some clinical data, but generally does not include detailed laboratory test results, comprehensive imaging data, physiological parameters, or specific causes of death.5
Operational Definitions
In Japan’s DPC system, medical diagnoses are recorded using standardized codes. Each diagnosis is typically assigned a 7-digit diagnosis code (DX code; shoubyoumei code) from the Japanese Standard Disease Name Master.8 Although this system maintains mappings to the ICD-10 for standardization, it functions as a distinct Japanese classification. Modifier codes (byoumeihuka codes) can be appended to DX codes for greater clinical detail, specifying aspects such as laterality or severity. Surgical and interventional procedures are classified using K-codes derived from the Japanese medical fee schedule. Collectively, these DX codes, modifiers, and K-codes provide the basis for DPC patient classification and reimbursement.
Building upon this system, we established specific operational definitions to serve as the analytical criteria in the present study for identifying relevant patient cohorts and procedures. These detailed definitions are provided in the supplementary materials. Supplementary Table 1 specifies CVD using both ICD-10 and Japanese DX codes. Supplementary Table 2 details the criteria for severity stratification. Supplementary Table 3 outlines the inclusion criteria and K-codes for surgical and interventional procedures. For FY2016–FY2021, reporting of the New York Heart Association (NYHA) classification was not mandatory, resulting in a significant amount of missing data. Furthermore, during the period FY2012–FY2015, acute aortic dissection was not registered as a distinct diagnostic entity, and thus, data for this specific condition are unavailable.
Ethical Consideration
The study protocol received approval from the Ethics Committee of the JCS (Reference No. 1-1-9). The JROAD-DPC registry used for this study consisted of data that had undergone anonymization processing at the participating medical institutions before submission. Consequently, no personally identifiable information was provided to the researchers, and no handling of personal information occurred during this study, thereby ensuring patient confidentiality. The original collected data are stored under strict management at the Open Innovation Center, National Cerebral and Cardiovascular Center, Japan.
Results
Participating Facilities
Figure 2 shows the annual number of institutions participating in JROAD and JROAD-DPC. Despite minor fluctuations, 1,516 medical institutions participated in JROAD in FY2022, of which 1,254 were DPC-eligible facilities. Among these, 860 facilities consented to provide DPC information and participated in JROAD-DPC. This represents approximately 70% of the cardiovascular specialist training facilities and affiliated training facilities participating in JROAD. It should be noted that the number of JROAD-DPC participating facilities increased from 610 in FY2012 to 860 in FY2022 (≈40%; Figure 2). This PP presents aggregated data from all participating facilities each year without adjustment for this increase.
Figure 2.
Number of participating facilities in the JROAD and JROAD-DPC (FY2012–FY2022). JROAD-DPC, Japanese registry of all cardiac and vascular diseases-diagnosis procedure combination.
Registry Characteristics
Table 1 presents trends in patient demographics and facility characteristics of the JROAD-DPC registry from FY2012 to FY2022. During this period, the number of participating facilities increased from 610 in FY2012 to 860 in FY2022, and the total number of registered patients more than doubled. The median patient age rose from 73.0 (interquartile range [IQR] 64.0–80.0) in FY2012 to 75.0 (IQR 65.0–82.0) years in FY2022, reflecting a significant demographic shift toward an older population. This was evident in the age distribution: the proportion of patients aged 60–69 years decreased from 22.8% to 15.8%, while the proportion of those aged 80–89 years increased from 22.9% to 26.8%. Notably, the absolute number of patients aged ≥90 years nearly quadrupled (from 32,532 to 120,799), and their proportion rose substantially from 4.8% to 7.8%. Meanwhile, the proportion of male patients slightly decreased from 63.8% to 61.2%. Facility characteristics remained consistent; the median number of hospital beds was stable, and the distribution of patients by hospital size showed little change, with the majority (≈41%) consistently treated in facilities with 450–749 beds.
Table 1.
