INTRODUCTION: Myocardial infarction (MI) is a major cause of premature death and disability in China, but available data on diagnostic accuracy, treatment and prognosis of MI cases remain limited. METHODS: The China Kadoorie Biobank enrolled 512 000 adults (mean age 51 years, 59% women) from 10 (five urban, five rural) areas between 2004 and 2008. Medical records were sought on 37 159 reported first incident ischaemic heart disease (IHD) cases occurring before 2018 for adjudication. Diagnostic accuracy and risks of recurrent MI, stroke, heart failure and all-cause mortality were assessed for reported and adjudicated cases of MI and all IHD. RESULTS: Among 19 135 adjudicated IHD cases, 10.2% (n=1948) had MI, with diagnostic accuracy of 98% for MI and 93% for all IHD. Use of guideline-directed (antiplatelet, lipid-lowering, antihypertensive and anticoagulant) medications in the hospital was high (93% MI; 83% all IHD), but differed by hospital tier and area. Use of coronary revascularisation was low (39% MI; 10% all IHD), even in higher tier hospitals. The 28-day case-fatality was sixfold greater for MI than all IHD (12.0% vs 2.0%) and higher in older cases and in women, residents in rural areas, but comparable by hospital tier. Disparities in 5-year mortality rates persisted for both MI and all IHD, by age, sex, area and hospital tiers. CONCLUSIONS: This large community-based prospective study demonstrated high levels of diagnostic accuracy for MI and all IHD in Chinese adults, but also identified disparities in treatment and outcomes that should be prioritised in future prevention strategies.