Chest Pain Units in Emergency Departments
A Report from the Short Term Observation Services Section Revised August 8, 1994
Acute myocardial infarction (AMI) is the leading cause of death in the adult population in the United States. Recent advances in the treatment of myocardial infarction have greatly reduced the mortality and morbidity from this disease. Yet, these treatments are time dependent and necessitate rapid initiation. For good outcome, it is desirable for the patient to recognize quickly the signs and symptoms of an AMI. For good outcome, it is desirable for the physician to diagnose quickly the patient with an AMI and initiate quickly the appropriate treatment of the patient with an AMI. For good outcome, it is desirable for the physician to identify those patients with an AMI having atypical signs and symptoms of this disease process.
In the past, patient outcome has not been ideal for the majority of patients who experience delays in the emergency department (ED) before receiving definitive therapy. Furthermore, a significant number of patients with AMI (4% to 13%) are not diagnosed by the emergency physician.(1-3) These AMI patients inadvertently released home have an 11% to 25% risk of dying from their MI.(2,3) These patients are the most expensive cause of malpractice litigation against emergency physicians, comprising 20% of dollars paid for malpractice.(4,5) Even when the emergency physician provides appropriate patient care patient outcome may be suboptimal since most patients delay a number of hours before seeking medical care for their chest discomfort. After this delay, substantial myocardial necrosis has occurred minimizing the impact of medical intervention.
The first area of concern is the delay in patient access to medical care. Many patients wait a significant amount of time after the onset of AMI symptoms before seeking medical care. The median delay from chest pain onset until the patient arrives in the ED is greater than 2 hours.(6,7) The mean delay is significantly longer due to a small number of people who wait 24 hours or longer before accessing care. If a patient with an AMI is treated within 70 minutes, the left ventricular damage can be minimized due to infarct abortion and mortality decreased to 1.6%.(8) If therapy is delivered after this time, mortality and morbidity from AMI rise dramatically, as significant irreversible myocardial damage has already occurred.
Secondly, after patients arrive in the ED, significant delays may occur before the physician makes the diagnosis of AMI and definitive care such as thrombolysis is implemented. Chest pain is one of the most common complaints of patients presenting to the ED comprising 5% of ED visits.(9) Yet only ten to fifteen percent of patients with chest discomfort have an AMI.(1-3) Most chest pain patients do not have significant disease. System problems in the organization of prehospital care and the ED can lead to significant delays in the administration of definitive care such as thrombolytics.(10-16) Delays in the diagnosis of AMI may also be increased due to gender biases, staff perception of a patient's pain as non cardiac, or lack of rapidly available serum markers for AMI.(17) Delays in administration of thrombolytics may lead to larger infarct size, increased number of complications, and ultimately higher mortality.(8)
The third area of concern is physician failure to correctly diagnose AMI in patients with atypical signs and symptoms. Four to thirteen percent of patients with AMI are released from the ED with false reassurance that they do not have coronary artery disease as a cause of their symptoms.(1-3) Many of these patients suffer complications from their AMI with 11% to 25% of these patients dying.(2,3) With these difficulties in diagnosing AMI, combined with complications from this serious disease process, it is not surprising that awards for patients with missed AMI constitute 20% of all dollars paid for malpractice in emergency medicine.(4,5)
Because of the failure in the traditional approach to patients with chest pain, many EDs have begun chest pain units (CPUs) with designated resources of personnel, protocol, space, and equipment for the patient presenting with chest pain. A 1992 survey of ACEP's leadership (n=319) showed that 9% of EDs had CPUs. Various names have been given to these programs: "Chest Pain ER", "Chest Pain Center", "Chest Pain Evaluation Unit", "Short Stay ED Coronary Care Unit", "ED Monitored Observation Bed", "Chest Pain Units".(18-21) These programs have developed approaches to improving the care of the patient with chest pain.
This document was developed by the Short Term Observation Services Section of the American College of Emergency Physicians to describe the components of these programs which are considered essential to a successful CPU program. Six common components have been identified in CPUs:
- Attack Program - an ED attack program to minimize delays in institution of therapy for an AMI (thrombolytics, nitrates, heparin, aspirin, etc.).
- Observation Program - an ED or hospital observation program which monitors and evaluates low risk patients to avoid the inadvertent release home of patients with AMI or unstable angina.