Trends in Patient Demographics in the JROAD-DPC Database (FY2012–FY2022)
| FY | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of participating facilities | 610 | 637 | 744 | 754 | 817 | 812 | 820 | 828 | 830 | 862 | 860 |
| No. of patients | 672,436 | 750,267 | 946,462 | 1,257,491 | 1,479,495 | 1,526,333 | 1,593,663 | 1,599,488 | 1,473,009 | 1,538,520 | 1,542,727 |
| Age, median (IQR), years | 73.0 (64.0, 80.0) |
72.0 (63.0, 80.0) |
73.0 (64.0, 81.0) |
73.0 (64.0, 81.0) |
73.0 (64.0, 81.0) |
73.0 (64.0, 81.0) |
74.0 (64.0, 82.0) |
74.0 (65.0, 82.0) |
74.0 (65.0, 82.0) |
74.0 (65.0, 82.0) |
75.0 (65.0, 82.0) |
| Male, n (%) | 429,236 (63.8) | 474,371 (63.2) | 598,780 (63.3) | 777,136 (61.8) | 915,688 (61.9) | 943,763 (61.8) | 981,249 (61.6) | 982,329 (61.4) | 908,352 (61.7) | 941,176 (61.2) | 944,789 (61.2) |
| No. of patients by age group | |||||||||||
| ≤29 years, n (%) | 22,090 (3.3) | 31,173 (4.2) | 39,630 (4.2) | 41,406 (3.3) | 51,027 (3.5) | 52,317 (3.5) | 56,624 (3.6) | 54,874 (3.4) | 47,335 (3.2) | 52,786 (3.4) | 51,124 (3.3) |
| 30–39 years, n (%) | 8,618 (1.3) | 10,126 (1.3) | 11,868 (1.3) | 16,230 (1.3) | 21,766 (1.5) | 21,768 (1.4) | 22,473 (1.4) | 22,499 (1.4) | 19,973 (1.4) | 21,742 (1.4) | 21,743 (1.4) |
| 40–49 years, n (%) | 25,452 (3.8) | 29,683 (4.0) | 36,700 (3.9) | 52,617 (4.2) | 66,656 (4.5) | 68,848 (4.5) | 70,688 (4.5) | 69,775 (4.4) | 63,190 (4.3) | 64,837 (4.2) | 62,236 (4.0) |
| 50–59 years, n (%) | 59,126 (8.8) | 65,420 (8.7) | 79,762 (8.4) | 108,078 (8.6) | 129,403 (8.8) | 132,961 (8.8) | 139,822 (8.8) | 140,878 (8.8) | 131,784 (9.0) | 138,433 (9.0) | 139,994 (9.1) |
| 60–69 years, n (%) | 153,082 (22.8) | 166,871 (22.2) | 205,822 (21.8) | 271,316 (21.7) | 323,454 (21.9) | 313,676 (20.7) | 308,177 (19.4) | 288,700 (18.1) | 250,107 (17.0) | 249,045 (16.2) | 244,121 (15.8) |
| 70–79 years, n (%) | 217,847 (32.4) | 240,068 (32.0) | 304,617 (32.3) | 387,590 (31.0) | 443,636 (30.0) | 460,711 (30.4) | 494,858 (31.2) | 510,799 (32.1) | 473,015 (32.2) | 487,659 (31.8) | 489,268 (31.7) |
| 80–89 years, n (%) | 153,684 (22.9) | 170,623 (22.7) | 219,358 (23.2) | 303,071 (24.2) | 356,522 (24.1) | 373,368 (24.6) | 393,612 (24.8) | 398,304 (25.0) | 376,483 (25.7) | 403,108 (26.3) | 413,439 (26.8) |
| ≥90 years, n (%) | 32,532 (4.8) | 36,291 (4.8) | 46,701 (4.9) | 69,747 (5.6) | 84,402 (5.7) | 91,218 (6.0) | 99,542 (6.3) | 106,497 (6.7) | 105,304 (7.2) | 117,128 (7.6) | 120,799 (7.8) |
| Hospital beds, median (IQR) | 480.0 (329.0, 639.0) |
466.0 (343.0, 647.0) |
473.0 (343.0, 642.0) |
487.0 (341.0, 658.0) |
482.0 (344.0, 658.0) |
486.0 (344.0, 658.0) |
499.0 (347.0, 656.0) |
500.0 (347.0, 651.0) |
489.0 (343.0, 656.0) |
486.0 (343.0, 654.0) |
496.0 (350.0, 656.0) |
| No. of patients by no. of hospital beds | |||||||||||
| ≤99 beds, n (%) | 11,289 (1.7) | 11,185 (1.5) | 14,479 (1.5) | 17,697 (1.4) | 15,939 (1.1) | 13,951 (0.9) | 17,869 (1.1) | 18,218 (1.1) | 20,806 (1.4) | 17,821 (1.2) | 13,529 (0.9) |
| 100–199 beds, n (%) | 34,985 (5.2) | 29,892 (4.0) | 44,135 (4.7) | 59,355 (4.7) | 68,736 (4.6) | 69,629 (4.6) | 72,383 (4.5) | 76,178 (4.8) | 69,494 (4.7) | 72,757 (4.7) | 80,687 (5.2) |
| 200–299 beds, n (%) | 68,709 (10.2) | 75,188 (10.0) | 81,032 (8.6) | 119,833 (9.5) | 135,028 (9.1) | 131,510 (8.6) | 137,679 (8.6) | 137,395 (8.6) | 129,646 (8.8) | 141,730 (9.2) | 128,033 (8.3) |
| 300–449 beds, n (%) | 183,365 (27.3) | 229,378 (30.6) | 295,980 (31.3) | 358,702 (28.5) | 437,738 (29.6) | 446,947 (29.3) | 449,745 (28.2) | 434,589 (27.2) | 415,767 (28.2) | 438,207 (28.5) | 429,422 (27.8) |
| 450–749 beds, n (%) | 266,926 (39.7) | 291,215 (38.9) | 386,116 (40.8) | 522,182 (41.5) | 608,883 (41.2) | 644,097 (42.2) | 673,361 (42.3) | 690,401 (43.2) | 603,131 (40.9) | 631,633 (41.1) | 636,301 (41.2) |
| ≥750 beds, n (%) | 107,162 (15.9) | 112,126 (15.0) | 124,720 (13.2) | 179,722 (14.3) | 213,171 (14.4) | 220,199 (14.4) | 242,626 (15.2) | 242,707 (15.2) | 234,165 (15.9) | 236,372 (15.4) | 254,755 (16.5) |
FY, fiscal year; IQR, interquartile range; JROAD-DPC, Japanese registry of all cardiac and vascular diseases-diagnosis procedure combination.
