By the year 2020, 2.5 million individuals older than age 65 will die annually in the United States, and approximately 40% of these deaths will occur in nursing homes (Brock Sc Foley, 1998; Lynn, 1998). Currently, 20% of all deaths occur in nursing homes (National Center for Health Statistics [NCHS], 1996), and 34% of all adults older than age 85 die in nursing homes (Riley, Lubitz, Prihoda, & Rabey, 1987). The majority (65%) of nursing home deaths are women (NCHS, 1996). One third to one half of nursing home residents die within 12 months of admission (Keay, Fredman, Taler, Datta, & Levenson, 1994).
Death is part of the nursing home culture, yet little research has evaluated the quality of nursing home care at the end of life (Buckingham, 1983; Keay et al., 1994; Lynn, Teño, et al., 1997). By virtue of their demographics and numerous comorbid conditions, older adults in nursing homes are frail, vulnerable, and socially isolated individuals. This supports the need to evaluate end-of-life care for this vulnerable population.
Although death is one of life's unequivocal outcomes, few researchers have described or evaluated endof-life care (Lynn, Teño et al., 1997). Thus far, research on dying (e.g., symptom management, satisfaction with care, quality of care, quality of life) has focused primarily on those with cancer diagnoses or on seriously ill individuals in hospital settings (American Medical Association Council on Scientific Affairs [AMA], 1996; Cleeland, Gonin, Hatfield, Edmonson, Blum & Stewart, 1994; Garvin & Chapman, 1995; Institute of Medicine [IOM], 1997; Lynn, Harrell, Cohn, Wagner, & Connors, 1997; SUPPORT Principal Investigators, 1995). Little is known about the needs and outcomes of care for those dying from chronic illnesses (e.g., congestive heart failure) that have more variable disease trajectories.
The lack of research on dying, especially for those with chronic illnesses, leaves a vast array of unanswered questions for clinicians. In particular, reliable information describing end-of-life care in nursing homes, where chronically ill, functionally impaired older adults are the majority of the nursing home population, is largely unavailable (Engle, 1998; IOM, 1997; Keay et al., 1994).
The purpose of this study is to describe the end of life in a nursing home from the perspective of residents who are chronically ill and declining, their family caregivers, and staff. This study addressed the following research questions:
* How do residents who are declining, family caregivers, and nursing home staff describe end of life in a nursing home?
* How do staff describe, recognize, and respond to older adults who transition from chronically ill to terminally ill?
* What contextual factors (e.g., nursing home regulations, staffing ratios, physician influence) facilitate or impede end-of-life care?
The growing number of older adults living with multiple chronic illnesses coupled with changes in the delivery of acute care (i.e., shortened hospital stays) is changing the complexion of the nursing home population. The following section will review the changing nursing home population, the difficulty identifying who is terminally ill, and the current indicators used to evaluate nursing home care.
Nursing Home Population
The demand for nursing home beds, acuity level of residents, use of medical technology in the nursing home setting, and residents* disability are all increasing (IOM, 1996). Typically, nursing homes have three types of residents (IOM, 1996):
* Those admitted for rehabilitation after an acute hospital stay, expected to be discharged back to the community.
* Those admitted terminally ill, expected to die within a short period of time.
* Those with multiple chronic conditions who are placed in the nursing home for the rest of their lives.
Individuals admitted for rehabilitation are expected to be discharged home with an improved functional status. Residents who arrive terminally ill typically have been admitted from an acute care institution and have short nursing home stays of days or weeks (IOM, 1996). For the largest number of nursing home residents, the chronically ill and functionally disabled outcomes are complex and length of stay is variable. Maintenance or improvement of one's functional status is an initial goal, but declining functional status and death are the inevitable outcomes.
Who is Terminally III?
The first step in measuring quality of care at the end of life is to understand who is dying. MucH; of clinical practice, including hospice care, has been based on the belief^ that one easily can identify who is dying or who is terminally ill (Teño & Coppola, 1999). Disease states that currently characterize dying are mostly degenerative ia nature, such as heart and vascular diseajÎE (AMA, 1996). Although a caffcer disease may last for years, cancer patients usually have an identifiable period of decline that heralds approaching death. This period of decline is what McCormick and Conley (1995) refer to as the terminal phase.
