(Circulation. 1996;94:143-150.)
© 1996 American Heart Association, Inc.
Articles |
Cost-effectiveness of a Coronary Care Unit Versus an Intermediate Care Unit for Emergency Department Patients With Chest Pain
and the Departments of Medicine and Community and Family Medicine, Dartmouth Medical School (A.N.A.T.), Hanover, NH; the Department of Medicine (L.G.), University of California San Francisco, School of Medicine; and the Section for Clinical Epidemiology, the Division for General Medicine and the Cardiovascular Division, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School (T.H.L.), and the Department of Health Care Policy, Harvard Medical School (S.U.), Boston, Mass.
Correspondence to Thomas H. Lee, MD, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background Guidelines are not available for which patients with acute chest pain should be admitted to the coronary care unit and which patients can be reasonably triaged to monitored beds in lower levels of care.
Methods and Results Clinical and resource utilization data from 12 139 emergency department patients with acute chest pain were used in a decision-analytic model to identify cost-effective guidelines for the admission to a coronary care unit versus an intermediate care unit for initially uncomplicated patients without other indications for intensive care. The probability of clinical complications and death were derived from data on age-specific subsets of the population. Resource utilization estimates were based on cost data from a subset of 901 patients and length of stay data for the entire cohort. The survival benefit associated with initial triage to the coronary care unit instead of an intermediate care unit was assumed to be 15%. In the baseline analysis for 55- to 64-year-old patients, the probability of acute myocardial infarction (AMI) at which the coronary care unit had an incremental cost-effectiveness below $50 000 per year-of-life-saved was 29%. Triage to the coronary care unit was somewhat more cost-effective in elderly patients because their higher early complication rate more than offset their shorter life expectancy.
Conclusions This analysis indicates that the coronary care unit usually should be reserved for patients with a moderate (21% or more, depending on the patient's age) probability of AMI unless patients need intensive care for other reasons. Clinical data suggest that only patients with ECG changes of ischemia or infarction not known to be old have a probability of AMI this high. Intermediate care units are appropriate for patients whose risks are not high enough for a coronary care unit to be cost-effective but too high for other alternatives to be recommended for safety and effectiveness.
Key Words: cost-benefit analysis • myocardial infarction • coronary disease
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The natural history of acute myocardial infarction is frequently characterized by life-threatening complications.1 As a result, triage of patients with acute infarction to coronary care units is generally accepted as the standard of care.2 However, since this diagnosis is difficult to exclude with certainty on the basis of data available during the emergency department evaluation,3 4 5 only about 30% of patients who are admitted to coronary care units for evaluation of acute chest pain are ultimately found to have myocardial infarctions.2 Even when low-risk patients who are admitted to coronary care units have completely uncomplicated courses, they usually have longer lengths of stay6 and receive more nursing care7 than comparable patients who are admitted to intermediate care units.
To reduce coronary care unit admissions for low-risk patients with acute chest pain, previous investigations have developed5 8 and prospectively validated multivariate algorithms for the prediction of acute myocardial infarction9 10 and acute ischemic heart disease11 12 from emergency department clinical data. Although these algorithms provide insight into the risk for these diagnoses for individual patients, their application has been hindered by uncertainty over what probability thresholds should be used to determine triage decisions.
We therefore combined clinical and resource utilization data from a large population of emergency department patients with acute chest pain and used these data in a decision-analytic model to identify cost-effective guidelines for initial admission to a coronary care unit versus an intermediate care (or stepdown) unit on the basis of probability of acute myocardial infarction. This analysis differs from previous studies13 by the use of actual resource utilization and clinical data from a prospective clinical cohort to identify optimal thresholds for triage to the coronary care unit on the basis of a patient's age and underlying probability of acute myocardial infarction. The findings are intended to be applied to patients who, at the time of initial presentation, did not already have life-threatening complications or other indications for intensive care.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Decision-Analytic Model
The decision-analytic model assessed expected survival and hospitalization costs for patients with suspected acute myocardial infarction at the time of presentation to the emergency department. The model tracked patients over the course of their hospitalization after admission to either the coronary care unit or the intermediate care unit (Fig 1
). Within the first 48 hours, patients could either have acute myocardial infarction diagnosed (ruled in) or excluded (ruled out), and they could either die or survive. Patients who survived the first 48 hours were classified according the level of severity (I to IV) of the worst complication that occurred during this period. Complications were grouped on the basis of clinical and empiric estimates of their influence on early and late mortality and resource utilization. Level I included no complication other than premature ventricular contractions. Level II complications included frequent premature ventricular contractions, junctional ectopic rhythm (transitory or sustained), paroxysmal supraventricular tachycardia, atrial fibrillation or flutter, pericarditis, and sinus bradycardia or pause requiring treatment. Level III complications included ventricular tachycardia, pulmonary embolism, and recurrent ischemia. Level IV complications included ventricular fibrillation, complete heart block, shock, cardiac arrest, Mobitz II heart block, pulmonary edema, and infarct extension.
|
During the remainder of the hospitalization, patients remained at risk for developing complications or dying. Expected resource utilization and life expectancies were calculated for patient subsets that were defined by their initial survival, acute myocardial infarction status, initial triage site, and worst complication during the hospitalization.
