Trial Design

The Acute Myocardial Infarction (CAMI) Registry: A national long-term registry- research-education integrated platform for exploring acute myocardial infarction in China Haiyan Xu, MD, PhD, a Wei Li, PhD, a Jingang Yang, MD, a Stephen D. Wiviott, MD, b Marc S. Sabatine, MD, MPH, b Eric D. Peterson, MD, MPH, c Ying Xian, MD, PhD, c Matthew T. Roe, MD, MHS, c Wei Zhao, MBBS, a Yang Wang, MS, a Xinran Tang, BS, a Xuan Jia, BS, a Yuan Wu, MD, a Runlin Gao, MD, a and Yuejin Yang, MD, PhD a , on behalf of the CAMI Registry study group , China; MA, and NC, USA

Background Acute myocardial infarction (AMI) has become a major cause of hospitalization and mortality in China. There has been limited data to date available to characterize AMI presentation, contemporary patterns of medical care, and outcomes in China. Aims The CAMI Registry is a national project with the objectives to timely obtain real-world knowledge about AMI patients and to provide the platform for clinical research, guide preventive measures and care quality improvement efforts in China. Methods and Progress The CAMI registry is a prospective, nationwide, multicenter observational study for AMI patients. The registry includes three levels of hospitals (representing typical Chinese governmental and administrative models) from all provinces and municipalities throughout Mainland China except Hong Kong and Macau. Sites were instructed to enroll consecutive patients with a primary diagnosis of AMI. Clinical data, treatments, outcomes and cost are collected by local investigators and captured electronically, with a standardized set of variables and standard definitions, and rigorous data quality control. Post-discharge patient follow-up to 2 years is planned. The CAMI Registry was launched in January 2013. A total of 108 hospitals have participated in the registry so far. As of September 2014, 26,103 patients with AMI were registered. Conclusions The CAMI registry represents a well-supported and the largest national long-term registry-research- education platform for surveillance, research, prevention and care improvement for AMI in China, the world's most populous nation. The broad representation of all provinces and different-level hospitals will allow for the exploration of AMI across diverse geographic regions and economic circumstances. (Am Heart J 2016;175:193-201.e3)

The mortality of acute myocardial infarction (AMI) has reperfusion, effective antithrombotic therapy, and inten- been declining over the past 30 years in US and European sive evidence-based medication into routine clinical countries.1–3 The significant improvement in outcomes is practice.4,5 concurrent with the translation and integration of Whereas in China, the morbidity of ischemic heart evidence-based AMI care including early myocardial disease including AMI has sharply increased and become a major cause of emergency medical care, hospitalization and death over the past few decades.6,7 This increase in part represents economic and medical progress that has From the aFuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy led to declines in infectious diseases and a shift to chronic b of Medical Science and Peking Union Medical College, Beijing, China, TIMI Study Group, diseases in China.8,9 However, increasing urbanization, Cardiovascular Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, and cDuke Clinical Research Institute, Duke University Medical Center, economic prosperity and western influences have also Durham, NC. increased consumption of animal fats, cigarette smoking, Clinical Trial Registration: http://www.clinicaltrials.gov Identifier: NCT01874691. and obesity, all leading to the marked increase in Submitted March 17, 2015; accepted April 9, 2015. cardiovascular diseases, especially in the younger and Reprint requests: Yuejin Yang, M.D., PhD, Department of Cardiology, Fuwai Hospital, 167 10,11 Beilishi Road, Beijing 100037, People’s Republic of China. rural population. The broad geography, unbalanced E-mail: [email protected] economy, different population, and varied life-styles in 0002-8703 China may bring about variation in prevalence and © 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). clinical characteristics of AMI, but epidemiologic nation- http://dx.doi.org/10.1016/j.ahj.2015.04.014 wide data across the whole country is lacking. American Heart Journal 194 Xu et al May 2016

