Abstract
Objective:
Compare clinical outcomes and costs in a study group of long-term acute care hospital (LTCH) patients with a control group of LTCH-eligible patients in an acute care hospital. LTCHs were created to provide post-acute care services not available at other post-acute settings. This is based on the premise that these patients would otherwise have stayed at acute care hospitals as high-cost outliers. The LTCH hospital is intended to deliver care to patients more efficiently, however, there are little documented clinical and financial data regarding the comparative clinical outcomes and costs for patients.
Methods:
Retrospective medical and billing record review of patients from the following groups: (1) LTCH study comprising patients admitted directly from an acute care hospital to the study LTCH and discharged from the LTCH from September 2004 through August 2006; (2) a control group of LTCH-eligible, medically complex patients treated and discharged from an acute care hospital in FY 2002. The control group was selected from approximately 500 patients who had at least one of the ten most common principle diagnosis DRGs of the study LTCH with >30-day length of stay at the referring hospital and met NALTH admitting guidelines.
Results:
Discharge disposition is an important outcome measure of the quality of care of medically complex patients. The in-hospital mortality rate trended lower and home discharge was 3 times higher for the LTCH study group than for the control group. As a possible result, SNF discharge of LTCH patients was approximately half that of the control group. Both mean patient cost per day and mean total cost per patient were significantly higher in the control group than in the LTCH study group.
Conclusions:
The patients in the LTCH study group had both better clinical outcomes and lower cost of care than the control group.
Introduction
Long term acute care hospitals (LTCHs) were created in the 1980s to address an unmet need of caring for medically complex patients with acute care needs in a post-acute care facility. At the present time, an LTCH can either be a freestanding hospital or a hospital-within-a-hospital (HwH), a separate and distinct entity located physically within another hospital or on its campusCitation1. An LTCH has the following requirements/characteristics: (1) compliance with all accreditations of acute care hospitals; (2) physician-supervised, interdisciplinary medical teams provide inpatient services for chronic diseases or medically complex conditions with an average length of stay ≥25 days; (3) specific admission and patient stay criteriaCitation1,Citation2. LTCHs are considered an integral component of the healthcare system providing a needed piece of the continuum of careCitation3.
LTCHs were created to provide post-acute care services not available at other post-acute settings, e.g. skilled nursing facilities. LTCHs, unlike short-term acute care hospitals, provide a programmatic approach of care similar to the rehabilitation model. This approach emphasizes physicians and staff who have extensive expertise in the care of specific patient populations. These programs require a critical mass of patients in order to efficiently provide care. Examples of patient programs are weaning from prolonged ventilation and nonsurgical management of complex wounds. LTCHs were to foster timely discharge from acute care hospitals leading to reduced Medicare costsCitation2, the assumption being that these patients would otherwise have stayed at acute care hospitals for an extended period of time as high-cost outliers prior to transfer to skilled nursing facilities (SNF)Citation3. However, questions regarding the comparative clinical outcomes and costs for patients in LTCHs have arisenCitation1,Citation4–9 with little documented clinical and financial data. To address those questions Federal Public Law 90–248 Section 402(a) was enacted to empower states to create LTCH demonstration projects. Connecticut General Statute Sec. 1 9a–6 1 7b required the Office of Health Care Access (OHCA), in consultation with the Departments of Public Health (DPH) and Social Services, to authorize LTCH demonstration projects. Although OHCA approved two demonstration projects, only a facility at St. Francis Medical Center (SFHMC) in Hartford, CT was establishedCitation10. The purpose of the demonstration project was to study clinical outcomes and cost effectiveness of patients in an LTCH setting. This report summarizes the results of the study with conclusions on comparative clinical outcomes and costs.
