Changes in Knowledge and Attitudes of Certified Nursing Assistants about Ethics of Treatment Choices for Nursing Home Residents with End-Stage Alzheimer's Disease

Material Information

Changes in Knowledge and Attitudes of Certified Nursing Assistants about Ethics of Treatment Choices for Nursing Home Residents with End-Stage Alzheimer's Disease
Copyright Date:


Subjects / Keywords:
Advance directives ( jstor )
Alzheimers disease ( jstor )
Caregivers ( jstor )
Certified nursing assistants ( jstor )
Death ( jstor )
Dementia ( jstor )
Hospice care ( jstor )
Nursing ( jstor )
Nursing homes ( jstor )
Palliative care ( jstor )
Greater Orlando ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Annette Kelly. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
Resource Identifier:
71313006 ( OCLC )


This item is only available as the following downloads:

Full Text




Copyright 2004 By Annette Kelly


This dissertation is dedicated to all Certified Nursing Assistants in nursing homes. Their care and compassion are an important part of our health care system.


iv ACKNOWLEDGMENTS I would like to acknowledge and thank the following individuals for their assistance in the completion of this dissertation. I am particularly grateful to my supervisory committee whose expertise and patience were critical to this accomplishment, and to Dr. Claydell Horne, supervisory committee chair, whose personal encouragement and interest in the topic were key to my completion. Dr. Horne has been a mentor as well as a critic, and has helped me keep the goal in sight as she walked each small step with me. Dr. Meredeth Rowe provided specific research insights as well as general direction to the project. Dr. Hossein Yarandi provided careful and detailed instruction on statistical methods, and was thorough and patient in assisting with the analysis of this research. Dr. Michael Marsiske provided an invaluable perspective that enriched the process and strengthened the project. I am grateful also for the assistance of others. Dr Marie Visscher encouraged me many years ago to continue to be interested in the big questions. Dr. Mary BearÂ’s enthusiasm for nursing in the community was contagious. Dr. Jeffrey DwyerÂ’s intelligent approach to questions of public policy enriched my perspective. I am especially grateful to my teachers and colleagues whose influence has guided me throughout my nursing career: especially the Sisters of Charity at St. VincentÂ’s Hospital School of Nursing in New York City, whose passion for excellence and zeal for justice have influenced my own life.


v I am grateful to my family. My love and thanks to my husband, Jerry, whose patience and willingness to listen to every detail have been amazing. Our children (Moira, Kathleen, Deirdre, Patrick and Amiee) have been enthusiastic and positive. Even my granddaughter, Hilary, has provided an occasional reference that I might have overlooked. My parents, Jim and Mae, one in memory and the other still very much in my daily life, are the embodiment of the persons I hope this research will respect and nourish in their last days.


vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 Role of Ethics...............................................................................................................3 Palliative Care...............................................................................................................5 Dying with AlzheimerÂ’s Dis ease in the Nursing Home...............................................6 The Family Experience...............................................................................................10 Staff Caregiver Perspectives.......................................................................................11 Hospice and AlzheimerÂ’s Disease..............................................................................12 Intervention Programs................................................................................................13 Purposes of the Study.................................................................................................15 Research Hypotheses..................................................................................................15 Research Questions.....................................................................................................16 Assumptions...............................................................................................................16 Delimitations...............................................................................................................16 Significance of the Study............................................................................................17 Definition of Terms....................................................................................................18 2 REVIEW OF THE LITERATURE............................................................................20 Ethics and Long-Term Care........................................................................................20 Advance Directives.....................................................................................................22 End-of-Life Care and the Health Care System...........................................................23 The SUPPORT Study..........................................................................................23 Effects of the Patient Self Determination Act (PSDA).......................................24 Cost of Care at the End of Life...................................................................................27 Measuring Quality in End-of-Life Care.....................................................................28 Dying in the Nursing Home........................................................................................31 AlzheimerÂ’s Disease...................................................................................................35


vii Advanced AlzheimerÂ’s Disease...........................................................................35 Dying with AlzheimerÂ’s Disease.........................................................................36 Hospice Care and AlzheimerÂ’s Disease..............................................................38 Hospice and Advanced Dementia.......................................................................40 Policy Issues and Hospice Care...........................................................................43 Barriers to Appropriate End-of-Life Care...........................................................45 Treatment Decisions at the End of Life......................................................................46 Hospitalization and hospice care.........................................................................50 Artificial Nutrition and Hydration (ANH)..........................................................51 End-of-Life Care in the Nursing Home Setting..........................................................56 The Direct-Care Worker......................................................................................56 An Emerging Culture of Care..............................................................................60 Intervention models.............................................................................................64 Policy Development....................................................................................................65 3 METHODOLOGY.....................................................................................................66 Research Design.........................................................................................................66 Subjects....................................................................................................................... 68 Measures.....................................................................................................................70 Protection of Hu man Subjects....................................................................................74 Procedures and Protocols............................................................................................74 Reliability Protocol..............................................................................................74 Subject Assignment.............................................................................................74 Education Program Protocol................................................................................75 Education/Support Protocol.................................................................................76 Post-Test Protocol...............................................................................................77 Data Collection Protocol.....................................................................................77 Statistical Analysis of Data.........................................................................................77 4 RESULTS...................................................................................................................80 Description of the Setting...........................................................................................80 Demographic Characteristics of the Sample...............................................................81 Hypothesis 1...............................................................................................................87 Hypothesis 2...............................................................................................................87 Hypothesis 3...............................................................................................................88 Hypothesis 4...............................................................................................................88 Other Findings............................................................................................................89 5 DISCUSSION AND RECOMMENDATIONS.........................................................93 Implications................................................................................................................93 Change in Knowledge about Ethics of Treatment Decisions in End-Stage AD.94 Change in Attitude about Treatment Choices in End-Stage AD.........................95 Additional Support Sessions................................................................................96 Level of Knowledge of AD.................................................................................96


viii The Certified Nursing Assistants................................................................................97 The Experience of Implementing Standing By You ....................................................99 The New Culture of Residential Long-Term Care...................................................100 Recommendations.....................................................................................................101 Future Research........................................................................................................102 APPENDIX A CERTIFIED NURSING ASSISTANT QUESTIONNAIRE...................................104 B DEMOGRAPHIC QUESTIONNAIRE....................................................................109 LIST OF REFERENCES.................................................................................................110 BIOGRAPHICAL SKETCH...........................................................................................123


ix LIST OF TABLES Table page 3-1 Study timeline..........................................................................................................68 3-2 Major Study Variables.............................................................................................78 3-3 Hypothesis testing....................................................................................................78 4-1 Description of the study settings..............................................................................80 4-2 Frequencies and percentages of th e sample characteristics (N = 60).......................81 4-3 Comparison of ethnicity and race by study sites......................................................82 4-4 CNA-Q means table (N = 60)..................................................................................85 4-5 Impact of the Education Intervention and Education plus Support Sessions on Level of CNA Knowledge about Ethics of Treatment Decisions and the Number of Preferred Attitudes Among CNAs about Treatment Choices for Persons with End-Stage AD..........................................................................................................88 4-6 Group comparison....................................................................................................90 4-7 Means table for knowledge about ethics..................................................................90 4-8 Means table for attitude s about treatment choices...................................................91 4-9 Means of three groups at Time 1 and Time 3..........................................................92 4-10 One-way ANOVAs to compare 3 groups at time 1 and time 3................................92


x LIST OF FIGURES Figure page 3-1 Research design........................................................................................................66 4-1 Mean subscale scores...............................................................................................86


xi Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHANGES IN KNOWLEDGE AND ATTITUDES OF CERTIFIED NURSING ASSISTANTS ABOUT ETHICS OF TREATMENT CHOICES FOR NURSING HOME RESIDENTS WITH END-STAGE ALZHEIMERÂ’S DISEASE By Annette Kelly December 2004 Chair: Claydell Horne Major Department: Nursing Nearly three-fourths of the 1.7 million residents in skilled nursing facilities have AlzheimerÂ’s disease or a related disorder; and most of these residents will die in the nursing home setting. It is estimated that more than 40% of deaths in the United States will occur in skilled nursing homes by 2020. End-of-life care decisions for severely cognitively impaired persons is challenging, because getting the dying personÂ’s current perspective on treatment choices is not possible, and because advance directives are very often missing, incomplete, or ignored in the nursing home setting. More than 90% of the direct care of persons with end-stage AlzheimerÂ’s disease is provided by Certified Nursing Assistants who have little or no information about (or input into) treatment decisions. We are still a health care system with a bias toward aggressive intervention, resisting the more compassionate and relational application of a palliative care approach. However, a relational, palliative-care approach has become the hallmark of excellent preferred Alzheimer care, as the disease progresses to the end stage.


xii Certified Nursing Assistants are moral agents, and should be considered responsible individuals in the delivery of appropriate care and support to persons with end-stage AlzheimerÂ’s disease. Everyday care for persons with end-stage AlzheimerÂ’s disease is founded on the dignity of the individual and demonstrated in respect for the individualÂ’s wishes and preferences. The ethi cal principles of autonomy, beneficence, and nonmaleficence can be lived out in the careful management of everyday ethical care. Standing By You (an ethics education and support program for Certified Nursing Assistants who care for persons with end-stage AlzheimerÂ’s disease) was developed and implemented to address this lack of ethics education. The researcher found significant changes in knowledge about the ethics of treatment decisions and a greater number of pr eferred attitudes about treatment choices in Certified Nursing Assistants following an education intervention. A support intervention, following the education component, did not show significant changes. The variables of religiosity, ethnicity, and number of years working in long-term care did not affect the rate of change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes about treatment choices for SNF residents with end-stage AD after the implementation of an education intervention.


1 CHAPTER 1 INTRODUCTION Almost five million Americans have AlzheimerÂ’s disease (AD), and well over 100,000 people die with AD in our country each year (AlzheimerÂ’s Association, 2004; Hebert, Scherr, Bienias, Bennett & Evans, 2003). Dying with AD or a related dementing illness can be a very difficult event for the family, for formal caregivers in the health care setting, and possibly for the individual person. We have not yet focused sufficient creative reflection on the complex issues of dying with AD. It is often a lonely death, prolonged and expensive. Family caregivers who are responsible for decision-making, as well as health care providers who guide and implement most care decisions, are too often not involved in addressing the real needs of persons dying with AD. The loneliness and dissatisfaction is felt (not necessarily quantifiable but intuited) and expressed by increasing numbers of families. Professionals call for a new perspective. A balance between burden and benefit should be considered, when weighing options in treatment. The prolongation of life, and often the absence of a peaceful death, is too often the result of aggressive and/or inappropriate interven tions along the way; an outcome of a fixed focus on individual autonomy, with limited attention to the equally important ethical principles that help us to do good, to do no harm, and to be fair. Such a poor death experience may also be the result of inadequate planning for this important time of life, and the result of a lack of institutional and societal support around dying. However, such a tragedy is more often due to a lack of everyday ethical decision making, conceptualized in this research as an ethic of care. The expense of this failing in care is twofold. The


2 tangible economic burden to family and government of nursing home cost, repeated hospital admissions, invasive procedures, a nd aggressive medication protocols is only one part. The human expense to the person dying with AD (at whatever level it can be realized) and to the family and formal caregivers must also be recognized. The persistent bias in the American health care system leans toward acute care and high-tech intervention rather than toward palliative care. Annually, more than 25% of nursing home residents experience a hospital admission (Brooks, Warshaw, Hasse & Kues, 1999). Fewer than 8% of those with a primary diagnosis of end-stage AD and their families experience the support of hospice care principles (US Department of Health and Human Services, 2004). Yet, AD is a terminal illness. The growing number of aging persons has increased our focus on the problems inherent in an aging society, not the least of which is the presence of AlzheimerÂ’s disease and related dementing illness. An anxiety exists that has influenced the increasing public acceptance of physician-assisted suicide. Indeed, there appears to be more worry and fear about the cognitive losses of dementia than the physical pain or impairment associated with other terminal diseases (Topinkova & Callahan, 1999). It is interesting to note that Dr. Jack KervorkianÂ’s first case, Janet Adkins, was in the early stage of AlzheimerÂ’s disease. Former President Ronald Reagan brought international attention to AD when he wrote his letter to the American people in 1994 announcing his AlzheimerÂ’s diagnosis, and as he progressed through stages of decline, until his death in June 2004. However, a major goal of Alzheimer research has been to find a cure or cause, with a less adamant focus on care and systems of caregiving. There is a long-term danger that the impending crisis of more people with the disease in a system too slow to respond with