Moreover, there are several challenges with AMI care research and education platform (Figure 1). The database and outcomes in China. The medical care patterns of AMI will be openly shared to all the participating investigators have changed over the past decades. Health care systems for analyses and research. The project is currently funded are in the process of the reform in China, but still not as one of the National Twelfth Five-year Science and optimizedforthedramaticincreaseinAMIinci- Technology Support Projects by Ministry of Science and dence.12,13 Patients in different areas and population Technology of China (Grant No. 2011BAI11B02). It is may get differential access to hospitals and medical care organized and conducted by Fuwai Hospital, National which may lead to disparate outcomes.14,15 In routine Center for Cardiovascular Diseases of China (NCCD). clinical practice, we found many patients delayed The organizational structure of the CAMI registry is presentation to hospital and some patients were mis- displayed in Figure 2. The principal investigator (PI) is diagnosed.16,17 Still some patients being correctly diag- responsible for all aspects of the project, including the nosed didn’t receive effective, timely and appropriate management and integrity of the design, conduct, and treatments.18 But limited comprehensive data across report of the research project. The Scientific Committee China is available to understand the roles of these factors. consists of experienced American cardiologists and clinical Furthermore, it is unclear whether the strategies accord- trialists and Chinese cardiology leaders. The members of ing to evidence-based medicine and guidelines from Scientific Committee help to assure the scientific and western countries are completely applicable and optimal rational design of the registry, and supervise data for Chinese patients, due to differences between China collection, registry operation, quality control, and data and western countries. analysis. The Data Monitoring Committee is in charge of To date, China, the most populous country in the world, data quality evaluation and supervision. The members of with broad geography, varied life-styles and wide differ- Executive and Steering Committee manage the design and ences in socioeconomic status, and disparate medical care, execution of the registry. The Clinical Support Team has not had a representative national long-term registry consists of eight clinical cardiologists who are trained to be program for the patients with AMI. Therefore, we designed responsible for medical aspects in the registry, including and launched China AMI (CAMI) registry, a nationwide supporting the investigators, answering medical hotline, registry of the hospitalized patients with AMI as an and assist in the supervising of the registry process and integrated research and education platform in China. on-site audits. Project managers are in charge of coordinat- ing and managing the operational process. The Data Management and Statistics Teams are managed by Medical Methods Research and Biostatistics Center, NCCD. The Analysis and Objectives of the CAMI Registry Publication Committee is responsible for reviewing and The specific aims of the CAMI Registry are to (1) serve approving research proposal, and overseeing the genera- as a national long-term hospital-based registry and tion of abstracts and manuscripts from the database. Each surveillance program for AMI to timely obtain real-world hospital has a local study group, including local PI, research information about clinical characteristics, medical care coordinator, investigators, research assistants, and local and outcomes of Chinese patients with AMI across quality control person. The members of committees and different provinces, prefectures and counties; (2) analyze teams are listed in online Appendix A. risk factors and clinical presentations among different patients across different regions, and trends over time, to Ethical review and patient privacy develop scientific preventive strategies for AMI, especial- This project was approved by the institutional review ly among individuals at high risk; (3) aid in the board central committee at Fuwai Hospital, NCCD of development of effective strategies and update of China. The CAMI registry study group maintains the guidelines applicable to Chinese patients and, thus, institutional review board–approved protocol that de- improve medical quality and outcomes for AMI through scribes methods used to protect the privacy of patients implementation into clinical practice; (4) establish a and maintain confidentiality of data collected. Written risk-adjusted evaluation system of cost-effectiveness and informed consent is to be obtained from eligible patient clinical pathways suitable for Chinese AMI patients in before registration. However, to avoid a registration bias different-level hospitals and across different regions; (5) which informed consent may bring and to make sure all set up a national platform and provide a valuable eligible patients being registered, investigators can nationwide infrastructure to advance research, future register the patients without providing informed consent multi-center clinical trials, physician and patient educa- but data is anonymous and omits private information. tion, and care quality improvement in China.

Overview and organization of the CAMI registry Site selection The CAMI registry is a prospective, nationwide, According to Chinese administrative divisions, there multi-center, and observational registry as an integrated are five practical levels of local government: province, American Heart Journal Xu et al 195 Volume 175

Figure 1

Design overview of the CAMI Registry. prefecture, county, township, and village, typical of the and urban. The vertical administrative relationship of Chinese vertical governmental and administrative model. three levels of hospitals can reflect the transfer status and In China, public hospitals were founded following this medical care system for AMI patients. structure and provide the majority of health care. In Chinese administrative partition, there are 27 provinces Patient population and inclusion criteria and 4 municipalities (directly under the central govern- Participating sites are required to enroll consecutive ment) in Mainland China (except Hong Kong and Macau AMI patients to the registry. Eligible patients must be as Special Administrative Regions). In every province, admitted within 7 days of acute ischemic symptoms with there are some prefectures, in each of which there are a primary clinical diagnosis of AMI, including ST-segment some counties. The CAMI registry includes 3 levels of elevation myocardial infarction (STEMI) or non–ST- hospitals that cover 27 provinces and 4 municipalities elevation myocardial infarction (NSTEMI). Final inclusion throughout Mainland China (Figure 3). In every province criteria must meet third Universal Definition for Myocar- or municipality, we invited one provincial-level academic dial Infarction (2012).19 According to the classification of hospital (first level) which is one of the largest central myocardial infarction, types 1, 2, and 3 and types 4b and hospitals in its own province or municipality. Then the 4c are included in the present registry. Types 4a and type local PIs of these hospitals as provincial cardiology 5 are not eligible for the CAMI registry. leaders recommended one to three prefectural-level hospitals (second level) in their own provinces or Data collection municipalities, and then 1 to 4 county-level hospitals The data elements collected in the CAMI registry were (third level) in the selected prefectures. These three developed and determined through the discussion of levels of hospitals reflect typical Chinese governmental principal investigator, clinical and research experts of the and administrative model in China and all these hospitals Scientific Committee and Executive and Steering Com- are among the largest or central hospitals within their mittee, with input from data managers, statisticians and regions. Thus, this list of 108 participating hospitals is to from CAMI registry investigators regarding the feasibility assure a broad representation of hospitals across three and burden of data collection. Standardized data collec- levels, with broad coverage of geographic region, rural tion, with particular consideration of Chinese patients American Heart Journal 196 Xu et al May 2016