- Outreach Program - an ED or hospital based community outreach program which educates the public to promptly seek medical care if they have symptoms of an AMI, such as chest pain, chest discomfort, shortness of breath, diaphoresis, syncope and risk factors for coronary artery disease, particularly smoking.
- Unit Design - an ED CPU which has a functional design for optimal chest pain evaluation. It includes appropriate cardiovascular monitoring equipment.
- Unit Staffing - additional ED physician and nursing staff sufficient to provide the additional services required for patients in the CPU.
- Unit Management - CPU management structure based on continuous quality improvement program principles to ensure quality patient care and proper utilization of ED resources.
Development of a New Paradigm for Chest Pain
The new paradigm for evaluation and treatment of patients with chest pain includes a multi-tiered approach. Attack programs minimize hospital delays in provision of rapid therapy. Only 40% to 50% of patients with AMI have clear evidence of an infarction by ECG on their initial presentation to the ED. The ED requires a protocol to ensure prompt evaluation and therapy for the AMI patient. Observation protocols in the ED minimize the risk of unintentional release of patients with AMI, as well as identify patients with AMI or acute myocardial ischemia, with initially normal or non-diagnostic ECGs and normal levels of cardiac serum markers such as CK. These patients require evaluation beyond the initial emergency department work up, but can be safely observed in EDs with a protocol driven diagnostic algorithm. Outreach programs address the patient hesitancy to seek medical care. Community members require education to avoid delay in seeking care for symptoms such as chest pain, chest discomfort, diaphoresis, syncope, or shortness of breath that can represent an AMI.
Attack Program - Minimize Delays in the ED
Up to 60% of patients with an AMI have a clinical presentation suggesting a high or moderate probability of myocardial infarction or unstable angina. These patients' initial history, physical finding, and ECG are consistent with AMI. The MITI trial has shown prompt intervention in these patients can lower mortality to 1.6% by aborting the infarction.(8) This window of opportunity starts closing one hour after the patient's symptoms begin.
CPUs include systems to minimize delays in instituting definitive therapy. Many factors have been demonstrated to contribute to the failure to provide prompt therapy: delays in obtaining an ECG, delays in decision making on instituting thrombolytics, delays in preparing and obtaining the medication, and asking for consultation in patients with clear evidence of AMI.(10-16) Many different strategies have been developed to address these problems including a consensus strategy by the National Heart Attack Alert Program Coordinating Committee.(17). Those measures which decrease "mean time delay to thrombolysis" are appropriate whether in the pre-hospital setting (ECG done by pre-hospital personnel), in the emergency department (centralization of thrombolytic equipment in a designated room or mobile cart), or in the hospital (protocols for decreasing delays in preparation and delivery of thrombolytic drugs to the bedside). The measure of the success of a hospital program is not just compliance with certain clinical paths, but improvement in the efficiency and time of delivery of quality of patient care.
Observation Program - Minimize ED Release of Undiagnosed AMI Patients
Many patients with AMI have an initial clinical presentation suggestive of low probability of AMI or unstable angina. The patient's history, physical examination, ECG and initial serum cardiac markers do not suggest an AMI. Despite an atypical clinical presentation, the physician seeks to avoid release of the patient with an initially unimpressive presentation of ischemic heart disease until observation/testing can provide a more definitive diagnosis.
Work from the Multicenter Chest Pain Study Group has shown that 50% to 60% of patients who present to the ED with chest pain require an extended evaluation if virtually all of the patients with AMI are to be identified.6 All the patients with atypical chest pain who are at risk of having an acute MI can not be evaluated in the hospital. It has been estimated to cost 1.5 million dollars per life saved to evaluate in the hospital a patient with a 5% probability of AMI.(7,22) The ED observation bed is a more feasible site to evaluate the patient with chest pain having a low probability of AMI. The cost for observation in the ED is typically 20% to 50% the cost of in hospital evaluation.(21,23,24) Serial cardiac enzymes and continuous cardiac monitoring can identify the patient with an acute myocardial infarction who has an atypical clinical presentation. The work of the Emergency Medicine Cardiac Research Group has shown nearly all of these patients can be diagnosed in the ED observation bed in a short time period with the use of presently available diagnostic tests such as serial rapid CK-MB analysis and serial ECGs with ST segment trend monitoring.(21,25,26)
In addition to evaluating low probability of MI chest pain patients who under the traditional approach have been discharged inadvertently from the ED, CPUs can be used to evaluate many patients with a low probability of AMI chest pain who have traditionally been admitted to the hospital. A third of chest pain patients admitted to hospital intensive care units do not have cardiac disease.(23) Most of these patients can be evaluated in CPUs at a savings of $2,000 per patient compared with in hospital admission. Nationwide, the savings to the health care system are substantial. One billion dollars per year would be saved for every 10% of ED chest pain patients (500,000 patients) who are evaluated in CPUs rather than hospital ICUs.