Methods
Setting
The study included patients who were admitted to the LTCH at SFHMC from a variety of area hospitals, however, the majority of patients were admissions from SFHMC. The LTCH at SFHMC is a 28-bed facility under the Hospital for Special Care (HSC) license and has been operational since September 10, 2004. The LTCH used physician-supervised, interdisciplinary medical teams to provide all of the services that are critical to the treatment and possible successful rehabilitation of long-term medically complex patients with acute care needs. These services included a variety of medical specialty services, especially pulmonary, infectious disease and physiatry, as well as physical, occupational, and speech therapy. Attending staff physicians were board certified and consultation services were available in most medical and surgical subspecialties.
Study design and analysis
The primary study was a retrospective medical and billing record review of consecutive patients admitted directly from an acute care hospital to the study LTCH and discharged from the LTCH during a 24-month period from September 2004 through August 2006. The following clinical outcomes were measured: (1) length of stay, (2) mortality, (3) home discharge, and (4) SNF discharge. The following cost outcomes were measured: (1) total cost per patient, (2) cost per patient per day. These data were subjected to the following statistical tests, as appropriate: (1) Z-test for the difference in true proportionsCitation11; (2) the chi-square test for independenceCitation11; (3) the paired t-test for the difference in the meansCitation11; (4) Mann–Whitney (nonparametric) test for the difference in the mediansCitation12; (5) sign (nonparametric) test for the median difference equal to zeroCitation12; (6) Wilcoxon (nonparametric) test for the median difference equal to zeroCitation12.
Control group selection process
All patients discharged from Saint Francis Hospital and Medical Center from fiscal year 2002 were reviewed by an LTCH physician, and approximately 500 patients were identified who had at least one of the ten most common principle diagnosis DRGs that were admitted to the institution the previous year and who had a greater than 30-day length of stay at the referring hospital. Subsequently, the NALTH admitting guidelines were applied to the group and 187 LTCH eligible patients arrived at.
Study group
Patients were excluded if they were not admitted from an acute care facility or had not been discharged by September 2006. Patients in the study group admitted to the LTCH were reviewed for eligibility prior to admission by an experienced nurse evaluator using NALTH admission guidelines. Those patients deemed eligible, based upon NALTH guidelines, were then reviewed by an LTCH physician and a final decision was rendered. The process by which patients with potential for admission to the LTCH was determined is complex and involves multiple variables. The admission process is, in part, guided by program availability, the requirement of a prolonged acute care stay, an expected benefit in 4–6 weeks, a firmly established diagnosis, bed availability, and adherence to NALTH admission guidelines.
The OHCA developed a discharge data layout for both the control group and the LTCH study group; the data included patient demographics, primary diagnosis, utilization of mechanical ventilation, major co-morbidities, resuscitation status, dates of admission and discharge, costs, and discharge disposition. As this was a non-interventional demonstration project developed with, approved by, and conducted in part by the Connecticut State OHCA and DPH and authorized by Connecticut and Federal law, it was determined to be exempt from full IRB review.
Results
Patient characteristics
There were 187 patients in the control group and 196 patients in the study group. The characteristics of the patients are shown in . Both groups had a mean age of greater than 60 years with identical numbers of co-morbidities. Interestingly, a majority of the subjects in the control group were female whereas males comprised a majority of the LTCH study group subjects. The majority of patients in both groups were ventilator-dependent, many had complex wounds and multiple medically complex co-morbidities requiring active treatment. All patients in both groups met NALTH admission guidelines.
Table 1. Patient characteristics.
Clinical outcomes
Three clinical outcomes could be compared between the control and the LTCH study groups, in-house mortality, home discharge and discharge to a SNF. As shown in , the in-hospital mortality rate trended lower for the HSC LTCH study group than for the control group, 13.3 vs. 17.6%; however this was not a statistically significant difference, p = 0.235.
Table 2. Clinical outcomes.