In contrast, non-cancer diagnoses are characterized by periods of relative stability interspersed with acute exacerbations, and death frequently occurs abrupdy (Teño & Coppola, 1999). Individuals with chronic illnesses are Hkely to die suddenly, never having been diagnosed as terminally ill (Lynn, Harrell, et al., 1997). Being identified as terminally ill allows one to realign priorities, mend relationships, and say goodbye (Bartholome, 1999). A terminal diagnosis also allows staff the opportunity to realign care priorities to assist residents in meeting important goals and achieving a peaceful life closure. Unfortunately, for many nursing home residents the chronic illness trajectory disguises the dying process, making if difficult to plan and provide end-of-life care,
Indicators of Care
There has been a growing emphasis on measuring indicators of nursing home care such as functional status, cognition, use of restraints, activity involvement, eating, and nutrition (Hawes et al-, 1997; Mor et al., 1997; Phillips et al., 1997; Zimmerman et al., 1995). However, these indicators, while important, may have little relevance to effective end-of-life care. In addition, the 1996 IOM repon documents several problems in nursing home care that have the potential to affect end-oflife care negatively.
These problems include the lack of, or negligible amount of, registered nurse time per resident, high staff turnover, lack of education of the primary staff caregiver (nurse aides), and the low level of physician involvement. Finally, nursing home deficiency citations continue in areas that could affect end-of-life care, such as preservation of resident dignity and resident care planning (IOM, 1996). Despite an emphasis on improving care, authors of the 1996 IOM report concluded "the quality of care in some nursing facilities still leaves much to be desired" (p. 140).
Probably the strongest indicator for the potential of poor end-of-life; care in the nursing home has been the lack of pain recognition, assessment, and management for residents. It is estimated that 40% to 80% of nursing home residents experience pain, and major sources of pain include chronic conditions such as low back pain, arthritis, and neuropathies (Ferrei!, Ferrell, & Osterweil, 1990; Ferrell, Ferrell, & Rivera, 1995; Pannelee, Smith, & Katz, 1 993). Of the residents reporting pain in one study, approximately 50% reported that pain interfered with ambulation, enjoyable activities, and sleep (Ferrell et al., 1990). Pain is underdetected in nursing home residents, especially the cognitively impaired (Keay et al., 1994). As cognitive impairment increases, the assessment and treatment of pain for these residents decreases (Ferrell et al., 1995; Parmelee et al., 1993).
Pain assessment and treatment also vary with age and ethnicity (Bernabei et al., 1998; Engle, 1998). Bernabei et al. (1998) reported that residents older than age 85 were less likely than a younger cohort (65 to 74 years) to receive even weak opiates or morphine for pain, and Black residents were less likely than White residents to receive any pain medication at all. Others report that when Black residents do receive pain medication, they are less likely to be assessed for adequate pain relief (Engle, 1998).
Few researchers have studied care at the end of life for those with chronic illness diagnoses in longterm care settings, particularly nursing homes. Although researchers have begun to examine quality indicators for nursing home care, these indicators target functional maintenance or improvement and may have little relevance for quality care at the end of life. In light of the research to date on pain management, one could logically conclude that quality endof-life care is lacking. In general, even the experience of nursing home residents who die has not been well documented (IOM, 1997).
DESIGN AND METHODOLOGY
Qualitative methods were used to gain understanding of the end of life in one nursing home. Qualitative inquiry is a valuable approach when one wants to achieve a holistic understanding of a multifaceted phenomenon in a context-bound setting (Patton, 1990). The qualitative approach allowed the research team to be discovery-oriented, open to a variety of personal perspectives, and close to the process of dying within the nursing home environment. The research team consisted of the principal investigator (PI) and three graduate students.
The setting was an urban nursing home comprised of 170 residents and 160 employees. Potential participants included older adults residing on three units, excluding special care dementia units and the assisted living unit. One unit was considered a lighter-care unit composed of 21 residents and typically staffed with one licensed staff member (licensed practical nurse or registered nurse) and one nurse aide. The other two units each ranged from 30 to 35 residents and were considered to have heavier-care residents. Staff defined heavier care as care for residents requiring a two-person transfer or assistance with eating. These two units, typically, were staffed with one licensed nurse, two nurse aides, and one medication aide.
To be eligible for the study, older adults had to be English-speaking, have cognitive capacity or a caregiver (family member) willing to participate, have a chronic illness diagnosis, and be recognized by staff as declining or closer to death. Because of the difficulty identifying who is terminally ill, staff were asked how they identified residents who were closer to death. Staff identified the criteria used for selecting the study sample as weight loss, recent hospitalizations, increase in falls, cognitive decline, personality change, and social withdrawal. For the purposes of this study cognitive capacity was defined as an:
* Awareness of person and place.
