EMERGENCY departments (EDs) represent a

Transcription

1 1024 PROCEEDINGS Gordon et al. SAFETY NET RESEARCH IN EM PROCEEDINGS Safety Net Research in Emergency Medicine: Proceedings of the Academic Emergency Medicine Consensus Conference on The Unraveling Safety Net JAMES A. GORDON, MD, MPA, JOHN BILLINGS, JD, BRENT R. ASPLIN, MD, MPH, KARIN V. RHODES, MD Abstract. A primary goal of the Academic Emergency Medicine Consensus Conference, The Unraveling Safety Net: Research Opportunities and Priorities, was to explore a formal research agenda for safety net research in emergency medicine. This paper represents the thoughts of active health services researchers regarding the structure and direction of such work, including some examples from their own research. The current system for safety net care is described, and the emergency department is conceptualized as a window on safety net patients and systems, uniquely positioned to help study and coordinate integrated processes of care. Key words: safety net; emergency department; research agenda. ACA- DEMIC EMERGENCY MEDICINE 2001; 8: EMERGENCY departments (EDs) represent a unique research consortium. Functioning at the interface of multiple systems, providers, and patients, the ED is a rich laboratory for research on the health care safety net. With more than 100 million ED visits annually, EDs provide not only a window on the health care needs of a diverse population, but also a window on the effectiveness of safety net systems. As the only provider mandated by federal law to provide universal health care, the ED is uniquely qualified to work at the interface of medicine, public health, and social work. With the advent of uniform data collection systems 1 and ED-based research networks, the ED can become an even more powerful tool in the development of efficient systems of care. From the Department of Emergency Medicine, Massachusetts General Hospital, Division of Emergency Medicine, Harvard Medical School, Institute for Health Policy, Massachusetts General Hospital/Partners HealthCare System, Boston, MA (JAG); Center for Health and Public Service Research, Robert F. Wagner Graduate School of Public Service, New York University, New York, NY (JB); Department of Emergency Medicine, Regions Hospital, HealthPartners Research Foundation, St. Paul, MN (BRA); and Section of Emergency Medicine, University of Chicago, Chicago, IL (KVR). Received June 20, 2001; accepted June 20, Address for correspondence and reprints: James A. Gordon, MD, MPA, Department of Emergency Medicine, Massachusetts General Hospital, 55 Fruit Street, Clinics 115, Boston, MA Fax: ; jgordon3@partners.org THE HEALTH AND PERFORMANCE OF THE SAFETY NET FOR VULNERABLE POPULATIONS: USING THE ED AS A RESEARCH TOOL (JOHN BILLINGS) There were no good old days for vulnerable patient populations. In the current health care environment even providers are at risk, and things may get worse still. To put it mildly, the performance of the safety net is less than swell. 2 But it doesn t have to be that way, especially if safety net providers such as the ED take an active role in studying and improving the system. The Performance of the Safety Net For Better or for Worse? In New York City (NYC) over the past decade, the number and pattern of preventable hospital admissions among vulnerable populations have remained constant. The advent of Medicaid managed care has added a new layer of access barriers and reimbursement dilemmas. Most Medicaid payers simply pay too little, and those who do pay well often unwittingly encourage an exodus of patients away from traditional safety net providers and toward more affluent private facilities. The Balanced Budget Amendment of 1997, cutbacks in federal Disproportionate Share (DSH) payments, and funding reductions for Federally Qualified Health Centers (FQHCs) place even further stress on a system that must support a growing number of uninsured patients.