The decision model was used to estimate expected costs and life-years remaining for patients aged 30 to 44, 45 to 54, 55 to 64, 65 to 74, and 
75 years admitted to the coronary care unit and intermediate care unit with varying underlying probabilities of acute myocardial infarction. Each age group was analyzed separately because the risks of complications and mortality differ among age groups, as does subsequent life expectancy. Expected costs and remaining life expectancy were used to compute incremental cost-effectiveness ratios for the coronary care unit relative to the intermediate care unit. Measured as cost per year-of-life-saved, the incremental ratio was defined as the net difference in costs divided by the net gain in life-years-saved for the coronary care unit relative to the intermediate care unit. We estimated the threshold probabilities of acute myocardial infarction at which the incremental cost-effectiveness ratios for the coronary care unit were below $100 000, $75 000, and $50 000 in 1992 US dollars. Both costs and life-years were discounted at a rate of 5% per year.
Data and Assumptions
To estimate probabilities of death and of minor, major, and life-threatening complications, the decision-analytic model used clinical data from patients who were enrolled from 1984 to 1986 in the Multicenter Chest Pain Study, a cooperative investigation of the clinical characteristics and outcomes of emergency department patients with acute chest pain.1 3 4 7 14 15 16 17 18 19 20 21 22 23 24 25 All patients who were 30 years or older and who came to a participating emergency department with a chief complaint of anterior, precordial, or left-sided chest pain unexplained by obvious local trauma or abnormalities on a chest roentgenogram were eligible for the study for up to three visits. The participating institutions were Brigham and Women's Hospital, Boston (which began enrolling patients in January 1984); Yale–New Haven Hospital, New Haven, Conn (December 1983); Danbury (Conn) Hospital (January 1984); Milford (Conn) Hospital, (June 1984); St Mary's Hospital, Waterbury, Conn (April 1984); University of Cincinnati (Ohio) Hospital (July 1984); and William Beaumont Hospital, Royal Oak, Mich (June 1985). Of 14 832 potentially eligible patients, 2140 did not consent to participate. For 434 patients who enrolled, follow-up was inadequate to allow the investigators, who were blinded to the patients' emergency department data, to determine whether a myocardial infarction had occurred. Other clinical data from the initial presentation or subsequent course that were necessary for these analyses were unavailable for another 119 patients. Therefore, a total of 12 139 patients contributed clinical and resource utilization data for this report.
Clinical data from the emergency department evaluation and subsequent clinical courses for these patients were collected prospectively according to a previously published protocol.10 Actual resource utilization data for a subset of 901 patients at one teaching hospital6 were used to assign median total costs for patient subsets. Age-specific probabilities of events were derived from logistic regression models, which included age in 10-year intervals after adjusting for whether the patient had an acute myocardial infarction.
To make the analyses both clinically relevant and tractable, the hospital course was divided into two segments: the first 48 hours and the remainder of the hospitalization. In our model, the first 48 hours would be spent in the original triage location unless subsequent clinical information would lead to early discharge or transfer to another level of care. For the remainder of the hospitalization, the patient's course, location, and resource utilization would be influenced primarily by events that occurred after admission.24
48-Hour Mortality
The initial triage decision for a patient would be expected to have its impact on outcomes during the first 48-hour segment of the patient's course; therefore, 48-hour mortality rates were determined for patient subsets. Of 1603 patients with acute myocardial infarction who were 30 years of age or older and admitted to the coronary care unit, 90 (5.6%) died within 48 hours of admission. Because in the younger age groups too few patients were available to make an accurate point estimate of mortality, a logistic regression model was used to estimate age-specific probabilities of mortality from acute myocardial infarction within the first 48 hours among patients initially admitted to the coronary care unit. The model included age in intervals, whether the patient had an acute myocardial infarction, and place of admission. The estimated 48-hour mortality rates for patients ages 30 to 44, 45 to 54, 55 to 64, 65 to 74, and 75 years were 3.1% (95% confidence interval [CI], 1.4% to 6.8%), 2.8% (95% CI, 1.5% to 5.3%), 4.6% (95% CI, 3.1% to 6.8%), 6.8% (95% CI, 4.9% to 9.5%), and 9.5% (95% CI, 7.0% to 12.9%), respectively.
