ACADEMIC EMERGENCY MEDICINE 2011; 18:e39–e44 © 2011 by the Society for Academic Emergency Medicine
For acutely ill patients, health care services are available in many different settings, including hospital-based emergency departments (EDs), retail clinics, federally qualified health centers, and outpatient clinics. Certain conditions are the sole domain of particular settings: stabilization of critically ill patients can typically only be provided in EDs. By contrast, many conditions that do not require hospital resources, such as advanced radiography, admission, and same-day consultation can often be managed in clinic settings. Because clinics are generally not open nights, and often not on weekends or holidays, the ED remains the only option for face-to-face medical care during these times. For patients who can be managed in either setting, there are many open research questions about which is the best setting, because these venues differ in terms of access, costs of care, and potentially, quality. Consideration of these patients must be risk-adjusted, as patients may self-select a venue for care based upon perceived acuity. We present a research agenda for acute, unscheduled care in the United States developed in conjunction with an Agency for Healthcare Research and Quality–funded conference hosted by the American College of Emergency Physicians in October 2009, titled “Improving the Quality and Efficiency of Emergency Care Across the Continuum: A Systems Approach.” Given the possible increase in ED utilization over the next several years as more people become insured, understanding differences in cost, quality, and access for conditions that may be treated in EDs or clinic settings will be vital in guiding national health policy.
When people become acutely ill, they can access or be referred to health care services in many different settings, including hospital-based emergency departments (EDs), retail clinics, federally qualified health centers, and outpatient clinics.1 Each of these settings is different with regard to access, cost of care, and quality and also differ by practical logistics such as resources available, expertise, and ability to provide longitudinal care. With national health reform enacted, determining the best setting for acute, unscheduled care must take into account these factors. The two settings that are the subject of this article are personal physicians and EDs, which account for 70% of all first-contact care in the United States.1
The patient-centered medical home (PCMH) model proposes that a patient’s personal physician be responsible for providing and coordinating all of the health care needs of an individual, including when an individual becomes acutely ill.2 According to the principles of the PCMH, the personal physician is to “… provide first contact, continuous and comprehensive care.”2 The principles also include enhanced access to care, where personal physicians are encouraged to have open scheduling and extended hours and use new forms of communication. It is this proposed model that forms one of the cornerstones of health care reform.
By contrast, EDs have a dual role. The primary purpose is stabilization of patients with acute, severe illness or injury. Few would argue that critically ill patients should not be managed in an ED setting or oppose the important role of EDs as rapid diagnostic units to differentiate whether severe illness or injury exists, for example, the evaluation of abdominal pain, which often requires advanced imaging to be immediately available. In an effort to better describe conditions that require emergent evaluation and management, a proposal for defining a set of “emergency care sensitive conditions” has emerged.3 The second role of EDs is based on their 24-hour-a-day, 7-day-a-week availability to provide off-hours health care services outside of traditional clinic hours. This fact underscores the safety net role that EDs play for the health care system—the place where anyone can seek medical attention regardless of his or her condition or ability to pay. This role is codified in the Emergency Medical Treatment and Labor Act of 1986, mandating that a medical screening exam be provided to any person presenting to an ED with an acute medical condition.4
It is in the latter role that controversy exists regarding the appropriate roles of outpatient clinics (particularly the PCMH model) and the ED, especially because a subset of patients can be appropriately managed in either setting. EDs generally charge higher fees than personal physicians, and ED providers tend to order more tests.5 While several factors account for this phenomenon, it may be due at least in part to the higher probability of serious disease in patients who self-select to present to the ED, as opposed to their personal physician, based upon their perceived acuity. However, if it is possible to reduce ED visits for conditions that could be treated in lower-resourced settings, payers (such as insurance companies) may reduce their costs of care, resulting in significant savings.
Patient-centered medical homes may offer improved access to longitudinal follow-up and preventive care. On the other hand, EDs are always open and may be more convenient, and while charges are higher, the marginal costs may be similar to those in the outpatient clinic setting.6 In addition, because EDs order more tests and have immediate access to resources such as advanced radiography, it could be argued that there is a lower likelihood of missing acute pathology.