2 ACADEMIC EMERGENCY MEDICINE November 2001, Volume 8, Number To make matters worse, mandatory managed care could actually happen, or Medicaid managed care rates could plummet. The economy could take a dip, and public providers could experience even more serious problems. In fact, the crown jewel teaching hospitals, often the core of safety net providers, could fare even worse under new federal funding guidelines. On the hopeful side, George W. Bush says he is going to provide more support. The federal state Children s Health Insurance Programs (CHIP) could begin to put a dent in the number of uninsured, and Medicaid managed care might improve substantially. But the President has already cut key safety net funding, and the status of managed care remains unclear. Using the ED to Gauge the Performance of the Safety Net. If the ED is the safety net for the safety net, we should be able to gather informative data on the performance of the safety net by looking at what s going on in EDs. Looking at our ED Pulse Project in NYC, for example, just over 40% of visits to the ED in 1998 were nonemergent. Approximately one-third of ED visits were emergent but could have been treated the same day in a doctor s office, while the remainder truly required ED care. The highest rates of nonemergent use of the ED (over 50%) could be found among Medicaid recipients and the uninsured in immigrant neighborhoods with serious barriers to primary care. Medicaid patients had the highest rates of nonemergent and primary-care treatable ED use, while commercially insured patients had the lowest. Many patients wait a considerable amount of time before heading to the ED. Uninsured and Medicaid patients wait the longest (more than half wait three days or more before going to the ED), while commercially insured patients wait just a little less. Regardless of the time to treatment or insurance status, however, most patients do not contact a health care provider before going to the ED. In fact, few patients in NYC are loyal to a single provider with whom they keep in touch. While just over half of newborns have no ambulatory doctor s office visit within the first month of birth, these infants are seen in the ED during the first year of life. More than half of new mothers will visit the ED during their pregnancy. The ED truly is the safety net for the safety net. The Safety Net Can Do Better: An Agenda for Change. While the performance of the safety net may be faltering, some providers are delivering very effective care. Patients of some hospital clinics use the ED at a much lower rate than others, producing a utilization rate comparable to that seen in private physician offices. Some of those hospital clinics are even efficient enough to meet their own expenses (which is quite an accomplishment). So how can we strengthen the safety net so that the ED is not always sitting there with a catcher s mitt? It s not going to be easy, and it s going to take some money money to support the primary care safety net, and money to ensure proper referrals, drug benefits, and other vital health services. And it s going to take better management. We must learn what patients want, how they use the system, and how non-health factors affect their choices. If a clinic visit takes twice as long as an ED visit, almost all patients will continue to go to the ED as a logical choice. Vulnerable populations will need coordinated multidisciplinary care and dedicated help in managing a complex system. A new attitude and commitment must begin. Hospitals have to get serious about primary care, and referral systems have to work to make vertical and horizontal integration actually work. Freestanding clinics have to coordinate care better with other parts of the system, and the ED has to be involved as a crucial component of the overall enterprise. Private physicians as well must be recognized as key components of the safety net Medicaid reimbursement of $11 per private office visit simply is not feasible. Finally, the strengthening of the safety net is going to require new knowledge. The ED can be a unique resource for studying how the entire system might improve. As a window on the safety net, EDs can monitor the performance of the entire system. Emergency departments see the patients who fall through the cracks, and can study where those patients come from, where they get resources, and who provides their primary care. Much like crash investigators from the National Transportation Safety Board, EDs can systematically investigate failures of the entire system. Why did the patient with diabetic ketoacidosis present to the ED (in some cases the ultimate preventable accident ) why wasn t that patient caught earlier with hyperglycemia and how can the system prevent such occurrences in the future? The ED can also be a window on the needs of vulnerable patients. Because the ED serves many of the most disadvantaged patients, it is an ideal setting from which to study such patient behavior and choices. The results of such inquiry will allow patients to truly partner in the design of effective systems of care. And by studying both the system and its patients, the ED can learn how to most effectively link its patients with needed health and welfare services. In this way the ED can actually help coordinate the care of its most vulnerable patients, and serve as a valuable tool for learning how to strengthen the safety net for everyone.