3 supportive, cost effective, and appropriately targeted services will become overwhelmed by the staggering problem and choose the slippery slope as the way out. Will assisted suicide or euthanasia proponents find a willing populace in the next, generation when faced with 14 to 17 million persons with AD? Role of Ethics Biomedical ethics has emerged as a discipline in the past 25 years, partly because of the development of more complex and difficult alternatives available to clinicians in health care professions. Biomedical ethics is based on four key principles, with several additional principles (Beauchamp & Childress, 2001) flowing from those main four. Autonomy : the right of an individual to privacy and self-direction. Beneficence : doing good for another. Nonmaleficence : to do no harm. Justice : the rights and responsibilities of all parties are observed. There is a growing concern within long-term care settings regarding the usefulness of these biomedical ethical principles in providing adequate guidance to everyday caregivers in chronic disease, particularly caregivers of persons who are cognitively incapacitated by AlzheimerÂ’s disease and similar disorders (Holstein, 2001). It is difficult to apply the principle of autonomy, for example, to an individual for whom making a currently purposeful, reasoned decision to act is a relative impossibility. While we can know on paper that a particular decision was made by an individual (as in an advance directive) we cannot really know what this cognitively impaired individualÂ’s current decision is about the same question. It is also difficult to honor the individuality of direct-care workers in a setting where institutional goals are often at odds with personal values. Certainly, just as other health care professionals, Certified Nursing A ssistants (CNAs) who provide almost all of


4 the direct nursing home care, possess moral agency. That is, they have the ability to take responsibility for their choices. In the daily care of persons whose need for compassionate accompanying care is enormous, these workers must be supported by a care system that recognizes each personÂ’s worth and dignity (worker as well as patient). Caplan and Kane (1990) described the morality of the mundane in a book on everyday ethics. The authors cautioned against too great a focus on the big life-and-death decisions, lest we come to believe that, in a setting where death is all around us, life at all costs is the ultimate goal. Feminist ethics recognize the moral world of relationship between caregiver (formal or family) and the person with AD, including responsibility, obligations, love, and concern (Holstein, 2001). Recognizing the moral agency of cognitively impaired nursing home residents (by honoring their wishes in daily discourse) enhances the dignity of both care receiver and caregiver, imbuing the relationship with value and humanity. AlzheimerÂ’s disease and similar disorders are above all relational, where the best treatment prescription is setting oneÂ’s self aside for the otherÂ’s needs. This resulting ethic of care is the lens through which we must address the enormity before us. Millions of people will die with advanced dementia in the coming decades, even as research brings us closer to cause, cure, and prevention. Focusing on an ethic of care does not abandon autonomy. In fact, an ethic of care requires that the needs and wants of the person dying with AD be the center of the picture. Using the same conceptual framework of personhood, even the direct-care worker has moral agency. All care staff are moral agents regardless of their level in the organizational hierarchy. Direct care staff ar e largely ignored by the system, not included


5 in the development or evaluation of the resident care environment. This gap creates a void in the delivery of ethically appropriate care. Wilmot, Legg, and Barratt (2002) conducted a focus group research study (n=12) specifically focused on feeding issues in end-of-life care. The key finding in the anal ysis of the focus group dialogue was staff membersÂ’ reliance on personal responsibility and commitment to provide care for the person as informed by their own values and life experiences. Palliative Care Palliation is a concept that embraces the idea of an ethic of care. The word comes from the Latin word pallium, which means cloak. Palliative care provides comfort like the image of a cloak that entirely encompasses and comforts the person. Palliative care is defined by the World Health Organization as the active total care of patients whose disease is not responsive to curative treatment. Control of pain, of other 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 (World Health Organization, 1990). Most often expressed in a hospice environment, formal palliative care is traditionally directed to patients dying with cancer, even though hospice reimbursement has been available to those with non-cancer diagnoses since 1994 . However, while Do Not Resuscitate (DNR) may be the guideline of choice, palliative care is rarely the formalized protocol for the management of life at the end of the long AlzheimerÂ’s journey. Living-Dying Trajectory There is a growing conviction among advocates for improvement in end-of-life care that palliative care ought to be part of the discussion in all health care. While it is


6 difficult to see this concept in healthy, young people, it is not as difficult to apply to aging persons with chronic disease, even though they may be relatively healthy. The concept of a living-dying trajectory requires that we examine quality-of-life issues while planning care; and requires that individual pr eferences for treatment options be included in every planning session. Dying with AlzheimerÂ’s Disease in the Nursing Home There are 1.6 million elders in our nationÂ’s 18,000+ nursing homes; approximately 50% are permanent residents; that is, they are not admitted for brief rehabilitation (US Department of Health and Human Services, 2004). Of more than 700,000 full-time equivalent employees providing direct care, 60% are CNAs (Strahan, 1999). Almost 50% of persons aged 85 and olde r have AD or a related disorder. By midcentury, the number of persons with AD or a related disorder will grow from the present 4.5+ million to more than 14 million (AlzheimerÂ’s Association, 2004). Over 70% of all nursing home residents have a dementing illness. Of all deaths in the United States in 1993, 20% occurred in nursing homes. This proportion is likely to rise to 40% by 2020 (Brock & Foley, 1998). Of all SNF deaths, 50-60% are residents who have a diagnosis of AD or a related disorder (Magaziner, German, Zimmerman, Hebel, Burton & GruberBaldini et al., 2000.) Of all new nursing home admissions, 30% die within 6 months. More than 25% of nursing home residents experience at least one hospital admission annually (Brooks et al., 1999). These numbers illustrate the enormity of the issues surrounding the topic of dying in a nursing home. As a terminal illness, AD is deceiving, because the incremental cognitive decline and its symptoms distract both family a nd staff caregivers from recognizing the dying process, often until quite late in the disease progression. It is not known what part of this


7 dying process the person with AD recognizes. It has been said that the oft-repeated plea of the person with AD (“Take me home!”) may represent an existential yearning for heaven or mother or final rest. Clearly, the task of recognizing and accompanying the nursing home resident with advanced AD on this final journey is a complex and challenging job and a very special mission. In the Institute of Medicine report on care at the end of life, an abiding belief was how difficult death in a nursing home often is (Field & Cassell, 1997). According to Mezey, Dubler, Mitty, and Brody (2002), we in the healthcare system “lack the moral, legal and ethical consensus as to when someone with dementia has begun the dying process, or how to manage symptoms related to pain and nutritional status in persons with dementia” (p.11). Special Care Units (SCUs) for persons with AD (which have developed as both marketing and care avenues for facilities) are designed to address the needs of ambulatory residents (or at least those mobile in a wheelchair) who require activity structure and who may be an elopement risk. In most cases, the resident’s decreasing ability to participate, or decreased need for exit security, will make him a candidate for SCU discharge to the main facility population. This discharge is usually not to a unit with an advanced AD focus or end-of-life mission. In fact, the discharge is often back to the rehabilitative/custodial environment, which is in complete denial of the meaning of escalating cognitive and functional loss in the advanced stage of AD. It has been said that the goal of care in the medical-model nursing home is the prevention of death (Volicer, 1997); thus, the incongruity of a terminal disease in a death-denying environment.


8 The Patient Self-Determination Act (PSDA) Focus on end-of-life decisions in nursing hom e residents is more than a policy issue or a merely academic concern. The Patient Self Determination Act (PSDA) went into effect in 1991. This law requires that any person admitted to a health care facility that receives Medicare or Medicaid be informed of the right to execute an advance directive, which is a declaration of individual preferen ces about care and treatment. The health care facility is required by law to note in the resident record the presence or absence of a directive and to educate its staff and the community about advance directives. The goals of this law are to foster autonomy in decision-making for persons in health care facilities; and to encourage community-wide, open communication about end-of-life issues (Mezey, Mitty, Rappaport & Ramsey, 1997). The PSDA is important in that it provides a legislated mandate to move nursing homes toward a more substantive engagement with residents, family, and staff on the subject of end-of-life care. The law requires th at persons be informed of their right to participate in their own medical decision-making and their right to complete written advance directives. Institutions are required to ask whether the patient has an advanced directive, and are required to have formal policies about directives. In a 1993 survey, however, only 20% of nursing home residents had a health care proxy, and 13% had a living will, although 51% had a DNR order (Teno, Lynn, Weenger, Phillips, Murphy, Connors et al., 1997). Only 8% included directives about hospitalization and artificial nutrition. The highest level of cooperation in this study was from family proxies of persons with dementia (88%), leading to the assumption that families in general might be willing to prioritize goals and discuss treatment levels if given the opportunity and


9 encouragement. In 1997, Bradley, Blechner, Walker, and Wetle reported that family members filled out most of the advanced directive forms in six Connecticut nursing homes, even where residents had decision-making capacity. In a 2000 report by Mezey, Teresi, Ramsey, Mitty, and Bobrowitz, only 51% of all nursing home residents across the country had an advance directive. While the impact of the PSDA on the numbers of persons completing directives is generally positive, there is little information about how capacity in cognitively impaired residents is measured, how often the directive is actually reviewed with the resident or proxy, or what constitutes broader staff education on these topics. It is also unknown whether the greater presence of advance directives actually affects treatment decisions (Mezey et al., 1997). Involvement in real needs of persons dying with AD should include awareness and communication, as well as implementation of any prior health care wishes of the patient. It should include careful monitoring of di sease progression and managing of end-stage symptoms, as well as reasonable and proactive decision making about aggressive or invasive care. Persons in the last stages of AD need comfort care and a supportive environment. They need pain and anxiety management, the presence of others, and opportunities to be connected in some positive way to the world they will soon be leaving. Caregivers (both family members and nursing home care staff) need comfort care and grief support themselves, as well as accurate information and training opportunities. It may be challenging, but still possible, to assess the state of well-being of a person in the early stages of AD. However, by the time the disease progresses to the end stage, both formal and family caregivers may be completely at a loss as to how to make a


10 difference in quality of life, given the limited communication ability and overwhelming physical support needs of the patient. We ignore the issues of dying with AD, because it is hard to measure success without traditional patient feedback. It is also difficult to move out of the medical model which is designed for diseases that have acute exacerbations periodically requiring vigorous intervention, followed by a period of relative stability or even semi-recovery. Alzheimer’s disease is more like a very long sloping road with an inevitable end. Can we provide a supportive environment for dying and a palliative care plan for someone who is currently unable to direct such an event, taking into account prior wishes and present realities? Can we change the culture of care in nursing homes to invite formal caregivers as well as family members into the dying experience in a meaningful way? The Family Experience Persons with AD die over a period of years rather than months (Sachs, 2000). Programs like Dr. Sachs’ Palliative Excellence in Alzheimer’s Care Efforts (PEACE) (a project designed to foster advance care planni ng, to address family caregiver issues, and to support a palliative approach through hospice care) have the potential to lead the way toward positive systemic change. Recognizing the losses that are part of the Alzheimer’s disease journey is not a onetime event; but requires acknowledgement of a series of losses, as symptoms worsen (Kuhn, 2001). Family caregivers often describe these loss events as surprising, because they wrongly believed they had already “lost the person.” The most overwhelming challenge for families is the decision to relocate a loved one to a nursing home. Kuhn (2001) described the mixed emotions of guilt, worry, and relief. This placement event is


11 so negative that it is devoid of the celebratory ritual that accompanies other life transitions like marriage, retirement, and even death. Current systems of care do not support people with long-term progressive disability and eventual death (Lynn, 2000). Lynn, a champion of improving end-of-life care, has called on clinicians and provider organizations to improve end-of-life practice through better family communications, improved understanding and treatment of symptoms, and advocacy for public policy change. A study by Steinhauser, Christakis, Clipp, McNeilly, McIntyre, and Tulski (2000) showed that families value pain and symptom management, and communication about what to expect in the dying process, as the most important elements in excellent end-of-life care. Staff Caregiver Perspectives Since the person with advanced dementia cannot advocate for himself, families are often slow to trust caregiving staff in any care setting. In the nursing home, where the medical model, liability concerns, and regulatory issues continue to frustrate our collective intuition about what care should really be like, the trust issue is palpable. In their Partners in Caregiving communication program, Pillemer, Hegeman, Albright, and Henderson (1998) found that nursing assistants said they do the work of caring for frail adults because it gives them an opportunity to help others, and that this was meaningful work to them. Contrasted with the poor public image of nursing homes, these values seem incongruous. Ortigara (2001) and others (Bargmann, 2001; Gwyther, 2001; Troxel & Bell, 1997; Thomas, 1996) are in the forefront of those who lead our thinking in creating models for the support and training of direct-care workers. The Pioneers for Culture Change in Long-Term Care movement, a national coalition of professionals and


12 providers committed to changing the culture of care in nursing homes, embodies the dual goals of resident and worker dignity (Pioneer Network, 2002). Nursing home regulations that hold facility administrators accountable for medication use, resident weight loss, a nd other indicators of a worsening prognosis, further frustrate the staffÂ’s ability to acknowledge the dying process (Keay, 1999). It may be difficult to know, in the long disease trajectory of persons with dementing illness, when the final stage is reached. With no systematic support to look closely at changes, such recognition seldom happens. Cultural differences between the nursing home and the hospice program also create barriers. However, benefits to patient, family, and staff are obvious when successful collaboration does exist (Keay, 1999). Both family and staff caregivers of persons with advanced dementia provide care without the reciprocity or gratitude that is often available from persons who are cognitively intact. This giving of self with little return can be an enormous personal burden. For staff, giving this kind of care every day to many people is a large task (Mezey, Miller, & Linton-Nelson, 1999). Family members do not really know what encompasses quality end-of-life care; and the current climate of rehabilitation and recovery in the nursing home does not openly provide such information (Mezey et al., 1999). An incongruity in the situation demands support and education for both family and staff caregivers on the topic of endof-life and related subjects. Hospice and AlzheimerÂ’s Disease Hospice concepts, palliative care, and comfort measures all make sense as approaches to care in end-stage AD. Mechanisms exist for applying hospice principles, yet fewer than 8% of all hospice patients have a primary diagnosis of dementia (US


