Providing Insights that Contribute to Better Health Policy
Insurance Coverage & Costs Access to Care Quality & Care Delivery Health Care Markets Issue Briefs Data Bulletins Research Briefs Policy Analyses Community Reports Journal Articles Other Publications Surveys Site Visits Design and Methods Data Files
Nonurgent Use of Hospital Emergency Departments
Statement of Peter J. Cunningham, Ph.D., Senior Fellow and Director of Quantitative Research
Before the U.S. Senate Health, Education, Labor and Pensions Committee Subcommittee on Primary Health and Aging
Hearing on "Diverting Non-urgent Emergency Room Use: Can It Provide Better Care and Lower Costs?"
Chairman Sanders, Sen. Paul and members of the Subcommittee, thank you for the invitation to testify about use of hospital emergency departments for nonurgent health problems. My name is Peter Cunningham, and I am a researcher and director of Quantitative Research at the Center for Studying Health System Change (HSC).
HSC is an independent, nonpartisan health policy research organization affiliated with Mathematica Policy Research. HSC also is the research arm of the nonpartisan, nonprofit National Institute for Health Care Reform, a 501(c) (3) organization established by the International Union, UAW; Chrysler Group LLC; Ford Motor Company; and General Motors to conduct health policy research and analysis to improve the organization, financing and delivery of health care in the United States (www.NIHCR.org).
I and other HSC researchers have conducted a number of studies documenting the increase in the use of hospital emergency departments, including for nonurgent health problems, and the problems of crowding at some emergency departments (EDs). We have examined how these trends affect and are affected by larger developments in the health care system, the reasons why people use emergency departments for minor ailments, and the potential for hospitals to shift some of their emergency department visits to primary care providers in the community.
Our goal at HSC is to inform policy makers with objective and timely research on developments in the health care system and the impact on people. We do not make specific policy recommendations. Our various research and communication activities may be found on our Web site at www.hschange.org.
There has been much concern over the past decade about what many believe is a national crisis of crowded and overloaded hospital emergency departments and the consequences for patient care and the ability of EDs to respond to both individual and mass-casualty emergencies. Contributing to the problem of ED crowding is a substantial increase in emergency department utilization among the U.S. population, which is often attributed to growing use of emergency departments for nonurgent health problems. As a result, many policy makers and health care providers believe that it is essential to shift emergency department use for nonurgent health problems to primary care providers in the community to relieve crowded emergency departments, lower the costs of care and improve the quality of care.
My testimony today will make the following key points:
The Evolving Role of Hospital Emergency Departments
Hospital emergency departments are a critical and indispensable component of the U.S. health care system. While their traditional mission is to provide trauma and emergency services for people in imminent danger of losing their life or suffering permanent damage to their health, the role of emergency departments has evolved over the past several decades. EDs are on the front lines of communities’ preparedness efforts and responses to natural disasters, other mass-casualty events, and public health emergencies arising from outbreaks of influenza and other communicable diseases.
And emergency departments have become the true provider of “last resort” for uninsured people and other patients who are unable to afford other medical providers in the community, largely as a result of the 1986 federal Emergency Medical Treatment and Labor Act (EMTALA) that requires hospitals to provide emergency screening and stabilization services regardless of patients’ ability to pay. Along with the fact that emergency departments are often the only medical facilities in a community that are open 24 hours a day, seven days a week, true emergencies comprise only a relatively small share of visits to emergency departments. Today, hospital emergency departments are a major source of primary health care in the community, treating a broad range of health problems that include many visits for minor ailments and other “nonurgent” conditions.
Use of Emergency Departments Still Relatively Rare
Americans made a total of 124 million visits to hospital emergency departments in 2008, the latest year for which data are available from the National Hospital Ambulatory Medical Care Survey (NHAMCS)—the most authoritative and cited source of information on emergency department utilization (see Table 1). Compared with other forms of ambulatory care use, however, use of hospital emergency departments is relatively rare, accounting for only 10 percent of all ambulatory care visits to medical providers. By contrast, Americans made 956 million visits to physician offices in 2008—representing 80 percent of all ambulatory care visits—and 110 million visits to hospital outpatient departments.