* Ability to respond to questions about basic needs in a consistent manner, confirmed by staff.
* Absence of a mental illness diagnosis.
For those residents who did not have cognitive capacity, formal interviews were conducted with family caregivers. Residents whose primary diagnosis was cancer were excluded to focus on residents with more variable disease trajectories. Staff participants included individuals in contact with older adults who are declining (e.g., social workers, registered nurses, licensed practical nurses, nurse aides, nursing home administrator, therapists, medical director, director of nursing, visiting hospice nurse).
Thirteen residents who were chronically ill and declining (10 women, 3 men) and 3 family caregivers participated in formal and informal interviews on multiple occasions (approximately 45 encounters) over a 4-month period. Interviews with residents on more than one occasion promoted rapport between the investigator and the resident, allowing for the exploration of personal reflections on declining health and approaching death. Thirty staff members participated in either formal, one-on-one interviews or focus group interviews. During the 4-month period, 6 of 13 residents died.
Data were collected by participant observation; in-depth, semi-structured, face-to-face, formal interviews; informal interviews; focus groups; and health record reviews following approval by the Human Subjects Committee at the University of Kansas Medical Center. Participant observation, by licensed and unlicensed personnel, included observing day-to-day care activities delivered to residents who were declining (e.g., feeding, ambulating, turning, medication administration); formal care planning meetings; and formal and informal communication among staff and between staff and the resident. Observations occurred at the nurse station and in hallways, residents' rooms, staff meeting rooms, the residents' lounge, and residents' dining areas.
After the research team was in the facility for approximately 1 month, informal interviews occurred naturally whenever a team member was present. Staff became more comfortable with the research team presence and spontaneously shared stories about their interactions with each other and residents during the dying process. Data obtained during early observations and informal interviews allowed the research team to refocus subsequent observation periods to address specific activities related to emerging patterns or themes, to identify key staff informants to participate in more in-depth interviews, to investigate conflicting information, and to validate information obtained in formal interviews.
Residents meeting the eligibility criteria and their family caregivers were invited to participate in the indepth, semistructured interviews. After staff identified the residents as declining or closer to death, the PI approached residents individually and invited them to participate in the study. For residents who did not meet the cognitive criteria, permission was obtained from the resident's surrogate decision-maker prior to study inclusion.
The cognitively impaired residents were included in the participant observation, and formal interviews occurred with the family caregiver. Formal interviews with residents and family caregivers began with broad, common questions, but the investigator spontaneously explored, probed, and asked more specific questions as the interview evolved. Resident interviews began with nonthreatening questions such as, "How has your health changed over the last year?" and moved to more specific questions such as, "What does your future look like?" and "What does death look like to you?"
The research team remained in the field for a sufficient length of time to build trust with key informants, gain familiarity with the environment, develop an understanding of the end of life on the three units, and analyze emerging themes and patterns. Two graduate students each spent a total of 70 hours in the home. The PI and one additional graduate student spent 8 hours per week in the facility over the 4-month period. Participant observation occurred during all three shifts.
Data collection and analysis was simultaneous and continuous. Analysis began with the first observation and continued with each additional interview and observation. Qualitative content analysis - the process of identifying, coding, and categorizing patterns as they emerge from the data - was used (Coffey & Atkinson, 1996; Patton, 1990). A qualitative software program, Q.S.R. NUD*IST (1994), was used to facilitate coding and categorization.
Data analysis occurred in three phases. First, data were transcribed and coded. Following transcription, each interview and observation was read to obtain a sense of the whole (Tesch, 1990). Reading for a sense of the whole provided the investigator with a gestalt from which to filter and analyze individual pieces of data. Codes, words, or phrases that organize pieces of data were derived from the data (Miles & Huberman, 1994). Definitions were written for each code. Categories were developed to organize coded data into meaningful clusters, and detailed descriptions were written for each category. There were 298 codes and 26 categories.
In the second phase, codes and categories were explored and reconceptualized to evaluate emerging patterns and themes. A clear and detailed description of each pattern was written to promote understanding and synthesis into a coherent whole. In the third phase, interpretation, the research team explored the patterns and themes in detail to generate tentative relationships and working hypotheses.
Credibility of the qualitative analysis was supported by searching for rival explanations, explaining negative cases, using peer debriefing, and using member checking. For example:
* The research team stayed in field a sufficient length of time explore all negative cases emerged.
* Peer debriefing occurred 2 weeks among members of research team and on four with the qualitative consultant.
