Hospice and Palliative Care: The Last Year in Review

INTRODUCTION

This article will review several recently published articles in the field of hospice and palliative care that may have practical application to the care of residents in long-term care facilities (LTCFs). Given that the one-year mortality of residents in nursing homes is 25% and that 66% of residents in 1999 remained in the nursing facility to die rather than being transferred to the hospital for the final days of their life,^1,2 a paradigm shift to comfort care that encompasses excellence in palliative and hospice care in LTCFs is critical to help residents to live each day as fully as possible until they die. Also, 20-25% of Americans ages 65 and older die in nursing homes while about 50% die in hospitals.³ Accordingly, physicians, mid-level practitioners, and other members of the LTCF health care team need to acquire attitudes, skills, and competencies that enable them to provide excellent end-of-life (EOL) care, whether residents have a terminal condition or not. Noteworthy articles were abstracted from a review of the following journals: Journal of the American Medical Association, Journal of the American Geriatrics Society, Journal of the American Medical Directors Association, Journal of Palliative Medicine, American Journal of Hospice and Palliative Care, and American Journal of Medicine. The World Health Organization’s concept of palliative care also guided this article’s focus.⁴ Palliative care has been defined as “…the active, total care of patients whose disease no longer responds to curative treatment. Control of pain, of other distressful symptoms, and of psychological, social, and spiritual problems is paramount. The goal of palliative care is the achievement of the best possible quality of life for patients and their families….”⁴

HOSPICE CARE: HIGHLIGHTS DURING THE LAST YEAR

Excessive Short Hospice Stays
A retrospective descriptive study by Miller et al⁵ on 46,655 nursing home residents enrolled in hospice reported that short hospice stays (7 days or less) had increased from 26% in 1994 to 34% in 1999, while the median stay had decreased from 26 days to 15 days and long stay (greater than 180 days) decreased from 17% to 10%. A short hospice stay was especially associated with admission to the long-term care facility (LTCF) from the hospital, primarily for the diagnoses of congestive heart failure, renal failure, or stroke in residents age 85 and older. With shorter stays on hospice (ie, when a late referral to hospice occurs), the Medicare savings may not be as great now as when the Medicare Hospice Benefit (MHB) became effective in 1983, which had been based on data from the National Hospice Study done in the late 1970s. The MHB was extended to residents of nursing homes in 1985. This study highlights that LTCFs need: (1) to be better prepared to effectively care for residents discharged from hospital and concomitantly referred to hospice upon admission to their facility who may have only days to 1 or 2 weeks to live, and (2) to better identify those residents with advanced disease in the terminal phase of life and to expeditiously refer these residents to hospice. For hospice care to be most effective, patients probably need to be on service at least several weeks or months. Physicians also need to know the general and disease-specific guidelines for hospice eligibility.

Hospice Usage by Minorities
A study by Greiner et al⁶ examined racial/ethnic variations in rates of hospice utilization by 11,291 deceased individuals who died of nontraumatic causes. Hospice use by Caucasians was 8.6% while that for African Americans was 5.5% (P < 0.001, 40% lower use). Economic and educational level and access (nonaccess) to care were not found to be determinants for this difference in hospice use. However, African Americans who had a living will had increased hospice use. The authors concluded that unrecognized systemic and individual stereotyping may be factors that limit referral to hospice services. Thus, it is essential to ask how these findings may potentially impact access to palliative and hospice care by African-American residents in LTCFs. Perhaps there is a need to more effectively communicate and explain the goals and philosophy of palliative/ hospice care in the LTC setting to minority residents and their families, which in turn will enable more informed choices relative to end-of-life care. This is vital given that many LTC facilities probably have minority populations that comprise the majority of residents in those facilities.