3 1026 PROCEEDINGS Gordon et al. SAFETY NET RESEARCH IN EM THE ED AS A WINDOW ON SAFETY NET SYSTEMS: CROWDING AND AMBULANCE DIVERSION (BRENT ASPLIN) Perhaps more than any other issue, the problem of ED crowding highlights system deficiencies that threaten safety net care. Although there are many potential research questions related to ED crowding, I would like to focus on three major areas. First, we must define ED crowding and work to identify its causes and consequences. This is a prerequisite to developing effective solutions. Second, the relationship between the federal Emergency Medical Treatment and Labor Act (EMTALA) and ED crowding must be explored. The unfunded mandates of EMTALA have created a funding vacuum that threatens the availability of quality emergency and inpatient care. Finally, we must evaluate the effect of crowding on patient safety and medical errors. The heightened national emphasis on patient safety by funding agencies, provider organizations, and patients should be leveraged to expand research opportunities and to improve the safety of ED care. Definitions, Causes, and Consequences of ED Crowding. The first step in an emergency medicine (EM) safety net research agenda is to establish a working case definition of ED crowding, one that can be reproducibly linked to outcomes of interest (e.g., patient morbidity and mortality, patient safety and medical errors, patient and provider satisfaction). While many emergency practitioners know what crowding feels like or can identify crowding when they see it, reliable metrics for crowding are still lacking. Consider the following three factors as a conceptual model for ED crowding: 1) input factors (what makes people present to the ED), 2) throughput factors (what happens to patients in the ED), and 3) output factors (what affects efficient patient disposition). 3 Input factors include any system or patient characteristic that leads to ED use. Examples include the capacity of ambulatory care providers, barriers to office-based care, and ED use as both a necessity and a convenience. Although nonurgent use of the ED is an important policy issue, it is not the cause of ED crowding. 4,5 Nonurgent ED use leads to crowding in the waiting room, not crowding in the treatment area patients with nonurgent problems are routinely triaged to wait while more urgent medical care is provided for the sickest. While input factors are largely beyond our control, throughput factors are often within our control. Throughput factors include the design of ED care systems, the use of evidence-based services, and the accessibility of information and technology. The study of throughput factors is an opportunity to examine our own contributions to the problem of ED crowding, and to develop interventions that improve the efficiency of ED care. Another issue that affects our ability to reduce throughput times for ED patients is the presence (or absence) of payment incentives for efficient care. For example, rapid stress testing of cardiac patients may be efficient (saves hospital time and money), but many insurers simply do not pay for it. Similarly, Medicare requires a three-day inpatient stay before transfer to a skilled nursing facility (SNF), when many ED patients would be better served by a direct transfer from the ED that avoids hospitalization altogether. While input and throughput issues clearly contribute to ED crowding, output problems are the primary determinants. For example, the availability of follow-up care often dictates ED output. If we cannot guarantee efficient primary care followup for the marginal patient with cellulitis or asthma (and often we cannot), that patient will be admitted to the hospital instead of being discharged. But an even greater problem and perhaps the single most important factor in the current ED crowding crisis is the inability to get our sickest ED patients into inpatient beds. This is a product of multiple factors, but when hospitals are forced to operate with such narrow margins in order to survive, their ability to handle fluctuations in the demand for inpatient beds is severely compromised. The practice of boarding admitted patients in the ED is the most frequently identified reason for ambulance diversion, and represents inadequate access to care for our sickest patients. There are some things that the ED simply cannot do as efficiently as the intensive care unit (ICU). It is a striking commentary that patients within the hospital system actually in the ED still have inadequate access to appropriate inpatient care. To the best of my knowledge, the research into this problem has yet to be done, including the baseline epidemiology of boarding inpatients in the ED (frequency, etiology, consequences, solutions). For example, would reducing our use of hospital telemetry beds affect the number of boarders in the ED? We simply don t know how current inpatient triage policies affect patient care and ED crowding. Research designed to answer these questions might lead to substantive improvements in the overall system. Other potential areas of research relating to patient boarding include workforce/staffing issues (e.g., nursing shortages), the effect of hospital payer mix/margin on inpatient capacity, and the ability of hospitals to discharge patients to skilled nursing facilities.