The survival benefit associated with initial triage to the coronary care unit instead of an intermediate care unit for patients with acute myocardial infarction was estimated from a previous multivariate analysis of Multicenter Chest Pain Study data.18 This study did not identify a statistically significant benefit from initial triage to the coronary care unit; however, the mid point of the 95% confidence interval in an analysis that adjusted for factors associated with initial triage decision indicated that the probability of death in patients with acute myocardial infarction might be 15% greater if they were admitted to the intermediate care unit instead of the coronary care unit. Therefore, our baseline assumption, which favors the coronary care unit, was that death would be 15% greater among patients with acute myocardial infarction who were admitted to an intermediate care unit. For example, the probability of death within 48 hours for a 60-year-old patient with an acute myocardial infarction but without major complications on presentation to the emergency department was assumed to be 5.27% if the patient was triaged to the intermediate care unit compared with 4.58% if the patient was triaged to the coronary care unit.
The probability of death within 48 hours for patients without acute myocardial infarction was assumed to be unaffected by age and admission to the coronary care unit or the intermediate care unit. Only nine predischarge deaths occurred among 10 364 patients without acute myocardial infarction in the Multicenter Chest Pain Study. Thus, the mortality for patients without infarction was estimated to be 0.1% (95% CI, 0.03% to 1.4%).
Rates of Complication
On the basis of analyses of empiric data from the Multicenter Chest Pain Study,21 the probabilities of each level of complication were assumed to be independent of place of admission. Among patients who survived the first 48 hours and had acute myocardial infarction in the Multicenter Chest Pain Study database, the most severe complication was a level I event for 20%, level II for 10%, level III for 47%, and level IV for 23%. For patients without acute myocardial infarction who survived the first 48 hours, the distribution of worst complication was level I, 79%; level II, 6%; level III, 13%; and level IV, 2%. Age-specific probabilities of the level of complication were estimated with an ordinal regression model,26 which included age and the presence of an acute myocardial infarction as predictors. For example, among 55- to 64-year-old persons surviving the first 48 hours, the probabilities of levels I to IV complications during the initial 48 hours were 27%, 10%, 49%, and 14%, respectively, for patients with acute myocardial infarction, and 80%, 7%, 12%, and 1%, respectively, for patients without myocardial infarction.
Resource Utilization
Median total costs were calculated from hospital charges by application of the Health Care Financing Administration's institution-specific cost-to-charge ratio to hospital charge data, which had been normalized to 1986 dollars, for patients admitted to Brigham and Women's Hospital. These estimates were then inflated to 1992 dollars using the general medical care component of the consumer price index.27 A previous analysis of these data6 indicated that age was not significantly associated with resource utilization after adjusting for other factors. Thus, median total costs were assigned to each patient outcome as defined by survival status, acute myocardial infarction status, initial triage location, and worst level of complication regardless of age.
In some outcome categories, only a small number of patients with full cost data were available for analysis (Table 1). Therefore, we used the entire cohort of 12 139 patients, stratified by survival status, acute myocardial infarction status, and whether or not any complications occurred during the hospitalization, to compare length of stay as a proxy of resource utilization between patients admitted to the coronary care unit and patients admitted to an intermediate care unit. We found that length of stay differed significantly between patients who were initially admitted to the coronary care unit and intermediate care unit only for those without level II-IV complications (Wilcoxon P<.001 for both acute myocardial infarction and nonacute myocardial infarction patients). Therefore, median total costs for patients with any complications were assumed to be the same whether they were admitted to the coronary care unit or the intermediate care unit (Table 1).
|
Life Expectancy
Predischarge mortality for patient subsets was derived from the Multicenter Chest Pain Study database. For nonacute myocardial infarction patients, posthospitalization survival and life expectancy were estimated using 1987 US Lifetables. For acute myocardial infarction patients, an annual excess risk of death was applied to the US Lifetable.27 To estimate the annual excess risk of death associated with acute myocardial infarction, the mortality observed at 1, 2, and 3 years after discharge in a cohort of 1956 patients in the Multicenter Chest Pain database28 was compared with mortality reported in the US Lifetable for persons of the same average age as the cohort. The resulting estimates of excess mortality were 4.9%, 3.6%, and 1.6% for patients 1, 2, and 3 or more years after discharge, respectively. Using lifetable analysis, we calculated both discounted and undiscounted life expectancies for the 30- to 44-, 45- to 54-, 55-to 64-, 65- to 74-, and 75-year-old age groups who were assumed on average to be ages 40, 50, 60, 70 and 80 years, respectively (Table 2). The average ages used for the youngest and oldest age groups reflect the mean age of persons with acute myocardial infarction in these age intervals.
|
Sensitivity Analyses
To determine the extent to which our results depend on the mortality for patients with acute myocardial infarction who were admitted to the intermediate care unit, we repeated the analysis for each age range under the assumption that mortality was either doubled to 30% or halved to 7.5% (versus 15% in the baseline analysis). In addition, we investigated the impact that the following parameters had on the probability of acute myocardial infarction at which the coronary care unit cost less than $50 000 per additional year-of-life-saved: mortality rates for patients with myocardial infarctions admitted to the coronary care unit; median costs; additional resource costs for patients with myocardial infarctions who were discharged from the hospital; and postdischarge life expectancies for patients with and without myocardial infarction.