Figure 2

CAMI Registry scientific and operational structure. EDC, Electronic data capture. and hospitals, includes patient demographics, clinical diagnosed with AMI and meet the inclusion criteria, the presentation, medical history, risk factors, triggering front page of electronic case report form (eCRF) must be factors, physical examination, laboratory and imaging filled out and submitted online within 24 hours from results, transfer facility therapies, reperfusion strategies, admission. Each patient will be assigned a unique ID medications, clinical events, and cost (Table I). All the identified through patient social ID number to ensure no information is collected using the standardized set of duplicate data input and track post-discharge follow-up variables and standard definitions (elements dictionary), and data query and revision. Site investigators are systematic data entry and transmission procedures, and required to collect all the data during the hospitalization rigorous data quality control. Definitions of elements and submit the completed eCRF upon the patient’s were based on the ACC/AHA Task Force on clinical Data discharge or death. Data input tracking, regular alerts, Standards and NCDR-ACTION-GWTG element dictio- rigorous data monitoring and queries are used to support nary.20–23 All variables were coded with CDISC, ICD-10, timely and accurate completion of the eCRF. MedDra, and WHO-DD to make them standardized. Data are collected, validated, and submitted through a Patient follow-up secure, password-protected, web-based electronic data When a patient is discharged, he or she receives an capture system (http://www.CAMIRegistry.org) by the educational brochure specialized for AMI patients to give local investigators at each participating site. Data the recommendations about healthy life style, medication collection during hospitalization and follow-up is per- and to arrange follow-up visits. Follow-up visits are formed by trained clinical cardiologists or cardiovascular planned at 30 days, 6, 12, 18, 24 months. The events fellows to ensure the accuracy and reliability of data. (including death, cardiovascular events, bleeding, and Element definitions are accessible to investigators auto- so on), medications, the reasons of medication discon- matically at the point of data entry. Once a patient tinuation, and cost of medication will be reviewed admitted to emergency department or in-patient unit is and collected either in person at clinic visit or by American Heart Journal Xu et al 197 Volume 175

Figure 3

Geographic distribution of hospitals throughout Mainland China (except Hong Kong and Macau). telephone call. The clinical events must be validated by educate and support the investigative sites. During these source documents. meetings, investigators can share their experience, learn methodology about clinical research, and receive com- Site start-up and investigator training mendation for outstanding performance in the registry. As the first step to participation, the invited sites were required to complete a survey form about the facilities of Data management and quality control their hospitals, the departments of emergency medicine The NCCD of China serves as the data entry system and cardiology, including cardiac care unit (CCU), provider and data coordinating center for the CAMI catheterization laboratory, surgical capabilities, bed registry. The database is stored and managed according to number in cardiology, and annual volume of AMI. The national information security protection law at Computer kick-off investigator meeting and first training workshop Informatics Center of NCCD. were held on August 11th, 2012. Before every site started, There are five steps to support data quality control in all investigators received detailed training on the proto- the registry: (1) Web-based data entry access is password- col, standard definitions of the elements, web-based restricted to trained and test-passed personnel at each software system, data collection and data entry proce- hospital. (2) Real-time automatic logic and range check dures. To participate, each site investigator must pass a on the completeness and validity of the data are registry exam for a test subject before accessing the live integrated in the data entry system to control basic data data-entry system. We performed a pilot registry to quality at the point-of-entry. (3) Data checks are provide the opportunity of practice for investigators and performed by the appointed and trained local cardiolo- improvement of eCRF in the last 2 months of 2012. gists responsible for the quality of data collection at their A web site accessible to registry sites provides own hospitals. Sites can access their registry data to education on definitions, diagnosis, and management; monitor local performance. (4) The Data Management current guidelines for AMI; and hotline discussion forum. Team regularly provides data quality checks and sends In addition, the regular teleconferences, annual face-to- queries for illogical, invalid or missing data elements to face conferences and regional meetings are held to participating hospitals to review and revise. (5) On-site American Heart Journal 198 Xu et al May 2016

Table I. The CAMI Registry data elements Table II. Characteristics of participating hospitals

Category Contents/example elements Hospital rank Provincial-level Prefectural-level County-level Total