Building upon the expanded registry coverage, we analyzed trends in the proportion and absolute number of each CVD to account for the substantial increase in total registered patients (Table 2). While the proportion of angina pectoris hospitalizations decreased markedly from 26.8% in FY2012 to 11.7% in FY2022, the absolute number of cases remained virtually unchanged (from 180,419 to 180,967) against the backdrop of a more than twofold increase in total registered cases. For HF, the proportion of cases followed a U-shaped trend, decreasing from 16.2% to 10.6% and then rising to 11.3%, while the absolute number of admissions steadily increased over time (from 108,665 to 174,588). In contrast, both the proportion and absolute number of atrial fibrillation/flutter cases increased throughout the study period (from 4.1% [27,315 cases] to 5.9% [90,546 cases]). The proportions of AMI and aortic dissection remained relatively stable, but their absolute numbers gradually increased in line with the overall growth of the registry.
Table 2.
Number of Cases for Each Cardiovascular Disease Registered in the JROAD-DPC Database (FY2012–FY2022)
| FY | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of patients | 672,436 | 750,267 | 946,462 | 1,257,491 | 1,479,495 | 1,526,333 | 1,593,663 | 1,599,488 | 1,473,009 | 1,538,520 | 1,542,727 |
| Angina, n (%) | 180,419 (26.8) | 185,776 (24.8) | 213,436 (22.6) | 219,330 (17.4) | 239,450 (16.2) | 232,168 (15.2) | 237,432 (14.9) | 221,993 (13.9) | 186,064 (12.6) | 188,224 (12.2) | 180,967 (11.7) |
| Unstable angina, n (%) | 41,435 (6.2) | 40,137 (5.3) | 41,196 (4.4) | 47,484 (3.8) | 48,989 (3.3) | 46,266 (3.0) | 44,616 (2.8) | 41,436 (2.6) | 37,489 (2.5) | 34,588 (2.2) | 32,235 (2.1) |
| AMI, n (%) | 35,824 (5.3) | 37,612 (5.0) | 42,483 (4.5) | 45,453 (3.6) | 49,568 (3.4) | 50,219 (3.3) | 52,035 (3.3) | 53,359 (3.3) | 53,029 (3.6) | 55,397 (3.6) | 58,080 (3.8) |
| AF/flutter, n (%) | 27,315 (4.1) | 34,853 (4.6) | 41,147 (4.3) | 47,450 (3.8) | 55,970 (3.8) | 64,674 (4.2) | 74,918 (4.7) | 80,548 (5.0) | 76,260 (5.2) | 84,886 (5.5) | 90,546 (5.9) |
| Cardiomyopathy, n (%) | 5,881 (0.9) | 6,697 (0.9) | 8,088 (0.9) | 8,457 (0.7) | 9,126 (0.6) | 9,993 (0.7) | 9,654 (0.6) | 9,642 (0.6) | 7,970 (0.5) | 8,156 (0.5) | 7,906 (0.5) |
| HF, n (%) | 108,665 (16.2) | 115,929 (15.5) | 131,641 (13.9) | 142,297 (11.3) | 156,205 (10.6) | 165,991 (10.9) | 171,487 (10.8) | 174,352 (10.9) | 168,590 (11.4) | 176,295 (11.5) | 174,588 (11.3) |
| Acute HF, n (%) | 77,100 (11.5) | 81,236 (10.8) | 89,029 (9.4) | 97,259 (7.7) | 107,480 (7.3) | 112,995 (7.4) | 116,284 (7.3) | 117,729 (7.4) | 115,584 (7.8) | 119,281 (7.8) | 117,314 (7.6) |
| Aortic dissection, n (%) | 10,563 (1.6) | 11,861 (1.6) | 13,848 (1.5) | 15,302 (1.2) | 17,356 (1.2) | 18,049 (1.2) | 18,626 (1.2) | 19,270 (1.2) | 18,814 (1.3) | 19,638 (1.3) | 20,876 (1.4) |
| Acute aortic dissection, n (%) |