Discharge disposition is an important outcome measure of the quality of care and rehabilitation of medically complex patients. In the LTCH study group, approximately 3 times as many patients were home discharged as in the control group; this difference was highly statistically significant, p = 0.000001 Home discharge comprised one of the following: self-care, home with assistance, or home with visiting nurse. Although a Zubrod performance status scoreCitation13 was not available for the control group, 32% of the LTCH patients had a score of 0–2 at the time of discharge, indicating a patient was fully active, or restricted in strenuous activity, or ambulatory and capable of self-care but not work. The percentage of discharged patients with a good functioning Zubrod score is consistent with the percentage of patients home discharged.
As a possible result of the higher proportion of home discharge, the percentage of LTCH patients discharged to a SNF was approximately half that of the control group, a highly statistically significant difference, p = 0.000001. The remaining approximately 1/3 of the patients in both groups were discharged to a variety of other facilities including a short-term acute care hospital, a hospice, a psychiatric hospital, or the HSC chronic unit.
Comparison of costs
As shown in , there were marked differences in the costs for patient care between the control and LTCH study groups. The costs for the control group were adjusted from the FY 2002 levels using the annual growth rate of the Medicare Cost Report 90th percentile cost per discharge for the intervening 4 years to obtain valid adjusted rates for the control group. The mean cost per day for the control group was $1,520 and $1,154 for the LTCH study group, the difference in the mean costs was highly statistically significant, p ≈ 0. This resulted in a mean total cost per patient over 1.5 times higher in the control group than in the LTCH study group, $59,103 vs. $36,626. The savings achieved by caring for patients in the HSC LTCH were $22,477 in total costs per patient or approximately $366 per day; these savings represented approximately 30% of the costs of the control group.
Table 3. Comparative costs control group care vs. LTCH group care.
Sixteen Connecticut acute care hospitals transferred patients to the LTCH during the 24-month study period; 58% were transferred from SFHMC. In a separate analysis of only these 114 SFHMC patients, the costs for their acute care prior to transfer to the LTCH were compared with the costs for their care in the LTCH. These charge comparisons are shown in . The mean length of stay in acute care was 19.5 days whereas the mean length of stay in the LTCH was 29.1 days. The mean daily charge per patient for LTCH care was approximately 50% lower than for acute care at SFHMC, a highly statistically significant difference, p ≈ 0. The savings generated by transferring each patient from SFHMC to the LTCH was $1,211 per day, a decline from $2,365 to $1,154. Thus, despite the LTCH length of stay being approximately 50% longer, the total cost of LTCH care was approximately 64% of the acute care cost.
Table 4. Comparative costs for acute care and LTCH care of SFMHC patients transferred from SFHMC acute care to LTCH care*.
Discussion
The results of the current study indicate that the patients in the LTCH study group had improved clinical outcomes. The in-hospital mortality rate trended lower than the control group and also significantly lower than the 25.2% mortality rate reported in a 2006 multi-center study published by the National Association of Long Term Hospitals (NALTH)Citation14, the only major published multicenter study. Although the majority of the LTCH patients were ventilator-dependent, the NALTH study patients were exclusively ventilator-dependent, thereby allowing inferences but not direct comparisons to be made between the current study and the NALTH study.
A second important clinical outcome is home discharge. Almost 1/3 of all LTCH study group patients were discharged to home and this was approximately 3 times that of the control group. This was slightly higher than that reported by NALTH for 2006Citation1Citation4. As stated previously, this was consistent with the proportion of LTCH patients with ‘good’ functionality at discharge. ‘Good’ functionality may also be the basis, at least in part, for the 1-year survival rate of 50.9% of the LTCH patients. An additional factor contributing to this higher 1-year survival rate may be the follow-up services provided to patients discharged from this LTCH, particularly support groups both for the people living with disability or illness and for their family members. These support groups provide a structured continuing education setting on disease management facilitated by emotional support and socialization. All of these parameters are highly indicative of improved clinical outcomes for the LTCH patients resulting in a significantly improved quality of life.