* Member checking, the clarification and validation of findings by the participants themselves (Lincoln & Cuba, 1985), was conducted the conclusion of data analysis. A member check was conducted with eight staff members (e.g., social worker, RN, aide, administrator, physician) and one resident. All members were in agreement with the data analysis with the exception the medical director, who denied that residents died in pain.
End of Life in a Nursing Home
Residents described death as a common experience and anticipated they would "die in the nursing home." Residents expressed a calm acceptance of death, saying, "I'm ready," "I'm gonna pass soon," "People come here to die, I know I am going to die." Most residents anticipated being reunited with friends and family through death. Although residents expressed an acceptance of death, the process of dying in this nursing home was described by both staff and residents as cold, lonely, and painful. Residents who were declining and were unable or unwilling to engage in social activities spent their days and nights alone. As one staff member stated:
I think dying is an extremely lonely process for most people in facilities. There are occasions where a family is involved.. .but by and large we don't see that available to a lot of nursing home residents.
There is a sense that staff withdraw from dying residents. One staff member said:
I think a lot of times the dying resident ends up a little bit neglected other than the 2-hour check because maybe it's too painful, too real, for staff to have to deal with.
Aides described the inability to provide individual attention. One aide said, "The last hour sometimes there is family here and sometimes not. I know we can't give no one-on-one, there aren't enough of us."
Dying was described as physically uncomfortable, with residents in pain and short of breath. Staff expressed anguish over watching residents die in pain, gasp for breath, or die alone. Observations confirmed that residents appeared uncomfortable and that staff lacked the skills to adequately assess and respond to physical discomfort. Conspicuously absent from the data was attention to spiritual and emotional needs either through the presence of a chaplain or from care staff.
Hospice services were viewed as a means of facilitating end-of-life care, by assisting with pain and symptom management; attending to spiritual, emotional, and social needs; supporting family; and most importantly, providing a person to manage the dying process. A hospice referral "gave permission" for the focus of care to change. For example, staff were more comfortable administering aggressive doses of narcotics and not pushing residents to ambulate or eat. Staff, family caregivers, and residents valued hospice, but there was tremendous staff conflict over the timing of the referral. Hospice services were rarely used in this nursing home, and even when used, referrals were late. The two residents in this study who were referred did not receive hospice care until 24 to 72 hours prior to death.
Staff Responses to Dying Residents
Long-term care, described by one staff member, is "palliative care.. .we're taking care of people at the end of their life." One nurse described the nursing home metaphorically by saying, "This is heaven's waiting room; they [residents] know what they are here for."
Staff expressed discomfort with death. There was limited evidence to suggest that nursing home staff attended to residents' changing physical and psychospiritual needs. Staff remarked that they were afraid of death. One staff member reflected on her coworkers and said:
I think they keep real superficial on this whole dying stuff and try not to let it get too close or acknowledge it too directly.
Another staff member said:
You have to accept death. And a lot of nurses and aides, well... they 've not accepted that, they're afraid of it. We're in a society where it is hard to accept death, for a lot of people.
The work environment was driven by tasks (e.g., bathing, ambulating, toileting) and death was treated as "just another task." One staff member said:
It's just another event in the day, another task, another thing that happens in the course of the day as opposed to an emotional and spiritual event that needs active participation on the part of the staff. Being with a dying resident is an amenity that the nursing home cannot afford the staff for.
Another staff member compared the task-oriented care that dying residents received to what she perceived would be a more appropriate focus, that is, psychological and spiritual care:
Residents are turned q 2, changed q 2, given their medicine, and that's it. It's been a very task-oriented response to the patient who is dying, to the resident who is dying without a lot of attention and loving support. Dying is largely a very spiritual process. It's not just a physical process.
Staff identified indicators of decline, namely, weight loss, falls, decline in activities of daily living, increase in infections, change in personality, decline in cognitive status, increase in hospitalizations, and disengagement. Although staff could describe indicators of decline, they were unable to predict the timing of death. Some residents died suddenly, and others rallied again and again. One daughter described her mother as having "nine lives" (she survived numerous acute exacerbations), whereas another resident said, "I ask the good Lord, 'Why am I still here?'"
Residents expressed a readiness to die. For example, one resident said, "I feel ready, and sometimes I ask the good Lord to take me." Residents were comfortable talking about death, but noted that staff did not initiate discussions regarding death. An interviewer asked, "Does the staff know that you are ready?" The resident responded, "I don't know, they don't ask, but my roommate and I, we talk about these things." Residents' perceptions of staff hesitancies were verified through direct observation of care, attendance at care planning conferences, and the lack of attention to roommate needs following death.