Patterns of Functional Decline at End of Life
A study by Lunney et al⁷ identified four patterns of functional decline in a community-based cohort age 65 and older that occurred during the cohort’s last year of life before death. This included sudden death (15%), terminal cancer (21%), organ failure in particular chronic obstructive pulmonary disease (COPD) and congestive heart failure (CHF) (20%), and frailty, essentially adult failure to thrive (20%). A fifth group classified as other (24%) demonstrated modest functional decline during the last year of life, 40% of whom had a primary diagnosis of ischemic heart disease. Only cancer decedents consistently demonstrated life expectancy of 3 months or less once decline in activities of daily living (ADLs) began. The authors concluded that it is difficult to determine—based on functional decline in ADLs—when to refer people to hospice who are manifesting progressive organ failure or frailty due to noncancer diagnoses. This study highlights the dilemma of identifying residents in LTCFs (many of whom have a pattern of slow functional decline) who may benefit from a more palliative philosophy of care, whether they have a terminal diagnosis or not. This study did show that individuals with a diagnosis of cancer, heart disease, stroke, diabetes, or hip fracture were found to be at higher risk of dying over the next 2 to 3 years. Perhaps such residents should be targeted for a more palliative type of care if this is congruent with the residents’ and their families’ philosophy and goals of care.

Residents Identified as at End of Life
Porock et al⁸ reported a descriptive, retrospective cohort study on all residents admitted to nonhospital-based LTCFs with hospice contracts in Missouri in 1999. On admission, 4.5% of residents met EOL criteria; one-half of whom were either on or started on hospice at the time of admission. Those residents defined as at EOL either had serious unstable health conditions or endstage disease. The EOL cohort had a median survival of 33 days (while only 5% of non-EOL residents had died by 33 days), while at 6 months, only 17% of EOL residents remained alive in the facility (only a few had been transferred to other facilities). In addition, 76% of EOL residents had died in the facility, in contrast to only 20% of non-EOL residents. The authors concluded that there is a vital need to prepare LTCF staff and physicians in order to provide quality palliative care to residents identified as at EOL. The study also showed that 66% of residents identified as at EOL had been admitted from the hospital and were more likely to have been diagnosed with terminal cancer, emphysema, or renal failure and to have had recent/ongoing IV therapy, tube feedings, chemotherapy, radiation therapy, or hemodialysis. Thus, it is important for physicians to carefully review residents’ advance health care directives soon upon admission to LTCFs in order to negotiate appropriate goals of care and to avoid treatments or repeated hospitalizations that would no longer provide benefit nor add quality life to residents at the end of life.

Needs at End of Life for Residents of Nursing Homes
Reynolds et al9 reported on the needs of residents dying in two nursing homes during a one-year period that was based on after-death interviews with facility nurses and nurse aides and family members, all completed within 3 months of the residents’ deaths. Sixty percent of deceased residents had resided in the nursing home less than one year, and 90% had died in the nursing home. “A good death” was reported by 58% of respondents. The need to improve communication on physician treatment decisions was cited by 36% of family members and 25% of facility staff. In 37% of deaths, family members could not identify the doctor in charge of their relative’s care. During the last 3 months of life, common physical needs in residents included pain control (86%), cleanliness (81%), relief of dyspnea (75%), incontinence care (59%), and amelioration of fatigue (52%). Emotional symptoms included depressed mood (44%), anxiety or agitation (31%), and loneliness (21%). Unmet needs included lack of emotional and spiritual support (30% of decedents), more care for personal cleanliness (23%), and better treatment of pain (19%). This study should help guide physicians and LTCF staff as to what distressful symptoms and needs must be addressed and better palliated for residents in nursing homes.