Patients with atypical signs and symptoms of cardiac disease who experience unstable angina are also at risk of being released home with false reassurance that no serious disease exists. These patients are at risk of developing an acute MI. A diagnostic approach is necessary to identify those who have coronary artery disease. Up to 70% of patients with AMI have had prodromal symptoms before their infarction.(27-31) Serial cardiac enzymes and ECGs often can not make the diagnosis of unstable angina. Upon discharge from the CPU the patient can be scheduled for a form of testing that is reliable for identifying ischemic heart disease. Many strategies are under investigation for fulfilling this role of identifying the patient who is at high risk of developing an AMI: thallium scan,(32-33) echocardiogram,(34) exercise stress testing,(35-36) dobutamine echocardiogram(37) or single proton emission computed tomography.(38)
Outreach Program - Minimize Delays Seeking Medical Care
With an appropriate attack program and observation programs, patient outcome may still be compromised if the patient delays many hours before seeking medical care. Most of the benefit from aggressive medical intervention in the patient with an acute MI is during the first hour or two of symptoms.(8) For multiple reasons most patients delay many hours before seeking medical care.(39-41) The median delay in seeking medical care is > 2 hours.(6,7) With these long patient delays in seeking medical care, even prompt medical intervention (less than 1/2 hour hospital delay in administering thrombolytics) is of limited effectiveness since most of the myocardial tissue has necrosed;(42-44) the window of opportunity for intervention has closed. If the patient delays seeking medical care until the myocardial tissue is non-viable, the physician can only treat the complications of AMI.
Thus, prevention of patient delay in seeking medical care has a potentially greater impact on improving patient outcome than application of expensive, complex medical care in the hospital. Outreach programs educate the public to seek medical help as soon as symptoms begin that could represent AMI, including chest pain, chest discomfort, syncope, and faintness.(45-47) Annually, 540,000 patients suffer an AMI and present to the hospital. They would benefit from earlier arrival at the hospital. Outreach programs also educate the public to seek medical help as soon as prodromal symptoms begin that could represent angina. Most AMI patients have had prodromal symptoms days to weeks before they suffer their AMI.(27-31) If these patients seek medical care at the time of their prodromal symptoms, and the physician identifies they are at risk of an AMI, then the AMI may be avoided. Multi-faceted approaches, such as the Early Heart Attack Care (EHAC) program, utilize a broad range of techniques, i.e., videotapes, deputization programs, and school programs, to educate and motivate the public to seek medical care before it is too late.(48)
Outreach programs also seek to improve outcome by educating the public on the value of modifying risk factors for coronary artery disease. Smoking has a profound effect on the development of coronary artery disease. Cessation of smoking can significantly reduce an individual's chance of developing an AMI.(49-50) Hypertension and hyperlipidemia are also treatable factors for the development of atherosclerosis and coronary artery disease. Proper medical management of hypertension(51-53) and hyperlipidemia(54-55) reduce the individual's risk of AMI. Exercise conditioning can also significantly reduce the risk of AMI.(56-57) Model educational programs have been developed by the American Heart Association to educate the public on these issues. The emergency physician and the ED can play an instrumental role in public education on reducing risk factors for AMI.
Chest Pain Unit Design
The CPU in the ED is designed for the optimal delivery of diagnostic and therapeutic modalities to patients with chest discomfort. The units have the appropriate diagnostic testing available to identify patients with an AMI and acute ischemic coronary disease. The CPU is equipped to resuscitate patients who become unstable.