However, the reality is that neither the ED nor primary care environments have adequate capacity to handle the entire volume of acute unscheduled care.1 For the foreseeable future, both primary care settings and EDs will be necessary to meet that need. Given this reality, we propose the following research agenda to explore the advantages and disadvantages of each of these settings, offering guidance to policy-makers in determining the optimum settings for acute, unscheduled care. The following research agenda was developed in conjunction with a conference funded by the Agency for Healthcare Research and Quality and hosted by the American College of Emergency Physicians in October 2009 titled, “Improving the Quality and Efficiency of Emergency Care Across the Continuum: A Systems Approach.” This is one of five research agendas that were produced from the conference. The other research agendas focused on health information technology,7 high-cost users of ED care,8 end-of-life care,9 and the use of the ED as a rapid diagnostic unit.10
Access to care
Much of our health care system outside the ED is currently poorly structured to meet acute unscheduled care demands. This may be reflected by the fragmentation of first-contact care,1 as well as the recent growth in the use of EDs in the United States, which has outpaced population growth.11 One of the foremost reasons is the accessibility of services—the ED is always open, while most personal physicians have limited available hours. While greater access to personal physicians has been associated with lower nonurgent ED use,12–14 it is sometimes difficult to obtain timely, convenient appointments with personal physicians, and personal physicians (or their staff) sometimes directly refer patients to the ED.15–19 High levels of ED use have also been associated with the lack of a personal physician, but many other factors also contribute to ED use rates.20–22 Patients may not have ready access to a PCMH due to inadequate or nonexistent health insurance coverage or due to government health insurance that pays lower fees to personal physicians;23 thus, when reimbursements are decreased, clinic visits decrease in Medicaid patients, while use of the ED increases.24 People trust the familiarity of the ED and feel that the more comprehensive ED resources are necessary when they are acutely ill.25 With the recent increase in use of diagnostic imaging, its greater accessibility in ED settings may be one of the reasons patients and personal physicians refer to the ED for both urgent health care needs and convenience-driven nonurgent needs for outpatient imaging services.26
From a patient’s perspective, there is competition between the use of the ED and primary care settings for certain conditions. While the distribution of complaints is different between EDs and outpatient settings, many of the same complaints are commonly cared for in both settings, although again, patients may self-select one venue of care over another based upon the perceived acuity of their illness or the perceived quality of care provided.23,27 The balance between the use of the ED and the personal physician for acute care needs varies between individuals and communities, likely related to how people choose to access health care services, which services are available locally, and how long people are required to wait before receiving them. Whether increasing patients’ use of primary care providers actually will correlate with decreased use of EDs or significantly change overall costs remains unclear.28–30
Studying patients with health insurance provided by Massachusetts health care reform, the Urban Institute and others suggest that there are major needs for health care after normal business hours.19 We consider this as an indicator for national health reform, much of which is based upon the Massachusetts model. While some of this care may be able to wait for normal business hours, there is clearly a need for expanded access to care. Given the 24/7 nature of EDs and the business hours of primary care settings, the time of care and place of care are certainly linked. In a more patient-centered paradigm, the needs of the patient would outweigh the desires of medical personnel to maintain daytime, weekday office hours.
In addition, because EDs are always open and an appointment is not needed, EDs can provide same-day service for patients with acute conditions when office appointments are unavailable or when clinic providers are unwilling to provide service for various reasons. On the other hand, 24/7 access in EDs may drive higher health care utilization because patients may improve clinically while waiting for outpatient clinic appointments and ultimately not require physician evaluation for conditions that seemed urgent to the patient initially.31,32 However, long waits for physician evaluation can also delay diagnosis and management of serious illness, resulting in clinical deterioration and poor outcomes.
Research Questions for Access to Care Between the ED and PCMH Settings
- 1. What are the differences in the utilization patterns for business hours versus nonbusiness hours care in EDs? Do these utilization patterns change based upon geography? What are predictors of patients seeking ED care versus personal physician care for ambulatory care sensitive conditions and other certain conditions that could be treated in either setting (such as an upper respiratory tract infection) during business hours? Excess utilization of EDs during business hours may be a marker of various modifiable factors, including access to primary care or even transportation. In addition, understanding the utilization of EDs during nonbusiness hours may provide information on system demands that will be required of the PCMH for that model to succeed.
