Abstract
Abstract
Background:
Although the provision of hospice emergency kits (HEKs) by home hospice agencies is thought to be widespread, little is known about their use, safety, and impact.
Objective:
This study evaluated HEK medication utilization, safety, diversion, and perceived impact.
Design:
Evaluation consisted of a retrospective patient chart abstraction and an anonymous questionnaire for home hospice nurses.
Setting/subjects:
Chart abstraction examined the computerized records of deceased veterans discharged to home hospice in 2009 (N=76). The questionnaire was completed by 78 home hospice nurses from 16 agencies.
Measurements:
Chart abstraction examined HEK medication utilization, symptoms addressed, and safety/diversion concerns. The hospice nurse questionnaire assessed their perceptions of HEK utilization, safety, and impact, including efficacy for preventing emergency department (ED) visits and hospitalizations.
Results:
Of patients who received an HEK, its use was documented in 50% of cases. The most used items were morphine concentrate and antibiotics. Nurses estimated that the HEK was utilized in 66.3% of cases, with the most frequently used medications being morphine, lorazepam, promethazine, and haloperidol. Fifty-nine percent of nurses felt HEKs were helpful 100% of the time (mean=84.2%; median=100% of the time) and 93% felt that an emergency department (ED) visit or hospitalization was avoided by having the kit in the home. Eighteen percent believed that medications in the kit are used by someone other than the patient.
Conclusions:
HEKs have value as a means to alleviate many symptoms that emerge predictably in home hospice patients and may avoid unwanted hospitalizations and ED visits.
Introduction
Often symptoms arise unexpectedly and if uncontrolled can prompt emergency department (ED) visits and hospitalizations, which may not be consistent with a patient's goals of care and may be avoidable. 3 Recent evidence indicates that 77% of patients who visit the ER during the last month of life are hospitalized, and 68% of these patients die in the inpatient setting. Although surveys consistently report that most Americans would prefer to die at home, 4 most will die in acute care hospitals or nursing homes. This highlights the importance of helping hospice patients and caregivers deal with emergent symptoms in the home. 5
There is growing interest in identifying and assessing programs, practices, and procedures that support home hospice caregivers during the patient's dying process and assist caregivers with symptom control and medication management in the home hospice setting.6–8 One such practice is the placement of hospice emergency kits (HEKs) in the home. These kits go by many names, including “emergency,” “relief,” or “comfort” kits. 9 HEKs were first described in the literature in 1998, when Petrin reported a rudimentary “symptom relief” kit for home hospice use. 10 In 2001, LeGrand and colleagues described a more robust kit to address symptoms common in the hospice population. 11 Although the provision of HEKs by home hospice agencies has been described as widespread, considerable variation exists in kit contents and practice. 9
Some benefits of HEKs have been reported in the literature. In a study of hospice recipients with heart disease, HEKs alone reduced the odds of ED visits by 67%, and the interaction of HEKs with the presence of a caregiver in the home reduced the odds of ED utilization by 75%. 12 A survey of hospice nurse managers reported they believed that medication kits helped patients to avoid ED visits. 9 Hospices in Maryland reported cost savings with HEKs, and the perceptions of hospice managers and clinicians were that HEKs prevented ED visits and increased patient/family satisfaction. 13 However, to our knowledge, no patient-centered outcomes of HEK use have been reported, nor are there patient-centered data to characterize the safety of the kits. Moreover, hospice nurses working in the home have not been surveyed regarding their experiences with HEKs.
With HEKs being provided selectively and used inconsistently by area home hospice agencies, the Birmingham VA Medical Center (BVAMC) developed an HEK that is provided to any enrolled veteran referred to a hospice program. The purpose of this study was to evaluate HEK medication utilization, safety, and diversion, and possible impact using retrospective chart abstraction. In addition, we surveyed hospice nurses regarding perceptions of HEK utilization, safety, and impact, including efficacy for preventing ED visits and hospitalizations.
Methods
Evaluation of the BVAMC HEK consisted of two parts: (1) a retrospective chart abstraction of the electronic medical records of deceased veterans who had been discharged to home hospice and (2) an anonymous questionnaire completed by home hospice nurses.