NA | NA | NA | NA | 12,819 (0.9) | 13,270 (0.9) | 14,009 (0.9) | 14,687 (0.9) | 14,347 (1.0) | 15,233 (1.0) | 16,509 (1.1) |
| Acute aortic dissection (Stanford A), n (%) |
NA | NA | NA | NA | 5,386 (0.4) | 5,706 (0.4) | 6,144 (0.4) | 6,572 (0.4) | 6,512 (0.4) | 6,926 (0.5) | 7,850 (0.5) |
| Acute aortic dissection (Stanford B), n (%) |
NA | NA | NA | NA | 4,854 (0.3) | 5,021 (0.3) | 5,316 (0.3) | 5,537 (0.3) | 5,372 (0.4) | 5,702 (0.4) | 6,124 (0.4) |
| Aortic rupture, n (%) | 2,630 (0.4) | 2,780 (0.4) | 3,055 (0.3) | 3,284 (0.3) | 3,432 (0.2) | 3,453 (0.2) | 3,692 (0.2) | 3,774 (0.2) | 3,599 (0.2) | 3,786 (0.2) | 3,678 (0.2) |
| Cardiac arrest, n (%) | 26,508 (3.9) | 29,798 (4.0) | 35,666 (3.8) | 34,943 (2.8) | 37,670 (2.5) | 39,691 (2.6) | 40,303 (2.5) | 41,052 (2.6) | 43,687 (3.0) | 47,274 (3.1) | 51,577 (3.3) |
| Pulmonary embolism, n (%) | 1,838 (0.3) | 5,841 (0.8) | 6,538 (0.7) | 6,503 (0.5) | 7,373 (0.5) | 7,359 (0.5) | 7,890 (0.5) | 8,372 (0.5) | 8,460 (0.6) | 9,249 (0.6) | 9,214 (0.6) |
| Primary pulmonary hypertension, n (%) |
454 (0.1) | 3,431 (0.5) | 3,743 (0.4) | 4,160 (0.3) | 4,618 (0.3) | 4,728 (0.3) | 4,919 (0.3) | 5,025 (0.3) | 4,682 (0.3) | 5,545 (0.4) | 5,764 (0.4) |
| Tetralogy of Fallot, n (%) | 603 (0.1) | 1,248 (0.2) | 1,408 (0.1) | 1,541 (0.1) | 1,783 (0.1) | 2,055 (0.1) | 1,997 (0.1) | 1,941 (0.1) | 1,851 (0.1) | 1,972 (0.1) | 1,970 (0.1) |
AF, atrial fibrillation; AMI, acute myocardial infarction; DX, diagnosis; FY, fiscal year; HF, heart failure; ICD, international classification of diseases; JROAD-DPC, Japanese registry of all cardiac and vascular diseases-diagnosis procedure combination.
Table 3 summarizes the annual number and proportion of major procedures for principal CVD. For patients hospitalized with angina pectoris, both the absolute number of percutaneous coronary intervention (PCI) procedures and their proportion among angina cases increased, with PCI rising from 64,055 (35.5%) in FY2012 to 79,073 (43.7%) in FY2022, despite total angina admissions remaining relatively stable. For AMI, the proportion of patients treated with PCI remained consistently high at approximately 80%, and the absolute number increased from 27,487 to 46,528 as the total number of AMI admissions grew. The most notable change was seen in atrial fibrillation/flutter: the number of percutaneous catheter ablations increased more than fivefold, from 11,181 (40.9%) to 64,014 (70.7%), driven by both a higher adoption rate and substantial growth in total atrial fibrillation/flutter hospitalizations. In contrast, for HF, the use of cardiac resynchronization therapy remained low, consistently accounting for <1% of hospitalizations. Regarding other surgical procedures, the proportion of coronary artery bypass grafting remained stable for both angina and AMI, whereas for acute aortic dissection, the use of open surgical stent–graft insertion more than doubled from 3.0% to 6.9%.
Table 3.