4 ACADEMIC EMERGENCY MEDICINE November 2001, Volume 8, Number The Emergency Medical Treatment and Labor Act. Both managed care and the Balanced Budget Act of 1997 have imposed economic constraints on staffing levels and availability of inpatient beds. Yet another major fiscal strain on hospitals and EDs can be found in EMTALA. The EMTALA statute guarantees stabilizing care, including hospitalization if necessary, for any patient who presents to a hospital ED. Though the intent of EMTALA is laudable, the statute has evolved into a hidden universal health care system for acute and catastrophic care. Most emergency physicians (EPs) agree that the services they provide under EMTALA are appropriate; however, because the EMTALA system is unfunded, hospitals must shift operating resources to cover the unreimbursed costs of EMTALA-mandated care. The costs of EMTALA and strategies for financing EMTALA services are discussed in another paper in this issue of Academic Emergency Medicine. 6 Patient Safety. Ever since the Institute of Medicine published its report on medical errors, 7 policymakers, government agencies, providers, patients, and the public have focused on patient safety as an important policy issue. A crowded ED represents a system that is operating beyond capacity, and constitutes a high-risk environment for medical errors. The research agenda for patient safety in EM is just beginning to evolve, but the study of ED crowding represents a logical component of a federally-funded strategy to improve patient safety. We need research to describe the basic epidemiology of medical errors, to understand the relationship between errors and ED crowding, and to propose solutions that alleviate ED crowding and enhance patient safety. For example, how do the rates of medical errors differ between crowded and noncrowded periods in the ED? What types of errors occur during periods of ED crowding and what system interventions reduce the likelihood of adverse events and errors? These questions should be part of EM s research agenda on ED crowding and medical errors. In summary, ED crowding is both our challenge and our opportunity as emergency practitioners. We must continue to demonstrate the core safety net role of the ED and to illuminate the importance of EMTALA as an unfunded mandate for ED-based universal care. Both present excellent opportunities to build a case for providing health insurance coverage for all citizens. Research addressing the causes and consequences of ED crowding must consider all facets of the problem, particularly the outflow problem of boarding ED patients. Baseline definitions and epidemiology are required. The costs of EMTALA must be quantified so that legislative funding remedies can be sought. And the inherent threat of crowding must be highlighted as a serious item on the national patient safety agenda. THE ED AS A WINDOW ON SAFETY NET PATIENTS: ACCESS AND PREVENTION (KARIN RHODES) Our patients should define our research agenda. As EPs and safety net providers, we need to study how we are doing at providing health care for our patients. One way of framing a patient-oriented research agenda is to measure what our patients need versus what they actually get. Do Our Patients Get What They Need? Our patients clearly have unmet needs when they come to the ED. In a multisite survey, 8 Lowenstein et al. reported that in addition to multiple other health risks, 23% of ED patients were at-risk drinkers (positive CAGE screen), 53% did not regularly use seat belts, and 48% smoked cigarettes. Many of them lacked the basic screening and preventive services for early detection and prevention of chronic disease. Despite the prevalence of unhealthy behaviors and chronic disease, practitioners do a fairly poor job of identification and referral for risky behaviors. In my own institution, a chart review of 355 consecutive nonurgent patients found very low rates of documentation for five major behavioral risks, all known to be common in the ED population: smoking, problem drinking, street drug use, high-risk sexual behavior, and domestic violence. Out of all the charts reviewed there was only one documented referral for domestic violence, and one for detoxification. Chart review certainly has its limitations, but it can provide an overview of the deficiencies in our care processes. Opportunities to Improve and Evaluate Preventive Care. A patient-oriented research agenda should be driven by a desire to improve the system to improve patient care. For example, simple changes such as placing check boxes or prompts on ED charts can improve screening and detection of substance abuse and domestic violence. Even better, patients can screen themselves for risky health behavior in the waiting room. We conducted pilot work using a self-administered computer health risk assessment in the ED waiting room, taking advantage of the long waiting times. Patients answered questions on a computer touch screen in a private setting. The program generated tailored health advice and referrals. We found that patients willingly disclosed even the most sensitive health risks, ranging from substance abuse, to injury-