13 Department of Health and Human Services, 2004). The Medicare requirement of a 6months-to-death prognostic indicator is diffi cult to apply to persons with AD Their disease course is not linear and they may actually live much longer with excellent and appropriate care. Current researchers are further clarifying some of the inequities in both scope and benefits under Medicare for end-of-life care (Buntin & Huskamp, 2002). Palliative care and comfort care require the understanding and commitment of all the players (family and formal) and often do not fit in the nursing homeÂ’s institutional culture of rehabilitation and intervention. Nursing homes are marketed nursing homes as places where people live; we have ignored the reality that most of the people who live there will die there. It is also true that families are often ambivalent about their own decisionmaking powers and may poorly understand the principle of surrogacy, or substituted judgment (Brenner, 1999), thinking that they are in some way hastening the death if they refuse inappropriate aggressive care. Although hospice enrollment of nursing home residents has increased, hospice is still not widely accepted in nursing homes. Biases still exist that hospice means giving up rather than doing all we can to provide comfort and support. There is little evidence of a real embracing of hospice philosophy in nursing homes, and little is known about the influence hospice has on the culture of care in a nursing home setting. Incorporating formal palliative care philosophy into nursing home care seems contradictory to the insistently rehabilitative marketing message preferred by most nursing homes. Intervention Programs The development of an education and support intervention designed to enhance knowledge and address attitudes about treatment decisions is supported by study results (Mezey, et al., 2000; Molloy, Guyatt, Russo, Goeree, OÂ’Brien, Bedard, et al., 2000;


14 Walker, Blechman & Gruman, 1998a). Ethics education programs, such as Decisions Near the End of Life (developed to educate professionals in hospitals and nursing homes in 1991) paved the way for education targeting paraprofessional staff. Additional research focused on educating the hospice team in the direct care of the dying, but did not include the long-term care setting, despite the very high death rate in that setting. Researchers survey the nursing home assistant population with a focus on an urban, stable staff. In reality, many in this work force are multicultural in values, language, and religion; and have a higher than 100% annual turnover rate. The Ethics Education and Support Program, developed for nursing a ssistants by Chichin, Burack, Olson & Likourezos (2000) includes a preliminary questionnaire session and three 1-hour education/discussion sessions, with handout materials for participants. In this study, the intervention group was compared to a control or non-intervention group. As expected, change in attitude about end-of-life care did not change dramatically in this short, focused program. However, significant findings from their study included increases in Understanding about the difference between euthanasia and allowing to die (61.4% in the control group versus 48.4% in the in tervention group said that not agreeing to a particular treatment was the same as committing suicide). Comfort level with resident refusal of life-sustaining treatment (23.7% in the intervention group versus 27.2% of the control group). Agreement with the belief that some treatments given to nursing home residents are too uncomfortable for the resident (46.6% of subjects in the intervention group versus 35.5% in the control group said that we sometimes give residents treatments that are too uncomfortable). Focusing on the nursing assistant in the nursing home is critical because of the close and personal role these staff members have in caring for the person dying with AD. Elements of everyday ethics, basic advance directives, information about AD, related comfort-care techniques, and follow-up support sessions are all necessary components of


15 a meaningful education and support program designed to effect culture change and move closer toward compassionate care of the dying. This two-part model forms the proposed intervention program, Standing By You. The values of personhood, individual dignity, and mutuality are foundational to the Standing By You program. The inherent dignity of the person with advanced dementia, as well as that of the staff caregiver, is enhanced when resident choice is honored and respected at the end of life, and when compassionate and informed comfort care are both encouraged and practiced. By fostering an atmosphere of mutuality, through communication of care preferences and disease information, the inherent dignity of both sides of the relationship is recognized. Dying with AD in a nursing home will become an even more frequent occurrence. A solution may lie in our ability and willingness to back up from the event (dying with AD in a nursing home) to explore the elements in the current setting, and to design and implement a new approach that will form the framework for a growing cultural change in how we view this event. It is hoped that this structured program , Standing By You, will become an important element of end-of-life nursing home care for persons with AD. Purposes of the Study 1. To measure and analyze the change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes about treatment choices for SNF residents with endstage AD after the implementation of an education intervention. 2. To determine if added support sessions make a difference in the degree of change in knowledge and attitudes compared to those who receive only the education intervention. Research Hypotheses 3. CNAs will have greater knowledge about the ethics of treatment choices in the care of persons with end-stage AD after the implementation of an educational intervention than before the intervention.


16 4. CNAs will have a greater number of preferred attitudes about treatment choices in the care of persons with end-stage AD after the implementation of an educational intervention than before the intervention. 5. CNAs who have participated in the educational intervention and additional support intervention will have a greater increase in knowledge about ethics of treatment choices in end-stage AD than before the support intervention and greater than those who did not receive additional support. 6. CNAs who have participated in the educational intervention and additional support intervention will have a greater increase in preferred attitudes about the treatment choices in end-stage AD than before the support intervention, and greater than those who did not receive additional support. Research Questions 7. Does ethnicity affect the rate of change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes a bout treatment choices for SNF residents with end-stage AD after the implementation of an education intervention? 8. Does religiosity affect the rate of change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes a bout treatment choices for SNF residents with end-stage AD after the implementation of an education intervention? 9. Does the number of years worked as a CNA in long-term care affect the rate of change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes about treatment choices for SNF residents with end-stage AD after the implementation of an education intervention? Assumptions 10. Research and medical ethics reflection identifies certain preferred attitudes regarding the ethics of treatment choices in the care of persons with end-stage AD. 11. CNAs (as direct caregivers in SNFs) have some informal influence over the end-oflife treatment decisions made by families for residents with end-stage AD. 12. The level of literacy required for CNA certification training provides some assurance that continuing staff training developed at a similar reading level is also appropriate. 13. The SNFs identified by using the inclusion criteria of bed number, non-profit status, and use of hospice are similar to each other in ways that were important to the study purposes. Delimitations 14. The subjects in this study were limited to Certified Nursing Assistants (CNAs) in skilled nursing facilities (SNFs). 15. CNAs participated in the intervention at different times in their workday; some were after work, and others were prior to work or on a day off. 16. This research was not conducted in a controlled laboratory setting; therefore, setting characteristics such as staff turnover were inevitable.


17 Significance of the Study The conceptual framework in this research was defined as an ethic of care, developed from the biomedical ethics pr inciples of autonomy, beneficence, and nonmaleficence, and influenced by feminist et hics. An ethic of care (specifically, the everyday ethical care of dying persons with end-stage AD) was conceptualized as care that is person-centered, that uses the principles of palliative care and comfort care, and that respects the spoken wishes or written advance directives of these dying persons at a time when they can no longer direct their own care. Since CNAs provided up to 90% of direct care in skilled-nursing facilities, they were increasingly viewed as critical members of the care team to educate and support. Legislation has mandated our attention to advance directives for end-of-life treatment choices for persons in health care facilities. However, persons with AD and related disorders, due to fluctuating capacity, are often deprived of the right to have their choices known and followed. Researchers found this inequity is often caused by lack of knowledge about advanced and end-stage diseas e on the part of the direct care staff caregiver; and lack of psychological and informational support for caregivers as they confront these significant ethical issues in their workplace. Hospice has become the household word for a good death. Yet, hospice is significantly underused for persons with AD because of nursing home regulatory issues and reimbursement requirements. In addition, not enough is known about the elements of providing quality comfort care to dying persons who are cognitively impaired. The intent of the researcher in this study was to show that providing appropriate education and support (based on the conceptual framework of an ethic of care, and the underlying principle of the dignity of persons) would result in an increase in knowledge


18 about the ethical issues surrounding the treatment needs and care of dying persons with end-stage AD. The study was intended to show that providing support would enhance the education component, making a case for ongoing supportive or coaching sessions for CNAs. The study was also intended to show that variables within the CNA population should cause us to plan this education and support differently, taking into account the cultural and religious influences that make the ethics of everyday care for dying persons so personally challenging. Finally, outcomes of the study might give voice to the countless direct-care staff that are the companions of cognitively impaired dying persons in our nursing homes. These voices can then state that ethical care is important to them and that they value their role as key providers of everyday compassionate care. It was also conceivable that this study would point to needed changes in the hospice/facility care environment and would play a role in the development of adjustments to the restrictive hospice care Medicare benefit. Definition of Terms Advance Directive: a legal document, such as a living will or a health care proxy that described an individualÂ’s future treatment wishes in the event of incapacity (Walker et al., 1998). Advanced/end-stage of AlzheimerÂ’s disease : characterized by the individualÂ’s inability to communicate intelligibly, inability to ambulate, and need for complete assistance in all activities of daily living. And related disorders: references to AD also included less common dementias such as Vascular Dementia, Diffuse Lewy Body Disease, PickÂ’s, CJD, Binswangers, and others.


19 Comfort care : that which strengthened and empowered sick persons in relation to their own bodies, or in contrast with their discomfort. Direct care staff caregivers : those whose role in the nursing home required frequent, direct, sustained contact with reside nts, mostly the Certified Nursing Assistant (CNA). Ethnicity: self-described as Hispanic or Latino or Not Hispanic or Latino Family caregiver : the person(s) responsible for arranging and/or monitoring care of the person without compensation. Hospice : a comprehensive program of supportive and palliative services for dying persons and their families that includes physical, psychological, social, and spiritual dimensions of care. Hospice Influence : a facility with a hospice contract for at least the previous year and with at least one active hospice enrollee in the previous 30 days. Moral agency : the ability to judge the goodness of a proposed action, to choose a course of action, and to accept responsibility for what occurred as a result of that choice. Palliative care : the active, total care of persons whose disease was not responsive to curative treatment (WHO, 1990). Religiosity : the importance of active participation in an organized religious group.


20 CHAPTER 2 REVIEW OF THE LITERATURE A major outcome of our present extended life expectancy is the prolongation of morbidity. The majority of older Americans can expect to die from the consequences of a chronic illness. This review of the literature will summarize current thinking in ethics and end-of-life care and will review recent system-wide influences on end-of-life care, quality of that care, and cost issues, especially as related to persons with advanced AlzheimerÂ’s disease (AD). The review will describe the advanced stage of AD and the role of the nursing home and Certified Nursing Assistants (CNAs) in care delivery. Despite a dearth of research on the subject, this review will examine existing studies addressing advance directives and hospice care in the nursing home environment. Current research on treatment choices for persons with advanced AD will be presented as well as new approaches to palliative end-of-life care emerging from research. The literature review is intended to develop a case for an intervention study addressing ethics education and support for CNAs as the primary staff caregiv ers of persons with end-stage AD in nursing homes. Ethics and Long-Term Care Ethical Principles For the informed patient with decision-making capacity, the principle of autonomy is one of the fundamental principles underlyi ng medical care choices. This right to selfdetermination should be balanced with the principles of beneficence, nonmaleficence, and justice. In modern life it is much more typical for death to come after a long, chronic,


21 and debilitating illness. In the United States 78% of people live past their 65th birthday and more than 75% of them will be affected by the top five, cancer, stroke, heart disease, lung disease, or dementia (Hogan, Lunney, Gabel, & Lynn, 2001). Services to the chronically ill older adult are still fashioned from the rescue medicine philosophy of previous generations, and do not serve this new population well (Lynn, 2000). In addition to the important goal of relieving unnecessary suffering, we are also facing the need to address enormous increases in the numbers of chronically ill and dying persons. One difficulty presented by chronic illness is that of knowing when the patient is no longer seriously ill, and is dying. Finucane (1999) described this as a transition between gravely ill and fighting death to terminally ill and seeking peace. It is upon this fragile point on the continuum that this work focuses. The ethical principle of autonomy can be problematic in nursing home care at the end of life, although, it is fundamental to the concept of advance directives. Researchers exploring these challenges call for reliance on an ethical framework that stresses dignity and personhood (Carter, 2002; Kitwood, 1997; Tauber, 2003). Even when residents can still participate in some decisions, researchers find that health care professionals gradually withdraw from residents the invitation to decide for themselves. It appears that in a three-year ethnographic study in four nursing homes such removal from decisionmaking autonomy is a generic feature of the nursing home setting (Shawler, Rowles, & High, 2001). Describing residents as being informed rather than consulted, this study highlights a key feature about the culture of institutional life that depersonalizes and disregards, thereby, setting the stage for the same kind of impersonal end-of-life care.


