Know Your Long-term Care Hospital FREE TO VIEW

Chronic critical illness is a significant component of every ICU clinician’s day. Although accounting for only 10% of all ICU admissions, chronically critically ill patients account for a large number of ICU-bed days and consume up to 40% of ICU resources.¹ Much of these resources are consumed beyond the seventh day of mechanical ventilation. Since 1983, the considerable burden of chronic critical illness on acute hospital ICUs has been relieved somewhat by the rapid growth of long-term care hospitals (LTCHs). These institutions manage a range of long-term care conditions such as complex wound care, but many of them admit patients requiring prolonged mechanical ventilation (PMV) after acute illness. LTCHs focus on successful weaning of patients requiring PMV by taking advantage of personnel who are trained and experienced in managing this unique patient population.² Ideally, patients undergo a rehabilitation-based approach to care that encompasses multidisciplinary input from expert nurses, respiratory therapists, nutrition specialists, physical, speech and occupational therapists, and experienced discharge planners.²³ Some centers also include psychiatrists, clinical psychologists, and palliative care specialists on the care team to help patients and families with their often complex emotional needs.

While transferring stable chronically critically ill patients from the acute ICU setting to the rehabilitation-based environment of an LTCH can potentially have important therapeutic benefits, transferring such patients can certainly have financial benefits for the acute hospital. Reducing length of stay of chronically critically ill patients will reduce reimbursement deficits that occur when the costs of their acute ICU care exceed their diagnosis related group (DRG) and outlier payments. Additionally, the ICU beds that would otherwise be utilized for weeks or months could be recovered for use by patients in more urgent need of ICU care, thus improving patient flow from emergency departments, operating rooms, and acute hospital wards. An analysis by the Medicare Payment Advisory Commission⁴ confirmed that costs for total hospitalization (acute hospital and, when relevant, LTCH) are lower for PMV patients managed in regions with access to LTCHs compared to similar patients managed in regions without access to LTCHs. These cost savings are most likely achieved by lower intensity nursing in the LTCH setting compared to the acute hospital ICU, and possibly through more efficient movement through the rehabilitation process. Under the initial Tax Equity and Fiscal Responsibility Act of 1982-based system of reimbursement and the current DRG-based system, the LTAC industry has remained a rather profitable enterprise. That, combined with a large and willing referral base, has fueled explosive growth of the industry. The number of LTCHs has grown 12% per annum since 1992, with more recent growth advancing at an even higher rate.⁴

The existence of LTAC hospitals has presented important opportunities to study outcomes of chronically critically patients and define their unique medical needs. However, the proliferation of LTAC hospitals has occurred with minimal standardization of structure or function. Individual institutions have developed according to different community needs or resources, varied interpretations of reimbursement policies, and differing business models. Therefore outcomes studies based within individual hospitals or hospital systems have produced quite varied results. For example, hospital mortality reported from single-center studies ranges from 10 to 50%.⁵ While approaches to care could explain part of these differences in outcome, patient selection and thresholds for discharge back to acute care hospitals when conditions worsen are more likely explanations. Recognizing that selection bias compromises studies of outcome in patients requiring PMV, a consensus conference⁶ published guidelines on standard definitions for the following: (1) patient selection: patients requiring > 21 days of mechanical ventilation without interruption of > 2 days; (2) successful weaning: remaining free of mechanical ventilation for > 7 consecutive days; and (3) relevant primary outcome: 1-year mortality.

Two multicenter prospective cohort studies⁷⁸ published in this edition of CHEST (see pages 76 and 85, respectively) by members of the National Association of Long Term Hospitals (NALTH) provide a wealth of information on the types of chronically critically ill patients managed by LTCHs and typical outcomes. By including consecutive patients admitted to centers that are linked only by a professional organization, the cohort provides a useful cross-section of LTCHs across the United States, thereby overcoming difficulties associated with comparing multiple single-center studies. While their population is defined by LTCH admission rather than number of days of mechanical ventilation, they included 1-year mortality as an outcome to account for differences in discharge practices among the different hospitals. They defined successful weaning as discharged alive and free of mechanical ventilation, a more stringent definition than recommended in the guidelines.