Hospice Eligibility for Patients with Dementia
Schonwetter et al¹⁰ studied the validity of the current Medicare hospice eligibility guidelines for dementia. Current hospice eligibility guidelines require a minimum Fast Stage 7C (Functional Assessment Staging Scale), which first and foremost must include the inability to ambulate without personal assistance due to the late stage of their dementia, as well as the inability to dress, urinary/fecal incontinence, speech ability that is limited to the use of a single intelligible word in an average day, and at least one of the following conditions in the preceding 12 months: aspiration pneumonia, upper UTI, septicemia, multiple stage 3 to 4 decubitus ulcers, recurrent fever despite antibiotics, and inability to maintain sufficient fluid and calorie intake as evidenced by either a 10% weight loss during the previous 6 months or a serum albumin less than 2.5 gm/dL. A retrospective chart review of 165 persons on hospice with the diagnosis of end-stage dementia found that 75% had died within 6 months of admission to hospice (25% died after 6 months), with a range in survival of up to 2 years. The authors acknowledged that a major problem in predicting death within 6 months was the uncertainty in the occurrence of acute and life-threatening medical conditions. Given these findings and the current Medicare hospice benefit eligibility guidelines for dementia, which require that a physician determine that a patient’s prognosis is 6 months or less if the dementia follows its natural course, the authors advocate that the MHB for dementia should be extended to patients who have a life expectancy of 12 months or less. In addition, the authors propose a palliative care benefit for those patients in late-stage dementia who do not meet the current MHB criteria.

In current physician practice, it is imperative to consider other factors and comorbities (physical, social, emotional and spiritual) when prognosticating life expectancy of 6 months or less in patients with dementia (often termed a “constellation of conditions”). In this study, anorexia and cachexia were significantly related to decreased survival (P<0.001) while the following individual predictors were not: age, dysphagia, dyspnea, presence of advance directives, and comorbid conditions. It should be noted that patients with late-stage dementia may more likely be eligible for hospice services under the diagnosis of adult failure to thrive that requires a low level of physical function (equal to or less than 40% on the Palliative Performance Scale (ie, the resident is mainly in bed and requires assistance with all ADLs), a body mass index below 22 kg/m2, and the patient is not responding to enteral feedings or declines parenteral feeding.

PALLIATIVE CARE: HIGHLIGHTS DURING THE LAST YEAR

Tube Feeding vs Hand Feeding
Mitchell et al¹¹ have reported a cost comparison between 11 tube-fed and 11 hand-fed LTCF residents with advanced dementia. This was a retrospective cohort study conducted at a 700-bed LTCF. Study residents must have lived more than 6 months as a condition to obtaining accurate costs of care in each group. This study showed significant dollar cost shifting from Medicaid to Medicare ($5200 vs $9400, respectively, P < 0.04) in those residents who were tube-fed, mainly due to higher costs associated with feeding tube replacement, emergency room visits, hospitalizations and physician services. Higher Medicaid costs were incurred for hand-fed than for tube-fed residents ($4200 vs $2400, respectively, P < 0.006). The author challenges the LTC providers to self-reflect on whether potential financial incentives for nursing facilities (ie, less staff time and higher Medicaid reimbursement for residents who are tube-fed) encourages the inappropriate placement of feeding tubes. Recent studies have shown that the use of feeding tubes in advanced dementia does not affect survival, and may in fact potentially decrease it.¹² Accordingly, physicians who care for residents with advanced dementia in nursing facilities must carefully evaluate the appropriateness of feeding tube placement when these residents begin to manifest weight loss; they must also understand their own personal beliefs and attitudes with regard to the use of artificially administered fluids and nutrition. Another study has shown that the odds ratio for the placement of feeding tubes in residents with advanced cognitive impairment is twice as likely to occur in nursing facilities that have a speech therapist for 35 hours or more per week.¹³

ACE Inhibitor Use in Congestive Heart Failure
Hutt et al¹⁴ assessed the association between outcomes of rehospitalization and mortality from CHF, severity of CHF exacerbation, comorbidity, and the process of care in residents of Medicare-certified skilled nursing facilities (SNFs). This was a retrospective study in 58 SNFs in 5 states, which entailed 156 residents who developed acute CHF among 4693 residents within the first 90 days of admission to the SNF. The 156 residents were 70% female, 90% Caucasian, average age 83, 34% previously residents of a nursing facility, and had had an average hospital stay of 20 days prior to discharge to SNF care. Two-thirds of these patients had a do-not-resuscitate (DNR) order in place, and almost 60% had required transfer back to the emergency room or hospital. Significantly, the odds ratio for mortality at 60 days was 0.31 for those CHF patients who were on angiotensin-converting enzyme (ACE) inhibitors. Accordingly, physicians should carefully evaluate each SNF resident with CHF as to the benefits/burdens of either starting or continuing an ACE inhibitor.