As with most, if not all, LTCHs and consistent with their original primary intended use, ventilator-dependent patients represented the largest patient population in our LTCHCitation8,Citation15,Citation16. As such, important clinical outcomes are weaning frequency and time to weaning. Although not measured in the control group in this study, the weaning frequency of the LTCH study group was greater than 56%, slightly higher than that reported by NALTH in 2006. In a previously published study both weaning frequency and time to weaning have been shown to be hallmarks of this study LTCHCitation17.
A secondary anticipated benefit resulting from the creation of LTCHs was to increase the access to acute care beds. Based on acute care hospital discharge data available to OHCA (data not shown), acute care discharges, patient days and intensive care days have been increasing concurrently since the LTCH began operation. However, there are insufficient data to attribute these increases to the addition of LTCH beds to the system in the area.
Comparing the current study group with the historic controlled, potential patient group is not ideal. This study is limited in analysis of the details of care provided to the control group. One can not know, nor predict, the use or impact of high-cost diagnostic modalities in that patient population. Historically and currently, the practice at the study LTCH, rarely uses high-cost diagnostic modalities because one of the admission guidelines is to have a firmly established diagnosis prior to admission.
A significant component to the lower cost of care at the study LTACH is the lack of high-cost centers, i.e. emergency room, delivery or surgical suites or an oncology service. These are well-known high cost centers that impact the high cost of care for all patients in an acute care facility. Additionally, the authors feel that the cost of care in the LTCH may be underappreciated due to the inherent differences in the provisions of care at the LTCH as compared to a short stay acute care hospital. The cost at this LTCH is all inclusive and encompasses physician charges for physician visits as the primary providers of care are employed physicians. Acute care hospitals’ overall costs do not usually include physician costs, thus making the potential cost savings larger than reported.
The data analyzed were mandated by OCHA as part of the requirement of the demonstration of project. The authors believe the analysis, not only provides evidence of the value of this study LTCH, but also of the value of LTACHs in the continuum of care.
It is important to note that improved clinical outcomes also may have indirect impacts on lowering the cost of healthcare. For example, home discharge is likely to have a lower cost to the healthcare system than discharge to a skilled nursing or other rehabilitation facility. Coupled with improved clinical outcomes and their possible indirect cost savings, there were significant direct cost savings achieved by transferring eligible patients to the LTCH. Both the mean cost per patient per day and the mean total cost per patient were significantly lower for the LTCH patients as compared to the control group. Similarly, in the separate paired analysis of acute care and LTCH care costs of FY 2005–2007 SFHMC patients transferred from acute care to LTCH care, the cost of LTCH care was approximately 50% of the acute care daily cost per patient. If SFHMC had cared for each patient for the entire mean 48.6 days of hospitalization at the mean daily rate of $2,365, it would have cost a total of $114,939 or $30,084 more per patient. This has the potential for significant cost savings on a per diem basis offering high complexity care at a lower cost center. The absolute per patient cost for the LTCH group is similar to that reported for all LTCHs in the NALTH 2006 studyCitation17. These direct and possible indirect cost savings have a major added value in the economics of healthcare delivery.
Although LTCHs have been functioning for over 20 years, there has been a need for a comparison of clinical outcomes and costs between patients in LTCHs and those in acute care hospitalsCitation1,Citation4,Citation5–10. This need is particularly acute in light of recent publicityCitation18 and also the federal health reform legislation. As an example, a recently published retrospective cohort studyCitation19 reported that from 1997 to 2006 both LTCH admissions and the cost of LTCH care increased approximately 300% while mortality remained unchanged. However, without a control group for comparison, it is impossible to draw any conclusions on the medical or cost benefits of LTCH care. A prospective study in which LTCH-eligible patients are randomly admitted to either an LTCH or an acute care hospital is not ethically possible.