22 Indeed, the standard for the application of ethical decision-making is substituted judgment, or recreating the patientÂ’s own choice, rather than best interest, which is what is best for the patient as judged by the proxy. Sullivan (2002) argued that patients should be allowed to choose what they are capable of choosing, and examined the danger of allowing the former autonomy of the persona with dementia to govern the personÂ’s current wishes. Claiming that autonomy is rarely present at the end-of-life due to illness and the very fact of impending death, Sullivan called for a change in community values and systems of care to a system that actually fosters a good death. Allen and colleagues (Allen, DeLaine, Chaplin, Marson, Bourgeois & Dijkstra et al., 2003), in a study of 78 nursing home residents in five facilities and their proxies, determined the need for engaging residents and families as well as staff in structured discussions about end-of-life care. The authors cautioned about the tendency to assume a lack of decisional capacity in individuals who are non-verbal. Advance Directives The Patient Self Determination Act (PSDA) of 1991 requires that persons seeking care in a health care organization that receives government funding be told of their right to execute an advance directive, which includes such documents as a living will, a durable power of attorney, or a health care proxy. After the PSDA, the number of persons with advance directives increased initially, and then leveled (Mezey et al., 1997). The spirit of the law, intended to address our understanding of end-of-life care and treatment choices, was abandoned to the letter of the law, leaving us with legislation that makes little difference besides paperwork. In order to describe how and where advance directives are summarized in a residentÂ’s record and how staff understand t hose directives, Cohen-Mansfield, Libin, and


23 Lipson (2003) analyzed data on 122 residents in a 550-bed nursing home. There was limited agreement on specific directives of individual residents when three locations for such information were examined: on the front of the residentÂ’s chart, inside the chart, and on the Minimum Data Set, a required resident documentation. Since almost 47% of decision making involving acute events during this study was handled by on-call physicians who did not personally know the residents, this lack of agreement on advance directives was found to be unacceptable and particularly so when a cognitively impaired resident is involved. The most common and widely used directives were found to be durable powers of attorney, living wills, and do-not-resuscitate (DNR) orders (CohenMansfield et al., 2003). End-of-Life Care and the Health Care System The SUPPORT Study The SUPPORT Study (Study to Understand Prognosis and Preferences for Outcomes and Risks of Treatment) was conducted in 1990, with subsequent phases and analysis of data in 1992 and 1993. Data were available for over 1,500 patients. More than 100 articles were published documenting this well-designed studyÂ’s disappointing findings. SUPPORT researchers found that, in acute care settings, physicians were often unaware of their patientsÂ’ wishes for care, severe pain was not at all unusual, and dying attached to a machine was unacceptably common (Lynn, 2000). SUPPORT Phase II was a controlled trial of an intervention intended to improve care given to seriously ill hospitalized patients. This phase, which was actually addressing key problem areas found earlier in the experimental group, did no better than the control group, and researchers drew the conclusion that simply providing information to physicians would not change the flawed process of end-of-life care in th ese targeted acute care settings. It was clear


24 that communication about resuscitation and prolonged mechanical ventilation was uncommon. In a later investigation that focused on identifying barriers to end-of-life communication, Hofmann and colleagues found that the majority of these same subjects also did not wish to discuss end-of-life issues, further limiting open dialogue (Hofmann, Wenger, Davis, Teno, Connors, & Desbiens et al., 1997). SUPPORT had several major outcomes, one of which was the development of the Last Acts Campaign, with funding from the Robert Wood Johnson Foundation, spurring wide community dialogue on end-of-life issues. Last Acts is a coalition of prominent professional and lay organizations which seeks to promote systemic reform of the behavior of physicians and other health care providers, payers of care, hospitals and nursing homes, and consumers in the area of end-of-life care. While SUPPORT did not focus on the nursing home setting, the study contributed to a heightened awareness about end-of-life care throughout the medical community. Effects of the Patient Self Determination Act (PSDA) Persons with AD will experience inevitable progressive cognitive decline. The nursing home resident population is at grave ri sk of losing autonomy and choice due to the prevalence of cognitive impairment as in AD. Advocates who seek improvement in advance care planning for nursing home residents call for a process of ongoing communication about end-of-life care, not merely a check-the-box approach to documentation (AlzheimerÂ’s Association, 2004; Forbes, Bern-Klug & Gessert, 2000; Teno & Lynn, 1996). A study conducted in four nursing homes in the Midwest, with careful attention to including diverse populations, held focus groups of family members of residents in the later stages of AD (Forbes et al., 2000). Five themes emerged to describe the experience of end-of-life decision making for these family members. Key


25 elements of those themes were overwhelming burden unrelieved by placement, undying guilt, violation of trust, pain of loss, and extreme changes in the caregiverÂ’s life. Although the downward trajectory of the disease was apparent, family members found it difficult to discuss end-of-life decisions, and some were completely unable to consider future treatment options. Focus on day-to-day care and activities remained a more comfortable and familiar place for families, and the lack of information from professionals confirmed this denial of the residentÂ’s deteriorating condition. Researchers summarized that family members were unable to conceptualize the dying process without assistance from professional caregivers, resulting in the fact that, even in the case of a specific advance directive, there were no assurances that the residentÂ’s wishes would be followed. In a study of over 9,000 seriously ill, hospitalized patients, Teno and colleagues interviewed or reviewed medical records to determine the effectiveness of written advance directives (Teno et al., 1997b). The time frame included two years prior to the PSDA and two years after the PSDA and included the Phase II data from SUPPORT in order to encompass these important large efforts to improve end-of-life decision-making. There were no significant differences in th e documentation about resuscitation decisionmaking and only 12% of patients with an advance directive talked with a physician when completing the document. Researchers concluded that simply increasing the frequency of advance directives was unlikely to have a meaningful effect on end-of-life care, and called for additional efforts to see that ongoing communication occur with persons who are dying and their families.


26 In a related study, Teno and colleagues also found that the advance care directives, even when part of the medical record, often did not guide medical decisionmaking beyond naming a surrogate (Teno, Licks, Lynn, Wender, Connors, & Phillips et al., 1997a). In nearly 50% of subjects in the study, care was inconsistent with advance directives. Other troubling factors included th e researchersÂ’ concern that a significant number of advance directive documents stated to exist were not actually part of the medical record, and that the wording of widely used advance directives documents was ambiguous and not helpful. In a random sample of Connecticut nursing homes, Walker, Blechman and Gruman (1998a) conducted structured telephone interviews with nursing home administrators and directors of nursing in 1991 and again in 1995 in order to determine the effect of the PSDA on such core elements as written advance directives, determination of advance directives, and iden tification of responsible staff. Significant increases in those three elements were reported between 1991 and the second wave in 1995. However, challenges observed in implementing the PSDA in nursing homes were further examined in a qualitative study by Walker, Bradley, Blechner, and Tetle (1998b). Three problems examined were the prevalence of cognitive disorders, the frequency of death, and the reality of the limited personal autonomy enjoyed by nursing home residents. Utilizing in-depth personal interviews, Walker and colleagues interviewed 19 health care professionals from the same five Connecticut facilities that were representative of a variety of disciplines. Experiences and perceptions were elicited from staffs that were closely involved with the implementation of the PSDA within their


27 organizations. A key finding was the pervasive presence of an institutional focus on compliance in the nursing home setting, which may well undermine the true goals of the PSDA that was intended to enhance resident autonomy. Additional findings included a strong focus on DNR Orders, the absence of a consistent approach to measuring resident capacity to make decisions, the lack of surrogate compliance with resident wishes, and serious limitations in the level of knowledge of end-of-life information among the staff who were expected to be guiding and supporting family discussions. It is also important to know more about the incidence and content of discussions about end-of-life issues between nursing home residents, their surrogates, and clinicians. Bradley, Peiris, and Wetle (1998) reviewed the documented discussions about future treatment decisions from 600 randomly se lected nursing home records during 1991 and 1994 in six Connecticut nursing homes. Although the number of nursing home residents discussing future treatment decisions increased 50% since the PSDA of 1990, 71.5% still had no documented discussions. Of the residents with documented discussions, 90% had only one initial discussion, and more than half of those discussed only life support, tube feeding, and ventilation without any broad di scussion of treatment preferences or proxy decisions. Researchers concluded that resident and family roles in end-of-life decisions were very limited. Cost of Care at the End of Life Financial burden for care at the end of life is enormous and the statistics are alarming. Medicare served more than 40 million beneficiaries in 2002, which will nearly double by 2030 (Centers for Medicare and Medicaid, 2002). While skilled nursing facility care cost Medicare $12.4 billion in 2001, Medicaid pays the bulk of nursing home care for those with dementia at a cost of $32,000 to $55,000 annually per person. Taylor


28 and Sloan (2000) found Medicare annual costs three times higher for persons with a claims-based diagnosis of AD. The 10-year out-of-pocket cost for care was estimated to be as high as $106,000 (Kinosian, Stallard, Lee, Woodbury, Zbrozek, & Glick, 2000). Nine billion dollars are spent annually prolonging the lives of people with advanced AlzheimerÂ’s disease using artificial nutrition and hydration (Post, Volicer, & Ortigara, 1999). While the financial burden may be staggering, the human cost of poor end-of-life care cannot be estimated. End-of-life researcher and crusader, Joan M. Teno, M.D., called for policy makersÂ’ attention to staff turnover and training, and regulatorsÂ’ attention to getting the right assessment information to improve care of the dying (Teno, 2002). Measuring Quality in End-of-Life Care In recent years, patient or consumer satisfaction scales have lost value because of the weak research methods that accompany some advertising claims about health care as well as other goods and services. Regardless of this erosion of meaning, the fact remains that patient satisfaction is the gold standard, and in end-of-life care it is the patientÂ’s expectations, wishes, and preferences that should take priority. The most important outcome of end-of-life care is how well h ealthcare providers meet the goals of the individual person and his family. The Toolkit of Instruments to Measure End-of-Life Care (TIME) was developed by researchers specifically to highlight the patient and family perspective and to be clinically meaningful in meeting the needs and wishes of dying persons (Teno, 1999). Widely disseminated via web sites, end-of-life research centers and the palliative care network, the Toolkit includes numerous validated and reliable measures of Quality of Life, Pain and Other Symptoms, Depression, Functional


29 Status, Survival Time and Aggressiveness of Care, Continuity of Care, Spirituality, Grief Assessment, and Caregiver and Family Experience Assessment. AlzheimerÂ’s disease or a related disorder adds another dimension to the task of measuring quality. There is intuitive acceptance of the health care providerÂ’s ability to affect quality of life for persons with dementing illness, but measurement is a challenge. Despite the difficulty, quality of life in dementia care has significance whether one is assessing the success of a service, judging the efficacy of an intervention, developing clinical guidelines, or concerned about maki ng end-of-life decisions. Typically, we rely upon either objective scales of function or proxy measures to measure quality of life in dementia care, believing that the affected individual cannot have a reliable and valid subjective response. The Dementia Quality of Life instrument (DQoL) (Brod, Stewart, Sands, & Walton, 1999) is one example of the effort underway to directly assess subjective quality of life in persons with mild to moderate dementia. The DQoL was found to be reliable by examination of item test-retest, internal-consistency, and 2-week scale test-retest reliability. In addition, construct validity was confirmed by correlation with the Global Deterioration Scale (GDS), a well accepted dementia-specific assessment tool (Reisberg, 1988b). Specific domains assessed using this instrument are self-esteem, positive affect and humor, negative affect, feelings of belonging, and sense of esthetics. The researchers recommended the use of well-accepted objective scales combined with the subjective areas directly assessed using the DQoL tool as a way to accomplish a comprehensive assessment. In order to address the issues of quality care in all settings, it is important to identify methods of assessment that address the challenges of measurement in a cognitively impaired subject.


