Percutaneous Coronary Intervention Facility Access and Clinical Outcomes Among Patients With Acute Myocardial Infarction.

Importance: The proliferation of percutaneous coronary intervention (PCI) facilities across the US has raised questions about whether expanding capacity improves outcomes in areas already served vs those without prior access. Objective: To evaluate the association between the opening of PCI-capable facilities and outcomes following acute myocardial infarction (AMI) in communities with and without baseline PCI access. Design, Setting, and Participants: This cohort study used California data from inpatient, emergency department, and ambulatory surgery encounters from January 1, 2011, to December 31, 2022, including all adults hospitalized with a primary diagnosis of AMI. Zip code communities were categorized by baseline (2011) access to PCI-capable facilities within a 30-minute drive. Exposure: Opening of a PCI-capable facility within 30 minutes of a community's zip code during the study period. Main Outcomes and Measures: Primary outcomes were (1) admission to a PCI-capable facility, (2) receipt of PCI during the index hospitalization, (3) unscheduled acute care visit within 30 days, and (4) all-cause mortality at 30 days and 1 year. Linear probability models with community and time-fixed effects estimated changes in outcomes following PCI facility openings separately for communities with and without baseline access. Results: Among 590 282 patients with AMI (157 435 younger than age 60 years [27%]; 222 773 female [38%]; 63 460 Asian [11%], 129 137 Hispanic [22%], 315 326 White [53%]), 500 778 (85%) lived in communities with baseline PCI access, and 89 504 (15%) lived in communities without; 18 805 patients (21%) without baseline access experienced a PCI opening. Openings in communities without baseline access were associated with a 30.0 percentage point (pp) increase (95% CI, 28.2 to 31.8 pp) in admission to PCI-capable hospitals and a 5.1 pp increase (95% CI, 3.3 to 7.0 pp) in PCI receipt, along with reductions in 30-day (-1.8 pp; 95% CI, -3.2 to -0.5 pp) and 1-year (-2.8 pp; 95% CI, -4.7 to -1.0 pp) mortality. In contrast, openings in communities with baseline access were associated with smaller gains in admission (1.5 pp; 95% CI, 1.1 to 1.9 pp), modest increases in PCI use (1.0 pp; 95% CI, 0.5 to 1.6 pp), decreases in unscheduled visits within 30 days (-0.7 pp; 95% CI, -1.2 to -0.2 pp), and no mortality change. Sensitivity analyses restricted to higher-volume PCI centers (50 annual cases or more) yielded consistent findings. Conclusions and Relevance: In this cohort study of 590 282 patients with AMI in California, opening new PCI facilities was associated with substantial improvements in communities without prior PCI availability, including a 53% relative increase in access to admission, 16% increase in PCI use, and 20% and 15% reductions in 30-day and 1-year mortality, respectively. In contrast, already served areas showed smaller benefits and no mortality improvement. PCI expansion should prioritize underserved regions to maximize population-level benefits and avoid inefficient duplication of services.