Emergency department use is also much less frequent than physician office visits on a per capita basis. There were 41 emergency department visits for every 100 Americans in 2008, compared to 320 physician office visits for every 100 Americans. About 84 percent of Americans visited a physician’s office in 2007, compared to 23 percent who visited a hospital emergency department.
Table 1 Use of Ambulatory Medical Care Services by the U.S. population, 1995-2008
Source: CDC/NCHS, National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, as reported in Health, United States, 2010
But Increases in Utilization Contribute to Crowding
Concern about the use of hospital emergency departments increased substantially over the past decade because of widespread reports of growing demand by patients and crowding at many emergency departments. Indeed, in a 2007 report, the Institute of Medicine described a growing national crisis of crowded emergency departments leading to delays in care for patients, ambulance diversions to other hospitals, and inadequate capacity to handle a large influx of patients from a public health crisis or mass-casualty event.
Increased crowding at emergency departments has a number of causes, and a 2003 U.S. Government Accountability Office report concluded that insufficient inpatient capacity—the inability of hospitals to move patients from the emergency department into inpatient beds—was a major factor. As a result of problems with “throughput,” emergency department patients are (1) waiting longer to be seen in the emergency department; (2) waiting longer to be admitted as an inpatient if necessary, and; (3) increasingly leaving the emergency department without being seen. Also, there has been an increase in hospitals diverting ambulances to other hospitals because of emergency department crowding.
Increased demand for emergency departments has exacerbated these problems. Between 1995 and 2008, visits to hospital emergency departments increased 28 percent, with much of the increase because of increased per person use—from 37 visits per 100 persons in 1995 to 41 visits in 2008 (see Table 1). However, physician office visits increased by an even greater amount between 1995 and 2008—37 percent—with per person use increasing from 266 visits per 100 persons in 1995 to 320 visits in 2008. Thus, increases in emergency department use over the past decade and a half reflect a more general increase in the demand for ambulatory care and must be understood in the broader context of changes in the health care system. As physician practices have become busier and patients have greater difficulty getting timely appointments with their physicians, some of the excess demand for ambulatory care is no doubt spilling over into emergency departments.
Privately Insured Patients Account for Most of the Increase in ED Volume
Also, while there is a common perception that emergency department crowding is driven primarily by increases in utilization by the uninsured, most of the growth in emergency department volume during this period was driven by insured people. For example, the share of emergency department visits classified as “self-pay” or “no charge”—mostly uninsured patients—actually decreased from 17 percent of visits in 1995 to 15 percent in 2008, despite the fact that the number of uninsured increased by 23 percent during this period. , , In contrast, the share of emergency department visits made by
The perception that the uninsured are responsible for the problems of emergency department crowding may be because uninsured people depend more on emergency departments for access to care. For example, more than one-fourth of all ambulatory care visits by the uninsured are in emergency departments, compared to only 7 percent for the privately insured and 17 percent for Medicaid enrollees. Even more striking is that uninsured people’s dependence on EDs for care has grown dramatically since 1995 when 16 percent of ambulatory care visits by the uninsured were in hospital emergency departments.
The increasing dependence on hospital emergency departments by the uninsured reflects an erosion in access to office-based physicians, as evidenced by declines in the percent of physicians providing any charity care during this period. Some physicians believe they are no longer able to afford to provide charity care because of financial pressures from payers, while others have much less time for charitable and volunteer activities because of the increased demand for care by privately insured patients.
Many observers have attributed increases in “nonurgent” use of emergency departments as a key driver of crowding at some EDs. However, defining a “nonurgent” ED visit is not straightforward and has been the subject of much debate and controversy. Estimates of the percent of emergency department visits that are for nonurgent health problems vary widely, from about half of all visits to less than 10 percent. The wide differences in estimates largely reflect differences in the assumptions made about the feasibility of shifting certain types of visits to a primary care physician’s office or clinic without harm to the patient.