The BVAMC hospice emergency kit
The BVAMC HEK was designed to manage the most common symptoms that emerge in the last days and hours of life, including pain, dyspnea, confusion, nausea, vomiting, anxiety, edema, infections, and excessive secretions. The specific medications and items in the HEK were chosen based on clinical experience and evidence derived from the available data on drugs prescribed to hospice patients in the last week of life 2 (see Table 1).
Antibiotic may vary depending on current formulary.
HEK, hospice emergency kit.
HEK development was coordinated with the pharmacy, medical staff, and support staff for the Computerized Patient Record System (CPRS). The HEK is ordered through an outpatient pharmacy order set to ensure that all items are ordered and delivered as a group by mail or during hospital discharge. A pharmacy consult is added as a safety check to assure that inappropriate medications are not dispensed and that substitutions are made in cases of allergy or other contraindications.
The medications are dispensed in a sealed bag with an information sheet that (1) itemizes the contents of the bag, (2) describes the symptoms for which the specific medication is indicated, and (3) instructs the patient or caregiver on the proper storage and use of the medications (i.e., place bag in the refrigerator; open and use only at the direction of the home hospice nurse).
Relationship with community hospice agencies
In the absence of its own independent hospice agency, the VA contracts with local hospice agencies to provide hospice care. Through a national network of VA Hospice-Veteran Partnerships (HVP), the VA has established hospice liaisons at each medical facility to coordinate the provision of hospice care in the community. 14 The selection of the private hospice agency is at the discretion of the veteran.
The BVAMC has agreements with local hospices to guide care provided to veterans. Uniform standing orders for veterans' care supersede the agency's standing orders. VA palliative care physicians maintain oversight as the attending physicians of record, but the hospice agencies provide the interdisciplinary care team. VA physicians provide 24-hour call to assist with problems outside the scope of the HEK and for symptoms not controlled with the HEK.
The home hospice agency is informed that the VA palliative care program should be informed or consulted regarding any change in status of hospice patients or initiation of the HEK medications. A designated VA hospice nursing coordinator receives and documents these reports in the veteran's electronic chart. Responses to these calls often involve the VA staff communicating with the veteran or family, as well as the hospice nurse, which would also be documented in the CPRS chart. Medicines related to the hospice diagnosis are provided by the hospice agency, with the exception of the VA HEK.
Chart abstraction
Retrospective chart abstractions were conducted of the CPRS charts of deceased veterans who had been referred to home hospice between January 1, 2009 and December 31, 2009. Using a standardized chart abstraction tool developed by the research team, we reviewed records to determine whether a HEK was dispensed and, if so, to identify any documented use of the HEK. Documented use included reports from the home hospice nurse or other provider related to the use of each of the medications in the HEK, such as requests for refills and any outcomes resulting from HEK use.
Chart abstraction items also included (1) dates of referral and death; (2) diagnosis for referral to home hospice; (3) location of death; (4) use of each of the HEK medications; (5) symptoms addressed by the HEK medications; and (6) notes on diversion, inappropriate use by veteran or family, or patient safety concerns.
Hospice nurse questionnaire
Nurses at 16 community hospice agencies were asked to complete a questionnaire designed to assess their opinions on the home use, hazards, and benefits of HEKs in general, and the BVAMC HEK in particular. The questionnaire included both structured and open-ended questions.
Questionnaires were distributed to the nursing coordinators of the community hospice agencies. Coordinators were asked to distribute the questionnaires to all of their clinical nursing staff. The nurses who chose to complete the questionnaire did so anonymously and returned it by mail using an individual preaddressed, stamped envelope.
Analysis
Chart abstraction items and questionnaire responses were tabulated and reported as counts, percentages, means, and medians. Open-ended questionnaire responses were tabulated and organized by themes.
Results
Chart abstractions
Chart abstractions were completed for 76 deceased veterans. Patients ranged in age from 42 years to 92 years (mean=69 years). Ninety-six percent of patients were male (n=73); 69.7% (n=53) were described as white or caucasian, and 30.1% (n=23) as African American or black. The average time from hospice referral to time of death was 70 days (range 2–957). The largest proportion of deaths occurred in the home or a nursing home (n=43). Other locations of death were the palliative care inpatient unit (n=15), other inpatient hospice (n=3), hospital (n=4), or not charted (n=11).