Their first study⁷ confirms that LTAC hospitals are diverse in their admission practices and approaches to care. Of the 19 hospitals that described facilities, 5 had ICU beds, Registered nurse staffing ranged from 3 to 10 patients per nurse, and respiratory therapist staffing ranged from 2 to 20 therapists per patient. The number of patients admitted to individual facilities ranged from 1 to 50% of total admissions. Consequently, outcomes reported in the second article⁸ are as diverse. Weaning success ranged from 42 to 83%, and LTAC hospital mortality ranged from 0 to 47% of admissions (25% overall). One-year mortality was 63%, which is within the range of other single-center studies of LTCH outcomes.⁵ Nearly 60% of patients were ambulatory after 1 year, although functional data were not available for 30% of patients known to be alive. Another recent study⁹ indicates that up to 68% of PMV patients have severe cognitive deficits 6 months after hospital discharge. This problem interferes with assessments of quality of life in longitudinal studies of PMV patients using standard instruments. Overly optimistic estimates of quality of life for a cohort will result if severe functional or cognitive limitations interfere with access to or completion of interviews.

These studies from the NALTH group should alert clinicians to the range of services provided to PMV patients by LTCHs. Clinicians should also be aware of the diversity in practices and outcomes among LTCHs and should familiarize themselves with the centers that are in their referral area. They should be learn what services are offered and whether the LTCH is suited to their patients’ needs. They should maintain high expectations for care that matches or exceeds the quality of PMV care in their own institution before arranging transfers.

One important question that arises from these results is whether all of the variation in outcomes can be attributed to selection criteria and transfer thresholds. Do variations in processes of care impact outcomes of these patients? The organization and practice variations evident in this cohort indicate that there is ample opportunity to determine optimum levels of staffing ratios and patient volume to provide best outcomes. Weaning protocols should be optimized and tested. Infectious complications such as ventilator-associated pneumonia, catheter-related urinary tract infections, and blood stream infections are quite common and contribute to morbidity, resource use, and death, so quality improvement initiatives directed at these issues are certainly indicated. The NALTH study was not designed to draw associations between variations in management and outcomes, but it provides a model of how multiple centers can pool resources (financial and intellectual) to design and implement cohort studies and clinical trials. The Center for Medicare and Medicaid Services is well aware of the cost burdens of chronically critically ill patients and how quality initiatives to standardize best practices across institutions would improve cost-effectiveness.⁴ Some significant commitment of resources in this direction would be wise.

There is great interest in the outcomes of the chronically critically ill patients since patients and caregivers need to know whether the heavy burden of illness and intervention that they are enduring is going to lead to an acceptable result. Is this cohort study the final word on outcomes in PMV? It clearly sets a standard for the LTCH population. But it must be kept in mind that the LTCH PMV population is a highly selected group. The advanced age of the population selects for worse long-term outcomes. However the current prospective payment system governing reimbursement of LTCH care discourages admissions of patients who are unlikely to wean from mechanical ventilation or who are likely to die in short periods. Most LTCHs are ill-equipped to manage ventilator-dependent patients who have hemodynamic instability or even renal failure. These factors bias LTCH cohorts toward more favorable outcomes. A few studies following patients requiring prolonged ventilation in the acute hospital setting have been published, but results are difficult to compare due to differences in enrollment criteria.¹⁰ Despite the growth of LTCHs, the large majority of patients with PMV are managed in the acute hospital setting due to limitations of access to post-acute care facilities. Therefore the acute ICU remains an important setting for studies of these patients, preferably using recommended standard definitions and outcomes.⁶ Cost-effectiveness, mortality and functional outcome prediction, and processes of care are important topics of interest in all settings. Chronically critically ill patients are ever present and growing in number.¹¹ Clinicians, investigators, and payors should remain invested in improving their care.