5 1028 PROCEEDINGS Gordon et al. SAFETY NET RESEARCH IN EM prone behavior, to depression and domestic violence. Moreover, patients taking the computerbased health risk assessment were more than twice as likely to remember health advice compared with a control group that received usual care (62% vs. 27%, respectively). 9 For many disadvantaged patients, the ED visit represents an important opportunity for screening and prevention activities, not only because the ED visit may be a teachable moment, but also because the ED visit may be the patient s only contact with the health and welfare system. 10 Although referrals to additional resources and primary care are helpful for some ED patients, such follow-up may be difficult to achieve. Research into effective opportunities for brief ED intervention thus becomes even more important, especially for those young healthy adults who don t regularly go to the doctor, but who could benefit from counseling on risky behavior. Perhaps acute episodic primary care in the ED is an effective strategy for such individuals, if accompanied by high-quality preventive care. How Are We Doing as Access Providers? We cannot simply identify patient health risks without also assuring needed follow-up and referral. No form of care in the ED will be effective if provided in a vacuum. While the ED is universally accessible, do we actually ensure access to timely acute care? Tracking waiting/throughput times and left without treatment trends may reveal gaps in the assumed theory of universal access to care in the ED. As health services researchers, we are in the position to propose and test creative cost-effective methods of addressing pressing access problems. For example, perhaps ED patients with chronic conditions could have appointments made for them, and those with serious medical conditions who require urgent follow-up could be visited by a discharge planner in the ED 11,12 (as is done on the inpatient ward). This could help ensure needed follow-up and track the effect of brief discharge planning on patient follow-up. To truly have an impact on health policy, though, we will need to document the effect of any intervention on actual health outcomes. My view of a patient-oriented research agenda for ED-based safety net research thus focuses primarily on access and prevention. We are sorely lacking in data that allow us to monitor the health status and behavioral risks of ED patients; identify the best linkage methods to needed services; track and measure access barriers to acute and followup care; provide efficient primary care and welfare services on site; and monitor the costs, quality, and outcomes of access and prevention initiatives. Only when we have good data for such items can EDs effectively advocate for resources needed to improve the system. THE ED AS A WINDOW ON SAFETY NET SYSTEMS AND PATIENTS: COORDINATION OF THE SAFETY NET (JAMES GORDON) Not only can the ED be a window on patients and systems of care, but the ED can serve as a bridge, helping its patients to more efficiently access the fields of medicine, public health, and social welfare. 13 Each of these broad fields has its own distinct niche within the safety net, often operating independently of the others. But the ED is uniquely situated to help coordinate patients with their multiple systems of care, and to track the effectiveness of such collaboration. As we describe the various systems and research areas below, keep in mind the properties of useful outcomes data in health services research: worth measuring; can be measured for diverse populations; understood by people who need to act; will galvanize action; will lead to improvements that are known and feasible; measurement over time will reflect results of action. 14 Emergency Medicine and Traditional Medicine. Emergency medicine has close ties to office-based physician practices, particularly primary care. Not only does it make intuitive sense to ensure proper primary care follow-up for our patients, but such efforts can have a significant impact on subsequent ED utilization. In one study, Medicaid children targeted for specific referrals from the ED used the ED less in the following six months than a control group who received usual care. 15 In addition to simple referrals, outreach efforts can also be quite effective. Simply assigning an ED worker to help facilitate primary care follow-up in NYC had a significant effect on decreasing subsequent ED visits. 16 Emergency departments also act as primary care providers not just of last resort, but for routine care such as tetanus immunizations. Research indicates that other kinds of routine primary care such as pneumococcal vaccinations in selected older populations can be appropriately provided by the ED Emergency Medicine and the Public Health System. Emergency medicine also has logical connections with the public health system. 20 For example, it makes sense to provide advice and referrals for those with substance abuse issues in the ED, and research has successfully targeted the ED for brief interventions during the teachable moment of an ED visit. 21,22 More extensive public health outreach has also been implemented from

6 ACADEMIC EMERGENCY MEDICINE November 2001, Volume 8, Number the ED, ranging from weapons surveillance systems in cooperation with local law enforcement, to comprehensive networks for tracking infectious disease outbreaks. 23,24 And many functions traditionally reserved for the public health community are being successfully carried out in the ED, such as screening for sexually transmitted diseases and handing out bike helmets. Emergency Medicine and the Social Welfare System. Emergency medicine can also be a liaison with the social welfare system. Emergency departments traditionally make referrals to shelters for the homeless, for example, but many EDs also employ social workers to provide active welfare outreach to disadvantaged patients. Studies of such outreach suggest that ED-based social services may have significant advantages not only for patients and staff, but also for the bottom line of the hospital. 