Number of Surgical and Interventional Procedures by Cardiovascular Disease (FY2012–FY2022)
| FY | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Angina | 180,419 | 185,776 | 213,436 | 219,330 | 239,450 | 232,168 | 237,432 | 221,993 | 186,064 | 188,224 | 180,967 |
| PCI, n (%) | 64,055 (35.5) | 66,356 (35.7) | 77,772 (36.4) | 82,050 (37.4) | 90,391 (37.7) | 88,539 (38.1) | 89,336 (37.6) | 87,311 (39.3) | 77,139 (41.5) | 80,601 (42.8) | 79,073 (43.7) |
| PTCA, n (%) | 8,260 (4.6) | 9,327 (5.0) | 12,514 (5.9) | 12,870 (5.9) | 15,097 (6.3) | 15,419 (6.6) | 16,224 (6.8) | 15,335 (6.9) | 12,921 (6.9) | 14,084 (7.5) | 14,642 (8.1) |
| Percutaneous coronary stenting, n (%) |
55,668 (30.9) | 57,867 (31.1) | 66,244 (31.0) | 70,336 (32.1) | 75,461 (31.5) | 72,875 (31.4) | 71,157 (30.0) | 67,132 (30.2) | 58,009 (31.2) | 58,397 (31.0) | 56,055 (31.0) |
| On-pump CABG, n (%) | 2,800 (1.6) | 2,871 (1.5) | 3,312 (1.6) | 3,174 (1.4) | 3,411 (1.4) | 3,388 (1.5) | 3,349 (1.4) | 3,345 (1.5) | 2,957 (1.6) | 3,224 (1.7) | 3,333 (1.8) |
| Off-pump CABG, n (%) | 4,038 (2.2) | 3,838 (2.1) | 4,060 (1.9) | 4,366 (2.0) | 4,242 (1.8) | 4,250 (1.8) | 4,389 (1.8) | 4,221 (1.9) | 3,553 (1.9) | 3,682 (2.0) | 3,597 (2.0) |
| AMI | 35,824 | 37,612 | 42,483 | 45,453 | 49,568 | 50,219 | 52,035 | 53,359 | 53,029 | 55,397 | 58,080 |
| PCI/PTCA/percutaneous coronary stenting, n (%) |
27,487 (76.7) | 29,164 (77.5) | 33,353 (78.5) | 36,117 (79.5) | 39,543 (79.8) | 40,039 (79.7) | 41,474 (79.7) | 42,602 (79.8) | 42,771 (80.7) | 44,476 (80.3) | 46,528 (80.1) |
| PTCA, n (%) | 3,589 (10.0) | 3,905 (10.4) | 4,783 (11.3) | 5,214 (11.5) | 5,927 (12.0) | 6,213 (12.4) | 6,906 (13.3) | 6,787 (12.7) | 10,471 (19.7) | 11,628 (21.0) | 12,835 (22.1) |
| Percutaneous coronary stenting, n (%) |
24,991 (69.8) | 26,595 (70.7) | 30,429 (71.6) | 32,893 (72.4) | 35,582 (71.8) | 35,947 (71.6) | 36,846 (70.8) | 37,287 (69.9) | 33,275 (62.7) | 33,909 (61.2) | 34,738 (59.8) |
| On-pump CABG, n (%) | 515 (1.4) | 506 (1.3) | 583 (1.4) | 624 (1.4) | 603 (1.2) | 582 (1.2) | 662 (1.3) | 580 (1.1) | 594 (1.1) | 591 (1.1) | 626 (1.1) |
| Off-pump CABG, n (%) | 392 (1.1) | 368 (1.0) | 407 (1.0) | 454 (1.0) | 432 (0.9) | 467 (0.9) | 450 (0.9) | 461 (0.9) | 425 (0.8) | 488 (0.9) | 442 (0.8) |
| AF/flutter | 27,315 | 34,853 | 41,147 | 47,450 | 55,970 | 64,674 | 74,918 | 80,548 | 76,260 | 84,886 | 90,546 |
| Percutaneous catheter ablation (transseptal and epicardial approach), n (%) |
11,181 (40.9) | 16,238 (46.6) | 20,288 (49.3) | 25,365 (53.5) | 32,161 (57.5) | 39,755 (61.5) | 48,488 (64.7) | 53,526 (66.5) | 52,181 (68.4) | 58,937 (69.5) | 64,014 (70.7) |
| Percutaneous catheter ablation (other approaches), n (%) |
2,915 (10.7) | 3,334 (9.6) | 3,175 (7.7) | 3,286 (6.9) | 3,540 (6.3) | 3,652 (5.7) | 3,623 (4.8) | 3,717 (4.6) | 3,159 (4.1) | 3,421 (4.0) | 3,398 (3.8) |
| HF | 108,665 | 115,929 | 131,641 | 142,297 | 156,205 | 165,991 | 171,487 | 174,352 | 168,590 | 176,295 | 174,588 |
| CRTP implantation, n (%) | 232 (0.2) | 226 (0.2) | 257 (0.2) | 323 (0.2) | 348 (0.2) | 413 (0.2) | 509 (0.3) | 618 (0.4) | 627 (0.4) | 613 (0.3) | 660 (0.4) |
| CRTP replacement, n (%) | 36 (0.0) | 30 (0.0) | 41 (0.0) | 48 (0.0) | 23 (0.0) | 45 (0.0) | 26 (0.0) | 34 (0.0) | 39 (0.0) | 48 (0.0) | 78 (0.0) |
| CRTD implantation, n (%) | 682 (0.6) | 641 (0.6) | 697 (0.5) | 681 (0.5) | 798 (0.5) | 860 (0.5) | 826 (0.5) | 929 (0.5) | 920 (0.5) | 946 (0.5) | 1,008 (0.6) |
| CRTD replacement, n (%) | 124 (0.1) | 142 (0.1) | 171 (0.1) | 141 (0.1) | 140 (0.1) | 115 (0.1) | 117 (0.1) | 163 (0.1) | 165 (0.1) | 150 (0.1) | 159 (0.1) |
| Acute aortic dissection | NA | NA | NA | NA | 12,819 | 13,270 | 14,009 | 14,687 | 14,347 | 15,233 | 16,509 |
| Aneurysmectomy, n (%) | NA | NA | NA | NA | 3,146 (24.5) | 3,287 (24.8) | 3,373 (24.1) | 3,525 (24.0) | 3,253 (22.7) | 3,544 (23.3) | 3,753 (22.7) |
| Open surgical stent-graft insertion, n (%) |
NA | NA | NA | NA | 380 (3.0) | 569 (4.3) | 696 (5.0) | 740 (5.0) | 913 (6.4) | 1,019 (6.7) | 1,140 (6.9) |
| Stent-graft insertion, n (%) | NA | NA | NA | NA | 584 (4.6) | 648 (4.9) | 691 (4.9) | 707 (4.8) | 733 (5.1) | 802 (5.3) | 761 (4.6) |