30 In a theoretical analysis of the challenges inherent in defining, measuring, and understanding the quality of life of nursing home residents, Kane (2003) posited that the dangerous solution might be that quality of life is too elusive for us to ever capture, leading us to give up. Solutions that call for individualized assessments of how to achieve quality of life, and individual decisions by direct care staff to help a resident meet his own goals for a good death, are counter to current long-term care structure. Even the quality of life of residents with cognitive impairment could be assessed by a combination of multiple direct observations and careful personal interview. Assessing quality of life for persons dying with advanced AD or a related disorder adds another level of complexity. One approach can be to use responses made by persons with mild to moderate dementia as proxies for those in the later stages (Brod et al., 1999). Volicer and Bloom-Charette (1999) present a model of quality of life in advanced dementia that includes three intersecting factors: medical issues, psychiatric symptoms, and meaningful activities, all of which must be considered. Various existing observational scales focus on one central domain and thus have limited applicability to the broad concept of quality of life (Hurle y, Volicer, Hanrahan, Houde, & Volicer, 1992; Lawton, 1994). More recently, researchers from the team at the Geriatrics Research Education Clinical Center at the EN Rogers Veterans Hospital, Bedford, MA, developed and validated three outcome measures of care for persons in the terminal stage of dementia: Satisfaction with Care at the End-of-Life in Dementia, Symptom Management at the End-of-Life in Dementia, and Comfort Asse ssment in Dying with Dementia (Volicer, Hurley, & Blasi, 2001; Volicer, Hurley, & Blasi, 2003). Initially, the team developed the


31 Caregiver Assessment of End-of-Life in Dementia Questionnaire using longstanding validated instruments: Katz Index of Activ ities of Daily Living, the Bedford Alzheimer Nursing Scale–Severity Subscale, and the Caregiver Burden Inventory. This research team also constructed an Index of Informal Assistance to measure the utilization on nonprofessional help. The three new instruments were then developed from this matrix of caregiver information. The instruments have been applied to the comparison of dementia end-of-life care in various settings in order to provide a more realistic method of deciding the most appropriate level of care for individuals. Dying in the Nursing Home Dying in a nursing home, while a common and perhaps expected outcome for a frail, functionally dependent person, is not well supported in regulation or reimbursement. Regulators often view compromised nutrition, weight loss, and immobility as evidence of poor care rather than symptoms of approaching death. In this type of environment creating an atmosphere of death acceptance and transitional experience is almost impossible. Indeed, SNFs have neither the organizational mission nor operational design to provide end-of-life care (Mezey et al., 2002). Current reimbursement systems largely ignore the tremendous labor-intensive care needs of the frail and cognitively compromised elder, a fact which makes the kind of care people need nearly impossible for families to find or institutions to provide. Acute episodes and aggressive procedures and interventions continue to form the backbone of reimbursed care. New programs to increase volunteers or improve environments must battle regulation and bureaucracy in order to succeed. In order to better understand terminal care in the nursing home setting, 11 focus groups were conducted to elicit the experience of 77 nursing home staff that care for the


32 dying. The subjects were nursing home staff members and physicians. Three areas of the study were the death event, a good versus a bad death, and what physicians and staff can do to promote a good death. Reported barriers to good care were found to be lack of training, regulatory emphasis on rehabilitation, and a resource-poor setting (Hanson, Henderson, & Menon, 2002). The definition of a good death included individualized care by a consistent and effective team of caregivers, comprehensive care planning that addressed prognosis and emotional preparation for death, and appropriate use of medical interventions. Kayser-Jones (2002) studied two SNFs (N=312 beds) over 30 months. Data included participant observation and interviews with residents (n=35), family (n=52), nursing staff (n=66), and physicians (n=36). Even ts that related to residentsÂ’ dying were analyzed, and care of 117 terminally ill residents was observed. Analysis focused on both the dynamics of providing care and the process of dying. Kayer-Jones concluded that multiple factors influenced dying and one of the most significant factors was cognitive impairment. Residents with impaired cognition were neglected. Comfort care was poorly defined and consisted of leaving the resident alone with basic needs unmet. This conclusion highlighted the need for additional research on the dying experience for residents with AD, as well as the need for application of the principles of excellent palliative and end-of-life care for SNF residents. Hospice in the nursing home. Since 1989, the rules for the Medicare Hospice benefit have include residents of nursing homes with the number of nursing home hospice enrollees rising to 17% of the Medicare benefit total in 1995. Yet, as recently as 1998, 70% of nursing homes had no hospice patients (Petrisek & Mor, 1999).


33 In a telephone survey of administrators of randomly selected nursing homes in Missouri, Parker-Oliver and Bickel (2002) addressed the experience with hospice, measuring benefit to residents particularly in the area of pain management, and administratorÂ’s knowledge about hospice regulation and reimbursement issues. Results indicated that community and physician awareness were a barrier to broad hospice use, that access to hospice in rural areas was an issue, and that pain management skills were still low in nursing home staff. Only 36% of respondents (56% of Administrators and 20% of Directors of Nursing) understood reimbursement and regulation of hospice services. Open-ended survey questions indicated a lack of clarity about how hospice services work in nursing homes. Staff complaints generally referred to the fact that hospice should take total care of the resident. In order to explore what nursing home phys icians and hospice staff thought about hospice care in the nursing home, Keay, Cranmer, and Hanson (2002) conducted a survey of nursing home, hospice, and hospital-affiliated personnel and physicians in five states. Using primarily dichotomous questions with two long-answer questions, researchers inquired about whether nursing homes provide d the same end-of-life care as hospice services, whether there were any perceived problems in communication between hospices and nursing homes, and what contributed to the effectiveness of hospices in nursing homes. The 250 responses were analyzed according to professional role, then transcribed and categorized. The responses from medical directors were similar to those of hospice personnel and nursing home personnel. Responders felt that hospice made a difference and that the nursing home could not do the same quality of care without hospice. It was agreed that there were issues with integrating hospice care into the homes, especially


34 around communication. These researchers concluded that it was unlikely that medical directors in nursing homes were a major barrier to the presence of hospice services. Goals of care. The goals of care for persons with advanced dementia in a nursing home are not well understood or agreed upon. In a pilot study, Katz, Sidell, and Komaromy (2001) developed a questionnaire intended to capture the issue of physicianpatient surrogate agreement on quality of life, functional status, and use of invasive procedures. Twenty paired sets of answers were included in the study, which showed significant disagreement between surrogates and physicians on the value of specific invasive diagnostic procedures such as bronchoscopy, but less disagreement about Percutaneous Endoscopic Gastrostomy (PEG) tube placement, even when the procedure would not improve functional condition. A palliative care approach to the goals of care was described by Head (2003) in eight guiding principles: individuals are value d, all are treated with respect and dignity, communication is always possible, challenging behavior is an attempt to communicate, the same human needs exist across the life span, small interventions can help, the environment and spiritual care are important, and the patient with his family is the unit of care. Full discussion of the pros and cons of any treatment decision are helpful to families, along with support for the decision once it is made. At the end stage of AD, the co-morbidities of aging such as diabetes and heart disease complicate the goals of care and must be part of the discussion. Special diets and multiple medications are examples of the impact of chronicity. The burdens of tr eating infections are rarely considered in the race to order antibiotic therapy, which incurs potential restraints, recurring phlebotomy for blood work, additional medication interaction, and probable diarrhea.


35 AlzheimerÂ’s Disease AlzheimerÂ’s disease is a cause of death in over 100,000 Americans annually (AlzheimerÂ’s Association, 2004). The cause or causes remain unknown although the patterns of neuron degeneration are becoming clearer. The greatest single risk factor in the development of AD is advanced age. In persons age 85 and older, 47% are estimated to have AD. Post (2001), a Case Western Re serve Medical School Professor and Ethicist with a commitment to the personal and societal challenges of dementing illness, suggested that AD be regarded as the epidemic of our time. Improved public health, medical insurance for the elderly, and a rescue medicine mindset have given us a longevity that prolongs the morbidity of chronic diseases like AD, and we find ourselves largely unprepared for the decision challenges of aggressive intervention and treatment choices. Advanced AlzheimerÂ’s Disease The advanced stage of AlzheimerÂ’s disease is marked by several key characteristics, including the inability to ambulate without significant help, verbalization reduced to a few words, not recognizing familiar people, difficulty swallowing, needing help with all aspects of daily living, and inc ontinence. In the terminal or end stage of advanced dementia, the resident is bedfast, essentially mute, and dysphagic and often experiences intermittent infections related to the latter. Common causes of death are septic complications from immobility including skin breakdown, urinary tract infection, or lung infection (Volicer, 2001). Measures of the physical decline of the person with advanced AD are more straightforward and well known than assessment of psychological well-being. Volicer and colleagues studied 57 individuals with AD in SNFs who exhibited agitation or withdrawal with the goal of measuring verbal and non-verbal cues


36 to well-being (Volicer et al., 1999). The scales used were as follows: the Lawton Positive Affect Scale, specific visual analog scales, di rect observation logs of patient interest, and the Short form of the Cohen-Mansfield Agitation Inventory. Using correlation analysis the research team identified three dimensions of well-being in advanced AD: engagement-apathy, happy-sad, and calm-agitation. The characteristic losses of AD are not always experienced in an orderly fashion; in fact, there is significant individuality despite our attempts to categorize a continuum. The Global Deterioration Scale (GDS) (Reisberg, 1988), a well-accepted seven-stage instrument, combined with the functional el ements of the Functional Assessment Staging Tool (FAST), provides a measurement standard for advanced dementia accepted by Medicare for enrollment in hospice care. However, Schonwetter, Han, Small, Martin, Tope, and Haley (2003) conducted a Florida study to assess the validity of the Medicare hospice guidelines governing the eligibility of persons with dementia. Following 245 hospice patients and an 80-patient validation sample, which died with dementia over a two-year period, the researchers found no significant relationship between the Medicare guidelines and six-month survival for either sample group. Dying with AlzheimerÂ’s Disease In the United States one in four persons dies in a nursing home, and it is predicted that by 2040 that ratio will be one in two (Brock & Foley, 1998). Even though the bulk of dementia care occurs in the community, 75% of persons with dementia eventually reside in a nursing home. Dementing illnesses present challenges to the formal caregiver to provide care differently. In a study comparing treatment given to advanced dementia patients and cognitively well patients in hospitals with hip fractures or pneumonia, researchers found that the persons with dementia were given the same degree of invasive


37 treatment (blood tests, x-rays, and catheters) but only half the amount of morphine for pain. There were no plans for palliative care approaches in the advanced dementia group and 90% had no evidence of discussion a bout treatment goals or withholding lifesustaining treatment (Morrison, 2000). The point of the study was not to recommend reduced treatment, but rather to advocate for the full discussion and evaluation of appropriate care in this population. In order to better understand the elements of a good death in the SNF setting, researchers used a triangulated research design to identify 100 family caregivers of persons with late stage AD residing in SNFs. A final sample of 57 whose family member had died while residing at the SNF, participated in the study. The structured telephone interview included both nominal data and qualitative short answer descriptions about the healthcare decisions made during the last three days of life (Bosek, Lowry, Lindeman, Burck, & Gwyther, 2003). A clinically significant 28% of caregivers did not believe the person with AD could experience a good death in a SNF. Yet, researchers concluded that several elements in the SNF could be used to provide a good death for these residents, such as aggressive pain and symptom management, careful education of family about what to expect, environmental adaptation, and hospitality to visiting family members. SNFs are not organized with the intenti on of providing end-of-life care, nor are they organized to provide advanced dementia care. A retrospective study was designed to identify how many of the subjects 65 years or older died with advanced AD (n=1,609) or advanced cancer (n=883) within one year of admission to SNFs in New York State (Mitchell, Kiely, & Hamel, 2004). In analyzing the comparison, 1.1% of those with advanced AD were expected to die within 6 months, and 71% of them died within that


38 time frame. These residents with advanced AD were more likely to have aggressive interventions such as tube feeding, lab tests, and restraints, and also to have co-morbid conditions such as decubitus ulcers (15%), constipation (14%), pain (12%), and dyspnea (8%). Researchers concluded that optimal palliative care was not available to these residents. Hospice Care and AlzheimerÂ’s Disease It has been almost 20 years since hospice care in the United States became a Medicare benefit, strengthening the concept and practice of palliative care for dying older persons. Yet, in 2001, the Center for Medicare Education reported that most Americans over the age of 45 do not think of hospice care as a choice for end-of-life care, 75% do not know this care can be provided at home or in a nursing home, and 90% do not know that Medicare has a hospice benefit. Of all patients who die, 15% receive hospice care, while only 2% who receive hospice care have dementia as a primary diagnosis (Ahronheim, Morrison, Morris, Baskin, & Meier, 2000). AlzheimerÂ’s disease is a terminal illness and persons in the later stages or endstage are ideal candidates for hospice care because aggressive medical care is generally not preferred. Almost 10 years ago in a large survey study, Luchins and Hanrahan (1993) found that 90% of family and professional caregivers thought hospice care an appropriate choice for end-of-life care, but only 13% knew that hospice was available to persons with dementing illness. In 1995, less than 1% of hospice patients had a primary diagnosis of dementia (Hanrahan & Luchins, 1995). Christakis and Escarce (1996) found that 35% of dementia patients with hospice care exceeded the six-month prognosis of death, the required enrollment criterion. Volicer and colleagues are among several expert groups of


39 professionals engaged in a continuing quest to present well-documented rationale for palliative care for persons with end-stage dementia (Volicer & Hurley, 1998). The Value of Hospice Care In his work with the dying, Byock (1996) proposed a complex developmental model of landmarks and tasks for the end of life. Elements in this model include a sense of completion with the worldÂ’s affairs, a sense of meaning about oneÂ’s own life, and acceptance of the completion of oneÂ’s own life. How does this relate to the dying person whose cognitive status is seriously diminished? The value of hospice is that the hospice philosophy does embrace even the cognitively im paired, providing the most essential care for each individual. To explore the beliefs of families regarding hospice care in the nursing home, Baer and Hanson (2000) surveyed the family members of all deceased nursing home hospice enrollees over a six-month period in North Carolina. From the families surveyed, 292 respondents (73% response rate) believed that nursing home hospice improved quality of care for symptoms and reduced hospitalizations for acute episodes. While this was not a dementia-focused study, it does illustrate the specific value hospice care can bring to the individual dying in a nursing home setting. While many agree that the hospice concepts of comfort and compassion are important in end-stage dementia, there remains a notable lack of consensus among health care professionals about which essential concepts of hospice are actually important for persons with end-stage dementia (Kovach, Wilson, & Noonan, 1996). There is a very high acceptance of hospice care as appropriate in end-stage dementia, but a very low