Carvedilol Use in Congestive Heart Failure
Krum et al¹⁵ have reported the results from the COPERNICUS study that has shown the benefit of carvedilol in decreasing death and hospitalization rates to have occurred as soon as 8 weeks after initiation of treatment with carvedilol in patients with an ejection fraction less than 25%. The starting dose of carvedilol was 3.125 mg bid and was uptitrated to 25 mg bid by 6 weeks if tolerated (60% of patients reached this target dose). This study challenges physicians to increase the use of beta-blockers (specifically carvedilol) in patients with late-stage and end-stage heart failure. Residents with CHF in long-term care facilities may benefit from use of beta-blockers, given that the increase in benefit can be seen within 2 -3 months of initiation. However, as with any medication or intervention considered in LTC residents, all individual risks and benefits must be carefully considered before deciding upon their use. This study reminds us all that the judicious use of carvedilol in select patients with CHF may reduce both premature deaths and transfers to the hospital, and improve the quality of life for these residents, all of which is within the tenets of good quality palliative and hospice care.

Vertebroplasty for Acute Osteoporotic Vertebral Fractures
Diamond et al¹⁶ performed a nonrandomized prospective cohort study in all patients who presented to the emergency room or were directly admitted to hospital with an acute osteoporotic vertebral fracture (50% had thoracic fractures and 50% lumbar). The control group consisted of those patients who chose conservative treatment while the intervention group consisted of those patients who agreed to percutaneous vertebroplasty. Both groups were considered for and/or received oral opioids, COX-2 inhibitors, hot packs, calcium supplementation, biphosphonates, and early mobilization. The mean length of hospital stay was 40% less in those patients who were treated with vertebroplasty (9 ± 6 days) than those treated conservatively (15 ± 6 days). Vitamin D deficiency was found in 60-80% of both groups (blood 25–OH vitamin D levels less than 30 nmol/L). At 24 hours after vertebroplasty there was a 53% reduction in pain score, a 29% improvement in ADL score, and 24% of patients were able to cease all analgesics, as compared to no improvement in the control group. However, when both groups were followed longitudinally, their clinical outcomes were similar at both 6 weeks and between 6 and 12 months. Thus, physicians should consider vertebroplasty in select high-functioning residents of LTC facilities in an attempt to provide rapid analgesia and/or maintain their high level of physical functioning. Benefits may occur up to 1-2 weeks (possibly up to 2-4 weeks) after fracture occurrence in residents who either are not responding to conservative treatment or have intolerable side effects to prescribed analgesics.

Hospice Care in Assisted Living and Residential Care
This study by Sloane et al¹⁷ reported that families were more satisfied with end-of-life care in assisted living (AL) care and residential care (RC) than that provided in nursing homes. Families in AL/RC settings perceived better staff answers to their questions (94% vs 80%), quicker symptom treatment (87% vs 70%), better family inclusion in making care decisions (98% vs 86%), and more thorough physician assessment (90% vs 70%). This study has important ramifications as more residents in AL/RC settings are remaining in these facilities to receive EOL and hospice care rather than transitioning to nursing facilities. In this study, 27% of residents in AL/RC settings received hospice care during their last month of life as compared to 16% in nursing home residents. The prevalence of moderate-to-advanced dementia was 67% in nursing home residents and 39% in AL/RC facilities. Thus, physicians who provide care to residents of AL and RC facilities must become proficient in providing good EOL care and hospice care. Physicians and staff in nursing homes must improve family communication, be more inclusive in making decisions, and provide more timely treatment to palliate distressful symptoms in residents.