Design of the CPU in the ED will vary between hospitals as many hospitals have different ED configurations. The area of the ED designated for the attack program should be designed to expedite administration of acute therapy to the patient with AMI. This may be a specifically designated area of the ED or a mobile cart with the appropriate equipment. The area of the ED designated for the observation program should be equipped to identify patients with ischemic coronary syndromes as well as AMI. This may be a designated unit or individual beds appropriately equipped and staffed. An ED with a small volume of patients with chest pain may provide services in usual ED beds while an ED with a large volume of patients with chest pain may have a separate physical site, separate nursing services, and a separate nursing station.
Cardiovascular monitors are necessary for all patients evaluated in CPUs. The American College of Cardiology guidelines for cardiovascular monitoring call for adequate human surveillance 24 hours per day with "the degree of human surveillance required being inversely proportional to the sophistication and reliability of cardiovascular equipment."(58) The CPU typically contains patients with low to moderate probability of AMI. A single nurse for each patient is not required: constant human surveillance at the monitor is not necessary. Monitors with advanced features such as arrhythmia alarms, arrhythmia memory, and ST segment trend monitoring are the rational choice for a CPU.
Several authoritative sources are available on the design of ED CPUs. The American Institute of Architects (AIA) Guidelines on Health Care Facilities includes a section with specifications for ED observation units.(59) This includes recommendations on the size and structure of the ED observation rooms. The ACEP textbook on ED Design has a chapter on the design of ED observation units(60) and a chapter on the design of "Chest Pain ED's".(61) Multiple model designs are reviewed and the relation between structure and function discussed. ACEP's practice management committee has published information on ED observation beds(62) and this has recently been updated by ACEP's Observation Section.(63) These guidelines emphasize the need to design the treatment areas to be consistent with the type of service offered. Treatment areas for patients with symptoms requiring non monitored beds (abdominal pain, asthma) are designed differently than treatment areas for patients with symptoms requiring monitored beds (chest pain, syncope).
Chest Pain Unit Staffing
Traditional staffing standards for emergency departments assume a standard mix of services provided in the ED. The number of staff required to provide services is proportional to the number of patients treated in the facility. For emergency physicians, the standard is one physician is required for every 4,000 to 4,500 patient visits to the emergency department per year. This corresponds with the emergency physician providing 22-24 minutes of service to the average patient.(64-67) The standard for nursing care is one nurse is required for every 1,500 to 3,000 patient visits to the emergency department per year.(68) This corresponds with the emergency nurse providing 38 to 76 minutes of service to the average patient. The staff are distributed on the day, evening, and night shift proportional to the frequency of patient arrival in the ED during those shifts. Most departments have the greatest influx of patients during the evening shift and the second half of the day shift. They will require substantial staffing during these periods of peak demand.
CPU services are not provided in ED's offering traditional ED services. Time studies have shown provision of observation services to ED patients more than doubles the amount of emergency physician service required per patient.(69) Emergency physician staffing beyond the traditional staffing standard is required to provide those services. Formulae are available to calculate staffing needs based on the amount of required physician service needs.(70) A similar magnitude of additional services is required from nursing and ancillary staff for ED observation patients. Formulae are available to quantify the additional service needs.(71-73) Another method of calculating staffing needs is the designated level of intensity of service, for example, one nurse for every four to six monitored observation beds.
Chest Pain Unit Management
Emergency departments with CPUs have a vigorous quality management and utilization program. The structure and function of the chest pain program is continuously reviewed. Problems are identified and addressed that relate to adding observation services to the emergency department mix of services. The American College of Emergency Physicians has published guidelines on the management of emergency department observation beds.(62-63) These guidelines recommend indicants for measuring outcomes and proper utilization of resources.
A continuous quality improvement (CQI) program is the recommended model for management of health care services.(74-76) Problems in patient care are identified, the patient care system analyzed, and changes are devised in the patient care system to address the problem. Outcome indicators are followed to quantify whether the quality of care is being improved, e.g., mean delay time until administration of thrombolytics, percent of MI's not diagnosed by the emergency physician, hospital mortality for patients with AMIs. The measurement of success of a program is the improvement in the indicator of good patient care services.
Quality management is inherent to the functioning of the CPU. Thom Mayer has described how the CPU and ED are "process-rich", service oriented environments.(77) The CPU should have a management plan which is aligned with the plan of the ED and the parent institution. The Juran model of "planning for quality" is well suited for the service environment.(78) In the Juran approach, the term "quality management" has three components, all of which bear directly on the CPU: quality planning, quality control, and quality improvement.