40 level of knowledge about hospices that actually do serve dementia patients (Hanrahan, Raymond, McGowan, & Luchins, 1999). In an early investigation of actual enrollment in hospice care, a key finding was the problem in predicting survival time in persons with advanced dementia. A long-term follow-up study by Luchins, Hanrahan, and Litzenberg was initiated in 1998, using participants in nine Midwestern hospice programs, to evaluate the six-month prognostic criteria for hospice enrollment. By accurately predicting correlates to survival time, this research will allow for appropriate referral and enrollment of persons with advanced dementia in hospice, and thus enhance their palliative care options and contribute to their quality of life. According to the Health Care Financing Administration (HCFA), one of the problems in end-of-life care for elders with advanced dementia is correctly ascribed to the fact the Medicare hospice benefit was too rigid and too quickly adopted (Vladek, 1999). The care and management of end-stage AD and related dementing illnesses is summarized in a monograph from the Hospice and Palliative Nurses Association entitled, “Treatment of End-Stage Noncancer Diagnoses” (Smith, 2001). Hospice and Advanced Dementia Kovach and colleagues (1996) focused their experimental research project on nursing home residents with end-stage AD. The intervention included creation of “care households” of sixto eight-bed areas within the facility and educating staff about hospice concepts and elements of dementia care. Inclusion criteria were a diagnosis of irreversible dementia in the severe stage with accompanying symptoms. The treatment group of 35 received the household approach including therapeutic activity, meaningful human interaction with familiar people, and protocols in place to appropriately address infection and other symptoms. The control group of 37 received usual facility care.


41 Groups were pre-tested using instruments specifically designed to assess behaviors in dementia (BEHAVE-AD) (Reisberg, 1988), the DS-DAT, designed to measure discomfort in dementia (Hurley et al., 1992), and a checklist of physical symptoms. While at pre-test there were no significant differences between the two groups, at 2 months post-test there was a statistically lower discomfort level in the treatment group. A fundamental goal of palliative care is comfort; therefore, this finding was viewed as a critical outcome. There were also fewer be havior problems, although, the finding did not reach significance. The study did not show a reduction in physical symptoms, which the researchers believe could be due to the fact that the intervention did not focus exclusively on these areas. Evaluation of family and staff response to the intervention showed the staff to be more positive than family members. Increased job satisfaction, a greater sense of involvement with residents, and examples of functional improvement in some residents were among the beneficial outcomes noticed by staff. Family members described little of this improvement. A multifaceted survey study was conducted to determine whether hospice/palliative care was viewed as appropriate and a preferable option for persons with advanced dementia (Luchins et al., 1998). In one element of the study, three groups were surveyed: physicians caring for elders; gerontologists in nursing, education, and social work; and families of persons with dementia. A mailing list from the Gerontological Society of America was utilized with a focus on physicians, gerontologists, and, also, representative Alzheimer Association Chapters in different regions of the country. Of the 1,408 respondents (a 61% response rate), most were in some way experienced with advanced dementia. Significantly, only 13% were aware of the fact that hospice covers end-stage


42 dementia care, even though most were in favor of palliative care approaches in advanced disease. While 70% of physicians favored the home as the site for hospice care, 56% of families were likely to choose institutional hospice care. The most important features of the hospice benefit to the families surveyed were home visits and round-the-clock availability of medical staff; expertise in providing care; and practical assistance with equipment, medication, and supplies. Although this benefit was not widely appreciated, respite care is critical to some families. National Hospice Organization Guidelin es for Hospice for Noncancer Diagnoses The National Hospice Organization (NHO) guidelines for identifying dementia patients as appropriate for the Medicare Hospice Benefit can be a challenge and often cause non-referral or late referral (National Hospice Organization, 2000). Hanrahan and colleagues (1999) conducted a study of 45 hospice patients who were enrolled over two years. Twenty-four patients met the NHO enrollment criteria as having reached or surpassed Stage 7C of the Functional Assessment Staging Tool (FAST). The NHO guidelines were validated in identifying a group of people whose survival time was less than 6 months; these patients died within an average of 4.1 months. However, almost half the 45 patients (n=20) were identified as nonordinal; that is, their progression was not in order since they reached FAST 7C in mobility but only 7A in verbal ability and could not be scored on the FAST because it required stepped progression. Only 30% of these patients died within 6 months, with 10.7 months as the median. It is interesting to note that the use of antibiotics as an intervention did not significantly influence survival time in either group.


43 In an analysis of criteria for hospice care in dementia, Luchins, Hanrahan, and Murphy (1997) found that the application of Stage 7C of the Functional Assessment Staging Tool (FAST) as the key indicator for hospice enrollment was appropriate in cases where decline followed the ordinal pattern of the FAST. The findings showed that nonordinal patients needed to be reassessed monthly in order to recognize the window of opportunity for enrollment under current guidelines. This study also demonstrated significantly increased survival time for patients whose care plans called for the use of medication for acute illness. Policy Issues and Hospice Care It appears that nursing homes have few st aff with palliative care experience, and the nursing home use of hospice care is limited. Aspects of health policy limit resident access to palliative care and actually emphasize restorative and rehabilitative care reimbursement (Zerzan, Stearns, & Hanson, 2000). Nursing home reform actually requires that the goal of care is to achieve or maintain the highest practicable physical, social, and psychosocial level of each resident. The Resident Assessment Instrument (RAI), including the nine-page Minimum Data Set (MDS,) was developed to address quality of care by focusing standard evaluation on areas that may indicate treatable reversible conditions. It is clear that the same signs of potentially reversible conditions such as weight loss, functional decline, nonparticipation in activities, and dehydration are also part of the anticipated decline in terminal illness. The current nursing home survey system is based on this perspective of restorative care, leaving out any indicators to address standards of palliative care. In addition, the Medicare skilled nursing benefit covers many skilled therapies and facilities may maximize this number of days covered by this reimbursement level. It is important to note that the lower reimbursement rates are


44 for intensive personal care, symptom management, and emotional care. Palliative care for the dying would fall into the latter category, carrying a financial disincentive for the facility. Choosing between the Medicare skilled nursing benefit and the Medicare hospice benefit is not financially neutral. Since the Medicare skilled nursing benefit covers room and board as well as care, there is an incentive to the facility and the client to choose the skilled nursing benefit. If the person is also a Medicaid beneficiary, the care choice is financially neutral. The room and board cost must be 95% of the usual rate and is passed through the hospice provider to the facility. The adequacy of the pass-through amount becomes the financial issue for the facility. Additional issues that create barriers to the use of hospice services include a pervasive lack of knowledge about hospice admission criteria among clinicians and a continuing disinclination on the part of physicians to acknowledge terminal status (Ferrera-Reid, 2004). In the mid 1990Â’s,the office of the Inspector General instituted a broad investigation of Medicare fraud known as Op eration Restore Trust, which included the hospice benefit. The focus was on providers with longer lengths of stay in hospice, those with higher rates on non-cancer diagnoses, and large numbers of nursing home enrollees. The nursing home area was so problematic that elimination of the benefit was suggested. Fraud and abuse were found, but there were considerable flaws also found in the method of evaluating these hospice payments. Nevertheless, the outcome has been a general dampening of enthusiasm for the nursing home hospice approach. Encouragingly, there is a more recent groundswell of interest in nursing home culture change, and palliative care in end-stage AD may be a strong part of that change.


45 Barriers to Appropriate End-of-Life Care In a national study of 1,429 nursing homes, utilizing responses from 400 randomly selected respondents, Moss, Braunschwei g, and Rubenstein (2002) described Best Practices and Least Good Practices in the area of terminal care. Least Good included staff training around dying residents with dementia and emotional support for staff. The Best Practices listed included keeping the resident comfortable, treating residents like family, and keeping in close touch with the residentÂ’s family. The researchers drew an overall conclusion that there was a need for nursing homes to establish a terminal care approach or palliative philosophy in the midst of caring for residents who may not yet be at the end of their lives. It is also important to make this a regular part of staff education. In a prospective ethnographic study, Kayser-Jones (2003) identified elements of the nursing home environment that were barriers to the general implementation of palliative care principles. One hundred seventeen terminally ill residents from two large nursing homes were observed in the study for a time period of between one day of terminal care to as long as 15 months, with a mean of 45 days. Observations ranged from several times a week, to several times a day when death was imminent, and focused on the dynamics of providing care during the dying process. A systematic qualitative research method was used. The physical environment around dying residents included clutter, noise, inadequate resources, blaring televisions, and pervasive loneliness. Inadequate staffing and a lack of supervision characterized the settings, as well as a lack of basic nursing care such as oral hygiene, bathing, and provision of appropriate food and fluids. Important themes in end-of-life care, according to these researchers, were privacy and space, in order to support the dying process for both residents and families. Lack of respect was demonstrated by the noise and clutter as well as the lack of basic care. Although not


46 generalizable due to the small sample size, this study is an alarming yearlong review of real care settings. A review of the last 10 years of literature by Blasi, Hurley, and Volicer (2002) revealed substantial barriers to high quality end-of-life care for persons with advanced dementia. The barriers include measurement issues, inappropriate interventions, poor symptom management, and policy issues. In spite of these barriers, there are some examples of high quality end-of-life care for persons with dementia. Individual long-term care sites have demonstrated changes in outcomes for dying residents, including increased meaningful advanced directives documentation and improvement in pain scales related to ongoing comfort care. As examples, Jacob Perlow Hospice in New York City, focused on individualized care (Brenner, 1998) , and the Bedford GRECC utilized a fivestage level of care model, each stage moving toward comfort care only (Volicer, 2001). Treatment Decisions at the End of Life In a study of data from 279 persons who died with AD and 24 controls without AD, researchers examined severity of dementia, use of systemic antibiotics, defined as aggressive care, and use of narcotic and non-narcotic pain medications and nasal oxygen, defined as palliative care. Fifty-three per cent of subjects with AD and 46% without AD received systemic antibiotics; 14% with AD and 38% without AD received narcotic pain medication. Severity of dementia was not f ound to be a factor. Researchers concluded that for persons with end-stage AD the use of systemic antibiotics is prevalent at the end of life and the under use of pain medication may be problematic (Evers, Purohit, Perl, Khan, & Marin, 2002). Hospitalization


47 Morrison and Siu (2000), conducted a prospective cohort study to examine survival rates in persons age 70 years and older with end-stage dementia who were hospitalized for hip fracture or pneumonia compared with those who were cognitively intact. The measures included mortality, treatments directed at symptoms and the use of distressing or painful interventions during the hospitalization. Persons with end-stage dementia and hip fracture had a 55% chance of dying within 6 months compared to 12% in the cognitively intact; the same comparison in pneumonia was 53% versus 13%. Of these 118 patients, those with end-stage dementia received as many burdensome procedures as the cognitively intact, in fact bladder catheterization and mechanical restraint were more common in those with dementia. Patients with dementia and hip fracture had significantly fewer morphine equivalents per day, and only 24% had a standing order for analgesics. In addition, only 7% of patients with end-stage dementia had made a decision to forego lifesustaining treatment other than CPR. In a retrospective chart review process utilizing both qualitative and qualitative methods, Travis, Bernard, Dixon, McAuley, Loving, and McClanahan, (2001) examined the life experiences of 41 nursing home residents who died over an 18-month period. While most subjects received some palliative care, such care was clearly arrived at after indecision and gaps in that care approach, were interspersed with acute care interventions or happened in the last few days of life. Authors highlighted the dilemmas of regulation and reimbursement, barriers to good palliative care heard from every corner of the health care arena, as well as the lack of advance directives that are clearly understood by surrogates who are committed to acting as such. More specific findings included high hospitalization rates and emergency room in the last year of life, slightly higher for those


48 residents with dementia. Researchers reco mmended placing the resident on a living-dying trajectory in order to more appropriately identify goals of care that would incorporate the benefits of palliative care throughout the care continuum. Lynn (2000) listed the conditions the nursing home would need to fulfill to provide care for an acutely ill resident: the ready availability of competent nurses, accessibility of various drugs and treatments and routine tests, belief in a good death supported by the team, and advance care planning that is reliable and competent. It seems clear that our current nursing home environm ents often do not provide these conditions, and so hospitalization during an acute care need is frequent. In a two-year study of 103 nursing homes in a national chain, Fried and Mor (1998) found that physically frail residents least likely to benefit from hospitalization during an acute episode were those most likely to be transferred. An advance directive in any format did not lead to a lower rate of hospitalization. Those with severe functional impairment were most likely to be transferred, as were residents with respiratory, genitourinary, and circulatory system problems. This sample had a lower than 1% rate of Do-Not-Hospitalize (DNH) orders. Joanne Lynn, M.D., nationally known advocate for palliative care, and Director of the Center to Improve Care of the Dying, Arlington, VA, believes that transfers between settings are a cause of a loss of empathy and trust (Lynn, 2000), contributing to a general lack of confidence in society. In an effort to better understand the process of making treatment decisions, Bottrell, OÂ’Sullivan, Robbins, Mitty, and Mezey (2001) conducted focus sessions with three groups of nursing home Directors of Nursi ng (DON) in three states. Participants represented 13 homes, seven non-profit and six for-profit, evenly divided between