Thus, from the patient’s perspective, the visit is certainly urgent or emergent, but it is unlikely to be classified as such based only on the physician’s diagnosis.
For this reason, the “urgency” of a hospital emergency department visit is best determined by the level of immediacy (in minutes) assigned upon arrival at the emergency department by triage staff. The National Hospital Ambulatory Medical Care Survey uses this information to determine the urgency of a visit, which includes five categories: (1) Immediate (patient needs to be seen immediately; (2) emergent (needs to be seen within 15 minutes upon arrival); (3) urgent (between 15-60 minutes); (4) semiurgent (1-2 hours) and nonurgent (2-24 hours). It is important to note that the immediacy with which a patient should be seen is unknown for about 16 percent of emergency department visits in the NHAMCS data for 2008, in part because some emergency departments either do not triage patients in this way or do not keep records of their triage decisions.
Based on this classification system, 4 percent of emergency department visits in 2008 (a total of 4.6 million visits) were visits in which the patient needed to be seen immediately; 12 percent were considered emergent; 39 percent were considered urgent; and 21 percent were semi-urgent (see Table 2). Only 8 percent of visits—a total of 9.9 million—were classified as nonurgent. Trends in the relative number of nonurgent visits have actually decreased slightly since 2000, when 10.7 percent of visits were classified as nonurgent. In sum, most visits to hospital emergency departments are neither true emergencies requiring that patients be seen almost immediately nor are they clearly nonurgent problems that could be addressed in other primary care settings.
The majority of visits that are considered urgent or semi-urgent reside in a gray area as to whether they could potentially be shifted to other primary care settings, such as freestanding urgent care centers or through same-day appointments with private practice physicians. While many conditions associated with these visits could likely be treated in other outpatient settings, it is not necessarily inappropriate for the patient to use the emergency department depending on the circumstances, such as the availability of other health care providers in the area, the time of day and day of the week when services are needed, and the affordability of these other providers based on a patient’s insurance status and ability to pay.
Two-thirds of all emergency department visits occur outside normal business hours—8 a.m. to 5 p.m., Monday through Friday, compared to only 5 percent of visits to office-based physicians and 11 percent of visits to hospital outpatient departments. Thus, increasing the number of primary care providers in the community who are available after normal business hours (i.e. in the evenings and on weekends) is essential for any effort to shift visits from the ED to other primary care providers in the community.
Table 2 Triage Status of Emergency Department Visits, by Expected Source of Payment, 2008
Triage status is based on the following classification:
Source: CDC/NCHS. National Hospital Ambulatory Medical Care Survey: 2008 Emergency Department Summary Tables (Table 7)
Most Nonurgent Visits Made by Persons with Insurance Coverage
As with emergency department visits overall, people with insurance coverage account for most nonurgent ED visits, with privately insured persons alone accounting for about one-third of nonurgent visits (computed from Table 2). Uninsured persons account for slightly less than one-fourth of all nonurgent emergency department visits, while Medicaid enrollees account for 29 percent. Nevertheless, the uninsured are more likely to use emergency departments for nonurgent health problems compared to the privately insured: visits for nonurgent health problems account for 12 percent of ED visits by the uninsured compared to 6 percent for the privately insured. Similarly, the uninsured are less likely to use emergency departments for true emergencies compared to privately insured persons: emergencies accounted for 12 percent of visits for uninsured persons compared to 17 percent for privately insured persons (see Table 2).
Immigrants Infrequent Users of Emergency Departments
Another common perception is that immigrants—particularly undocumented immigrants—are responsible for much of the crowding in emergency departments. Although the National Hospital Ambulatory Medical Care Survey has limited information on race/ethnicity and immigration status, other studies call into question the extent of the problem that emergency departments have treating undocumented immigrants. Recent immigrants—in the United States for five years or less—are less likely to use emergency departments (9%), compared both to immigrants who have been in the United States for 20 years or more (19%), as well as native-born Americans (22%). In addition, an analysis of variation in emergency department use across communities showed that communities with high levels of emergency department use had fewer Hispanic noncitizens compared to communities with low levels of emergency department use.