Patient ID number, date of birth, sex, height, weight, Number of 31 45 32 108 demographics race, education, occupation, insurance, marriage hospitals Clinical presentation Symptoms, time of presentation, triggering factor Number of beds 124 (102,288) 83 (50,122) 47 (36,65) NA Initial medical First medical contact, transfer, in cardiology, contact electrocardiography, cardiac status median (IQR) Medical history Hypertension, diabetes, hyperlipidemia, smoking, CCU, n(%) 31 (100%) 43 (96%) 25 (78%) 99 (92%) and risk factors alcohol, family history, prior cardiovascular Coronary 31 (100%) 42 (93%) 14 (44%) 87 (81%) disease, prior revascularization, chronic kidney catheterization disease, lung disease, peptic ulcer, cancer Lab, n (%) Emergent reperfusion Thrombolysis, primary and for STEMI rescue PCI, complications Staged/Selected Selected PCI, complications, CABG revascularization Table III. Baseline characteristics of patients (N = 26103) Revascularization PCI, reason for emergent PCI, complications Variables % for NSTEMI Medications Antiplatelets, heparin, statin, β-blocker, other medications used one week before Age, y, median (IQR) 63 (53, 72) AMI, during hospitalization and on discharge Female sex 26 Lab results Cardiac biomarkers, lipid, creatinine, Ethnic Han 96 hemoglobin A1C, BNP, LVEF Medical insurance or employee payment 93 In-hospital outcomes Death, heart failure, reinfarction, arrhythmia, Hospital type cardiac arrest, cardiogenic shock, Provincial 32 mechanical complication, stroke, bleeding Prefecture 54 Discharge Discharge status, in-hospital County 14 2 duration, cost, medications BMI, kg/m , mean ± SD 24 ± 10 Follow-up Vital status; clinical events such as death, heart Current smoker 44 failure, MI, revascularization, Medical History readmission; medications; cost Hypertension 52 Diabetes 20 CAD, Coronary artery disease; MI, myocardial infarction; PCI, percutaneous coronary Known hyperlipidemia 8 BNP LVEF intervention; , brain natriuretic peptide; , left ventricular ejection fraction; Peripheral artery disease 0.7 CABG , coronary artery bypass graft. Prior MI 8 Prior stroke 10

BMI, Body mass index; MI, myocardial infarction. audits by trained auditors and cardiologists at Clinical Support Team are planned to go through and review medical records of patients drawn randomly from each β-blocker during the hospitalization after AMI were participating hospitals and check the accuracy of data. high (N90% for first four medications). The unadjusted in-hospital mortality rates were 4.9% for STEMI and 4.2% Progress to date for NSTEMI. The CAMI registry was launched on January 1, 2013. A total of 108 three levels of hospitals (including 31 provincial-level hospitals, 45 prefectural-level hospitals, Discussion and 32 county-level hospitals) from 27 provinces and 4 The CAMI registry represents a unique, large-scale, municipalities throughout Mainland China have partici- national, contemporary and long-term registry project for pated in the registry. The geographical distribution of AMI patients to establish a registry-research-education these hospitals is seen in Figure 3. The names, location, platform for surveillance, multicenter clinical research, and local PIs are listed in online Appendix B. The translational medicine, prevention, and care improve- characteristics of participating hospitals according to ment in China. three levels are displayed in Table II. Through the CAMI registry, comprehensive and con- Through September of 2014, a total of 26103 patients tinuous understanding of AMI epidemiology, real-world with AMI have been registered. Table III showed the practice care, outcomes, and cost in China obtained in baseline characteristics of these patients. The median age time and comparison with other countries will be getting was 63 years. There were approximately one quarter of possible. For example, from the preliminary data, we can women (26%). About 44% were current smokers, 52% see that the proportion of female patients with AMI was had hypertension, and 20% had a history of diabetes. lower than that in the United States (26% vs 36% Figure 4 illustrates that the use of aspirin, P2Y12 receptor respectively) and the prevalence of diabetes was also inhibitor, statins, heparin/fondaparinux, ACEI/ARB, lower (20% vs. 30%, respectively). But smoking is much American Heart Journal Xu et al 199 Volume 175