This hesitancy was vividly depicted by a care planning conference for a newly admitted resident with endstage cardiomyopathy. During this conference, discussions focused on increasing functional status and weight, as well as monitoring medications. Noticeably absent was any discussion surrounding end-of-life goals or treatment preferences. Furthermore, one resident described the lack of attention to the passing of his wife, saying, "We had a double bed when we first moved in. After she passed they took it out. I slept in it one night and then they moved things around and brought in a single bed. It took the props out from under me."
Contextual Factors Affecting Endof-LifeCare
One dominating pattern and five themes (Miles & Huberman, 1994) emerged to describe factors that facilitate or impede end-of-life care. Conflict emerged as a dominating pattern pervasive to all five themes: communication, quality of life, staff education, teamwork, and work environment. Moreover, themes were interactive and dynamic, making the presentation of each theme in isolation difficult.
Conflict emerged in a variety of ways because residents, staff, and caregivers were struggling with or fighting for what they believed would enhance a resident's quality of life. Conflict was fueled by a lack of communication, little teamwork, little or no educational preparation in geriatrics and the dying process, and demands of the work environment. Conflict had multiple foci. For example, within the nursing home, conflict was evident among disciplines, among shifts, and surrounding resident care decisions. Also, conflict was fueled by nursing home industry system dynamics, for example, the medical model focus of care, advance directives, die Minimum Data Set Plus (MDS+) care planning system, state and federal regulations, and reimbursement policy.
The work environment of nursing homes has been strongly influenced by state and federal regulations, the need for reimbursement, and the MDS+ care planning system. The focus of the MDS+ on maintaining or improving physical functioning created tremendous conflict in this nursing home. Conflict arose between the provision of active rehabilitation services and a perceived need for comfort care. Staff described residents who were actively engaged in rehabilitation days before death. Rehabilitation staff believed they were enhancing these residents' quality of life with therapy, while nurses questioned the residents* quality of life when they returned from therapy too weak to participate in activities or eat their meals.
Social service staff expressed concern that residents were not given time to focus on individual needs and preferences because of the energy spent in rehabilitation and also believed that residents did not have voice in choosing between rehabilitation or comfort care. Each discipline was concerned about residents' quality of life, but the lack of teamwork coupled with the inability to communicate effectively resulted in staff evaluating resident's quality of life within their own disciplinary framework.
The desire for resident quality of life caused not only interdisciplinary conflict, but also intense intrapersonal conflicts when deciding between therapy and comfort care. One therapist shared her difficulty in understanding whether a resident would be appropriate for rehabilitation or comfort care:
There are struggles because you don't know which direction the person will go. You have to go with your gut and be in touch with your patient when it comes time that they start talking about letting go. They start talking about death, about dying, about seeing ones who have gone before - those are the easy ones. The tough ones are the people that you don't know very well. They come from an acute care setting and you don't know how far they will progress. Do you push them? Some you just get over the humps and then they get better. Or are they going to go over the hump the other way and say goodbye?
In other words, it is obvious that some residents are going to die, but for others it is not clear - they may get better, they may not. Another therapist explained her conflict this way:
And then you've got people who won't live long, maybe 6 months or !ess. They just want to be left alone. You know you feel pretty cruel making someone get up every day doing all this stuff when in reality you know they probably don't have that long to live. You know you should be providing comfort care more than rehab.
All individuals, when entering the nursing home, are offered the opportunity to complete an advance directive. The purpose of an advance directive is to enhance communication and decision-making regarding end-oflife treatment preferences. Conflicts arose when residents lacked advance directives, when advance directives were unclear, or when family, staff, or residents had differing interpretations of the same advance directive. Advance directives invariably stated that life prolonging measures could be stopped when there "is a terminal illness, with no hope of recovery, declared by two physicians." This type of advance directive was impossible to enact given that residents suffered from multiple chronic illnesses, such as strokes, congestive heart failure or chronic obstructive pulmonary disease, which are not easily recognized as terminal conditions. Staff could recognize residents as declining, but were hesitant to label a resident as terminal until days or hours before death.
The inability to label someone as terminal disguised dying and resulted in a lack of communication regarding individual needs and personal preferences. This was confirmed by observing day-to-day care practices and MDS+ care planning conferences. For example, discussions concerning endof-life preferences were never observed during day-to-day care or MDS+ conferences for a resident with congestive heart failure (New York Heart Association Stage 4). When asked about care focused toward the resident's end-of-life preferences, one staff member said, "Well, she could live another year." The work environment, influenced by the MDS+ care planning process, created a focus on preserving and promoting functional status. This care planning assessment tool lacks the language and triggers for palliative care planning, thus decreasing the opportunity for communication regarding end-of-life treatment preferences.