Chart abstractions indicated that HEKs were distributed to 68 of the 76 patients (89.5%) referred to home hospice. Documented reasons for not dispensing a HEK included patient desire to avoid additional cost, patient having ongoing substance abuse, and patient refusal of hospice services upon returning home.
Of those patients who received a HEK, its use was documented in 50% of cases (34/68; see Table 2). The most used item was morphine concentrate, followed by the antibiotic. Promethazine was the only medication for which there was no documentation of use. In 12 cases, documentation indicated that the kit was used on more than one occasion.
HEK, hospice emergency kit.
Although it was not explicitly stated, in 20% of patients (n=15) there was documentation suggesting that the HEK may have facilitated a more symptom-controlled death at home and may have prevented transfer to a hospital. An example of this was a case in which a patient who lost the ability to take his pain pills two days before death was transitioned to morphine concentrate and died at home. For two patients there was documented concern of inappropriate use of medications. There was no documentation regarding possible diversion.
Hospice nurse questionnaire
The questionnaire was distributed to 160 nurses at the 16 community hospices agencies. Seventy-eight questionnaires were returned for a 49% response rate. The majority of nurses (66%) had more than three years of hospice experience. While working as a home hospice nurse, 90% had cared for a patient with some sort of HEK in the home; 77% had cared for a veteran with a BVAMC HEK in the home.
On average, nurses estimated that at least some of the medication in the HEK was utilized for 67% of patients. Every item in the kit was reported as used. Most frequently reported were morphine, lorazepam, promethazine, and haloperidol (see Table 3). Fifty-nine percent of nurses felt HEKs were helpful 100% of the time (mean=84.2%; median=100% of the time). Eighty-seven percent of nurses reported that they believed HEKs reduced caregiver/patient anxiety, and 93% felt that an ED visit or hospitalization was avoided by having the kit in the home. Eighteen percent of nurses felt that medications were used by someone other than the patient.
M, mean; Md, median.
The survey also included an open-ended section where hospice nurses were given an opportunity to enter comments regarding their perception of and experience with HEKs. Nurses' comments about the HEK were overwhelming positive, focusing on its benefit to patients and family members and its utility to hospice staff (see Table 4). Nurses' comments stressed the importance of HEKs for managing symptoms that emerge in the night, on weekends and holidays, in a crisis situation, or in rural areas where pharmacy assistance is not immediately available. The hospice nurses' comments also revealed that they are not just passive recipients of kits but have knowledge and experience with relevance to the choice of medications for the kit. Two nurses noted their hospice program had stopped providing their own HEKs as a cost saving measure within the capitated hospice system. There were no instances where nurses gave negative comments.
ER, emergency room; HEK, hospice emergency kit.
Discussion
This study is the first to report on documented HEK medication use. Remarkably, 50% of the patients who had an HEK dispensed had documentation of its use in the electronic chart, which is likely an underestimate considering that this measure depended on the nurse reporting its use to the VAMC. For many patients the HEK was reported as being used on separate dates, many of them months apart. Most home hospice nursing calls to report HEK use were also calls requesting refills on HEK items.
The most commonly used medication as reported to the VAMC was the morphine concentrate, not surprising given the frequency of pain and dyspnea at the end-of-life. This medication can be used sublingually when patients lose the ability to swallow as their condition declines. The National Home and Hospice Care Survey, most recently in 2007, similarly reports that a “narcotic analgesic” is the most commonly prescribed medication in the last week of life. 2
Interestingly, the second most commonly reported medication used from the HEK was the antibiotic. Although we were not aware of any other kits containing an antibiotic, we included a flouroquinolone in the HEK, because infection or infectious symptoms are common at end-of-life. It was our clinical experience that coming to the hospital or ED for antibiotic therapy was a common request from patients and family, who understandably would not want to forgo antibiotics if they could improve both quality and quantity of life. Some providers may not view infection as an emergency to be treated at home and may hold to the opinion that antibiotics should be ordered only as needed. However, our experience is that patients and family do consider infectious symptoms to be emergent and that prompt treatment is indicated and enabled by the availability of an antibiotic in the HEK.
