IMPORTANCE Prior observational studies suggest that quality of care improvement (QCI) initiatives can improve the clinical outcomes of acute coronary syndrome (ACS). To our knowledge, this has never been demonstrated in a well-powered randomized clinical trial. OBJECTIVE To determine whether a clinical pathway-based, multifaceted QCI intervention could improve clinical outcomes among patients with ACS in resource-constrained hospitals in China. DESIGN, SETTING, PARTICIPANTS This large, stepped-wedge cluster randomized clinical trial was conducted in nonpercutaneous coronary intervention hospitals across China and included all patients older than 18 years and with a final diagnosis of ACS who were recruited consecutively between October 2011 and December 2014. We excluded patients who died before or within 10 minutes of hospital arrival. We recruited 5768 and 0 eligible patients for the control and intervention groups, respectively, in step 1, 4326 and 1365 in step 2, 3278 and 3059 in step 3, 1419 and 4468 in step 4, and 0 and 5645 in step 5. INTERVENTIONS The intervention included establishing a QCI team, training clinical staff, implementing ACS clinical pathways, sequential site performance assessment and feedback, online technical support, and patient education. The usual care was the control that was compared. MAIN OUTCOMES AND MEASURES The primary outcome was the incidence of in-hospital major adverse cardiovascular events (MACE), comprising all-cause mortality, reinfarction/myocardial infarction, and nonfatal stroke. Secondary outcomes included 16 key performance indicators (KPIs) and the composite score developed from these KPIs. RESULTS Of 29 346 patients (17 639 men [61%]; mean [SD] age for control, 64.1 [11.6] years; mean [SD] age for intervention, 63.9 [11.7] years) who were recruited from 101 hospitals, 14 809 (50.5%) were in the control period and 14 537 (49.5%) were in the intervention period. There was no significant difference in the incidence of in-hospital MACE between the intervention and control periods after adjusting for cluster and time effects (3.9% vs 4.4%; odds ratio, 0.93; 95% CI, 0.75-1.15; P = .52). The intervention showed a significant improvement in the composite KPI score (mean [SD], 0.69 [0.22] vs 0.61 [0.23]; P < .01) and in 7 individual KPIs, including the early use of antiplatelet therapy and the use of appropriate secondary prevention medicines at discharge. No unexpected adverse events were reported. CONCLUSIONS AND RELEVANCE Among resource-constrained Chinese hospitals, introducing a multifaceted QCI intervention had no significant effect on in-hospital MACE, although it improved a few of the care process indicators of evidence-based ACS management.