One of the most striking areas for conflict was the different interpretations of the same advance directive. For example, an 88-year-old resident with advanced Parkinson's disease had a directive that stated she would forego antibiotics. The staff interpreted this as a directive to withhold antibiotics for a urinary tract infection, whereas the daughter said, "But if she gets a common litde infection that if treated would not be life threatening, we're not going to let it just go and go and go until it is systemic. That would be murder." Although advance directives guide care, the lack of ongoing communication between the staff and this resident's caregiver fostered conflict regarding treatment.
Most staff believed reimbursement policies fueled the use of rehabilitation services. Staff expressed resentment that palliative care was not an equal priority. Palliative care, as viewed by some staff, was less important and in competition with rehabilitation and a medical model focus. One staff member said, "This is what we do [provide palliative care], but we arc afraid to admit it. You get in trouble if you mention the ?' [hospice] word." Palliative care was a term used by staff, but not formally defined. When asked, staff remarked that palliative care meant caring for dying residents. Staff readily admitted they had not received any formal education related to death and dying and were uncomfortable with many interventions such as pain and symptom management.
The changing nursing home population, higher acuity with greater technological use, has created an environment in conflict with itself. Some staff view the nursing home as a resident's home and expressed frustration over care provided [as staff describe] in a medical model. Using a medical model prompted treatments such as blood tests on an individual with an ongoing gastrointestinal bleed who requested no additional treatment, or hospitalizations for 94-year-old and 96-year-old residents with end-stage disease processes. One staff member said, "The medical director is so aggressive. Where in all of this do you back off and say, 'This is their home-they wouldn't be doing all of this stuff if they were in their own home/ even given the same progression of their disease."
Conflict also emerged in the lack of teamwork. Aides described themselves as, "The lowest man on the totem pole," saying, "We are the silent voice," and "We have no input into the care pian, they [nursing staff] just say, 'You get on down the hall and do what you got to do."* Poor communication and the lack of teamwork was demonstrated by staff behavior in the following situations:
* Out of anger, aides withheld information from the charge nurse.
* The medical director told social service staff that acknowledging and reporting resident's pain was not their job responsibility.
* Staff perceived they had to go behind the medical director's back (to the family) to get a hospice referral.
As one staff member said:
There doesn't seem to be this real team concept of what everybody has to offer. You know, which really is a variety of different perspectives in taking care of a very complex thing [dying residents],
The lack of staff education regarding end-of-life care, a work environment that fosters a task-oriented approach to care, staff desire for and interpretation of residents' quality of life, and the lack of teamwork and communication created multiple opportunities for interpersonal and intrapersonal conflict regarding resident care issues. These factors influenced end-of-life care and left residents, family caregivers, and staff members with many unmet needs. As one staff member reflected about caring for dying residents:
I know that's a hospice realm and I know that's a social work realm. But I also think that everybody that deals with anybody in this setting needs training [education] in that. And I don't think you get that in school. If you're going to work in this environment you need it. Communication is the key. It takes everybody [teamwork].
Five themes as well as the dominant pattern of conflict influenced end-of-life care for these nursing home residents. The results of this study illustrate where and how problems within the nursing home industry, this particular nursing home, and between staff and residents influence and challenge care provided to dying residents.
The nursing home industry has dynamic factors influencing care. Care delivered within a medical model, the dominant mode of care delivery for most nursing homes (Keay de Schonwetter, 1998), promoted disagreements regarding the aggressiveness of care for declining residents in this study. A medical model of care delivery coupled with a care planning system that focuses on functional maintenance and improvement fueled interdisciplinary and intrapersonal conflicts. The MDS+ care planning process omitted important aspects of palliative care, such as attention to symptom control and psychospiritual needs.
The demands of a tightly regulated industry and fears of retribution reinforced the task-oriented approach of this nursing home. As noted in the 19% IOM report and supported by this study, the educational preparation of the major care provider (nurse aides) was very low. And finally, the passage of the Patient Self-Determination Act was an effort to ensure that health care providers (including nursing homes) acknowledged and honored patients' preferences for end-of-life care (Omnibus Budget Reconciliation Act, 1990). Advance directives, for this particular nursing home, were unable to guide care because of the inability to label a resident as terminally ill and the lack of ongoing staff-resident communication related to end-of-life preferences.
