Geographical Distribution of Emergency Department Closures and Consequences on Heart Attack Patients

We develop a conceptual framework and empirically investigate how a permanent emergency department (ED) closure affects patients with acute myocardial infarction (AMI). We first document that large increases in driving time to closest ED are more likely to happen in low-income communities and communities that had fewer medical resources at baseline. Then using a difference-in-differences design, we estimate the effect of an ED closure on access to cardiac care technology, treatment, and health outcomes among Medicare patients with AMI who lived in 24,567 ZIP codes that experienced no change, an increase of <10 minutes, 10 to <30 minutes, and ≥30 minutes in driving time to their closest ED. Overall, access to cardiac care declined in all communities experiencing a closure, with access to a coronary care unit decreasing by 18.64 percentage points (95% CI -30.15, -7.12) for those experiencing ≥30-minute increase in driving time. Even after controlling for access to technology and treatment, patients with the longest delays experienced a 6.58 (95% CI 2.49, 10.68) and 6.52 (95% CI 1.69, 11.35) percentage point increase in 90-day and 1-year mortality, respectively, compared with those not experiencing changes in distance. Our results also suggest that the predominant mechanism behind the mortality increase appeared to be time delay as opposed to availability of specialized cardiac treatment