- 2. How do waiting times for appointments, accessibility, and distance to primary care offices drive demand for ED services? What are the risk-adjusted differences in outcomes between patients with time-sensitive conditions who wait to be seen by a personal physician versus those who come to the ED? A potential indicator condition would be acute appendicitis, where prolonged times to evaluation and treatment are associated with higher perforation rates. Better definitions of specific emergency care sensitive conditions that encompass emergencies such as acute myocardial infarction and severe trauma, along with potentially serious conditions such as chest pain or abdominal pain, might foster further research in this area. Risk adjustment is necessary to account for patient perceptions of their own acuity and self-selection to a venue of care.
Costs of care
One of the major arguments promoting the use of the personal physician (PCMH) clinics for acute care rather than EDs is that clinic care is less expensive, of similar quality, and more “coordinated” because the same provider can follow patients longitudinally. This argument most often considers the charges paid for services, as opposed to the actual costs of care. With significant unused “standby” capacity, it would seem that the marginal cost of caring for additional patients in the ED would be minimal. This is in contrast to primary care providers without similar “standby” capacity. While few studies have considered the cost of ED care, at least one study ranks the marginal cost of ED care on par with the primary care setting.6 Another study has called this conclusion into question, finding that there are no economies of scale in larger EDs.33 However, these studies did not account for the costs of ED patients going elsewhere; costs in the PCMH might change considerably if ED visits were diverted to that setting, likely resulting in higher acuity patients and significantly greater opportunity costs to patients (e.g., time away from work given the restricted hours of most primary care offices).
Even if ED marginal costs are only slightly greater than those of personal physicians, it remains undeniable that the charges for an ED visit are often significantly greater than those associated with a primary care visit for the same condition. Much of this might be due to the fact that in EDs, in addition to the costs of staffing, there are large fixed facility costs associated with maintaining readiness, advanced technology, and space to care for patients 24/7. The magnitude of those fixed costs relative to marginal costs is magnified in the rural setting, where the costs per patient may be considerably higher given the necessary standby capacity to care for emergencies and the relatively few visits that fill any ED standby time.
Also, with easier access to advanced services and a greater probability that their patients have more severe disease that those seen elsewhere, emergency physicians may be more likely to use testing to promptly risk stratify patients otherwise assumed to be at low-risk. In addition, the costs from the patients’ perspective in terms of the value of convenience, ability for just-in-time comprehensive evaluations, and other opportunity costs such as missed work have not directly been studied.
Higher charges for similar conditions also likely stem from the economics of emergency care, which involves considerable cost-shifting. EDs also operate under the unfunded Emergency Medical Treatment and Active Labor Act (EMTALA) mandate, resulting in a subsidy for unreimbursed care by payers, governments (through tax dollars), or others who utilize the ED. Outpatient clinics are not subject to the EMTALA mandate and may choose to provide limited un- or underreimbursed care. Cost shifting in this setting is likely to be significantly less than in an ED.
With national health reform, it is likely that many more patients will become insured, although it is uncertain if their insurance will be adequate to reimburse the costs of care in either venue. However, as EDs likely have greater exposure to un- and underreimbursed patients, the argument for continued cost-shifting in EDs could be weakened.
Research Questions for Costs of Care Between the ED and PCMH Settings
Comparative research of the costs of ED and PCMH care for ambulatory care sensitive conditions will likely be complex and require close attention to self-selected versus directed patient populations (who may have different probabilities of serious disease) and the marginal costs of care in each setting, while accounting for the fixed costs of additional infrastructure and labor (if necessary) and the opportunity costs to patients, their employers, and the economy as a whole from receiving care in each venue.
- 1.What are the true costs of providing care for risk-adjusted ambulatory-care sensitive conditions in the PCMH and in the ED? Is the necessary diagnostic workup more costly in the ED due to the greater likelihood of disease the in ED population? Would such costs be shifted to the PCMH model should patients be directed to that venue? What are those costs from a patient-centered standpoint (e.g., opportunity costs such as fewer numbers of visits vs. scheduled visits)?