Within this particular nursing home, the lack of communication, little teamwork, and insufficient education, as well as a task-oriented, medically driven environment, negatively influenced the dying experience for many residents. Although staff expressed a positive desire for resident quality of Ufe, the lack of communication and teamwork led to unresolved conflict surrounding care decisions. Hospice care and family caregiver support positively influenced the dying experience by providing additional resources to the dying resident. Unfortunately, only two residents in this study received hospice care.
The majority nursing ïiome residents are there because of what Engle (1998) and McCormick and Conley (1995) call the chronic living-dying phase. For the majority of nursing home residents, death is an inevitable outcome and most will die without a terminal diagnosis. Within this nursing Home, staff and residents alike were influenced by the contextual factors. Residents were vulnerable to inadequate care at the end-of-life because of:
* A lack of educational preparation for death and dying within and across disciplines.
* A hesitancy of staff and some family members to talk openly about approaching death.
* A lack of discipline-specific leadership to enhance communication and promote problem recognition and solving.
* Differing resident expectations related to care at the end of life.
Nursing home staff were uncomfortable with death and did not ask important questions to facilitate individualized end-of life-care, such as:
* "How do you want to live out your remaining life?"
* "What do you want your death to look like, be like?"
* "What gives your life meaning on a day-to-day basis?"
IMPLICATIONS FOR PRACTICE AND FUTURE RESEARCH
The in-depth nature of this study highlighted numerous problems in this setting. For this particular nursing home, several approaches can be taken to address the problems identified by this study. These include educational programs aimed at conflict resolution, death and dying processes, communication processes, and care of residents near the end of life.
Development of institutional policies and methods to provide end-of-life care, either through the use of hospice or a focus on palliative care, is warranted in this nursing home. Team-building activities, particularly through the development of end-of-life programs, must be facilitated.
Future research needs to evaluate process variables such as pain and symptom management and advance care planning with important outcomes such as quality of life and resident satisfaction. In-depth explorations of additional nursing homes, especially those that achieve positive outcomes, will help identify models of effective end-of-life care and provide needed information to develop intervention studies aimed at improving end-of-life care. Nursing homes that obtain positive end-of-life outcomes could be used to illustrate problem-solving strategies for overcoming some of the system barriers illustrated in this study and be used to inform policy-making to improve this important component of care.
The dying process is not addressed or acknowledged for many nursing home residents. Dying residents deserve attention to their physical, emotional, and spiritual needs. Lack of attention to the dying process leaves little opportunity to prepare for death, to obtain spiritual and psychological growth, and to complete important life tasks.
- American Medical Association Council on Scientific Affairs. (1996). Good care of the dying patient. Journal of the American Medical Association, 275, 474-478.
- Bartholome, W.G. (1999). More lessons of the angel of death: A five-year course with a terminal illness. KU M ed, 49(2), 16-17.
- Bernabé I, R., Gambassi, G., Lapane, K., Landi, E, Gatsonis, C., Dunlop, R., Lipsitz, L., Steel, K., & Mor, V. (1998). Management of pain in elderly patients with cancer. Journal of the American Medical Association, 279, 1877-1882.
- Brock, D.B., & Foley, DJ. (1998). Demography and epidemiology of dying the U.S. with emphasis on deaths of older persons. The Hospice Journal, 13,49-60.
- Buckingham, R. W. (1983). Hospice in a longterm care facility: An innovative pattern of care. The Journal of Long-Term Care Administration, 11, 10-14.
- Cleeland, C.S., Gonin, R., Hatfield, A.K., Edmonson, J.H., Blum, R.H., Stewart, J.A. (1994). Pain and its treatment in outpatients with metastatic cancer. New England Journal of Medicine, 330, 592-596.
- Coffey, A.,& Atkinson, P. (1996). Making seme of qualitative data. Thousand Oaks: Sage.
- Engle, VF. (1998). The experience of livingdying in a nursing home: Self-reports of black and white older adults. Journal of the American Geriatrics Society, 46, 1091-1096.
- Ferrell, BA, Ferrell, B.R., & Ostcrweil, D. (1990). Pain in the nursing home. Journal of the American Geriatrics Society, 38, 409-41 4.
- Ferrell, B.A., Ferrell, B.R., & Rivera, L. (1995). Pain in cognitively impaired nursing home residents. Journal of Pain and Symptom Management, 10, 591-598.
- Garvin, J., & Chapman, C.R. (1995). Clinical management of dying patients. Western Journal of Medicine, 163, 268-277.