CABG, coronary artery bypass grafting; CRTD, cardiac resynchronization therapy defibrillator; CRTP, cardiac resynchronization therapy pacemaker; FY, fiscal year; NA, not available; PCI, percutaneous coronary intervention; PTCA, percutaneous transluminal coronary angioplasty.
Figure 3 presents the distribution of cases by disease-specific severity classifications alongside the annual trend in the total hospitalizations for each condition. For angina hospitalization (including unstable angina), the proportional distribution across CCS classes remained broadly stable over time, whereas total angina hospitalizations gradually declined with a marked drop in FY2020 (Figure 3A). For unstable angina, the number of hospitalizations showed a modest rise around middecade, followed by a plateau rather than a consistent yearonyear increase (Figure 3B). In AMI, the Killip distribution was largely stable throughout; a small uptick in Killip IV was observed in FY2020, while total AMI admissions did not substantially decrease that year and continued an overall upward trend (Figure 3C). For HF, total hospitalizations increased over time with a temporary dip in FY2020; however, detailed severity trend analysis was limited by substantial missing NYHA data in FY2016–FY2021, with improvement in FY2022 (Figure 3D).
Figure 3.
Annual trends in case distribution by disease-specific severity classification. (A) Annual distribution of angina hospitalization cases (including unstable angina) stratified by Canadian Cardiovascular Society (CCS) functional classification. (B) Annual distribution of unstable angina hospitalization cases stratified by CCS functional classification. (C) Annual distribution of acute myocardial infarction cases stratified by Killip classification. (D) Annual distribution of heart failure hospitalization cases stratified by New York Heart Association (NYHA) classification. Each panel presents a 100% stacked bar chart showing the proportion of cases for each severity class by year. The black line and dots indicate the total number of cases each year (right Y-axis). Unclassified and missing data are also shown.
Discussion
A key strength of the JROAD-DPC is its extensive coverage and scale, which yields an exceptionally large sample size that provides robust statistical power to evaluate outcomes and practice patterns.3 Importantly, the diagnostic coding has been validated for selected JROAD-DPC conditions (e.g., AMI and HF).9 Together, the breadth and quality of the JROAD-DPC data afford researchers a comprehensive “real-world” view of CVD care, ranging from patient demographics and treatments to in-hospital outcomes and costs on a national scale.3,10 Studies leveraging the JROAD-DPC have yielded important insights into the quality of care and health outcomes. Analyses of this registry consistently demonstrate substantial between-hospital variation in acute CVD care and its consequences.3,11 Moreover, the registry’s large scale enables exploration of complex comorbidities and public health trends.12 That such findings could be derived from routine claims data attests to the strength of the JROAD-DPC as a research foundation for identifying nationwide gaps and opportunities in CVD care.
The clinical significance of the severity distributions presented in this study is a prime example of the registry’s utility in interpreting these trends. For angina, the proportional distribution of CCS classes remained remarkably stable despite a sharp decline in total hospitalizations in 2020. This suggests that the reduction in admissions, likely influenced by the COVID-19 pandemic13 and evolving guidelines following the ISCHEMIA trial,14,15 primarily affected patients with less severe conditions who could be managed with optimal medical therapy, leaving a consistent clinical profile for the cohort requiring inpatient care. In contrast, for AMI, the stability of the Killip classification distribution occurred alongside a steady increase in total admissions, which suggest that the registry has become progressively more comprehensive in capturing the full spectrum of AMI severity nationwide without significant selection bias, reinforcing its value as a robust platform for outcomes research.