Figure 4

Medication use of patients during the hospitalization in AMI patients. more common than in the USA (current smoker: 44% vs. gies and, thus, inform the general public regarding 35%, respectively).24 We will get information on medical primary prevention, early recognition, and response to performance and adherence to evidence-based AMI. The second is the clinician-level effort through care medications at the patient level, clinician level, and quality feedback and scientific support. We will be able to system level in different-level hospitals across China. The encourage and assess the adoption and implementation analyses on registry data have the potential to help guide of guideline recommendations and evidence-based ther- to optimize safer and more effective therapeutic apeutic strategies, aiming to translate rapidly and directly strategies in clinical practice. into clinical practice and improve care quality across This registry not only acts as an observational study and different-level hospitals. For hospital administrative-level a care quality improvement effort for AMI but also has education, it could help the administrative departments vast potential as a scientific resource to advance medical of hospitals or health authorities improve the system research in patients with AMI.24,25 We encourage all the quality of care for patients with AMI through the analyses investigators to share and utilize the database for clinical of the system factors affecting clinical outcomes. research and educational programs. It can also play the There are several strengths of the CAMI registry. One is role in evaluation for the safety and efficacy of new drugs that we selected a diverse representing group of hospitals and devices in real-world practice for the patients with (including provincial-level, prefectural-level, and county- AMI, with the possibility of nested clinical trials.25,26 level hospitals, typical vertical model) covering all the Furthermore, this registry could contribute important provinces and municipalities across Mainland China, thus scientific questions for further clinical research. It could making the population in the registry representative of provide the metrics on clinical end point events used in different regions, different economic levels, and different the design of clinical trials through longitudinal prognos- access to medical resources. These hospitals can ade- tic evaluation of Chinese patients with AMI . The quately reflect the performance of the Chinese medical registered patients could be a pool of selected study system. The second strength is the direct access to subjects for clinical trials, and the collected information information during hospitalization and follow-up in this could be shared through direct linking to trials. The prospective registry and data collection directly per- hospitals could be the pool of selected sites according to formed by cardiologists or cardiovascular fellows, ensur- their performance in the registry. Thus, this program will ing accuracy and reliability of data. Another advantage of build a registry-research platform and research network the registry is that 2 years of follow-up can provide a for myocardial infarction. dynamic and long-term observation on the change of The CAMI registry may serve as a resource to educate medication use, care, and outcomes for patients with patients, physicians, and administrative personnels. myocardial infarction over time for years to come. Finally, Based on the registry, we hope to establish an education we also collect the costs during hospitalization and after platform aimed to facilitate efforts to decrease the discharge to analyze cost-effectiveness and burden for prevalence of AMI and improve the quality of AMI patient AMI patients. care. For patient-level education, we hope to propose The potential limitation is that any registry may not be scientific preventive measures and precautionary strate- fully representative of all patients and all hospitals. American Heart Journal 200 Xu et al May 2016