The linchpin of the Juran trilogy is the quality planning arm. This focuses the quality management process by asking questions about the customer's (patient's) wants and needs. Who are our customers? What services do our customers need? How satisfied are our customers with the services which we are currently providing? How do we develop services that are responsive to those needs? How do we plan to build the processes and systems that will generate those services? How do we get those plans off of the drawing board and into operation? Unfortunately, most of us do not ask our patients and other customers often enough what they think and feel about the service we are providing. We assume that we know, but all too often we are incorrect. A patient may leave the CPU, vowing never to return, based on dissatisfaction with any number of factors, such as noise levels, food quality, or the perception of how much the staff cared about the patient as an individual. These issues are not generally measured by our performance and outcome indicators. A survey instrument can be of great help in this area. Properly designed and analyzed, a satisfaction survey tool can give us customer feedback that is very powerful.(79)
The quality control arm of the trilogy is the process used in traditional "quality assurance". Indicators of quality of patient care are chosen carefully balancing what we would like to measure with what we need to measure. There are dozens of processes involved in each aspect of patient care in a CPU. It would be impossible to examine on a continuing basis all aspects of the program. Juran has found there are a "vital few" processes which have the greatest overall impact on patient care. The Joint Commission on Accreditation of Healthcare Organizations suggests high-risk, high-volume, high-cost, and problem prone processes are the "vital few" for health care organizations.(80) A focus on these areas in the quality planning process is both feasible and likely to significantly improve patient care. An example for the CPU would be "mean ED delays in administration of thrombolytics". Thirty minutes has been set as a standard for one of the "vital few" indicators.(17) The ability to meet the standard is a measure of the success of the program in meeting its goal of providing quality patient care.
The quality improvement process embodies the program's response to a problem detected in the quality control portion of the program. When an indicator is worsening, e.g., increasing delays in provision of thrombolytics, the program has a process to address this identified problem. On a practical basis, this means forming "quality improvement teams".
When the quality control arm of the CPU's quality management program identifies a problem, a team is brought together. The team is comprised of the workers who know and understand the process being analyzed. There is interdisciplinary teamwork with team meetings based on mutual respect, openness, and the assumption there are system problems to address, not individuals to punish. Quality control studies have shown 85% of problems are "system problems" and not "people problems".(81) Thus, the team's purpose is not the traditional goal of "quality assurance" to identify and accuse an individual of failure of will, intention, alacrity, or intelligence. A variety of specialized group processes and statistical tools are utilized. The team moves through a quality improvement cycle addressing the problem.(82)
Thus with quality planning, quality control, and quality improvement processes functioning in a mutually supportive manner, the ED Chest Pain Unit program evolves to meet the patient's most pressing and vital needs. The reduction of morbidity and mortality from AMI is important not only to us as individuals and physicians, but also to the nation, as this disease takes its toll in health care dollars and lost productivity. Implementation of a CPU within the ED, with its six components (the attack program, the observation program, the outreach program, unit design, unit staffing and unit management) is a new approach to achieving this reduction.
- Tierney WM, Fitzgerald J, McHenry R, et al: Physicians' estimates of the probability of myocardial infarction in emergency room patients with chest pain. Med Decis Making 1986:6;12-17.
- Rouan GW, Hedges JR, Toltzis R, et al: Chest pain clinic to improve the follow-up of patients released from an urban university teaching hospital emergency department. Ann Emerg Med 1987:16;1145-1150.
- Goldman L, Cook EF, Brand DA, et al: A computer protocol to predict myocardial infarction in emergency department patients with chest pain. N Engl J Med 1988:318;797-803.
- Rogers TT: Risk management in emergency medicine. Dallas, Emergency Medicine Foundation - American College Emergency Physicians, 1985.
- Karcz A, Holbrook J, Burke MC, et al: Massachusetts emergency medicine closed malpractice claims: 1988-1990. Ann Emerg Med 1993:22;553-559.
- Cooper RS, Simmons B, Castaner A, et al: Survival rates and prehospital delay during myocardial infarction among black persons. Am J Cardiol 1986:57;208-211.
- Maynard C, Althourse R, Olsufka M, et al: Early versus late hospital arrival for acute myocardial infarction in the Western Washington thrombolytic trials. Am J Cardiol 1989:63;1296-1300.