Beyond ischemic heart disease, one of the most striking findings within our registry was the more than fivefold increase in percutaneous catheter ablations for atrial fibrillation/flutter. This PP cannot establish causality, but several concurrent trends in Japan may provide context for this observation. First, the aging of the population is associated with a rising number of patients with atrial fibrillation.16 Second, the evolution of ablation technology could be a contributing factor. Such advancements may have contributed to the procedure’s standardization and broader adoption across a variety of institutions.17 Finally, there has been a significant nationwide expansion in the provision of this procedure. The total number of catheter ablation procedures and the number of hospitals performing them in Japan have consistently increased over the past decade,18 and this overall growth in procedural capacity is a key element in understanding the increased number of ablations recorded in our registry.
Looking ahead, the JROAD-DPC registry offers a valuable springboard for future investigations and subsequent improvements in CVD health. Its continued expansion over time will allow researchers to monitor longitudinal trends and the impact of emerging challenges in near real-time.10 Notably, the platform has already been used to assess the system-wide effects of the COVID-19 pandemic on cardiac care in Japan,10,13,19 demonstrating the agility of big data to inform public health responses. Additionally, comparative and international studies are anticipated to place Japan’s CVD outcomes in global context.
Study Limitations
When interpreting the results from JROAD-DPC data, careful consideration must be given to several limitations inherent to the characteristics of DPC data. One significant limitation is that the data does not capture medical care provided at other institutions. Because JROAD-DPC data is collected individually from each facility and contains only digitized and coded medical procedures performed within that institution, it excludes any treatment information from other facilities. For instance, when patients have been receiving oral medication therapy before admission and bring medications such as aspirin that are relevant to our analysis, that information is not reflected in the data. While the extent to which this affects prescription rates remains unclear, hospitals that receive many patients through referrals may show apparently lower prescription rates. Therefore, interfacility comparisons should be interpreted with appropriate caution and considered only as reference points.
Another important limitation concerns the absence of information regarding clinical decision-making processes, particularly the reasons why standard treatments were not implemented. Although standard care represents treatment principles based on scientific evidence and expert consensus that are considered effective for most patients, there are circumstances where medication should be avoided due to comorbidities or deteriorated physiological function, or where patients themselves decline standard treatment. The medical procedures captured in the data merely represent the outcomes of such clinical decisions, and the underlying reasoning process is not necessarily evident from the data alone. This represents a fundamental limitation of coded electronic data and addressing it would require examining medical records that include information from other institutions.
Furthermore, the medical procedure coding system sometimes lacks the clinical information necessary for comprehensive understanding of actual practice patterns. The DPC survey data utilized in this study represents diagnoses and medical procedures through billing codes designed primarily for reimbursement claims. Because the primary purpose of these codes is billing rather than clinical documentation, they may not fully capture all clinically relevant information. Although the DPC survey data itself is not directly related to reimbursement, we must acknowledge that the data may not comprehensively represent all medical procedures performed.
While extensive, JROAD-DPC represents approximately 70% of the cardiovascular specialist training facilities and their affiliates participating in JROAD, so generalizability to the remaining non-participating centers may be limited. Furthermore, participation expanded over time; therefore, absolute counts reflect both clinical volume and registry coverage and should not be interpreted as national incidence. Specific analyses focusing on broader representativeness are beyond the scope of this design paper.
Finally, formal validation studies can provide confidence in the registry’s diagnostic accuracy for major CVD,9 but it is important to acknowledge the possibility of other data errors, a limitation common to all large administrative databases. The DPC data consists of standardized medical information that has improved in accuracy through continued utilization, but the potential for billing omissions and coding errors persists. Consequently, some patients counted as not receiving standard care may have received such treatment that was not recorded in the data. As the utilization of this data continues to expand, we can expect further improvements in data accuracy over time.
Conclusions
The JROAD-DPC registry serves as a large-scale, nationwide platform providing valuable real-world insights into CVD care in Japan. Its design enables robust research on clinical trends, quality of care, and outcomes. Despite inherent limitations common to administrative datasets, JROAD-DPC is an essential resource for generating evidence to inform and advance Japan’s national strategies for overcoming CVD.
Disclosure
S.Y., S.K., T.M., Y.K.B., and Y.F. are members of Circulation Reports’ Editorial Team. The authors declare that there are no conflicts of interest.
IRB Information
The present study was approved by the Ethics Committee of the Japanese Circulation Society. Reference number: 1-1-9.
Supplementary Files
Supplementary Table 1. Supplementary Table 2. Supplementary Table 3.
Acknowledgments
We thank all facilities participating in the JROAD-DPC investigation.
Data Availability
The deidentified participant data will not be shared.
References
- 1. Yasuda S, Miyamoto Y, Ogawa H.. Current status of cardiovascular medicine in the aging society of Japan. Circulation 2018; 138: 965–967. [DOI] [PubMed] [Google Scholar]
- 2. Japanese Circulation Society/Japan Stroke Society.. The Five-Year Plan for Overcoming Cardiovascular Disease 2016 (in Japanese). https://www.j-circ.or.jp/five_year/files/five_year_plan.pdf (accessed October 2025).