Low ED use by recent immigrants reflects the fact that they are much less likely to use health care of any type, including physician office visits, primarily as a result of high uninsured rates and a lack of access to care. As with the uninsured, recent immigrants tend to rely on emergency departments to a much greater extent when they do use health care compared to native-born Americans, which may contribute to the perception that they are “flooding” local hospital emergency departments. Crowding of emergency departments by immigrants may occur in some isolated circumstances, such as in communities along the border with Mexico or areas that have seen a recent surge in immigration, but it is not a major contributor to hospital emergency department crowding nationally.
Lack of Primary Care Access Not the Reason for Emergency Department Use
It is not the case that people who use emergency departments for nonurgent health problems have no source of primary care they could use instead. One study found that among all people visiting the emergency department for nonurgent health problems, two-thirds reported they had a regular source of medical care at a physician’s office. Only 3 percent reported that the ED was their usual source of care, while 15 percent reported they did not have any usual source of care. In addition, people who use the ED for nonurgent health problems tend to have greater use of physicians in other ambulatory care settings over a one-year period. This strongly suggests that use of emergency departments for nonurgent problems does not reflect lack of access to other primary care providers for most patients, although it is a much more important reason for uninsured patients.
Capacity Constraints Contribute to Higher Emergency Department Use
As noted previously, it is possible that greater capacity constraints in the ambulatory care system are shifting some of the excess demand for ambulatory care from physician offices to emergency departments. Many experts and policy makers have been concerned about physician shortages—particularly of primary care physicians—resulting in some patients having greater difficulty finding physicians that are close to their home or work, scheduling same-day appointments with their primary care physician, and physicians being able to spend adequate time with patients.
In examining differences between communities with low levels of ED use and communities with high levels of ED use, I have observed that communities with high levels of ED use tend to have greater capacity constraints among office-based physicians, as reflected in longer average appointment waiting times for patients and a greater number of visits per physician in the community. This suggests that as demand for medical care increases over time and the capacity of office-based physicians is squeezed, some of the excess demand for ambulatory care will spill over to hospital emergency departments.
At the same time, many patients prefer to use hospital emergency departments even if they believe that their health problem could have been handled by a primary care physician outside of the emergency department. The greater convenience of hospital emergency departments relative to primary care providers is among the most important reasons for using EDs, especially the fact that they are open 24 hours a day and seven days a week, and that they can “walk-in” to the emergency department at their own convenience rather than scheduling an appointment. The greater convenience of emergency departments is especially important for people who are unable or unwilling to take time off from work to see a physician.
What is less clear is whether patient preferences for the emergency department will continue given the increased crowding at many facilities and the longer wait times. The total amount of time that patients spend in the emergency department—including time spent waiting as well as for examination and treatment—has increased from 45 percent of visits lasting 2 or more hours in 2001 to 60 percent of visits in 2008. Other research has shown that patients’ satisfaction with their visit to an emergency department decreases rapidly the longer they wait to be seen. For example, two-thirds of patients who waited 15 minutes or less to be seen by a medical provider in the emergency department reported that the thoroughness of their exam was very good or excellent. However, positive ratings of their visit dropped to 46 percent for patients who waited between 30 and 60 minutes, and 28 percent for patients who waited more than an hour.
Hospitals Expanding Emergency Department Capacity
At the same time, many emergency departments have been expanding capacity to meet increased demand. More than one-fourth of emergency departments in 2008 had expanded their capacity in the previous two years, and 28 percent had plans to expand their capacity in the next 2 years. Emergency departments serving a large volume of patients (50,000 or more per year) were much more likely to be expanding capacity compared to emergency departments serving smaller volumes of patients.
Expanding the capacity of hospital emergency departments appears to conflict with a widely held view that emergency departments are money losers for hospitals – i.e. they generate insufficient revenue from billings to cover the costs. In this view, hospitals should be reluctant to expand emergency department capacity and be eager to look for ways to decrease their volumes by shifting patients to other sources of primary care when appropriate.