49 independent and corporate models. Bed capacity ranged from 59 to 300. These DONs were all women with an average of eight years experience in nursing. The focus questions were aimed toward residents who were dying and not enrolled in hospice. Researchers were exploring the elements that might influence a DONÂ’s decision-making about whether and when to hospitalize residents. The factors considered were residentÂ’s status, staff perception of the institutionÂ’s policies, nursing home resources, interactions with the various players in the decision, nursesÂ’ knowledge of patient and family, and nursesÂ’ personal beliefs about what constitutes appropriate care. Themes emerging from the focus groups were considered to be important in understanding transfer decisionmaking. Each DON in the study believed that their staff could provide better care for dying residents than the hospital staff and that it was the nurseÂ’s assessment of a condition change that precipitated most transfers. Researchers concluded that the transfers occurred more as the result of family decisions than resident wishes, perhaps due to the high prevalence of cognitive impairment, and that resident wishes were never separately identified. Additional findings included the interaction of nurses with physicians is a major determinant in transfer, as is family dysfunction during a crisis; and also, the number of professional nursing staff available to provide for sick residents impacts decisions to transfer. Responses differed in the matter of staffing levels; New York nurses cited staffing levels as an importa nt factor in transfer decisions, Ohio nurses denied its relevance, and California nurses did not address the topic at all. The presence of hospice services in the facility seemed to enhance nursesÂ’ ability to talk with families about end-of-life care. Even so, there was a divided response to the need to have hospice services available. Fear of litigation did surface as a factor in transfer decisions.


50 Recommendations included consistent discussions with families about short and longterm goals for care, improved assessment and communication of clinical information to physicians, and staff training on the early communication of palliative care principles to families. Hospitalization and hospice care. In a retrospective five-year study of deceased hospice patients, including 9,202 nursing home enrollees compared with 27,500 non-hospice-enrolled residents, Miller, Gozalo, and Mor (2001) found that 24% of hospice residents, compared with 44% of the non-hospice residents, were hospitalized in the last 30 days of life. In addition, when taking into account all the non-hospice deceased residents (N = 226,469), those in facilities where hospice was not involved at all had the highest hospitalization rate, while facilities where there was even a low to moderate use of hospice services had a lower hospitalization rate. Researchers drew the conclusion that the philosophy of palliative care had an effect on the end-of-life care practices of the institution. In a study designed to determine which features of hospice care would be most important to caregivers of persons with moderate dementia (n=15) and severe dementia (n=30), Casarett, Takasaka, Karlawish, Hirs chman, and Clark (2002) found that avoiding hospital admission and aggressive treatment was the most valuable among other highly valued services. This study goal was to determine ways to market hospice services to caregivers, but the result confirmed the importance of focusing on the aggressive interventions that concern family caregivers.


51 Artificial Nutrition and Hydration (ANH) Most Americans say they do not want to be tube fed if terminally ill, yet tube feeding of individuals with severe dementia persists. In a study as long ago as 1990, Volicer, Rheaume, Riley, Karner, and Glennon reported the successful re-establishment of natural oral feeding in five of six tube fed individuals with AD. Data regarding the effectiveness of ANH, including improved survival and wound healing or preventing aspiration, do not support the widespread implementation that exists. Tube feeding does not decrease the risk of aspiration in people with swallowing problems, is not considered a palliative measure, and was found to provide no benefit and actually may increase the risks of aspiration pneumonia, serious infection, and other adverse effects (Finucane, 1999; Finucane, Christmas & Travis, 1999). That tube feeding will eliminate aspiration, is a myth. In addition, tube feeding is done as a clinical procedure rather than a relational experience depriving the person of the taste and smell of food, as well as the contact of caregivers and other residents. Current research calls for more awareness of the burdens associated with tube feeding; the true pathology associated with terminal dehydration, and more open discussion of the medical, legal, and ethical aspects of foregoing this intervention. Complications of tube feeding include diarrhea, cramping, stoma infections, and pain from obstruction (Post, 2001; Post et al., 999). In a journal commentary response to an article that recommended percutaneous endoscopic gastrostomy (PEG) over nasoentric tube (NET) in patients with dementia having feeding difficulties, Volicer (2002) urge d the use of careful hand feeding as the preferred alternative. Hand feeding allows fo r human interaction, food taste preferences, and a more normal mealtime environment.


52 The issue of “starving the patient” is often mentioned to or by family members, and can be a strong encouragement to give proxy consent for feeding tube placement. Dying cancer patients who are not being tube fed report a lack of hunger as the body shuts down its processes (Post et al., 1999). It is well accepted in the commentary on foregoing lifesustaining treatment, that feeding tubes ar e a medical procedure and, as such, are governed by the same principles as other medical procedures. This means that competent persons, who no longer wish to have medical intervention, may refuse or withdraw treatment (Meisel, 1995), and those who no longer have decision-making ability, and when the standard surrogate election has been satisfied, may also refuse or withdraw legally. According to Meisel (1995), actual practice does not reflect this legal reality, and families as well as formal caregiving staff have difficulty implementing such a directive. In a large descriptive, cross-sectional study including almost 5,000 cognitively impaired nursing home residents, Gessert, Mo sier, Brown, and Frey (2000) reported that feeding tube decisions appeared to be influenced by nonclinical factors such as nonwhite race, urban location, age under 86 years, male gender, greater dependency, and absence of a living will. Meier, Ahronheim, Morris, Baskin-Lyons, and Morrison (2001) investigated the survival rates of persons with advanced dementia who were hospitalized and had tube feeding procedures implemented. Ninety-nine persons met the criteria for admission to the study; the major finding was that the feeding tube made no difference in survival rates. The study also showed that being African American and residing in a nursing home were the highest predictors of feeding tube placement in advanced dementia.


53 In a thoughtful article on the current thinking about feeding tubes and advanced dementia, Gillick (2000) offered a succinct review including the fact that there is no evidence that feeding tubes prevent aspiration. It is also reasonable to assume, relying on advanced cancer patients for information, that not eating when one has stopped desiring or being able to process food, is not painful. The morality of withholding nutrition and hydration, which often strongly influences families to do all they can to preserve life, is described as a balance between benefit and burden. More recently, Gillick (2002) presented current information on the issues families face when deciding how to address eating problems, outlining the question in end-st age dementia as a question about the path of dying rather than a choice between treating or not treating. Gillick pointed out the reality that irreversible eating problems in advanced AD signal the arrival of the final stage. Aspiration, and possible resulting pneumonia, raises the question about the appropriateness of antibiotic therapy as a treatment, since antibiotics are not the simple remedy we have come to think they are, especially in elders. Antibiotic Treatment A series of studies conducted a decade ago by the research team at the E.N. Rogers Veterans Hospital, Bedford, MA, (Hurley, Volicer, Pempusheski, & Fry, 1995; Hurley, Bottino, & Volicer, 1994) provided a foundation for more quantifiable research questions comparing the benefit and burden of antibiotic therapy in persons with advanced AD. Determining the appropriate drug to treat infection requires invasive tests, and evaluation of symptoms relies on observable or measurable signs rather than self-report. In addition, researchers found that the use of antibiotics in a person with advanced AD does not extend life, and in fact, may cause diarrhea, immune response suppression, adverse drug


54 reactions, and drug resistance (Volicer, Hu rley, Fabiszewski, Montgomery, & Volicer, 1993). There was no difference in observable patient comfort between the treatment group and those whose symptom management included antipyretics and analgesics. A decreased immune function and difficulty in diagnosing infection, as well as incontinence and aspiration, are all causes of intercurrent infection in dementia (Post et al., 1999). In another early study Hurley and colleagues compared the progression of AD in three groups of patients: those who developed a fever and received aggressive care, those with a fever who received palliative care, and those who did not develop a fever (Hurley et al., 1992). The purpose of the study was to explore the relationship between antibiotic therapy for the fever episodes and the progression of AD for these subjects. From an ethical perspective, withholding antibiotics could be unethical if such therapy would also minimize the AD symptoms. One hundred thirteen subjects met criteria that included probable AD, nursing home placement, survival of at least 3 months from admission, and informed surrogate consent. Fever episode was defined as greater than 100 degrees Fahrenheit rectally for at least 24 hours or antipyretic treatment for as long, fever not caused by a noninfectious complication (dehydration or impaction), and a fever from which the patient survived. Forty-nine persons developed a fever episode and 61% of those persons received aggressive treatment . The Bedford Alzheimer Nursing Severity Subscale (BANS-S) developed by Volicer (1997) was used because it successfully measured disease progression into the advanced stage. In this study, the BANS-S scores indicated that the entire sample exhibite d significant disease progression, not one group over another. As the conclusion, palliative treatment of infection was not associated with an acceleration of the progression of AD in this study; therefore, the researchers


55 concluded that aggressive treatment did not prevent decline. There are limitations to a study of this type; random assignment would have been unethical, and the subjects were all men and were somewhat younger than typical nursing home residents. There was not significant difference in age between the three groups, nor was there a difference in disease severity across the groups. The decisions related to fever treatment were made by surrogates who were acting on behalf of the known prior wishes of the person. In combining these results with their previous work, Hurley, Volicer, and Blasi (2000) concluded that the comfort of the febrile patient with advanced AD was served by palliative care without worsening the course of AD, and valuable medical care resources were saved. Researchers have concluded that it is possible to keep a person with endstage dementia that has an infection comfortable without using antibiotics (Post et al., 1999). Such non-aggressive treatments include acetaminophen for comfort and fever, fluid-containing foods like popsicles and cool compresses. Making Treatment Decisions Cohen-Mansfield and colleagues (2003) surveyed physicians and nurse practitioner decision-makers (N=9) to describe the decision process for 70 residents of a large non-profit nursing home. Hospitalization was the most frequent treatment considered and chosen. Family members were involved in 39% of decisions and nursing home staff were involved in 34%. The most important considerations self-reported by these decision makers were the residentsÂ’ quality of life, relative effectiveness of the treatment options, and the wishes of the family.


56 End-of-Life Care in the Nursing Home Setting The Direct-Care Worker Almost a million long-term care workers in the United States are employed as nursing department staff, that is, they have some responsibility regarding care for residents. The National Nursing Home Survey data indicate that over 65% of these nursing employees are Certified Nursing Assistants (CNAs). The work is hazardous, challenging, and often thankless. The national average hourly wage is $8.16, with many areas of the country barely meeting the minimum wage laws. At issue in long-term care settings in many states is an average turnover rate near 100% for CNAs, which is both costly and fragments care. Staffing levels are required to be sufficient without clear definition about what that means in most st ates. In Florida, mandated staff ratios were considered for repeal before they were even fully instituted. Many CNAs work part time or contingent without benefits, and some wo rk double shifts to cover for staff shortages (Gregory, 2001). Many work two jobs. The CNA is the backbone of the care plan, and he or she is at the heart of providing quality of care and quality of life for persons with advanced AD in nursing homes. Empowerment of CNAs in long-term care settings was found to be related to appreciation and supportive feedback from supervisors, others having faith in their work and trust in their judgment, and feelings of self-fulfillment (Campbell, 2003). This indepth interview study included five CNAs. It would be impossible to generalize the findings, but the outcome supports what we in tuitively know about direct-care workers in long-term care. Moss, Moss, Rubenstein, and Black (2003) identified the metaphor of “family” as an apt descriptor of the expressions and fee lings of SNF direct care staff regarding the


57 deaths of residents in their care. In a multistate ethnographic study of grief and loss, these researchers conducted in-depth interviews with 26 direct care providers in SNFs. Feelings were described in family language, staff described a deep sense of satisfaction when fulfilling these unassigned roles with dying residents, and they used the language of loss and bereavement in their interviews. Feelings included a lack of clarity in how to express feelings in front of residentsÂ’ families as well as a fear of being overwhelmed by loss, especially if personal family loss had been recent. There was no effort in this study to control for inclusion of cognitively impaired residents, and no conclusions were drawn specific to that population. The issue of education and training is a critical one. Using focus group methodology, Ersek, Kraybill, and Hansberry (2000) explored educational needs and concerns regarding end-of-life care of licensed staff and CNAs in nursing homes. Between five and 16 staff participated in each group with a total of 39 CNAs and 15 nurses that represented a diverse group both culturally and in their level of experience. Major needs and concerns of CNAs centered on licensed staffÂ’s unwillingness to consider their observations, dealing with family anger and criticism, uncertainty about how to address questions regarding the dying process, uncertainty about the CNAÂ’s role in discussing the goals of care with families, lack of sufficient time to provide care, distress at the death of residents, and concern a bout residents who die alone. There was strong concern from CNAs about feelings of conflic t between familiesÂ’ and residentsÂ’ goals for care, and a lack of knowledge about the goals of hospice and palliative care. CNAs also expressed confusion over their own attachment issues and the lack of facility support for them. CNA participants expressed their understanding of hospice as nothing will be done


