Association for Public Policy Analysis & Management

In the past decade the U.S. has seen a striking increase in the number of Emergency Department (ED)
visits. From 2003-2010, the number of ED visits grew by 15.9%, from 113.9 million to 129.8 million, with
the majority of the growth occurring in the last three years. Unprepared for such a drastic increase in patient
volume, EDs became overcrowded with wait times increasing 25%, from 46.5 minutes to 58.1 minutes.
Simultaneously, EDs were financially burdened with uncompensated care costs growing 57.8% between
2003-2010. In response to this “crowding crisis,” the American College of Emergency Physicians (ACEP)
created a task force to identify the cause of ED overutilization. The ACEP deemed the source of the crowding
crisis to be the poor and uninsured. Lacking access to other forms of healthcare, the poor and uninsured
are relying more heavily on EDs for medical care, which often goes unpaid. Compounding the issue,
asymmetric federal regulation prevents EDs from diverting the poor and uninsured to more appropriate,
and cost efficient, healthcare providers.

 A common policy proposal addressing this issue is the expansion of public insurance to cover the uninsured.
However, these expansions are typically not limited to the previously uninsured; crowding-out also
occurs. In reality, there will be two (potentially competing) effects of expanding the eligibility of public
insurance. There will be an expansionary effect from the change in behavior from consumers going from
uninsured to covered by public insurance, but there will also be a crowding-out effect from the change in
behavior from consumers switching from private insurance to public insurance. In this paper, we examine
how the prior insurance status of the newly eligible population affects the changes in emergency department
usage and congestion following an expansion of public health insurance. That is, we examine how the expansionary
and crowding-out effects of a public insurance expansion impact emergency department usage
and congestion.

 Using hospital-level data from the 2013 and 2014 California Office of Statewide Health Planning and
Development (OSHPD), we employ a difference-in-differences estimation with continuous treatment effects
to compare ED usage and congestion between counties with varying crowd-out and uninsured rates before
and after the implementation of the ACA Medicaid expansion in California. We find that while ED usage
decreases, total ED congestion increases following the Medicaid expansion. Specifically, we find that total
ED visits in California decrease by 18,622 visits but ambulance diversion hours in California increase by
906. These results are driven by the crowding-out effect. That is, we find that the crowding-out effect
dominates the expansionary effect.

 This paper contributes to the literature as a general connection between the crowd-out literature and
the literature examining the healthcare utilization of the previously uninsured. Additionally, this paper
contributes to the literature by examining the efficacy of the Medicaid expansion of the ACA. If a goal of
expanding public insurance is to reduce ED utilization and congestion (thereby controlling healthcare costs
and increasing quality), it is imperative to understand the interaction of expansionary and crowding-out
effects.