However, when the overall financial status of many hospitals is considered, emergency departments generate more revenue for the hospital than they lose, mainly by serving as a conduit for inpatient admissions. Researchers at the University of Southern California estimated that by closing the emergency department, a hospital would lose one-third or more of its inpatient admissions, which would cost the hospital much more than the savings generated by closing the emergency department. Seen in that context, it is not surprising that many hospitals are expanding their emergency departments, not only to relieve crowding because of increased demand, but also as a way to generate more revenue from inpatient admissions.
Efforts to expand emergency department capacity and volume also suggest that many hospitals perceive few incentives or benefits to shift nonurgent care from their emergency department to primary care settings. Even if an emergency department visit does not result in an inpatient admission, nonurgent emergency department patients may require inpatient care or other hospital services in the future, in which case the assumption is that the patient would continue to use the same hospital to receive these more “profitable” services. Hospitals will especially encourage privately insured, Medicare and sometimes even Medicaid patients to use their emergency departments, as these patients generate revenue for the hospital. Many hospitals are likely to be much more selective about the patients they are willing to shift to primary care settings, focusing especially on their uninsured patients to decrease their uncompensated care costs.
Cost Savings from Reducing Nonurgent ED Use Likely to be Modest
About $47.3 billion was spent on emergency department visits in 2008, accounting for 4 percent of all health care expenditures received by the U.S. population during that year, according to the Medical
It is possible that shifting nonurgent emergency department visits to community health centers (CHCs) could result in greater savings than comparable visits to private office-based physicians. Research has shown that the availability of CHCs in an area is associated with lower rates of hospital emergency department use, particularly among the uninsured. There is some evidence that CHCs provide care more efficiently and at lower cost compared to private physician practices, perhaps because the large volumes of patients CHCs see permit greater economies of scale in the cost of patient care. Also, the
Cost Savings for the Uninsured and Medicaid Likely to Be Greater
While the cost savings to the health care system of shifting care out of the emergency department to Community Health Centers may be minimal, the cost savings to uninsured patients could be considerable. The average cost of an emergency department visit for uninsured persons was $1,203 in 2008, of which half is paid out-of-pocket. Nonurgent visits are likely to be less costly for the uninsured – as they are with the general population – but they may still be responsible for a bill of several hundred dollars or more. By contrast, community health centers typically charge patients on a sliding scale – the fee amount increases along with their incomes – and typically ranges from $20 to $60 per visit.
It should also be noted that most hospitals have policies that allow their charges to be waived or reduced based on the patient’s ability to pay, including for visits to hospital emergency departments. For poor or low-income patients, hospitals often use a sliding-scale method similar to that used by community health centers to determine the patient’s responsibility, and charges are often waived for the poorest uninsured patients. Thus, depending on the hospital’s charity care policies and the patient’s income, an uninsured person could pay little or none of the charge, or they could be responsible for most or all of the charge of the emergency department visit. However, hospitals sometimes limit the effectiveness of their charity care policies by failing to advertise them or making
Shifting nonurgent emergency department visits to community health centers and other sources of primary care could generate greater cost savings for the Medicaid program. Medicaid enrollees have the highest rates of emergency department use compared to persons with private insurance, Medicare or who are uninsured, and Medicaid enrollees account for more than one-fourth of nonurgent visits to the emergency department. Because Medicaid patients already comprise a large proportion of patients at community health centers—and they tend to live in areas where CHCs are located—programs designed to shift nonurgent care from EDs to CHCs may have greater potential to generate cost savings in the Medicaid program than for private payers, Medicare or even hospital uncompensated care costs from caring for the uninsured.
Finally, improvements in continuity of care, patient satisfaction and care coordination between primary care providers and specialists that can be facilitated by community health centers and other primary care providers can also increase cost savings to the Medicaid program, primarily by reducing redundant and unnecessary use of health services.