However, we believe our registry will reflect the routine trends in management and outcome in the reperfusion era. J clinical practice of AMI care in China because of Cardiovasc Risk 2001;8:21-9. geographical and hospital-level diversity. In addition, 6. ZhangXF,HuDY,DingRJ,etal.Status and trend of cardio-cerebral- the data collection burden for investigators may be the vascular diseases mortality in China: data from national disease greatest barrier to the registry that may lead to some surveillance system between 2004 and 2008. Article in ChineseZhonghua Xin Xue Guan Bing Za Zhi 2012;40:179-87. enrollment bias. We have carefully considered each 7. Li J, Li X, Wang Q, et al. ST-segment elevation myocardial infarction in element to limit the burden and have project managers, China from 2001 to 2011 (the China PEACE-Retrospective Acute coordinators, and auditors to assist in the registry. Myocardial Infarction Study): a retrospective analysis of hospital data. Lancet 2015;385:441-51. 8. Yang G, Wang Y, Zeng Y, et al. Rapid health in China, 1990–2010: Conclusions findings from the global burden of disease study 2010. Lancet This registry represents the largest Chinese national 2013;381:1987-2015. registry-research-education platform for surveillance, 9. Yang G, Kong L, Zhao W, et al. Emergence of chronic research, prevention and care quality improvement for non-communicable diseases in China. Lancet 2008;372:1697-705. AMI. The broad representation of different-level hospitals 10. Cao CF, Ren JY, Zhou XH, et al. Twenty-year trends in major cardiovascular risk factors in hospitalized patients with acute in all provinces will allow for the exploration of AMI in myocardial infarction in Beijing. Chin Med J (Engl) 2013;126: China, a geographically wide and economically diverse 4210-5. and the most populous country in the world. 11. Hu SS, Kong LZ, Gao RL, et al. Outline of the report on cardiovascular disease in China, 2010. Biomed Environ Sci 2012;25:251-6. 12. Yip WC, Hsiao W, Meng Q, et al. Realignment of incentives for Conflicts of interest health-care providers in China. Lancet 2010;375:1120-30. None. 13. Yip WC, Hsiao WC, Chen W, et al. Early appraisal of China's huge and complex health-care reforms. Lancet 2012;379:833-42. 14. Gao R, Patel A, Gao W, et al. Prospective observational study of acute Acknowledgements coronary syndromes in China: practice patterns and outcomes. Heart We are very grateful to the TIMI Study Group and the 2008;94:554-60. 15. Liu CY, Lin YN, Lin CL, et al. Cardiologist service volume, Duke Clinical Research Institute for their contributions in percutaneous coronary intervention and hospital level in relation to the design, conduct, and data analyses. We would like to medical costs and mortality in patients with acute myocardial thank all of the members of Scientific Committee, Data infarction: a nationwide study. QJM 2014;107:557-64. Monitoring Committee, and Executive and Steering Com- 16. Peng YG, Feng JJ, Guo LF, et al. Factors associated with prehospital mittee for their contribution to the CAMI registry. We also delay in patients with ST-segment elevation acute myocardial want to thank all of the study investigators and coordinators infarction in China. Am J Emerg Med 2014;32:349-55. for their great work and all of the colleagues of Medical 17. Ranasinghe I, Rong Y, Du X, et al. System barriers to the Research & Biometrics Center and Information Technology evidence-based care of acute coronary syndrome patients in China: Center. Thanks are given to professor Yiqing Song for his qualitative analysis. Circ Cardiovasc Qual Outcomes 2014;7: 209-16. help in the manuscript preparation and writing. We also 18. Bi Y, Gao R, Patel A, et al. Evidence-based medication use among appreciate the financial support from the Ministry of Science Chinese patients with acute coronary syndromes at the time of hospital and Technology of China (Grant No. 2011BAI11B02). discharge and 1 year after hospitalization: results from the Clinical Pathways for Acute Coronary Syndromes in China (CPACS) study. Am Heart J 2009;157:509-16. References 19. Thygesen K, Alpert JS, Jaffe AS, et al. The Writing Group on behalf of 1. Rogers WJ, Frederick PD, Stoehr E, et al. Trends in presenting characteristics the Joint ESC/ACCF/AHA/WHF Task Force for the Universal and hospital mortality among patients with ST elevation and non-ST Definition of Myocardial Infarction. Third universal definition of elevation myocardial infarction in the National Registry of Myocardial myocardial infarction. Circulation 2012;126:2020-35. Infarction from 1990 to 2006. Am Heart J 2008;156:1026-34. 20. Cannon CP, Battler A, Brindis RG, et al. American College of 2. Yeh RW, Sidney S, Chandra M, et al. Population trends in the Cardiology key data elements and definitions for measuring the incidence and outcomes of acute myocardial infarction. N Engl J Med clinical management and outcomes of patients with acute coronary 2010;362:2155-65. syndromes. A report of the American College of Cardiology Task 3. Laribi S, Aouba A, Resche-Rigon M, et al. Trends in death attributed to Force on Clinical Data Standards (Acute Coronary Syndromes myocardial infarction, heart failure and pulmonary embolism in Writing Committee). J Am Coll Cardiol 2001;38:2114-30. Europe and Canada over the last decade. QJM 2014;107:813-20. 21. ACCF/AHA 2011 Key Data Elements and Definitions of a Base 4. Gibson CM, Pride YB, Frederick PD, et al. Trends in reperfusion Cardiovascular Vocabulary for Electronic Health Records. A Report of strategies, door-to-needle and door-to-balloon times, and in-hospital the American College of Cardiology Foundation/American Heart mortality among patients with ST-segment elevation myocardial Association Task Force on Clinical Data Standards. J Am Coll Cardiol infarction enrolled in the National Registry of Myocardial Infarction 2011;58:202-22. from 1990 to 2006. Am Heart J 2008;156:1035-44. 22. Cannon CP, Brindis RG, Chaitman BR, et al. 2013 ACCF/AHA key data 5. Barakat K, Wilkinson P, Suliman A, et al. Changing face of acute elements and definitions for measuring the clinical management and myocardial infarction in east London: a prospective cohort study of outcomes of patients with acute coronary syndromes and coronary artery American Heart Journal Xu et al 201 Volume 175

disease: a report of the American College of Cardiology Foundation/ quality improvement for acute myocardial infarction. Circ Cardiovasc American Heart Association Task Force on Clinical Data Standards Qual Outcomes 2009;2:491-9. (Writing Committee to Develop Acute Coronary Syndromes and Coronary 25. Peterson ED, Roe MT, Chen AY, et al. The NCDR ACTION Artery Disease Clinical Data Standards). Circulation 2013;127:1052-89. Registry-GWTG: transforming contemporary acute myocardial 23. http://www.NCDR.com. infarction clinical care. Heart 2010;96:1798-802. 24. Peterson ED, Roe MT, Rumsfeld JS, et al. A call to ACTION (acute 26. Udell JA, Wang TY, Li S, et al. Clinical trial participation after coronary treatment and intervention outcomes network): a national myocardial infarction in a national cardiovascular data registry. effort to promote timely clinical feedback and support continuous JAMA 2014;312:841-3. American Heart Journal Xu et al 201.e1 Volume 175