Because cost-effectiveness is a relative concept, the cost-effectiveness of the coronary care unit versus an intermediate care unit for patients with chest pain can only be assessed in comparison to the cost-effectiveness of commonly accepted medical practices.29 The cost-effectiveness of several interventions used to treat or prevent coronary heart disease have cost-effectiveness ratios below $50 000. Therefore, we use the value of $50 000 per additional year-of-life-saved as a benchmark in the reporting of results from sensitivity analyses.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Baseline Results
Using the assumption that there is a 15% relative increase in mortality when patients with acute myocardial infarction are admitted to the intermediate care unit instead of the coronary care unit, we found that the costs per year-of-life-saved for triage to the coronary care unit varied markedly, depending on the age of the patient and the emergency department probability of acute myocardial infarction (Fig 2
). For 55- to 64-year-old patients with an emergency department probability of infarction of 1%, the cost per year-of-life-saved was $1.4 million, but when the probability of infarction was 99%, the cost per year-of-life-saved was $15 000. For 55- to 64-year-old patients, the cost per additional year-of-life-saved was less than $100 000 if the probability of infarction was more than 15%; less than $75 000 if the probability of infarction exceeded 20%; and less than $50 000 if the probability of infarction exceeded 29%.
|
Admission to the coronary care unit was generally more costly for younger patients. The coronary care unit had a cost-effectiveness comparable to other accepted medical interventions (less than $50 000 per year-of-life-saved)30 when the initial probability of acute myocardial infarction was greater than 57% among patients 30 to 44 years of age. This threshold probability for acute myocardial infarction dropped to 21% among patients aged 65 to 74 years. Cost-effectiveness ratios for admission to the coronary care unit instead of intermediate care unit for specific patient subsets are presented in Table 3.
|
Sensitivity Analyses
The probability of myocardial infarction at which initial triage to the coronary care unit instead of an intermediate care unit was cost-effective varied markedly with the assumption that the survival benefit was associated with the coronary care unit for patients with acute myocardial infarction. If the intermediate care unit is associated with half of the baseline estimate for the relative increase in mortality for patients with acute myocardial infarctions (7.5% instead of 15%), the threshold probability of acute myocardial infarction at which the cost per year-of-life-saved for the coronary care unit fell below $50 000 increased to approximately twice the baseline estimate. In contrast, when the relative increase in mortality was doubled for patients with acute myocardial infarction who were triaged to the intermediate care unit (30% instead of 15%), the cost per year-of-life-saved for the coronary care unit was less than $50 000 among patients whose probability of acute myocardial infarction was more than 15%.
When the age-specific estimated probabilities of death for patients with acute myocardial infarction were varied, higher mortality rates favored the coronary care unit as the initial triage location. Thus, when mortality was at the upper end of the confidence interval, the threshold probability at which admission of 55- to 64-year-old persons to the coronary care unit cost less than $50 000 per year-of-life-saved fell to 20% (versus the baseline threshold of 29%). When mortality was assumed to be at the lower end of the confidence interval, this threshold probability rose to 44%. Varying the mortality rate for patients without myocardial infarction across its 95% confidence interval (0.03% to 1.4%) did not change the threshold probabilities for admission of patients to the coronary care unit.
When median cost estimates were varied across their interquartile ranges, the threshold probability of acute myocardial infarction at which the coronary care unit cost less than $50 000 per additional year-of-life-saved ranged from 23% to 45% for 55- to 64-year-old patients. Overall, differences in costs had less of an impact on the cost-effectiveness of the coronary care unit than did differences in mortality among acute myocardial infarction patients.
When both lower cost estimates (25th percentile) and higher mortality rates (upper 95% CI) within 48 hours of admission for acute myocardial infarction patients were used, the threshold probability of acute myocardial infarction at which the coronary care unit cost less that $50 000 per additional year-of-life-saved remained above 12% for all age groups (Fig 3). This threshold probability ranged from 12% or above in 65- to 74-year-olds to 19% in 45- to 54-year-olds. When the relative benefit for the coronary care unit was doubled to 30% (versus 15% baseline) in conjunction with lower cost estimates and higher mortality estimates, the threshold probability for admission of patients to the coronary care unit ranged from 6% in 65- to 74-year-olds to 9% in 45- to 54-year-olds.
|
Because survivors of acute myocardial infarction may incur greater healthcare costs after discharge from the hospital than do patients without myocardial infarction, we repeated the analysis assuming that patients with acute myocardial infarction generated increased costs of $20 000 to $40 000 during their remaining years of life. This change had minimal effects on the cost-effectiveness of the coronary care unit. For example, when the additional costs were $40 000, the threshold probability for admission to the coronary care unit changed from 29% to 32% for 55- to 64-year-old patients. Similarly, varying the life expectancy of patients with myocardial infarction to equal that of patients without acute myocardial infarction had little effect on the findings.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This analysis indicates that the cost-effectiveness of the coronary care unit as the initial level of care for emergency department patients with acute chest pain is highly dependent on their probability of acute myocardial infarction and also varies by their age. Initial triage to the coronary care unit was generally more cost-effective for older patients who had higher age-specific mortality rates. The probability of acute myocardial infarction above which admission of patients to the coronary care unit instead of an intermediate care unit cost less than $50 000 per year-of-life-saved was 23% for patients aged 65 years or more in the baseline analysis and exceeded 12% in all age groups even when low cost estimates and high mortality estimates for acute myocardial infarction were used simultaneously.
