Study population

Our institution served as the centre of the World Health Organisation Monitoring Trends and Determinants in Cardiovascular Disease (WHO MONICA) project in China from 1984 to 1993.¹⁶ After accomplishment of the WHO MONICA project, we established the BCDSS using personal identification information by linking records routinely collected in the Beijing Hospital Discharge Information System (HDIS) and Beijing Vital Registration Monitoring System (VRMS).¹⁷

MI morbidity refers to the total MI events, including nonfatal or fatal events.⁶ Nonfatal events were identified by the International Classification of Diseases, Tenth Revision (ICD-10) codes I21 to I22 (acute MI and subsequent MI) in the principal discharge diagnoses, which refer to MI events in patients who survive for at least 28 days after hospital admission. Fatal events were identified by ICD-10 codes I20 to I25 for the underlying causes of death and were further classified as in-hospital or out-of-hospital deaths based on the site of death. The same deaths identified in both HDIS and VRMS were linked to 1 case. We had previously validated the diagnosis of MI in the BCDSS.¹⁸ The positive and negative predictive values of an acute MI diagnosis in the HDIS compared with the WHO MONICA criteria were 94.4% and 96.1%, respectively.

The system identified 435,775 MI events among permanent residents aged ≥ 35 years from January 1, 2007, to December, 31, 2019. Multiple steps were taken to avoid repeatedly counting MI events.¹⁹ The patients were deemed to have experienced a sequential course of care if rehospitalised or transferred on the day after discharge (n = 7649). Patients were deemed unlikely to have had an acute MI and were excluded if they were discharged alive with a duration of hospitalisation of less than 1 day and were not rehospitalised or died on that day (n = 3714). Patients with a record of any subsequent rehospitalisation or death within 28 days (n = 7516) were considered to have had 1 MI event according to the WHO MONICA protocols, and the onset time of the first event and the outcome within 28 days were retained. Information on sex and age was obtained for each MI event, and 2 records with missing values for sex were excluded. In total, 416,894 MI events were included (Supplemental Fig. S1).

The characteristics of the MI events were further identified. Types of MI were identified according to the principal discharge diagnoses using ICD-10 codes I21.0 to I21.3, I22.0, I22.1, and I22.8 for ST-segment elevation myocardial infarction (STEMI), and I21.4 for non-ST-segment elevation myocardial infarction (NSTEMI). Comorbidities were identified according to secondary discharge diagnoses (up to 7 diagnoses) using ICD-10 codes I46 to I49 for arrhythmia, I50 for heart failure, J44 to J45 for chronic obstructive pulmonary disease, I10 to I15 for hypertension, and E10 to E14 for diabetes. Payment methods (urban health insurance or public reimbursement/other) were used to represent socioeconomic status. Reperfusion therapy was identified according to International Classification of Diseases, 9th Revision, Clinical Modification Operations and Procedures codes 00.66, 36.01, 36.02, 36.05, 36.06, and 36.07 for percutaneous coronary intervention and 36.1 for coronary artery bypass grafting.