Gains in Quality of Care May Be Greater Than Cost Savings
Shifting ED use for nonurgent problems to primary care providers in the community is likely to have even more important implications for the quality of care. ED use for nonurgent health problems is associated with greater fragmentation and discontinuity of care with the patients’ primary care physicians and other medical providers they use. Studies have found that communication and coordination of care between EDs and primary care physicians tends to be haphazard and generally poor, which is exacerbated by a lack of shared information systems that could facilitate communication. The lack of coordination and continuity between EDs and other providers in the community often leads to duplicative testing and other redundant utilization, complicates appropriate follow-up care, and increases the risk of medical errors.
Shifting ED use to primary care physicians may also increase patient satisfaction with care. According to one survey, more than three-fourths of patients with scheduled appointments at a doctor’s office gave positive ratings about the thoroughness of the exam and the physician’s willingness to listen. By contrast, only about half of ED patients gave such positive assessments. Thus, patients may be motivated to go to the ED because of greater convenience and the availability of after-hours care but not necessarily because they believe the ED provides better quality of care.
Shifting ED visits for nonurgent health problems to primary care providers in the community is a necessary step for broader efforts in the health care system to create “patient-centered medical homes.” This would not only improve the quality of care by ensuring that patients have a primary care physician to see for their nonurgent health problems and coordinating care with specialists and other providers, but it is also likely to generate additional cost savings by reducing unnecessary or redundant utilization.
National Center for Health Statistics. Health, United States: With Special Feature on Death and Dying. Hyattsville, Md. (2011).
Stussman, Barbara J., “National Hospital Ambulatory Medical Care Survey: 1995 Emergency Department Summary,” National Center for Health Statistics Advance Data From Vital and Health Statistics; No. 285, Hyattsville, Md (1997).
The estimates in this paragraph are computed from published reports by the National Center for Health Statistics based on the 1995 and 2008 National Hospital Ambulatory Medical Care Survey and the 1995 and 2008 National Ambulatory Medical Survey.
Cunningham, Peter, and Jessica May, A Growing Hole in the Safety Net: Physician Charity Care Declines Again, Tracking Report No. 13, Center for Studying Health System Change, Washington, D.C. (March 2006).
McCaig, Linda F., and Nghi Ly, “National Hospital Ambulatory Medical Care Survey: 2000 Emergency Department Summary,” National Center for Health Statistics Advance Data From Vita and Health Statistics, No. 326, Hyattsville, Md. (April 2002).
Cunningham, Peter, and Samantha Artiga, How Does Health Coverage and Access to Care for Immigrants Vary by Length of Time in the U.S.,” Kaiser Commission on Medicaid and the Uninsured, Washington, D.C. (June 2009).
O’Malley, Ann S., et al., Rising Pressure: Hospital Emergency Departments as Barometers of the Health Care System, Issue Brief No. 101, Center for Studying Health System Change, Washington, D.C. (November 2005).
Agency for Healthcare Quality and Research, “Emergency Room Services – Median and Mean Expense Per Person with Expense and Distribution by Source of Payment, 2008,” Summary Data Table From the Medical Expenditure Panel Survey – Household Component.
Ibid. and Agency for Healthcare Quality and Research, “Emergency Room Services – Median and Mean Expense Per Person With Expense and Distribution by Source of Payment, 2000,” Summary Data Table From the Medical Expenditure Panel Survey – Household Component.
McRae, Thomas, and Robert D. Stampfly, An Evaluation of the Cost-Effectiveness of Federally Qualified Health Centers Operating in Michigan, Institute for Healthcare Studies at Michigan State University (October 2006).
Michelle M. Doty, et al., Enhancing the Capacity of Community Health Centers to Achieve High Performance: Findings From the 2009 Commonwealth Fund National Survey of Federally Qualified Health Centers, The Commonwealth Fund, New York (May 2010).
Agency for Healthcare Quality and Research, “Emergency Room Services – Median and Mean Expense Per Person with Expense and Distribution by Source of Payment, 2008,” Summary Data Table From the Medical Expenditure Panel Survey – Household Component.
Staiti, Andrea, et al., Balancing Margin and Mission: Hospitals Alter Billing and Collection Practices for Uninsured Patients, Issue Brief No. 99, Center for Studying Health System Change, Washington, D.C. (October, 2005).