Data from the Multicenter Chest Pain Study indicate that the probability of acute myocardial infarction for emergency department patients with acute chest pain is 21% or greater only when the ECG shows ST-segment or T-wave abnormalities that are suggestive of ischemia and that are not known to be old.25 In this cohort, 76% of patients who had ST-segment elevation of 1 mm or more in two or more leads had acute myocardial infarction. Patients who had ST-segment depression or T-wave inversion had a 28% probability of myocardial infarction. Thus, the baseline analysis indicates that for patients who do not have other indications for intensive care, initial triage of patients with chest pain to the coronary care unit is unlikely to be cost-effective in the absence of ECG changes (Table 3
).
Although elderly patients' lower postdischarge life expectancy increases estimates of cost per year-of-life-saved for the elderly, we found that initial triage to the coronary care unit was generally more cost-effective for older patients because their higher in-hospital mortality rates imply a greater absolute benefit from the coronary care unit than in younger patients. Results for 65- to 74-year-old patients and those 75 years of age and older were remarkably similar. Triage to the coronary care unit, however, was slightly more costly for the patients over age 75 years than for 65- to 74-year-old patients, because their higher 48-hour mortality rate did not completely outweigh their lower life expectancy.
These findings are based almost entirely on prospectively collected clinical and detailed resource utilization data from the Multicenter Chest Pain Study, which included three university hospitals and four community hospitals. The use of these extensive data allowed the probability of death and other outcomes among patient subsets to be estimated directly instead of extrapolated from findings of studies of different populations. Although the performance of this analysis within one extensive database had several advantages, our findings are subject to the limitations of these data. Long-term survival data were obtained only for a subset of 1956 patients, and cost data were available only for 901 patients from one hospital. However, length of stay data were available for the entire cohort, and these data support the cost assumptions used in the baseline analysis.
Since these data were collected, use of thrombolytic therapy and emergent revascularization have become more common for patients with acute myocardial infarction. However, these developments are unlikely to influence the findings of the analysis, since these interventions are not directed at the moderate risk population for whom admission to the coronary care unit is an issue and because access to invasive revascularization procedures usually is not based on initial level of care. Another important innovation is the development of coronary observation units and emergency department chest pain units in which patients with a low risk of acute myocardial infarction can undergo rapid, low-cost evaluations. However, these units are directed at a population with a lower risk for acute myocardial infarction than the patients for whom admission to the coronary care unit is an issue. The availability of such units would be expected to make the cost-effectiveness of the coronary care unit even less attractive for low-risk patients.
Among the few assumptions that had to be used, the most critical was the survival advantage conferred by admission to the coronary care unit instead of an intermediate care unit on patients with acute myocardial infarction. Although it seems logical that the additional personnel and equipment available in a coronary care unit provide some benefit, no data have demonstrated an advantage from the higher level of care for patients who are not already in need of intensive care interventions at the time of the emergency department evaluation. Furthermore, it seems unlikely that randomized trials of this question will be performed.
We therefore assumed that mortality was 15% higher for patients with infarction who were admitted to an intermediate care unit, which was the midpoint of the 95% confidence interval derived from a previous multivariate analysis of the potential beneficial impact of the coronary care unit.22 This assumption favors the coronary care unit as an initial level of care in this analysis. If the true benefit of the coronary care unit is somewhat less, then the threshold probability for admitting patients to this level of care would increase.
These findings are consistent with and extend the results of a previous cost-effectiveness analysis of this issue,11 which was based on clinical data from published studies and charge data from 15 patients. The prior study also concluded that the coronary care unit was not cost-effective for patients with a probability of acute myocardial infarction ranging from 5% to 20%. In our baseline analysis, the coronary care unit did not become reasonably cost-effective until the probability of acute myocardial infarction exceeded 19%. If the increased mortality associated with admission of patients with infarction to an intermediate care unit is less than 15%, then intermediate care may be a cost-effective alternative to the coronary care unit for patients with an even higher initial probability of infarction. Our current analysis found that triage to the coronary care unit was cost-effective for patients with a probability of acute myocardial infarction in the 12% to 19% range only when low costs and high 48-hour mortality (upper bound of the 95% confidence interval) estimates were used for patients with acute myocardial infarctions. Furthermore, only when the relative benefit of the coronary care unit was doubled (30% versus 15%) in conjunction with lower cost and higher mortality estimates did the threshold probability fall below 12%. Even under these extreme assumptions, initial triage to the coronary care unit was not cost-effective for patients with a probability of acute myocardial infarction less than 6%.