- Weaver WD, Cerquerin M, Halstrom AT, et al: Prehospital initiation versus hospital initiation of thrombolysis: the Myocardial Infarction Triage and Intervention trial (MITI). JAMA 1993:270;1211-1216.
- Cochrane DG, Allegra JR, Graff LG: Epidemiology of observation services, in Observation Medicine, (L Graff editor), Boston, Andover Medical Publishers, 1993, p 37.
- Ornato JP: Role of the emergency department in decreasing the time to thrombolytic therapy in acute myocardial infarction. Clin Cardiol 1990:13;V48-52.
- Gonzalez ER, Ornato JP, Jones LA, et al: Hospital delays in thrombolytic therapy: A Multicenter prospective assessment of critical factors (abstract). J Am Coll Cardiol 1991:17;1593.
- Kereiakes DJ, Weaver WG, Anderson JL, et al: Time delays in the diagnosis and treatment of acute myocardial infarction: a tale of eight cities. Am Heart J 1990:120;773-780.
- Sharkey SW, Brunette DO, Ruiz E, et al: An analysis of the time delays preceding thrombolysis for acute myocardial infarction. JAMA 1989:262;3171-3174.
- Perry G, Wrightons WN, Hood L, et al: Delays to thrombolysis in the treatment of myocardial infarction. J R Coll Physicians Lond 1993:27(1);19-23.
- McGrath RB: Thrombolysis in acute myocardial infarction: a hospital emergency room experience. Indiana Med 1993:86(1);38-41.
- Birkhead JS: Time delays in provision of thrombolytic treatment in six district hospitals. Br Med J 1992:305(6851);445-8.
- National Heart Attack Alert Program Coordinating Committee, 60 minutes to treatment working group: Emergency department: rapid identification and treatment of patients with acute myocardial infarction. Ann Emerg Med 1994:23;311-329.
- DeLeon AC, Farmer CA, King G, et al: Chest pain evaluation unit: a cost-effective approach for ruling out acute myocardial infarction. Southern M Journal 1989:82;1083-1089.
- Gaspoz JM, Lee TH, Cook EF, et al: Outcome of patients who were admitted to a new short-stay unit to "rule-out" myocardial infarction. Am J Cardiol 1991:68;145-149.
- Bahr RD: Chest Pain ER - An Idea Whose Time Has Come. J of Cardiovasc Management 1991:Sept/Oct.
- Gibler WB: Chest pain evaluation in the emergency department: beyond triage. Amer J Emerg Med 1994:12;121-122.
- Ting HH, Lee TH, Soukup JR, et al: Impact of physician experience on triage of emergency room patients with acute chest pain at three teaching hospitals. Amer J Med 1991:91;401-408.
- Wears RL, Sergio L, Hernandez JD, et al: How many myocardial infarctions should we rule out? Ann Emerg Med 1989:18;953-963.
- Fineberg HV, Scadden D, Goldman, et al: Care of patients with low probability of acute myocardial infarction: cost effective alternatives to coronary care unit admissions. N Engl J Med 1984:310;1301-1307.
- Graff LG, Zun LS, Leikin J, Gibler WB, et al: Emergency department observation beds improve patient care: Society for Academic Emergency Medicine debate. Ann Emerg Med 1992:21;967-975.
- Hoekstra JW, Gibler WB, Levy RC, et al: Emergency department diagnosis of acute myocardial infarction and ischemia: a cost analysis. Academic Emergency Medicine 1994 (In Press).
- Gibler WB, Young GP, Hedges JR, et al: Acute myocardial infarction and chest pain patients with non diagnostic ECG's: serial CK-MB. Ann Emerg Med 1992:21;504-512.
- Gibler WB, Lewis LM, Erb RE, et al: Early detection of acute myocardial infarction in patients presenting with chest pain patients and non diagnostic ECG's: serial CK-MB sampling in the emergency department. Ann Emerg Med 1990:19;1359-1366.
- Harper RW, Kennedy G, DeSanctis RW, Hutter AM: The incidence and pattern of angina prior to acute myocardial infarction: a study of 577 cases. Am Heart J 1979:97;2;178-183.
- Kouvaras G, Bacoulas G: Unstable angina pectoris as a warning symptom before acute myocardial infarction. Quarterly J of Med, New Series:1987:64;244;680-684.