Incidence of DVT with Use of Megestrol
Kropsky et al¹⁸ determined the incidence of deep vein thrombosis (DVT) in users of megestrol by performing a retrospective chart review over a 3-year period in a 816-bed nursing home. Those taking megestrol numbered 246, had an average age of 87, and had received 400 mg of megestrol per day. There was a 4.9% incidence of DVT in megestrol users, and they had been taking it for an average of 183 days. This is in contrast to a 0.8% base incidence of DVT reported in nursing home residents and a 0.1% incidence in community-based populations. Unfortunately, this study did not report the incidence of DVT in residents who did not receive megestrol acetate. At this time, it is probably safest to limit megestrol acetate as an appetite stimulant in nursing home residents to no more than 4-8 weeks use at a dosage of 400-800 mg a day. Other appetite stimulant medications such as steroids (prednisone or dexamethasone) or mirtazapine are alternatives. Controversy exists as to the efficacy of megestrol acetate as an appetite stimulant, as many clinicians believe that the weight gain seen with megestrol may be due to either increased body fat or water retention, with no effect on reversing the protein catabolism that commonly occurs in the advanced stages of disease.

Olanzapine for Intractable Nausea
Jackson and Tavernier¹⁹ reported several case studies on the use of olanzapine to control nausea and vomiting in patients who had had multiple pathophysiologic mechanisms as a cause to their nausea and vomiting, which were resistant to treatment with traditional antiemetics. Patient diagnoses included liver cancer with brain metastases, colorectal cancer, stroke, COPD, multiple myeloma, and Alzheimer’s disease with gastroparesis and theophylline toxicity. Patients required careful dose titration of olanzapine over 7-10 days in the range of 2.5 mg to 7.5 mg administered at bedtime. The authors conclude that olanza- pine offers a safer alternative as an antiemetic than the traditional butyrophenones and phenothiazines, which are more likely to cause QT interval prolongation and extrapyramidal side effects. Although the cost of olanzapine is considerable (a 30-day supply of 2.5 mg tablets is $150; 5-mg tablets, $180; and 7.5-mg tablets, $195), it may have a significant therapeutic benefit in a variety of conditions associated with nausea and vomiting as opposed to other traditional pharmacologic treatments that have failed (eg, dexa- methasone, metoclopramide, lorazepam, haloperidol).

Evidence-Based Prescribing at End of Life
Bain et al²⁰ performed a restrospective analysis in 4752 hospice patients who received either temazepam or zolpidem over a 6-month period for the treatment of insomnia. After a review of the literature and the prescribing patterns for each drug in their hospice, the investigators concluded there was no evidence to support zolpidem as superior in efficacy to temazepam for the treatment of insomnia, although zolpidem is much more costly. Thus, they recommended temazepam as a first-line drug for the treatment of insomnia and zolpidem as a second-line drug. This study alludes to the use of Medication Use Guidelines (MUGsTM) in hospice as a means to better control medication costs under a fixed Medicare hospice daily rate that has not increased in several years. Physicians should be aware that more and more hospices are developing medication formularies (some of which may be disease-specific rather than applied to all hospice diagnoses) that are based on both drug cost (savings) and medication efficacy reported in the literature.

SUMMARY

This article has reviewed several studies reported in the literature during the last year that may potentially have a significant impact on physicians’ clinical practice of palliative and hospice care in various long-term care settings such as assisted living, residential care, and skilled and intermediate nursing facilities. Physicians, nurses, other practitioners, and facility staff must develop unbiased beliefs and attitudes regarding end-of-life care and effectively determine appropriate treatment options based on risks and benefits. They must communicate these options to residents and residents’ families and support residents and their families in their choices for care (including that of no treatment), with the ultimate goal of effectively palliating distressful pain and non-pain symptoms that occur in residents with progressive late-stage disease, whether the condition is terminal or not. This article has strived to sensitize clinicians to the interface between LTC settings and hospice and palliative care, and to relevant issues/factors related to the provision of that care. This must all take place in the context of improving resident comfort and dignity so that residents can live every day to the fullest and experience a peaceful death free of pain and suffering.