Conclusions
The results of this LTCH demonstration project show that both improved clinical outcomes and significant cost savings were achieved by the transfer of eligible patients to the LTCH. Extrapolation of these findings suggest that LTCHs, which are the primary venue of care for the medically complex patient requiring extended hospitalization, provide quality medical care with improved clinical outcomes and do so with a significantly lower cost to the healthcare system. Advances in critical care coupled with the aging of the US population are predicted to result in a significant increase in the incidence of patients with chronic diseases and disabilities. As a result, there will likely be an increased demand for LTCH care and LTCHs may be viewed as an integral, if not requisite, component in the healthcare continuum.
Transparency
Declaration of funding:
The work was performed at the Hospital for Special Care and the Connecticut Department of Public Health Office of Healthcare Access with no external funding.
Declaration of financial/other relationships:
None declared.
Acknowledgments:
Statistical analysis provided by Daniel T. Larose, PhD, Professor of Statistics, Central Connecticut State University.
References
- Medicare Payment Advisory Commission. Defining long-term care hospitals. 2004. Available at www.medpac.gov/publications%5Ccongressional_reports%5CJune04_ch5.pdf
- Munoz-Price LS. Long-term acute care hospitals. Clin Infect Dis 2009;49:438-443
- Murer CG. In abeyance: the status of LTACH. Rehab Manag 2008;21:39-41
- Ohio Medicaid Quality Monitor. Understanding post-acute recovery facilities. 2007;8:1-4
- Carson SS, Bach PB, Brzosowski L, et al. Outcomes after long-term acute care. Am J Respir Crit Care Med 1999;159:1568-1573
- Medicare Payment Advisory Commission. Medicare payment policy. Post-acute care providers: an overview of issues. 2007. Available at www.medpac.gov/chapters/Mar07_Ch03.pdf
- Dematte JE, D’Amico D, Donnelly HK, et al. Risk assessment for inpatient survival in the long-term acute care setting after prolonged critical illness. Chest 2003;124:1039-1045
- Nayar P. The impact of Medicare’s prospective payment system on staffing of long-term acute care hospitals: the early evidence. Health Care Manag Rev 2008;33:264-273
- Eskildsen MA. Long-term acute care: a review of the literature. J Am Ger Soc 2007;55:775-779
- Report to the Public Health Committee State of Connecticut. Long Term Acute Care Hospital Demonstration Projects. March 2007. Available at www.ct.gov/ohca/lib/ohca/publications/2007/ltachdemonstrationreport_final.pdf
- Larose DT. Discovering Statistics. New York, NY: W.H. Freeman, 2010
- Weiss, NA. Introductory Statistics, 8th edn. Reading, MA: Addison-Wesley, 2008
- Zubrod CG, Schneiderman M, Frei E, et al. Appraisal of methods for the study of chemotherapy of cancer in man: comparative therapeutic trial of nitrogen mustard and triethylene thiophosphoramide. J Chron Dis 1960;11:7-33
- Scheinhorn DJ, Hassenpflug MS, Votto JJ, et al. Ventilator-dependent survivors of catastrophic illness transferred to 23 long-term care hospitals for weaning from prolonged mechanical ventilation. Chest 2007;131:76-84
- Lundberg JA, Noll ML. The long-term acute hospital: a new option for ventilator-dependent individuals. AACN Clin Issues Crit Care Nurs 1990;1:280-288
- Scheinhorn DJ, Chao, DC, Stearn-Hassenpflug M, et al. Outcomes in post-ICU mechanical ventilation: a therapist-implemented weaning protocol. Chest 2001;l19:236-242
- Scheinhorn DJ, Hassenpflug MS, Votto, JJ, et al. Ventilation outcomes study group. post-icu mechanical ventilation at 23 long-term care hospitals: a multicenter outcomes study. Chest 2007;131:85-93
- Berenson A. Senate panel to investigate deaths at long-term care facilities. New York Times March 8, 2010
- Kahn JM, Benson NM, Appleby D, et al. Long-term acute care hospital utilization after critical illness. JAMA 2010;303:2253-2259