- Master AM, Dack S, Jaffe HL: Premonitory symptoms of acute coronary occlusion; a study of 260 cases. Ann Int Med 1941:14;1155.
- Bahr RD: Emphasizing chest pain as a symptom of heart attack. JAMA 1993:270;20;2435-6.
- Solomon HA, Edwards AL, Killip T: Prodromat in acute myocardial infarction. Circ 1969:60;463-71.
- Henneman PL, Mena IG, Rothstein RJ, et al: Evaluation of patients with chest pain and nondiagnostic ECG using thallium-201 myocardial planar imaging and technetium-99m first-pass radionuclide angiography in the emergency department. Ann Emerg Med 1992:21;545-550.
- Mace SE: Thallium myocardial scanning in the emergency department evaluation of chest pain. Am J Emerg Med 1989:7;321-328.
- OH JK, Miller FA, Shub C, et al: Evaluation of acute chest pain syndromes by two-dimensional echocardiography. Its potential in the selection of patients for reperfusion therapy. Mayo Clin Proc 1987:62;59-66.
- Mark DB, Shaw L, Harrell FE, et al: Prognostic value of a treadmill exercise score in outpatients with suspected coronary artery disease. N Engl J Med 1991:325;849-853.
- Tsakonis JS, Shesser R, Rosenthal R, et al. Safety of immediate treadmill testing in selected emergency department patients with chest pain. Am J Emerg Med 1991:9;557-559.
- Marcovitz P, Armstrong W: Accuracy of dobutamine stress echocardiography: correlation with coronary lesion severity as determined by quantitative angiography. J Am Coll Cardiol 1992:19;1197.
- Quinones M, Verani M, Haichin R, et al: Exercise echocardiography versus TI single-photon emission computed tomography in evaluation of coronary artery disease. Circ 1992:85;1026.
- Alonzo A: The impact of the family and lay others on care seeking during life threatening episodes of suspected coronary artery disease. Soc Sci Med 1986:22;1297-1311.
- Wielgosz ATJ, Nolan RP, Earp JA, et al: Reasons for patients' delay in response to symptoms of acute myocardial infarction. Can Med Assoc J 1988:139;853-857.
- Dracup K, Moser DK: Treatment seeking behavior among those with signs and symptoms of acute myocardial infarction. Heart Lung 1991:20;570-574.
- Tiefenbrunn AJ, Sobel BE: Timing of coronary recanalization: paradigms, paradoxes, and pertinence. Circ 1992:85;2311-2315.
- Ho MT, Eisenberg MS, Litwin PE, et al: Delay between onset of chest pain and seeking medical care. The effect of public education. Ann Emerg Med 1989:18;727-731.
- Blohm M, Herlitz J, Schroder U, et al: Reaction to a media campaign focusing on delay in acute myocardial infarction. Heart and Lung 1991:20;6;661-666.
- Herlitz J, Hartford MN, Blohm M, et al: Effect of a media campaign on delay times and ambulance use in suspected acute myocardial infarction. Am J Cardiol 1989:64;90-93.
- Bahr RD: Access to early cardiac care: chest pain as a risk factor for heart attacks, and the emergence of early cardiac care centers. Maryland Med J 1992:41;133-137.
- Rosenberg L, Kaufman DW, Helmrich SP, et al: The risk of myocardial infarction after quitting smoking in men under 55 years of age. N Engl J Med 1985:313;1511.
- Hermananson B, Omenn GS, Kronmal RA, et al: Beneficial six-year outcome of smoking cessation in older men and women with coronary artery disease. N Engl J Med 1988:319;1365.
- European Working Party on High Blood Pressure in the Elderly: Mortality and morbidity results from the European Working Party on High Blood Pressure in the Elderly trial. Lancet 1985:1;1349.
- SHEP Cooperative Research Group: Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. JAMA 1991:265;3255.
- Kannel WB, Sorlie P, Castelli WP, et al: Blood pressure and survival after myocardial infarction: the Framigham Study. Am J Cardiol 1980:45;326.
- Rossouw JE, Lewis B, Rifkind BM: The value of lowering cholesterol after myocardial infarction. N Engl J Med 1990:323;1112.
- Silberg JS, Henry DA: The benefits of reducing cholesterol levels: the need to distinquish primary from secondary prevention. Med J Australia 1991:155;665.