- 2.How will cost-shifting change with near-universal insurance of health reform? If cost-shifting is lessened, will the overall costs of care for treating conditions that could be treated in either setting decrease as more patients seek care with personal physicians, while charges for emergency conditions increase with less ability to cost shift?
- 3.Given the relatively high fixed costs and low marginal costs of EDs, what metrics should be used to determine the optimal ED capacity for a population relative to the capacity of the PCMH for managing acute unscheduled care? How do these metrics change with geography (i.e., rural vs. suburban vs. urban)? How do the costs of care vary with differing practice models, and how do these models vary with geography? How can the models of care for ED capacity and the PCMH be optimized over time to meet the needs of the individual and the community?
Quality of care
When it comes to severe illness, there is no question that patients are best served in EDs, which have more resources, immediate availability to advanced procedures and technology (including airway management and comprehensive radiology, laboratory, and pharmacy services), and more expeditious access to specialty care and inpatient beds. Physician specialty training for primary care versus emergency medicine is significantly different and in part dictates the patient assessment and management paradigm in each setting based upon expected case mix and acuity. Redistribution of patients between PCMH and ED settings may dictate a significant shift in training curricula to produce optimal quality and cost outcomes.
For patients with potentially serious conditions, EDs have the ability to rapidly risk-stratify conditions such as respiratory distress, abdominal pain, and chest pain. Limited data suggest that specialists are more likely to follow recommended guidelines and have better outcomes than primary care physicians, both from a cost and from a quality standpoint, particularly in the care of emergency conditions such as acute stroke and acute myocardial infarction.34–36 When patient care needs are less urgent, there is considerably less clarity regarding the most appropriate care site, because of the many lenses that this choice can be viewed through. For example, in minor acute illness such as urinary tract or upper respiratory tract infections, the quality of ED care may be similar to that provided by personal physicians. For other conditions, however, the picture is mixed.37 Some studies have found that patients identifying most of their care with a personal physician, as opposed to a specialist, have lower cost and mortality,38 although patients who have a regular primary care physician may have significantly different health and socioeconomic characteristics than those who do not.
Research Questions for Quality of Care Between the ED and PCMH Settings
What are the differences in the rates of test use, quality of care, and outcomes for risk-adjusted acute conditions cared for in geographically diverse ED and personal physician settings? Studies in this area may focus on both urgent and less urgent conditions with defined quality markers, such as guideline adherence in patients with symptoms of potential acute coronary syndrome (urgent) and the overuse of antibiotics for antibiotic-insensitive conditions (less urgent).
Neither primary care (e.g., patient-centered medical homes) nor the ED is capable of independently managing all acute, unscheduled care in the United States at this time. While various levers (e.g., copayments) might be used to influence patient choice as to the setting of care, relatively little is known about the comparative costs and quality of care provided in each setting. Until these issues are further elucidated, we suggest that it is premature to favor one setting over the other. Indeed, other settings presently account for a significant percentage of acute unscheduled care and, although small, may represent additional cost-effective venues.
We have presented a number of research questions generated by the acute, unscheduled care workgroup of the “Improving the Quality and Efficiency of Emergency Care Across the Continuum: A Systems Approach” conference. In dividing our discussion into the three major health care areas of access, cost, and quality, we do not mean to create a false trichotomy—our purpose is to provide a clear set of research questions. Certainly there will be overlap within these areas. In a patient-centered model, for example, patients may highly value the 24/7 convenience of an ED visit with complete testing, as opposed to a protracted series of visits with a patient-centered medical home and separate visits for laboratory and radiographic studies. A comparative cost/benefit analysis for such a scenario, and others pertaining to ambulatory sensitive conditions, would provide a more objective foundation for policy decisions.
At this time, we suggest that comprehensive, objective comparative research into the patient-centered medical home and ED models for acute, unscheduled care is necessary to inform continuing efforts in national health reform. We believe the answers to these research questions will help inform the optimal organization of the health care system to deliver quality, cost-effective acute unscheduled care for a population.