Our findings do not imply that patients without ECG changes are appropriate for discharge or even admission to a nonmonitored bed. This analysis is restricted to an evaluation of the cost-effectiveness of the coronary care unit in comparison to intermediate care facilities, which also have central ECG monitoring and adequate personnel to detect and treat in-hospital complications rapidly and effectively and to perform cardiopulmonary resuscitation should a patient have a cardiac arrest. The availability of flexible nursing staffing plans at some institutions may blur the distinction between the coronary and intermediate care units; in such settings, the incremental costs associated with placing a patient in a unit called the coronary care unit might be less than assumed in this analysis.
The translation of these findings into clinically meaningful terms is that for patients with chest pain who do not have conditions such as uncontrolled ischemia, relative hypotension, substantial heart failure, active arrhythmias, or other complications requiring intensive care, the coronary care unit generally is not a cost-effective initial level of care for patients without ECG changes that are suggestive of ischemia. This recommendation is based on the cost-effectiveness of the coronary care unit compared with other accepted medical interventions.31 32 It is also supported by clinical investigations, which have shown that patients without such changes on their initial ECGs are unlikely to have life-threatening complications,12 24 33 34 and even some patients with ECG ischemia are at relatively low risk of developing complications that require intensive care.24
This analysis has not addressed the threshold for admission to a monitored setting versus discharge directly from the emergency department. Future investigations should seek to identify cost-effective management strategies for patients with lower risks of acute myocardial infarction and complications.
|
Acknowledgments |
|---|
This study was supported in part by grants from the Agency for Health Care Policy and Research (RO1-HS06452 and 1-PO1-HS06431-02), the National Center for Health Services Research (HS -5927), and the John A. Hartford Foundation, New York, NY (83102-2H).
Received February 19, 1995; revision received January 11, 1996; accepted January 22, 1996.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Lee TH, Rouan G, Weisberg MC, Brand DA, Acampora D, Stasiulewicz C, Walshon J, Terranova G, Gottlieb L, Goldstein-Wayne B, Copen D, Daley K, Brandt AA, Mellors J, Jakubowski R, Cook EF, Goldman L. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol. 1987;60:219-224.[Medline] [Order article via Infotrieve]
2. Lee TH, Goldman L. The coronary care unit turns 25: historical trends and future directions. Ann Intern Med. 1988;108:887-894.
3.
Lee TH, Cook EF, Weisberg MC, Sargent RK, Wilson C, Goldman L. Acute chest pain in the emergency room: identification and examination of low-risk patients. Arch Intern Med. 1985;145:65-69.
4.
Lee TH, Weisberg M, Cook EF, Daley K, Brand DA, Goldman L. Evaluation of creatine kinase and creatine kinase-MB for diagnosing myocardial infarction: clinical impact in the emergency room. Arch Intern Med. 1987;147:115-121.
5. Tierney WM, Roth BJ, Psaty B, McHenry R, Fitzgerald J, Stump DL, Anderson FK, Ryder KW, McDonald CJ, Smith DM. Predictors of myocardial infarction in the emergency room patients. Crit Care Med. 1985;13:526-531.[Medline] [Order article via Infotrieve]
6. Udvarhelyi IS, Goldman L, Komaroff AL, Lee TH. Determinants of resource utilization for patients admitted for evaluation of acute chest pain. J Gen Intern Med. 1992;7:1-10.[Medline] [Order article via Infotrieve]
7. Lee TH, Cook EF, Fendrick AM, Shammash JB, Wolfe EP, Weisberg MC. Impact of initial triage decisions on nursing intensity for patients with acute chest pain. Med Care. 1990;28:737-745.[Medline] [Order article via Infotrieve]
8. Baxt WG. Use of an artificial neural network for the diagnosis of myocardial infarction. Ann Intern Med. 1991;115:843-848.
9. Goldman L, Weinberg M, Weisberg M, Olshen R, Cook EF, Sargent RK, Lamas GA, Dennis C, Wilson C, Deckelbaum L, Fineberg H, Stiratelli R. A computer-derived protocol to aid in the diagnosis of emergency room patients with acute chest pain. N Engl J Med. 1982;307:588-596.[Abstract]
10. Goldman L, Cook EF, Brand DA, Lee TH, Rouan GW, Weisberg MC, Acampora D, Stasiulewicz C, Walshon J, Terranova G, Gottleib L, Kobernick M, Goldstein-Wayne B, Copen D, Daley K, Brandt AA, Jones D, Mellors J, Jakubowski R. A computer protocol to predict myocardial infarction in emergency department patients with chest pain. N Engl J Med. 1988;318:797-802.[Abstract]
11. Pozen MW, D'Agostino RB, Mitchell JB, Rosenfeld DM, Guglielmino JT, Schwartz ML, Teebagy N, Valentine JM, Hood WB Jr. The usefulness of a predictive instrument to reduce inappropriate admissions to the coronary care unit. Ann Intern Med. 1980;92:238-242.
12. Pozen MW, D'Agostino RB, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary-care-unit admission practices in acute ischemic heart disease. N Engl J Med. 1984;310:1273-1278.[Abstract]
13. Fineberg HV, Scadden D, Goldman L. Care of patients with a low probability of acute myocardial infarction: cost-effectiveness of alternatives to coronary-care-unit admission. N Engl J Med. 1984;310:1301-1307.[Abstract]
14. Rouan GW, Lee TH, Cook EF, Brand DA, Weisberg MC, Goldman L. Clinical characteristics and outcome of acute myocardial infarction in patients with initially normal or nonspecific electrocardiograms. Am J Cardiol. 1989;64:1087-1092.[Medline] [Order article via Infotrieve]
15. Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, Goldman L. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med. 1987;106:181-186.
16. Lee TH, Cook EF, Weisberg MC, Brand DA, Rouan GW, Goldman L. Candidates for thrombolysis among emergency department patients with acute chest pain: the Multicenter Chest Pain Study Experience. Ann Intern Med. 1989;110:957-962.
17. Solomon CG, Lee TH, Cook EF, Weisberg M, Brand DA, Rouan GW, Goldman L. Comparison of clinical presentation of acute myocardial infarction in patients older than 65 years of age to younger patients: the Multicenter Chest Pain Experience. Am J Cardiol. 1989;63:772-776.[Medline] [Order article via Infotrieve]
18. Beamer AD, Lee TH, Cook EF, Brand DA, Rouan GW, Weisberg MC, Goldman L. Diagnostic implications of the circadian variation of the onset of chest pain. Am J Cardiol. 1987;60:998-1002.[Medline] [Order article via Infotrieve]
19. Cunningham MA, Lee TH, Cook EF, Brand DA, Rouan GW, Weisberg MC, Goldman L. The effect of gender on the probability of myocardial infarction among emergency department patients with acute chest pain. J Gen Intern Med. 1989;4:392-398.[Medline] [Order article via Infotrieve]
20. Lee TH, Short LE, Brand DA, Jean-Claude YD, Weisberg MC, Rouan GW, Goldman L. Patients with acute chest pain who leave emergency departments against medical advice. J Gen Intern Med. 1988;3:21-24.[Medline] [Order article via Infotrieve]
21. White LD, Lee TH, Cook EF, Weisberg MC, Rouan GW, Brand DA, Goldman L. Comparison of the natural history of new onset and exacerbated chronic ischemic heart disease. J Am Coll Cardiol. 1990;16:304-310.[Abstract]
22. Fiebach N, Cook EF, Lee TH, Brand DA, Rouan GW, Weisberg M, Goldman L. Outcomes of patients with myocardial infarction who are initially admitted to stepdown units: data from the Multicenter Chest Pain Study. Am J Med. 1990;89:15-20.[Medline] [Order article via Infotrieve]
23. Lee TH, Juarez G, Cook EF, Weisberg MC, Rouan GW, Brand DA, Goldman L. Ruling out acute myocardial infarction: a prospective multicenter validation of a 12 hour strategy for patients at low risk. N Engl J Med. 1991;324:1239-1246.[Abstract]
24. Goldman L, Cook EF, Lee TH, Rouan G, Brand DA. Prediction of subsequent major events in emergency department patients with acute chest pain. N Engl J Med. In press.
25. Pearson SD, Goldman L, Garcia TB, Cook EF, Lee TH. Physician response to a prediction rule for the triage of emergency department patients with chest pain. J Gen Intern Med. 1994;9:241-247.[Medline] [Order article via Infotrieve]
26. McCullagh P. Regression models for ordinal data (with discussion). J R Stat Soc B. 1980;42:109-142.
27. US Department of Labor. Bureau of Labor Statistics. Consumer Price Index Detailed Report, May 1992. Government Printing Office; 1992.
28. Lee TH, Ting HH, Shammash JB, Soukup JR, Goldman L. Long-term survival of emergency department patients with acute chest pain. Am J Cardiol. 1992;69:145-151.[Medline] [Order article via Infotrieve]
29. Weinstein MC. From cost-effectiveness ratios to resource allocation: where to draw the line? In: Sloan FA, ed. Valuing Health Care: Costs, Benefits and Effectiveness of Pharmaceuticals and Other Medical Technologies. New York, NY: Cambridge Press; 1995.
30. Weinstein MC, Stason WB. Cost-effectiveness of interventions to prevent or treat coronary heart disease. Ann Rev Public Health. 1985;6:41-63.[Medline] [Order article via Infotrieve]
31. Goldman L, Gordon D, Rifkind B, Hulley SB, Detsky AS, Goodman D, Kinosian B, Weinstein M. Cost and health implications of cholesterol lowering. Circulation. 1992;85:1960-1968.[Abstract]
32. Detsky AS, Naglie IG. A clinician's guide to cost-effectiveness analysis. Ann Intern Med. 1990;113:147-154.
33. Brush JE Jr, Brand DA, Acampora D, Chalmer B, Wackers FJ. Use of the initial electrocardiogram to predict in-hospital complications of acute myocardial infarction. N Engl J Med. 1985;312:1137-1141.[Abstract]
34. Slater DK, Hlatky MA, Mark DB, Harrell FE Jr, Pryor DB, Califf RM. Outcome in suspected acute myocardial infarction with normal or minimally abnormal admission electrocardiographic findings. Am J Cardiol. 1987;60:766-770.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  J. M. May, W. P. Shuman, J. N. Strote, K. R. Branch, L. M. Mitsumori, D. W. Lockhart, and J. H. Caldwell Low-Risk Patients With Chest Pain in the Emergency Department: Negative 64-MDCT Coronary Angiography May Reduce Length of Stay and Hospital Charges Am. J. Roentgenol., July 1, 2009; 193(1): 150 - 154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Hoffmann, F. Bamberg, C. U. Chae, J. H. Nichols, I. S. Rogers, S. K. Seneviratne, Q. A. Truong, R. C. Cury, S. Abbara, M. D. Shapiro, et al. Coronary computed tomography angiography for early triage of patients with acute chest pain: the ROMICAT (Rule Out Myocardial Infarction using Computer Assisted Tomography) trial. J. Am. Coll. Cardiol., May 5, 2009; 53(18): 1642 - 1650. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Hendel Is Computed Tomography Coronary Angiography the Most Accurate and Effective Noninvasive Imaging Tool to Evaluate Patients With Acute Chest Pain in the Emergency Department?: CT Coronary Angiography Is the Most Accurate and Effective Noninvasive Imaging Tool for Evaluating Patients Presenting With Chest Pain to the Emergency Department: Antagonist Viewpoint Circ Cardiovasc Imaging, May 1, 2009; 2(3): 264 - 275. [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Takakuwa and E. J. Halpern Evaluation of a "Triple Rule-Out" Coronary CT Angiography Protocol: Use of 64-Section CT in Low-to-Moderate Risk Emergency Department Patients Suspected of Having Acute Coronary Syndrome Radiology, August 1, 2008; 248(2): 438 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Hoffmann, J. T. Nagurney, F. Moselewski, A. Pena, M. Ferencik, C. U. Chae, R. C. Cury, J. Butler, S. Abbara, D. F. Brown, et al. Coronary Multidetector Computed Tomography in the Assessment of Patients With Acute Chest Pain Circulation, November 21, 2006; 114(21): 2251 - 2260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Swap and J. T. Nagurney Value and Limitations of Chest Pain History in the Evaluation of Patients With Suspected Acute Coronary Syndromes JAMA, November 23, 2005; 294(20): 2623 - 2629. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Christenson, G. Innes, D. McKnight, B. Boychuk, E. Grafstein, C. R. Thompson, F. Rosenberg, A. H. Anis, K. Gin, J. Tilley, et al. Safety and efficiency of emergency department assessment of chest discomfort Can. Med. Assoc. J., June 8, 2004; 170(12): 1803 - 1807. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Goldman and A. J. Kirtane Triage of Patients with Acute Chest Pain and Possible Cardiac Ischemia: The Elusive Search for Diagnostic Perfection Ann Intern Med, December 16, 2003; 139(12): 987 - 995. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Snider, M. Papaleo, S. Beldner, C. Park, D. Katechis, D. Galinkin, and A. Fein Is Telemetry Monitoring Necessary in Low-Risk Suspected Acute Chest Pain Syndromes?* Chest, August 1, 2002; 122(2): 517 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Junker, J. E. Zimmerman, C. Alzola, E. A. Draper, and D. P. Wagner A Multicenter Description of Intermediate-Care Patients* : Comparison With ICU Low-Risk Monitor Patients Chest, April 1, 2002; 121(4): 1253 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. H. Lee and L. Goldman Evaluation of the Patient with Acute Chest Pain N. Engl. J. Med., April 20, 2000; 342(16): 1187 - 1195. [Full Text] [PDF]
|
|
|
|
|
|
C. A. Polanczyk, K. M. Kuntz, D. B. Sacks, P. A. Johnson, and T. H. Lee Emergency Department Triage Strategies for Acute Chest Pain Using Creatine Kinase-MB and Troponin I Assays: A Cost-Effectiveness Analysis Ann Intern Med, December 21, 1999; 131(12): 909 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Moon, A. DeAnda Jr, G. T. Daughters II, N. B. Ingels Jr, and D. C. Miller Effects of mitral valve replacement on regional left ventricular systolic strain Ann. Thorac. Surg., September 1, 1999; 68(3): 894 - 902. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||