Appendix A. Members of committees Yao, Fengying Chen, Likun Ma, Tao Guo, Bin Li, Gesang Luobu, Wei Li. Clinical Support Group: Principal investigator: Xiaojin Gao, Xuan Zhang, Yunqing Ye, Qiuting Dong, Yuejin Yang Rui Fu, Xinxin Yan, Hui Sun, Peiyuan He Scientific Committee: Project managers: Stephen D. Wiviott, Marc S. Sabatine, Eric D. Peterson, Matthew T. Roe, Ying Xian (USA) Rongfu Li, Hong Wang, Chen Jin Runlin Gao, Zaijia Chen, Weifeng Shen, Wenling Zhu, Website and EDC Team: Luhua Shen, Chen, Yaling Han, Yude Chen, Shuigao Jin, Dongfeng Gu, Manlu Zhu (China) Wei Zhao, Jing Yuan, Ting Zhou, Yue Sun, Hui Li, Data Monitoring Committee: Lihua Zhang Data Management Team: Shubin Qiao, Bo Xu : Executive and Steering Committee Yi Sun, Xiaomeng Li, Juan Li Ruiyan Zhang, Xiaozeng Wang, Weiming Li, Daowen Monitor Team: Wang, Jiyan Chen, Xianghua Fu, Jianan Wang, Zheng Wan, Chuanyu Gao, Hongwei Li, Yitong Ma, Yang Xuan Jia, Xiaoru Cheng, Guofang Qiao, Jing Li, Xin Zheng, lianqun Cui, Bao Li, Zuyi Yuan, Yin Liu, Huang Shaobin Jia, Biao Xu, Lianglong Chen, Shenhua Zhou, Statistics Team: Xiaoshu Cheng, Lang Li, Tianhe Yang, Yongjian Wu, Yuan Wu, Hongbin Yan, Jinqin Yuan, Shijie You, Min Yang Wang, Xinran Tang American Heart Journal 201.e2 Xu et al May 2016

Appendix B. Full list of hospitals in the China AMI registry

Hospital Province/Municipality City PI

Fuwai Hospital Beijing Beijing Yuan Wu Beijing Friendship Hospital Beijing Beijing Hongwei Li Beijing Tongren Hospital Beijing Beijing Changlin Lu Beijing Daxing Hospital Beijing Daxing Shujun Cao Beijing Mentougou Hospital Beijing Mentougou Dezhao Wang Beijing Pinggu Hospital Beijing Pinggu Guanglin Wei Beijing Yanqing Hospital Beijing Yanqing Jianbing Wang Jiaotong University Ruijin Hospital Shanghai Shanghai Ruiyan Zhang Shanghai 10th Hospital Shanghai Shanghai Yawei Xu Shanghai Fengxian Hospital Shanghai Fengxian Zengyong Qiao Medical School General Hospital Tianjin Tianjin Zheng Wan Tianjin Baodi Hospital Tianjin Baodi Yanjun Cao Medical School 2nd Hospital Chongqing Chongqing Yaohui Yin Haerbin Medical School 1st Affiliated Hospital Harbin Weiming Li Qiqihaer 1st Hospital Heilongjiang Qiqihar Shuqing Wang Tailai Hospital Heilongjiang Tailai Gang Ma Shuihua 1st Hospital Heilongjiang Shuihua Yongchen Cai 1st Hospital Jilin Changchun Yang Zheng Tonghua Central Hospital Jilin Tonghua Xuxia Zhang Huinan County Hospital Jilin Huinan Hongyan Guo Shenyang Northern Hospital Shenyang Xiaozeng Wang Fushun Central Hospital Liaoning Fushun Ling Sun Xiuyan County Hospital Liaoning Xiuyan Jianhua Wu Neimonggu Medical College 1st Affiliated Hospital Hohhot Fengying Chen Chifeng Hospital Inner Mongolia Chifeng Ronghai Man Aohan Hospital Inner Mongolia Aohan Yanjie Li Medcial School 2nd Affiliated Hospital Hebei Shijiazhuang Xianghua Fu Qinhuangdao 1st Hospital Hebei Qinhuangdao Qingshen Wang Qinhuangdao 2nd Hospital Hebei Changli Liying Zhang North-China Oil-administration General Hospital Hebei Renqiu Xiaoli Gao Changzhou Hospital Hebei Changzhou Yali Hu Hengshui Hardison Hospital Hebei Hengshui Qun Zheng Cardiovascular Hospital Shanxi Taiyuan Bao Li Changzhi Hospital Shanxi Changzhi Yuping zhang Tunliu Hospital Shanxi Tunliu Yaohong Dong Provincial Hospital Henan Chuanyu Gao Linzhou Hospital Henan Linzhou Zhoushun Qin Hospital Henan Changyuan Guorui Hou Central Hospital Henan Xinxiang Lingling Liu Yanjin Hospital Henan Yanjin Shifeng Ren hospital Henan Ye County Dezhou wang Pindingshan 2nd Hospital Henan Pindingshan Xianting Luan Prefecture Hospital Henan Anyang Hui Liu People’s Hospital Henan Puyang Liping Ma Xihua Hospital Henan Xihua Chuntong Wang Xi’an Jiaotong University 1st Hospital Shan’xi Xi’an Zuyi Yuan Central Hospital Shan’xi Weinan Junnong Li Jiuquan Hospital Jiuquan Yaofeng Yuan Jinta Hospital Gansu Jinta Huide Liu Medical College General Hospital Ningxia Yinchuan Shaobin jia Wuzhong Hospital Ningxia Wuzhong Xianghong Luo University Affiliated Hospital Qinghai Yin Liu Qinhai Cardiovascular Hospital Qinghai Xining Pinfa Liu Xining 1st Hospital Qinghai Xining Xianning Zhao Prefectural Hospital of Qinghai Qinghai Gonghe Bao Ma Medical College 1st Affiliated Hospital Xinjiang Urumchi Yitong Ma American Heart Journal Xu et al 201.e3 Volume 175

