Global Crowding: Opportunities for Regionalization in Emergency Care

In this issue of Academic Emergency Medicine, Cha et al.1 remind us that emergency department (ED) crowding is not limited to the United States. Although the structure of health care systems varies considerably across the world, issues with throughput and capacity appear to be universal in emergency care. There is growing evidence that ED crowding and boarding of inpatients are associated with adverse patient outcomes and increased health care costs.2 Several countries, including the United Kingdom, Australia, and parts of Canada, are addressing this issue with limits on the time patients can remain in the ED. In the United States, the Centers for Medicare and Medicaid Services (CMS) has proposed using ED throughput time and boarding times as new quality measures for EDs.3 The authors of this paper from Korea used a new approach to address delays at a large tertiary referral center’s ED. Their intervention was called the Independent Capacity Protocol (ICP). Under the protocol, emergency physicians took greater responsibility for patient dispositions, particularly for patients who were assigned to the “emergency ward,” a 30-bed unit managed by emergency medicine, where patients waited for admission to the hospital. Prior to the initiation of the ICP, patients stayed in the emergency ward for an average of 4.5 days. The ICP goal was to limit the time patients waited in the emergency ward to 48 hours. With the assistance of a 24/7 transfer coordinating nurse, the percentage of patients transferred from the emergency ward to other community hospitals nearly tripled (from 2.9% to 8.2%), and the average emergency ward length of stay (LOS) decreased to 3.1 days. At first glance, many U.S. readers may wonder what this article has to do with their practice. Although ED-based observation units are common in the United States, the average LOS for most of these units is well below 24 hours. Even though boarding is a tremendous problem for many U.S. hospitals, even the most challenged systems do not board medical admissions in the ED for an average time that exceeds 3 days. Furthermore, the average total hospital LOS for most inpatients in the United States is not much longer than the time the patients in this study were spending in the emergency ward (i.e. before their “true admission” even began). Aside from some interesting insights into the emergency care system of another country, what lessons can we learn from this article here in the United States? We believe the answer to that question is the concept of regionalization in emergency care. Rather than focus solely on the resources available at the tertiary center, the ICP approach reached out to community hospital partners throughout the region. Even though the throughput times were beyond what we would consider acceptable in the United States, the basic accomplishments of the ICP are noteworthy. The protocol was safe, it distributed the demand for inpatient care throughout a region, and it decreased both ED LOS and boarding times at the tertiary ED. Rather than dismiss the approach because it has features that do not apply in our system, we believe the findings warrant further exploration as one additional option for addressing ED crowding in the United States. In its series on the Future of Emergency Care,3 the Institute of Medicine (IOM) envisioned a coordinated, regionalized, and accountable emergency care system. The prominent call for regionalization in the IOM report reflects the fragmentation of U.S. emergency care today. Providers across the emergency care continuum (from rural prehospital care through tertiary centers) are often unfamiliar with the capabilities of all the facilities within their region. Furthermore, there is little regional coordination of emergency care, with notable exceptions in a few geographic regions and with trauma care in most states.4 Enhanced regionalization of emergency care would not only help direct patients to centers of excellence for highly specialized care, it also could help address two of the most pressing problems in U.S. emergency care today: shortages of on-call specialists and ED boarding of inpatients, particularly at large referral centers. There is a long list of barriers to improving regionalization of emergency care. Financial incentives top the list, and this may explain why the transfer of admissions from tertiary centers to community hospitals is uncommon in the United States outside of integrated delivery systems. Because large urban referral centers often care for a disproportionate share of patients who are uninsured or covered by government programs, they are caught in a financial Catch-22 when it comes to transferring admissions to community hospitals that are not jointly owned by their system. If a potential transfer patient is uninsured or on Medicaid, the referral center faces potential public criticism and/or an EMTALA investigation for dumping patients based on payer status. The recent firestorm regarding the University of Chicago’s attempts to transfer patients out of its ED to community hospitals was in part driven by this criticism.5 On the other hand, if a potential transfer patient at a referral center has private health insurance, it is financially difficult for many tertiary hospitals to give up the margin for that admission, particularly because that margin is the only way they can cover the uncompensated care expenses of other patients. Ideally, a regionalized system would utilize centers of excellence in specific diseases to care for the high-risk, complex patient while more routine care is distributed to secondary hospitals. Regionalization could improve the quality of care provided to patients and decrease costs by reducing redundant technology. However, to achieve these outcomes, the financial and regulatory disincentives for regionalized care must be addressed. The current health care reform efforts hopefully will decrease the number of uninsured patients in the United States, but none of the proposals would create a uniform source of payment for hospitalized care. Therefore, the financial disincentives for regionalized distribution of admissions are likely to remain, even if reform efforts succeed in expanding health insurance coverage. Potential solutions include an expansion of integrated delivery systems, cost-sharing agreements between tertiary centers and community hospitals, and subsidies for uncompensated care from transfers that alleviate capacity bottlenecks at urban public hospitals. In addition to the financial barriers, EMTALA concerns will need to be addressed so that regional transfers can proceed without the threat of an EMTALA investigation. On June 2, 2010, Academic Emergency Medicine will host a consensus conference on regionalization of emergency care in the United States (http://www.saem.org/saemdnn/Meetings/AnnualMeeting2010/AEMConsensusConference2010June2/tabid/1349/Default.aspx). This represents an excellent opportunity to bring together experts to help realize the IOM’s vision for a coordinated, regionalized, and accountable system. Although many of the sessions at this conference are likely to focus on the movement of patients from community providers to centers of excellence for specialized care, we hope that some discussion occurs regarding the transfer of patients from tertiary referral centers to community hospitals. Ultimately, a regionalized system should take advantage of all the resources in the system, and this cannot occur unless we are open to the movement of patients both toward and away from tertiary centers. We commend Cha and colleagues for the success of the Independent Capacity Protocol in Korea1 and look forward to reports of similar programs in the United States. Ultimately, we should consider every available strategy to alleviate ED crowding and boarding. Although it is unfortunate that crowding and boarding is a global problem, it is important that we continue to search the globe for potential solutions.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.