Adjusting to an Increase in Demand: The Impact of the Affordable Care Act Medicaid Expansions on Emergency Department Wait Times

The complex relationship between health insurance status and emergency department (ED) use is of great importance to health policy makers. On the one hand, insurance coverage reduces the out-of-pocket cost of going to the ED, which could lead to more frequent visits. On the other hand, if the ED has been used as a substitute for outpatient care by the uninsured, insurance coverage might shift care to office settings and reduce ED visits. This complexity makes it challenging to anticipate the impact of insurance coverage on the timeliness of ED services provided, especially at acute care hospitals. Depending on the patterns of health care utilization that result from a health insurance provision shock and hospital ability to adjust capacity, hospitals might see increased wait times from a rise in emergency department visits, which exacerbates existing delays, or they may adopt new technologies for more timely provision of care, or wait times may decrease if there are fewer patients. Previous analysis on the impact of health insurance on ED wait time is fairly scarce, partly due to a lack of national data. In particular, research on this question requires data on wait times that represent sizable numbers of patients and are consistently recorded over time. We use ED wait time data now publicly available in Hospital Compare, and previously unused in the literature. The paper is the first to examine the impact of the Affordable Care Act’s Medicaid expansion on ED wait times in a national, hospital-level analysis. We further assess whether any impact on wait times is moderated by the pre-expansion insured rate of the area surrounding the hospital, theorizing that areas with greater uninsured would likely be more impacted by the expansion itself. We use the CMS Hospital Compare data from 2012-2016, which provide quarterly measures of ED wait times for all hospitals in the US. We employ a difference-in-differences approach, in which we compare the ED outcomes of hospitals in expansion states with non-expansion states before and after the of policy change. We further run a triple-difference model that assesses the impact of ACA-induced changes in county-level insurance rate. All models control for hospital-level characteristics from the AHA annual survey data. Our preliminary results imply that, with the parallel trend assumption held, the average number of minutes spent before admitted for inpatient care in ED in states that expanded Medicaid is on average 8 minutes (or 3 percent) higher after January 2014 as compared to the states that did not expand. The impact is also significant when different measures of wait time in emergency department are employed that reflect inpatient care vs. general care. There is no evidence, however, that the impact varies linearly with the baseline un-insurance rate observed before the time of expansion, contrary to our expectations. In continuing work, we will examine the validity of the DDD assumptions, and merge in ED operation characteristics, state-year data on the number of ED visits to understand mechanisms of insurance-related ED changes.