- 3. Yasuda S, Nakao K, Nishimura K, Miyamoto Y, Sumita Y, Shishido T, et al.. The current status of cardiovascular medicine in Japan: Analysis of a large number of health records from a nationwide claim-based database, JROAD-DPC. Circ J 2016; 80: 2327–2335. [DOI] [PubMed] [Google Scholar]
- 4. Nishi M, Miyamoto Y, Iwanaga Y, Kanaoka K, Sumita Y, Ishihara M, et al.. Hospitalized patients, treatments, and quality of care for cardiovascular diseases in Japan: Outline of the nationwide JROAD investigation. Circ J 2025; 89: 1081–1086. [DOI] [PubMed] [Google Scholar]
- 5. Hayashida K, Murakami G, Matsuda S, Fushimi K.. History and profile of Diagnosis Procedure Combination (DPC): Development of a real data collection system for acute inpatient care in Japan. J Epidemiol 2021; 31: 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Yasunaga H.. Updated information on the Diagnosis Procedure Combination data. Ann Clin Epidemiol 2024; 6: 106–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Campeau L.. Letter: Grading of angina pectoris. Circulation 1976; 54: 522–523. [PubMed] [Google Scholar]
- 8. Hatano K, Ohe K.. Information retrieval system for Japanese Standard Disease-code Master Using XML Web Service. AMIA Annu Symp Proc 2003; 2003: 859. [PMC free article] [PubMed] [Google Scholar]
- 9. Nakai M, Iwanaga Y, Sumita Y, Kanaoka K, Kawakami R, Ishii M, et al.. Validation of acute myocardial infarction and heart failure diagnoses in hospitalized patients with the nationwide claim-based JROAD-DPC database. Circ Rep 2021; 3: 131–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sugimoto T, Mizuno A, Yoneoka D, Matsumoto S, Matsumoto C, Matsue Y, et al.. Cardiovascular hospitalizations and hospitalization costs in Japan during the COVID-19 pandemic. Circ Rep 2023; 5: 381–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Nakao K, Yasuda S, Nishimura K, Noguchi T, Nakai M, Miyamoto Y, et al.. Prescription rates of guideline-directed medications are associated with in-hospital mortality among Japanese patients with acute myocardial infarction: A report from JROAD-DPC study. J Am Heart Assoc 2019; 8: e009692. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Nakai M, Iwanaga Y, Sumita Y, Wada S, Hiramatsu H, Iihara K, et al.. Associations among cardiovascular and cerebrovascular diseases: Analysis of the nationwide claims-based JROAD-DPC dataset. PLoS One 2022; 17: e0264390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kimura M, Matoba T, Nakano Y, Katsuki S, Sakamoto K, Nishihara M, et al.. Impact of the coronavirus disease 2019 (COVID-19) pandemic on the severity and the mortality of acute myocardial infarction in Japan: Analysis from the JROAD-DPC database. Circ Rep 2024; 6: 191–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Maron DJ, Hochman JS, Reynolds HR, Bangalore S, O’Brien SM, Boden WE, et al.. Initial invasive or conservative strategy for stable coronary disease. N Engl J Med 2020; 382: 1395–1407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Nakano S, Kohsaka S, Chikamori T, Fukushima K, Kobayashi Y, Kozuma K, et al.. JCS 2022 Guideline Focused Update on diagnosis and treatment in patients with stable coronary artery disease. Circ J 2022; 86: 882–915. [DOI] [PubMed] [Google Scholar]
- 16. Kodani E, Kaneko T, Fujii H, Nakamura H, Sasabe H, Tamura Y, et al.. Prevalence and incidence of atrial fibrillation in the general population based on National Health Insurance Special Health Checkups: TAMA MED Project-AF. Circ J 2019; 83: 524–531. [DOI] [PubMed] [Google Scholar]
- 17. Kanaoka K, Nishida T, Nishioka Y, Myojin T, Kubo S, Soeda T, et al.. The impact of hospital case volume on the outcomes after catheter ablation for atrial fibrillation according to the ablation technology. J Cardiovasc Electrophysiol 2022; 33: 1394–1402. [DOI] [PubMed] [Google Scholar]
- 18. Inoue T, Kuwabara H.. Regional variation in the use of catheter ablation for patients with arrhythmia in Japan. J Arrhythm 2021; 37: 22–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Sugimoto T, Mizuno A, Yoneoka D, Matsumoto S, Matsumoto C, Matsue Y, et al.. Hospitalizations for cardiovascular diseases during the early stage of the COVID-19 pandemic in Japan. Circ Rep 2022; 4: 353–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1. Supplementary Table 2. Supplementary Table 3.
Data Availability Statement
The deidentified participant data will not be shared.