- O'Connor GTJ, Buring JE, Yusuf S, et al: An overview of randomized trials of rehabilitation with exercise after myocardial infarction. Circ 1989:80;234.
- Oldridge NB, Guyatt GH, Fischer ME, et al: Cardiac rehabilitation after myocardial infarction. Combined experience of randomized clinical trials. JAMA 1988:260;945.
- Jaffe AS, Atkins JM, Field JM, et al: Recommended guidelines for in-hospital cardiac monitoring of adults for deterction of arrythmia. J Am Coll Cardiol 1991:18;1431-1433.
- American Institute of Architects Committee on Architecture for Health: 1992-1993 Guidelines for construction and equipment of hospitals and medical facilities. American Institute of Architects Press, Washington, D.C., 1993.
- Keeler ML, Donovan MR: Observation Areas, in Emergency Department Design (ed. L Riggs Jr.), American College of Emergency Physicians, Dallas, 1993, p 185-197.
- Donovan MR, Keeler ML. The Chest Pain ED, in Emergency Department Design (ed. L Riggs Jr.), American College of Emergency Physicians, Dallas, 1993, p 277-283.
- American College of Emergency Physicians Practice Management Committee: American Collge of Emergency Physicians. Management of Observation Units. Ann Emerg Med 1988:17;1348-1352.
- Short Term Observation Services Section of the American College of Emergency Physicians (Brillman J, Mathers-Dunbar L, Graff L et al: Management of Observation Units (Revised 1992). Ann Emerg Med (Submitted for Publication).
- Hannas RR: Staffing the emergency department. Hospitals 1973:47;83-86.
- van de Leuv JH: Physician staffing. In van de Leuv JH (ed): Management of Emergency Services. Aspen Publications, Asheville, Aspen, 1987, p 16-28.
- Cue F, Inglis R: Improving the operations of the emergency department. Hospitals JAHA 1978:52;110-119.
- Holley JE, Kellerman AL, Andrulis DP: Physician staffing in the emergency departments of public teaching hospitals: a national survey. Ann Emerg Med 1992:21;53-57.
- Hospital committee of the American College of Emergency Physicians: Role of emergency department nurses in Emergency Department Organization and Management, 2nd edition, Jenkins AL (editor), CV Mosby Co., St. Louis, 1978, p 35.
- Graff LG, Wolf S, Dinwoodie R, et al: Emergency physician workload: a time study. Ann Emerg Med 1993: 22;1156-1163.
- Graff LG, Radford MJ: Formula for emergency physician staffing. Am J Emerg Med 1990:8;194-199.
- Krized MB, Chernow PM: Implementing a patient classification system in the emergency deprtment. Emerg Nurs Rep 1986:1;1-7.
- Schulmerich SC: Developing a patient classification system for the emergency department. J Emerg Nursing 1984:6;298-305.
- Schulmerich SC: Converting patient classification data into staffing requirments for the emergency department. J Emerg Nursing 1986:12;286-90.
- Kritchevsky SB, Simmons BP: Continuous quality improvement: concepts and applications for physician care. JAMA 1991:266;1817-1823.
- O'Leary DS, O'Leary MR: From quality assurqnce to quality improvement: the Joint Commision on Accreditation of Healthcare Organizations and Emergency Care. Emergency Medicine Clinics of North America 1992:10;477-492.
- Ellwood P: Outcome management: a technology of patient experience. N Engl J Med 1988:318;1549-1556.
- Mayer T: Industrial models for continuous quality improvement. Emergency Medical Clinics of North America 1992:10;523-547.
- Juran JM: Juran on Planning for Quality. New York, the Free Press, 1988.
- Steiber SR, Krowinski WJ: Measuring and Managing Patient Satisfaction. USA, American Hospital Association Publishing, Inc., 1990.
- Joint Commision on Accreditation of Healthcare Organizations: Development and Application of Indicators in Emergency Care. Oakbrook Terrace, Illinois, Joint Commision on Accreditation of Healthcare Organizations Publications, 1991.
- Deming EW: Out of Crisis. Cambridge, Massachusetts Institute of Technology Press, Center for Advanced Engineering Study, 1989.
- Mayer TA: Quality Improvement in Emergency Medicine. In Schwartz G, Mayer TA, Cayten G, et al (eds), Principles in Practice of Emergency Medicine, 3rd edition. Philadephia, Lea and Febiger, 1992.