(continued)

Hospital Province/Municipality City PI

Changji Hospital Xinjiang Changji Mao Wang Fukang Hospital Xinjiang Fukang Shiming Gao Urumchi Friendship Hospital Xinjiang Urumchi Hang Lu Provincial Hospital Shandong Jinan Lianqun Cui Taian Central Hospital Shandong Taian Huanyi Zhang Xintai Hospital Shandong Xintai Hongyan Zhang Gulou Hospital Nanjin Biao Xu Jiangsu North Hospital Jiangsu Yangzhou Shenghu He Xuzhou 1st Central Hospital Jiangsu Xuzhou Qiang Fu Jiangyan Hospital Jiangsu Jiangyan Shihai Shen Provincial Hospital Anhui Hefei Likun Ma Fuyang Hospital Anhui Fuyang Bin Ning Taihe Hospital Anhui Taihe Jili Fan University 2nd Affiliated Hospital Zhejiang Hangzhou Yong Sun Taizhou Enze medical Center Zhejiang Taizhou Lijiang tang Taizhou Hospital Zhejiang Linhai Danlei Xu Medical College Union Hospital Fujian Fuzhou Lianglong Chen Xiamen Heart Center Fujian Xiamen Yan Wang Fuqing Hospital Fujian Fuqing Ping chen Longyan 1st Hospital Fujian Longyan Kaihong Chen Tongji Hospital Wuhan Daowen wang Jinzhou 1st Hospital Hubei Jinzhou Shuixian peng Tianmen 1st Hospital Hubei Tianmen Shuping Wan Gong’an Hospital Hubei Gongan Laxi Zhang Xiangya 2nd Hospital Shenhua Zhou Xiangtan Central Hospital Hunan Xiangtan Jianping Zeng Xiangxiang Hospital Hunan Xiangxiang Chonglun Zhou Ya’an Hospital Ya’an Haibo zhang Zigong 1st Hospital Sichuan Zigong Dechao Zhong Danleng County Hospital Sichuan Danleng Yuquan Xiao Medical College 1st Affiliated Hospital Guangxi Lang Li Beihai Hospital Guangxi Beihai Hai Zhu Hepu Hospital Guangxi Hepu Meisheng Lai Nanchang Universuty 2nd Affiliated Hospital Nanchang Xiaoshu Cheng Pingxiang Hospital Jiangxi Pingxiang Junming Ye Shangli Hospital Jiangxi Shangli Qishou Liu Cardiovascular Hospital Guizhou Guiyang Tianhe Yang Zhunyi 1st Hospital Guizhou Zhunyi Zhengqiang Yuan Honghuagang Hospital Guizhou Honghuagang Chengyuan Zhao Pan County Hospital Guizhou Pan Xianwen Jiang Provincial Hospital Guangdong Guangzhou Jiyan Chen Guangzhou Traditional Chinese Medical College 1st Affiliated Hospital Guangdong Guangzhou Wei Wu Jiangmen Hospital Guangdong Jiangmen Gaoxing Zhang Heshan Hospital Guangdong Heshan Haiyuan Mai Kunming Medical College 1st Affiliated Hospital Kunming Tao Guo Yunnan St. John’s Hospital Yunnan Kunming Yi Li Chuxiong People’s Hosptal Yunnan Chuxiong Xiaoming Liu Yao’an Hospital Yunnan Yao’an Jinlong Xu Tibet People’s Hospital Tibet Lahsa Gesang Luobu Hainan Provincial Hospital Hainan Haikou Bin Li Sanya Hospital Hainan Sanya Tiansong Wang Wenchang Hospital Hainan Wenchang Dong Wang