- Hawes, C., Mor, V, Phillips, C.D., Fries, B.E., Morris, J.N., Steele-Friedlob, E., Greene, A.M., & Nennsriel, M. (1997). The OBRA1987 nursing home regulations and implementation of the resident assessment instrument: Effects on process quality. Journal of the American Geriatrics Society, 45, 977-985.
- Institute of Medicine. (1996). Nursing staff in hospitals and nursing homes: Is it adequate? (G.S. Wunderlich, EA. Sloan, & C.K. Davis, eds.) Washington, DC: National Academy Press.
- Institute of Medicine (1997). Approaching death: Improving care at the end of life. (MJ. Field & CK. Cassel, eds.) Washington, DC: National Academy Press.
- Keay, T.J., Fredman, L., Taler, G.A., Datta, S., & Levenson, S.A. (1994). Indicators of quality medical care for the terminally i11 in nursing homes. Journal of the American Geriatrics Society, 42, 853-860.
- Keay, T.J., & Schonwetter, R.S. (1998). Hospice care in the nursing home. American Family Physician, S7, 491-494.
- Lincoln, Y.S., & Cuba, E.G. (1985). Naturalistic Inquiry. Newbury Park, CA: Sage.
- Lynn, J. (1998). Recent insights and new models of reform for care at the end of life. Paper presented at Visiting Professor Lecture, University of Kansas Medical Center, Center on Aging, Kansas City, KS.
- Lynn, J-, Harrell, F.E., Cohn, E, Wagner, D., & Connors, A.F. (1 997). Prognoses of seriously ill hospitalized patients on the days before death: Implications for patient care and public policy. New Horizons, 5, 56-61.
- Lynn, J., Teño, J.M., Phillips, R.S., Wu, A. W, Desbiens, N., Harrold, J., Ciaessens, M.T., Wenger, N., Kreling, B., & Connors, A.E (1997). Perceptions by family members of the dying experiences of older and seriously ill patients. Annals of Internal Medicine, 126, 97-106.
- McCormick, T.R., & Coniey, BJ. (1995). Patients' perspectives on dying and on the care of dying patients. Western Journal of Medicine, 163, 236-243.
- Mues, M.B., & Huberman, A.M. (1994). Qualitative data analysis (2nd ed.). Thousand Oaks, CA: Sage.
- Mor, V, Intrator, O., Fries, B.E., Phillips, C., Teno, J., Hiris, J., Hawes, C., & Morris, J. (1997). Changes in hospitalisation associated with introducing the resident assessment instrument. Journal of the American Geriatrics Society, 45, 1002-1010.
- National Center for Health Statistics. (19%). Vital Statistics of the United States, 1992, vol. 2: Mortality, Pan A. Washington, DC: Public Health Service.
- Omnibus Budget Reconciliation Act of 1990 (OBRA-90), Title IV, section 4026. Congressional Record. October 26, 1990: 12638.
- Parmelee, P.A., Smith, B-, & Katz, LR. (1993). Pain complaints and cognitive status among elderly institution residents. Journal of the American Geriatrics Society, 41, 517-522.
- Patton, M.Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage.
- Phillips, C.D., Morris, J.N., Hawes, C., Fies, B.E., Mor, V, Nennstiel, M., & Iannacchione, V. (1997). Association of the resident assessment instrument (RAI) with changes in function, cognition, and psychosocial status. Journal of the American Geriatrics Society, 45, 986-993.
- Q.S.R. NUD*IST [Computer software], (1994). Los Aptos, CA: Aladdin Systems.
- Riley, G., Lubitz, J., Prihoda, R., & Rabey, E. (1987). The use and costs of medicare services by cause of death. Inquiry, 24, 233-244.
- SUPPORT Principal Investigators. (1995). A controlled trial to improve care of seriously ill hospitalized patients: The study to understand prognosis and preferences for outcomes and risks of treatments (SUPPORT). Journal of the American Medical Association, 274(20), 1591-1598.
- Teno, J.M., & Coppola, K.M. (1999). For every numerator, you need a denominator: A simple statement but key to measuring quality of care for the "dying." Journal of Pain and Symptom Management, 77,109-113.
- Tesch, R. (1990). Qualitative research: Analysis types and software tools. New York: Falmer.
- Zimmerman, D.R., Karon, S.L., Arling, G., Clark, B. R., Collkins, T, Ross, R., & Sainfort, F. (1995). Development and testing of nursing home quality indicators. Health Care Financing Review, 16, 107-127.