The results of this study demonstrate that home hospice nurses are familiar with HEKs, have used them, and find them helpful the majority of the time. Over 90% of home hospice nurses believed that the presence of a HEK could avert an ED visit and/or prevent a hospital admission, presumably because symptoms could be adequately treated at home. While it is not within the scope of this study to determine if the hospice nurses' assessments of HEK efficacy were an accurate representation of patient and caregiver experiences, these findings point to a hospice culture that is positively disposed toward the use of HEKs.
Notwithstanding the positive attitudes of hospice nurses toward the HEK, it is important to insure that the symptoms that led to the consideration of transfer out of the home were, in fact, considered adequately controlled by use of the HEK. In 20% of cases there was documentation in the medical record indicating that the HEK facilitated an at-home death and averted an unwanted transfer at the end of life. If true, this represents an improvement in quality of care, as well as a potential cost savings to the health care system.
For patient safety reasons and liability, family members are not allowed to assist with medication administration and most patient care in the hospital setting. Ironically these very same patients, when discharged to home with home hospice, must rely on caregivers for medications and other treatments.
Lau and colleagues explored the notion of medication management among family caregivers 15 and in subsequent research found considerable variation in caregiver comfort in execution of home medication management for hospice patients, although use of HEKs were not addressed specifically. 16 It is possible that administering medications such as morphine in a much less controlled environment could generate anxiety and stress.17,18 Alternatively, caregivers may feel much more comfortable when they are empowered to control distressing symptoms common during the dying process. Nurses in this study reported that the presence of the HEK reduced the frequency or severity of patient/family anxiety.
Concerns related to HEKs are the potential use of the kit by the family without supervision of the hospice nurse and the potential for use of the kit by the nurse without adequate physician oversight. This study suggests that unsupervised use by caregivers may occur in a small number of cases. Home hospice nurses are likely to recommend and use medication from the kit without physician supervision. They may report use of the medication after the fact, or ask for assistance if symptom control is not achieved. Administering similar medications would be monitored much more closely in a hospital or nursing home setting. However, for a skilled and experienced home hospice nurse, such practices may be safe in most cases. Diversion of medication was reported, but was not common. Since home hospice nurses routinely monitor home medications as part of their visit in the home, the misappropriation of HEK drugs is not likely to go unnoticed.
Limitations
The limitations of this study include its retrospective nature and its reliance on the recall and perceptions of hospice nurses. Thus, patient/family outcomes and perceptions of the benefits or problems associated with HEK utilization were not measured directly. There is the possibility that nurses may have overestimated the benefit of the kit for the patient, because they like that the kit makes their job easier by reducing doctor phone calls, medication pickup, and additional home visits. The high rate of nurse endorsement of the kits may also be partially attributable to a response bias, such that those nurses with favorable opinions of the HEK were more likely to respond than those who were less pleased with the HEK. Chart abstraction data were based on second-hand and possibly incomplete information, which may have resulted in underreporting. Finally, there was no control group, thus we cannot determine what the true rate of ED visits and hospitalizations would have been in the absence of having a HEK in the home.
The authors are conducting a prospective study to evaluate the BVAMC HEK more directly by enrolling hospice patients and caregivers when the HEK is dispensed and monitoring its use over time. This will allow us to capture more fully the use of the HEK, determine its effectiveness for symptom control as a patient-centered outcome, and better evaluate concerns related to safety, supervision, complications, diversion, and caregiver stress.
Conclusions
In conclusion, HEKs appear to have value as a means for timely alleviation of the many symptoms that emerge predictably in the home hospice patient in the last few weeks, days, or hours of life. Further research is needed to examine the impact of HEKs on symptom control and caregiver outcomes, as well as the cost-effectiveness of placing an HEK in the home.
Footnotes
Acknowledgments
This researh was supported by the Birmingham/Atlanta Geriatric Research, Education, and Clinical Center (GRECC). The authors thank Donna Hix, RN, the Hospice and Palliative Care Coordinator for the BVAMC, and Shanette Granstaff, MPH, for analytic support.
Author Disclosure Statement
No competing financial interests exist.