25 Perhaps the ED should formally take on some of the roles of the traditional welfare office 12 should food stamps be handed out to malnourished children and families in the ED?; should insurance eligibility be determined in the ED?; should a city welfare worker be stationed in the ED? Pilot studies of insurance outreach, for example, suggest that simply providing insurance applications and referrals in the ED may be quite effective in helping uninsured children to obtain health care coverage. 26 In summary, there are many opportunities to study the safety net through research in the ED. The ED can serve as a window on safety net systems, providing a unique perspective on issues of crowding and funding. It can also serve as a window on safety net patients, particularly related to issues of access and prevention. And it can be a useful window on the interaction between safety net systems and patients, useful for research into how best to coordinate the entire safety net process. Stuart Altman, Chair of the Institute of Medicine committee that produced the report America s Health Care Safety Net: Intact but Endangered, indicated that the committee was struck by the dearth of reliable and consistent data that can be used to accurately assess, measure, or compare the changing status of the safety net. 2 We hope the ideas presented here will help ED researchers to take a leading role in efforts to advance the field. References 1. National Center for Injury Prevention and Control. Data Elements for Emergency Department Systems, Release 1.0. Atlanta, GA: Centers for Disease Control and Prevention, Lewin ME, Altman S (eds). America s Health Care Safety Net: Intact but Endangered. Institute of Medicine. Washington, DC: National Academy Press, Asplin BR, Camargo CA Jr. A conceptual model of ED crowding. Work-in-progress, Derlet RW, Richards JR, Kravitz RL. Frequent overcrowding in U.S. emergency departments. Acad Emerg Med. 2001; 8: Kellermann AL. Deja vu. Ann Emerg Med. 2000; 35: Fields WW, Asplin BR, Larkin GL, et al. The Emergency Medical Treatment and Labor Act as a federal health care safety net program. Acad Emerg Med. 2001; 8: Kohn LT, Corrigan JM, Donaldson MS (eds). To Err Is Human: Building a Safer Health System. Institute of Medicine. Washington, DC: National Academy Press, Lowenstein SR, Koziol-McLain J, Thompson M, et al. Behavioral risk factors in emergency department patients: a multisite survey. Acad Emerg Med. 1998; 5: Rhodes KV, Lauderdale DS, Stocking CB, Howes DS, Roizen MF, Levinson W. Better health while you wait: a controlled trial of a computer-based intervention for screening and health promotion in the emergency department. Ann Emerg Med. 2001; 37: Gordon JA, Chudnofsky CR, Hayward RA. Where health and welfare meet: social deprivation among patients in the emergency department. J Urban Health. 2001; 78(1): Goldberg RM, Bernstein E, Anglin D, Cotler M, Hayne R, Travnitz R. Health promotion and disease prevention in the emergency department. In: Bernstein E, Bernstein J (eds). Case Studies in Emergency Medicine and the Health of the Public. Sudbury, MA: Jones and Bartlet Publishers, 1996, pp Gordon JA. The hospital emergency department as a social welfare institution. Ann Emerg Med. 1999; 33: Gordon JA. Emergency care as a safety net [letter]. Health Aff. 2000; 19(2): Chrvala CA, Bulger RJ (eds). Leading Health Indicators for Healthy People Final Report of the Institute of Medicine Committee on Leading Health Indicators for Healthy People Washington, DC: National Academy Press, 1999, p Grossman LK, Rich LN, Johnson C. Decreasing nonurgent emergency department utilization by Medicaid children. Pediatrics. 1998; 102(1 pt 1): Ling LJ, Cooke JS, Kornfeld E. New models for emergency and ambulatory care at academic health centers part I: New York City. Acad Emerg Med. 1995; 2: Stack SJ, Martin DR, Plouffe JF. An emergency department-based pneumococcal vaccination program could save money and lives. Ann Emerg Med. 1999; 33: Irvin CB, Wyer PC, Gerson LW, for the SAEM Public Health and Education Task Force Preventive Services Work Group. Preventive care in the emergency department, part II: clinical preventive services an emergency medicine evidencebased review. Acad Emerg Med. 2000; 7: Rhodes KV, Gordon JA, Lowe RA, for the SAEM Public Health and Education Task Force Preventive Services Work Group. Preventive care in the emergency department, part I: clinical preventive services are they relevant to emergency medicine? Acad Emerg Med. 2000; 7: Gordon JA, Goldfrank LR, Andrulis DP, D Alessandri RM, Kellermann AL. Emergency department initiatives to improve the public health. Acad Emerg Med. 1998; 5: Bernstein E, Bernstein J, Levenson S. Project ASSERT: an ED based intervention to increase access to primary care, preventative services, and the substance abuse treatment system. Ann Emerg Med. 1997; 30: Bernstein E, Goldfrank L, Kellermann A, et al. A public health approach to emergency medicine: preparing for the twenty-first century. Acad Emerg Med. 1994; 1: Kellermann AL, Bartolomeos KK. Firearm injury surveillance at the local level: from data to action. Am J Prev Med. 1998; 15(3 suppl): Talan DA, Moran GJ, Mower WR, et al. EMERGEncy ID NET: an emergency department-based emerging infections sentinel network. The EMERGEncy ID NET Study Group. Ann Emerg Med. 1998; 32: Gordon JA. Cost benefit analysis of social work services in the emergency department: a conceptual model. Acad Emerg Med. 2001; 8: Gordon JA, Dupuie TA. Child health insurance outreach through the emergency department: a pilot study. Acad Emerg Med. 2001; 8: