SEARCH

SEARCH BY CITATION

Keywords:

  • cardiac arrest;
  • resuscitation;
  • outcomes;
  • cost of care

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Objectives:  Care provided to patients who survive to hospital admission after out-of-hospital cardiac arrest (OOHCA) is sometimes viewed as expensive and a poor use of hospital resources. The objective was to describe financial parameters of care for patients resuscitated from OOHCA.

Methods:  This was a retrospective review of OOHCA patients admitted to one academic teaching hospital from January 2004 to October 2007. Demographic data, length of stay (LOS), and discharge disposition were obtained for all patients. Financial parameters of patient care including total cost, net revenue, and operating margin were calculated by hospital cost accounting and reported as median and interquartile range (IQR). Groups were dichotomized by survival to discharge for subgroup analysis. To provide a reference group for context, similar financial data were obtained for ST-segment elevation myocardial infarction (STEMI) patients admitted during the same time period, reported with medians and IQRs.

Results:  During the study period, there were 72 admitted OOCHA patients and 404 STEMI patients. OOCHA and STEMI groups were similar for age, sex, and insurance type. Overall, 27 (38.6%) OOHCA patients survived to hospital discharge. Median LOS for OOHCA patients was 4 days (IQR = 1–8 days), with most of those hospitalized for ≤4 days (n = 34, 81.0% dying or discharged to hospice care). Median net revenue ($17,334 [IQR $7,015–$37,516] vs. $16,466 [IQR = $14,304–$23,678], p = 0.64) and operating margin ($7,019 [IQR = $1,875–$15,997] vs. $7,098 [IQR = $3,767–$11,138], p = 0.83) for all OOHCA patients were not different from STEMI patients. Net income for OOCHA patients was not different than for STEMI patients (–$322 vs. $114, p = 0.72).

Conclusions:  Financial parameters for OOHCA patients are similar to those of STEMI patients. Financial issues should not be a negative incentive to providing care for these patients.

ACADEMIC EMERGENCY MEDICINE 2010; 17:612–616 © 2010 by the Society for Academic Emergency Medicine

Survival after out-of-hospital cardiac arrest (OOHCA) is poor nationally, with most research estimating survival to discharge of 5%–10%.1,2 Many papers document that approximately 40%–60% of patients who are resuscitated in the field and are admitted to the hospital survive to discharge.3–5 A number of authors have advocated for specialty “cardiac arrest centers” that provide interventions that are known to improve outcomes. Therapies such as resuscitative hypothermia for comatose patients post–cardiac arrest,6,7 or percutaneous coronary intervention (PCI) for those arrests associated with ST-segment elevation myocardial infarction (STEMI), have been shown to improve outcomes in these patients,8 yet are not widely available. Tertiary care centers for emergency care to trauma, stroke, and STEMI patients have all been supported in the literature and have been shown to improve outcomes.9–15 The economics of trauma centers in particular have been shown to be an important issue in addressing whether these centers are available to provide care for patients.16–18

A common perception among physicians who provide postresuscitative care is that patients who survive to admission languish in hospitals for extended periods, consume an inordinate amount of resources, and are a poor use of resources. Perceptions of poor outcomes serve as a barrier to aggressive care for these patients. Perceptions of high costs and unfavorable margins could be significant barriers to administrators and physicians wishing to develop specialty centers for cardiac arrest patients. A number of studies have addressed cost-effectiveness of cardiac arrest care from a societal perspective.19–21 However, there is little work that addresses the costs of care from the hospital perspective.

To attempt to place financial issues regarding cardiac arrest patients into an administrative perspective, we sought to compare the cost of hospital care provided to OOHCA admissions to another common cardiac emergency, STEMI, which is commonly viewed as an appropriate and valuable use of resources.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Study Design

This was a retrospective records review to evaluate hospital financial outcomes for OOHCA patients. The institutional review board of the participating institution approved the study.

Study Setting and Population

The sample included a cohort of patients who were resuscitated after OOHCA and admitted to a single academic community hospital from January 2004 to October 2007. The cohort selected for convenience consisted of patients from one of six communities who were participating in an OOHCA trial. To provide a reference group for context of our findings, similar financial data were obtained for a cohort of STEMI patients admitted to the same institution during that same period. The cohort of STEMI patients were those who presented to the emergency department (ED) during the study period with an initial diagnosis of STEMI. All patients were transported directly to this institution (no transfers). The institution is a large (1,061 licensed beds in 2007) academic hospital that treated 118,000 emergency patients in 2007. The institution is an American College of Surgeons verified Level 1 trauma center, has an aggressive interventional cardiac care program, and routinely cares for OOHCA patients that are resuscitated and transported to it. Patient care provided to patients during this study was not evaluated, but can best be described as (except for therapeutic hypothermia) standard care for OOHCA patients postresuscitation, with intensive care unit support and interventional therapy (e.g., PCI for postarrest STEMI patients) being provided. As part of its emergency medical services (EMS) base station function, the ED maintains an ongoing database of OOHCA patients, which includes data on patients who survive to hospital admission and subsequent discharge. Patients who were OOHCA survivors and subsequently diagnosed as STEMI were included as cardiac arrest patients.

Study Protocol

Records of OOHCA and STEMI patients who survived to hospital admission were retrospectively reviewed and coded for demographics and insurance type, as well as hospital and facilities costs. We recorded survival to hospital discharge and length of stay (LOS). Financial parameters for each patient, including total hospital and facilities costs (direct and indirect), net revenue, and net operating margin, were abstracted and calculated from the hospital’s cost accounting system, in a method similar to that described in depth by Henneman et al.22 Costs were calculated using Sunrise ESPI Software (Eclipsys Corporation, Atlanta, GA). Costs fall into one of eight categories: capital, contracted services, medications, equipment (including maintenance/lease), labor, general, surgical implants, and medical/surgical supplies. They fall into one of four types (fixed direct, variable direct, fixed indirect, and variable indirect). Direct hospital and facilities costs are actual fixed and variable costs associated with direct patient care. Net revenue is defined as payments received from the payer. Net operating margin are the funds remaining when direct costs have been deducted from net revenue. Net income is defined as the remainder of total costs minus net revenue. Given that much of the inpatient care to the study patients was provided by private practice physicians, physician professional costs were not calculated and not included in hospital costs.

Data Analysis

Categorical variables were examined using a chi-square test where appropriate (expected frequency > 5); otherwise the Fisher’s exact test was used. All continuous variables were examined using the Wilcoxon signed rank test, as none of these variables were normally distributed. Median and interquartile range (IQR) or 90th percentiles are reported for these variables. The Type 1 error rate was 0.05. We adjusted for subgroup analysis by using a p-value of 0.025, Each of the two disease groups (OOHCA and STEMI) was dichotomized by survival to hospital discharge (survivors) for subgroup analysis. Because there is no prior literature addressing this topic, we were unable to perform an a priori power analysis. For purposes of analysis, patients discharged to hospice care were coded as not being discharged alive.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

During the study period, there were 81 OOHCA patients who survived to hospital admission from the study communities. Of those, 72 had complete identifiers for study inclusion. There were 404 STEMI patients during the study period.

The majority of OOHCA patients were male (41, 56.9%) and had public insurance (42, 58.3%). There were no overall significant differences between OOHCA and STEMI patients with respect to sex, age, insurance type, or LOS. The median LOS of all OOHCA patients was 4 days, with most (34, 81.0%) of those shorter than 4 days either dying or being discharged to hospice care. Overall, 27 (38.6%) survived to hospital discharge and four additional patients were discharged to hospice care. Summary demographic data are presented in Table 1, which identifies that these populations are demographically similar.

Table 1.    Demographics of Patients Resuscitated From Cardiac Arrest Versus Sustaining STEMI
 OOHCASTEMIp-value
  1. LOS = length of stay; OOHCA = out-of-hospital cardiac arrest; STEMI = ST-segment elevation myocardial infarction.

All casesn = 72n = 404 
 Age, median in yr (90th percentile)64 (81)65 (83)0.33
 Male, n (%)41 (56.9)244 (60.4)0.58
 Insurance: Medicare/Medicaid, n (%)42 (58.3)222 (55.0)0.59
 LOS, median in days (90th percentile)4 (18)4 (11)0.98
Survivorsn = 27n = 378 
 Age, median in yr (90 percentile)56 (81)64 (83)0.12
 Male, n (%)21 (77.8)231 (61.1)0.08
 Insurance: Medicare/Medicaid, n (%)12 (44.4)201 (53.2)0.38
 LOS, median in days (90th percentile)9 (22)4 (11)0.0001

Overall OOHCA and STEMI patients were not significantly different with respect to financial parameters (Table 2). However, for the subset of patients who survived to hospital discharge, OOHCA patients generated significantly higher charges and net revenue, which may be attributable to the significantly longer LOS in the OOHCA group compared to the STEMI group (median [90th percentile] = 9 [22] vs. 4 [11], p = 0.002). There was no significant difference in operating margin between groups. There was, however, net income which was lower in the cardiac arrest survivor group compared to STEMI patients who survived to hospital discharge.

Table 2.    Inpatient Financial Parameters of OOHCA and STEMI Patients
 OOHCASTEMIp (Wilcoxon)
  1. LOS = length of stay; IQR = interquartile range; OOHCA = out-of-hospital cardiac arrest; STEMI = ST-segment elevation myocardial infarction.

All Casesn = 72n = 404 
 Total charges (IQR)$61,823 ($25,056 to $154,696)$53,493 ($41,652 to $78,222)0.68
 Net revenue17,334 (7,015 to 37,516)16,466 (14,304 to 23,678)0.64
 Operating margin7,019 (1,875 to 15,997) 7,098 (3,767 to 11,138)0.83
 Net income−322 (−9,411 to 7,349)114 (−5,303 to 4,583)0.72
Survivorsn = 27n = 378 
 Total charges (IQR)$168, 583 ($69,470 to $257, 184)$52,848 ($41,187 to $75,490)<0.001
 Net revenue40,362 (17,971 to 59,068)16,439 (14,369 to 22,920)<0.0001
 Operating margin7,841 (3,553 to 19,051) 7,035 (3,801 to 10,945)0.28
 Net income−6,665 (−19,365 to 2,993)206 (−5078 to 4,510)0.01

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

These data document that, overall, caring for resuscitated OOHCA patients and STEMI patients has similar financial implications for this hospital. Both groups of patients had similar financial parameters of net revenue, operating margin, and net income. For the subset of OOHCA patients who survived to hospital discharge, there was increased net revenue for OOHCA patients and a nonsignificant decrease in net income compared to STEMI patients. The costs and LOS for OOHCA patients are not normally distributed, with a subset of the group with short LOS (and most commonly early death) and another population with longer LOS and increased costs. The relatively small numbers of survivors in the OOHCA cohort and their large variances tempers the conclusions that can be reached for this subset of patients. These data identify the need for further work that describes and analyzes the cost of providing care to this more complex population.

These clinical entities are fundamentally different, have different approaches to clinical care, and have widely divergent outcomes. However, from an administrative perspective, they must both be evaluated as financial entities, which require resources, and program support if the hospital wishes to provide care, or in the circumstance where expansion of services are contemplated, must be economically viable or meet the mission of the institution. Regionalization of emergency care has become an important issue in the United States. Lurie et al.23 have advocated for the development of Level 1 cardiac arrest centers, hospitals that specialize in postresuscitative care for OOHCA survivors. A number of other leaders in emergency medicine have recently advocated for the development of “shock-trauma centers,” which provide aggressive care for postarrest and other types of critical care patients. In 2009 alone, the Society for Academic Emergency Medicine, the National Association of EMS Physicians, and the Institute of Medicine have all convened conferences to address regionalization of emergency care. An understanding of the economics of such a movement must be understood.

Previous authors have evaluated the cost-effectiveness of interventions by EMS personnel to improve cardiac arrest survival.21,24,25 Little work, however, has been done to evaluate the value and cost of in-hospital postarrest care. Many providers perceive that resuscitation efforts for this population are often futile, and an increased understanding of clinical outcomes, LOS, and cost is important if physicians are to advocate for aggressive inpatient cardiac arrest care. In one analysis by Hamel, the cost of care for nontraumatic coma was not cost-effective from a societal perspective with a range of cost $87,000 to $140,000 per quality-adjusted life-year.26 However, this work focused on patients after Day 3 of coma, which does not reflect a substantial proportion of the postarrest population.

The concept of postcardiac arrest, or “shock-trauma” centers, requires data documenting the value of such hospitals. Work by Carr et al.,27 Liu et al.,28 and others has documented variation in survival between hospitals postresuscitation, raising the question as to whether specialty hospitals may result in improved outcomes. More robust data that address both the cost and the benefit of this equation are needed if physicians are to successfully advocate for specialty centers.

Limitations

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

These data cover a period during which therapeutic hypothermia was initiated as a research protocol at the study hospital (introduced March 2005). Although no charge was made for this therapy, it may have altered the LOS and outcome of patients. Similarly, this study was conducted during an interventional out-of-hospital cardiac arrest trial. Survival and LOS may have been affected by this trial as well and may be reflected in costs. Differences in costs for all patients are affected by multiple factors, including payer mix, outlier status, diagnostic-related group codes, and contractual arrangements between the hospital and payer. This number of variables limits our ability to analyze differences of costs between subgroups, such as survivors versus nonsurvivors. Our analysis using patients discharged to hospice as not being survivors could have affected our results and could be managed differently. Although unlikely given our ongoing surveillance systems, we cannot exclude the possibility that the most common type of OOHCA survivor (sudden arrest, single defibrillatory shock, STEMI on echocardiogram) may have inadvertently been classified as not being a cardiac arrest. This sort of patient, with a number of procedures, less morbidity, and shorter LOS, would have affected these data. Finally, despite the duration of the study and the size of the institution, our number of admitted cardiac arrest patients is relatively small. This limits our statistical power to measure differences between groups and to perform a robust subgroup analysis. It also limits the ability to assess the external validity of our observations.

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

In our population, we identify that the cost of caring for patients resuscitated from out-of-hospital cardiac arrest is similar to the cost of caring for STEMI patients. Hospitals that care for these patients do not bear an undue financial burden in providing care, and financial issues should not be a disincentive in providing care to them.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References