58 for the person or as essentially giving up. CNAs expressed great distress at the limiting or withdrawal of certain treatments. Genera lly, CNAs felt unappreciated by licensed staff and families and their knowledge of resident preferences was disregarded. Issues of symptom management in residents with demen tia were significant. This study identified a need for discussion of ethical issues regarding end-of-life care. In a study of long-term care staff perceptions of institutional end-of-life care, Zimmerman, Sloane, Hanson, Mitchell, and Shy (2003) found the need for more staff education, a higher ratio of CNAs to residents, and more inclusion of volunteer roles were ranked highest. Researchers concluded th at staff desires to provide the highest quality care. The National Board for Certification of Hospice and Palliative Nurses has a specialty certification for nursing assistants who provide palliative care in various settings. This certified nursing assistant has more patient contact than any other discipline, that is, 80% to 100% of direct patient care. The Family Perspective In deciding on long-term care placement, family members generally believe they are abandoning their loved one and have failed. Staff in nursing facilities, especially CNAs, often feel like they are the family, since their contact with residents is so intense. Issues of attachment can enhance care for the resident, while at the same time the resulting true bereavement at the residentÂ’s death can have a negative effect on staff. Wilson and Daley (1998) interviewed eleven family members who had experienced the loss of a relative in a nursing home. The caring behaviors of staff and availability of spiritual support was most important to these families, illustrating one aspect of the potential for positive staff-family relationships. Addressing the use of certain kinds of


59 treatment interventions demands more purposeful decision-making and may result in conflict. The physician and family, for example, make decisions that the nurse and care staff must fulfill. Sometimes clear decisions are present such as DNR, but other aggressive interventions like gastrostomy tube placement requiring surgery may be approved without discussion or even review of the advance directives of the person. Responding to weight loss or nutritional risk indicators can often bypass the necessary decision-making discussion. To describe family perceptions of care at the end of life, researchers interviewed 461 family caregivers of older adults who died from chronic diseases (Hanson, Danis, & Garrett, 1997). Almost half died in hospitals, 28% in nursing homes, and 25% at home. Informants varied in their relationships with the deceased. While 46% of deaths involved a capable adult patient, in 23% of deaths, family informants believed there was no discussion with patient or family about treat ment options. The absence of discussion was not related to whether or not the decedent had a power of attorney or living will. Seventyeight percent of family members believed their loved one experienced pain, and 12% said more should have been done to address the pain situation. Ninety-one percent of comments about hospice were positive; negative comments included the high expectation hospice had of family members, and the poor training of some hospice workers. The proportion of comments was lowest for nursing home care (51%). Negative comments about terminal care centered on poorly trained or inattentive staff and disinterested and non-communicative physicians. Rarely did families focus on specific aggressive care decisions, but rather wanted more honesty about prognosis and quality of life.


60 Researchers concluded that it was not a family priority in this sample to gain control over decision-making, but rather a need for humane treatment and compassion. Juozapavicius and Weber (2001) conducted reflective individual in-depth interviews with 20 caregivers of persons with AD who died, in order to explore caregiver issues and to better understand transition out of the caregiver role. The interviews were organized into five sections: demographics, beginning the caregiving role, the caregiving experience including regrets and barriers, change of role at death, and advice to others. Important to note was the common reflection that the nursing home experience was negative, left them still feeling guilty, and while some staff were helpful, overall there was a lack of caring and attention. Researchers concluded that the guilt can lead to selfdoubt and can leave caregivers unresolved, angry, and conflicted. In a telephone survey of 121 nursing homes in Michigan, 99% of facilities did not provide materials to the family caregiver on either the grieving process or community bereavement resources. Seventy-six percent had no plan to refer to appropriate grief counseling when asked. An Emerging Culture of Care In 2000, an informal group of nursing home care advocates working together for more than a decade, formally organized the Pioneer Network for Culture Change in Long-Term Care, dedicated to the way resident s and staff live and work together (Pioneer Network, 2002; Fagan, Williams & Burger, 1997). The group was comprised of health care professionals, corporate provider representatives, long-term care advocates like the AlzheimerÂ’s Association, and policy makers dedicated to bringing about systemic change. A key early impact of the movement was a positive effect on the training and retention of direct care staff through the de velopment of CNA Associations and training seminars. Principles of individualized care and comfort care were incorporated into


61 Pioneer philosophy. Currently several states have “Pioneer movements” which are organized around the goals of personhood, dignity, and innovative best practices. Another emerging concept studied was a pre-hospice program, designed to provide transition into a formal hospice program (Hayley, Muir, Stocking, Houghan, & Sachs, 2001). The goal of this program was to address the fact that the average length of stay in hospice was 36 days, and 15% of patients were enrolled for less than 1 week before death, in a program designed to meet the needs of enrollees for up to 6 months (180 days). This pre-hospice program was delivered through a home health agency and was intended to prepare patients and their families for eventual hospice admission. The only criteria for admission to the preprogram were a terminal diagnosis and an interest in additional palliative support. There was no prognostic requirement. The study was done retrospectively in order to better understand the determinants of those who transferred into the hospice program and those who did not . Key findings from a chart review of 123 patients in a two-year period included, 31% transferred to hospice and 30% died while still enrolled in the pre-hospice program. Those who transferred were older, had secondary insurance, lived alone, and had some psychosocial issues upon admission. The outcome, only about 1/3 transferred to hospice from a program designed to facilitate such transfer, was important. Researchers concluded that the younger group may have had resistance to hospice from family members, and may have wanted in some way to keep their options for treatment open. The current challenges of providing hospice care under Medicare guidelines have been shaped by fraud issues, and so skew the goals and directions of good palliative care programs. This research had a limited sample, 30% of its original number fell out of the scope of the study, and there was little formal charting


62 of the decision-making process within the families. However, in the final analysis, effective end-of-life care was provided outside of hospice. Half of the group cared for outside of hospice eventually converted to formal hospice care. Five hospice households created within three nursing homes were the subject of a study by Wilson, Kovach, and Stearns (1996). The households were designed to accommodate eight residents with end-stage dementia, providing care in five main areas: comfort, quality of life, dignity, support for family, and support for staff. Indicators in all study areas were established and provided the framework for the model. There was no additional staff engaged even though the edu cational component for existing staff was intensive. Another example of palliative care in the nursing home is the collaboration between Genesis ElderCare and the University of Pe nnsylvania School of Nursing (Tuch, Parrish, & Romer, 2002). This three-year intervention study in six nursing homes in Maryland, intended to integrate palliative care expertise into the nursing home setting, was supported by a grant from the Robert Wood Johnson FoundationÂ’s Promoting Excellence at the End of Life Initiative. As a partnership between academia and industry, researchers used an experimental design with three matched pairs of nursing homes: one with residents receiving usual care, one with sta ff receiving an educational intervention, and the third receiving education plus the intervention of a palliative care team. In addition to the education, all four of the intervention homes received the support of a nurse palliative care expert. Ongoing data collection and c ontinued training of new staff made the partnership with the University critical. Existing evaluation tools were used, some were modified, and eventually the project focused on three measurable areas: pain


63 management, advance care planning, and psychosocial issues. Data are being collected around the concept of describing all deaths, so the death event is examined from every aspect to analyze the effects of the interventions. Barriers include staff turnover, fluctuating from corporate leadership, the thr eat of regulatory scrutiny, and a general lack of knowledge about palliative care in long-term care settings. The regulatory issues have been addressed proactively by directly educating the regulators about care issues and the palliative plans. Additional issues currently being addressed are the entire process of integrating the nursing home into the community and finding ways to encourage staff stability. The most successful piece seems to be the palliative care team model, and that is the area chosen for replication even before all the study data are completed. The highest importance was given to the need to have committed, knowledgeable professionals leading the project consistently to ensure buy-in and to enhance culture change. Community based nursing home diversion programs, like pilot examples in Florida and the PACE (Program for All-inclusive Aid to the Elderly) Programs in several states, provide a combined model of care which recognizes that acute care, rehabilitative care, preventive care, and palliative care co-exist on a continuum. These models are designed to keep people out of nursing homes, and do not address the institutional issues that affect end-of-life care. MediCaring, another program designed to respond to individual needs rather than bureaucratic eligibility criteria, is currently being piloted, and may provide a more broad applicability to the dying. This concept would provide capitated, comprehensive palliative care based on functional status rather than pre-defined disease groups (Lynn,


64 Wilkinson, Cohn, & Jones, 1998; Meier & Morrison, 1999). Nevertheless, the structure of health care services continues to be under great scrutiny and debate. Intervention models It is widely recognized that, while CNAs provide the bulk of care, they are seldom included in care discussions and rarely in treatment decisions. In a study by Burack and Chichin (2001), 35 CNAs who worked for a large urban nursing home took part in a structured support group experience. In groups of four to eight, these CNAs met for five one-hour sessions in facilitated discussion about their experience of caring for dying nursing home residents. A short questionnaire during session two included a Likert Scale measure of care of the dying and feelings about such care. A final questionnaire at the end of the study provided evaluation of the support sessions. Results of this investigation included the fact that these CNAs found it mo re distressing to work with residents who refused a feeding tube. The CNAs did not think these decisions were honoring the residentÂ’s wishes, but rather those of the fam ily. In spite of this, the CNAs did not object to working with dying residents or with those who were receiving comfort care only. The evaluations of the support sessions were very positive. However, since the participants were selected by their supervisors, specifically for this project, researchers recommended broader inclusion for the program and a program with a longer duration. Mezey and colleagues (1999) called for the education and training of direct care nursing home staff in the area of end-of-life care, palliative care, and spiritual support; and also urged that ways be found to better incorporate these workers into the health care team. In Oregon, a model known as the Oregon Practice Improvement cluster (OPIC), was formed to improve pain management throughout the community by creating a


65 consortium of nursing homes, board and care homes, and home health agencies working with senior nurse consultants (Mezey et al., 1999). The model included expert consultation, regular information sharing, continuous quality improvement, broad education, liaison with state agencies, and outcomes evaluation. In a study comparing hospitalization rates of residents in three Ontario nursing homes against matched residents in three other homes over a four year period, researchers found that those in facilities with formal advance directive protocols and end-of-life staff education had almost half the hospitalization rate and a significantly lower resource use in dollars (Molloy et al., 2000). Policy Development Focus will continue on the issue of end-of-life care in nursing homes. Current efforts are yielding results. The joint advisory board of the US Department of VeteranÂ’s Affairs and the AlzheimerÂ’s Association participated in focus groups, conducted a review of existing data and meetings with experts, as well as a national survey of caregivers of persons with AD who had died. A consensus policy recommendation was promulgated which included statements supporting the appropriateness of early palliative care, need for expanded access to dementia-specific care, an earlier focus on advance care planning, the need for training of acute care staff, and the enormous need for research-based practice in end-of-life care for persons with AD (Hurley et al., 2000).


66 CHAPTER 3 METHODOLOGY This research study was designed to measure and analyze the change in CNAs' knowledge about ethics of treatment decisions and attitudes about treatment choices for skilled nursing facility (SNF) residents with end-stage AD after the implementation of an educational program, and to measure and analyze any change after additional support sessions. Chapter 3 includes the following sections: research design, subjects, measures, procedures and protocols, and statistical analysis of the data. Research Design This study was a pre-test/post-test design with two levels of intervention. Subject CNAs were randomized into one of two treatment groups. CNAs in up to three similar SNFs were randomly assigned within each SNF either to the Education Group (EG) or to the Education/Support Group (ESG). The EG participated in the education program only (X1) and the ESG received the education program plus two structured support sessions (X2). Pre-testing occurred at the beginning of the education program for all subjects (O1). The first post-test occurred at approximately 2 weeks after the education for all subjects (O2). Two structured support sessions occurred for the ESG following the first post-test at weekly intervals. All subjects received a second post-test in Week 6 (O3). Figure 3-1. Research design R O1 X1 O2 X2 O3 R O1 X1 O2 O3


67 Intervention The intervention used in this study was an educational presentation called Standing By You, a 90-minute presentation using a didactic classroom-style method with lecture, overhead transparencies, and illustrative wall posters. Topics included 17. Characteristics of end-stage AD 18. Responding to the needs of the person with end-stage AD 19. Ethical principles 20. The decision-making process 21. Advance directives 22. Treatment choices for NH residents at the end of life 23. Comfort care for NH residents in end-stage AD The education program was presented in a classroom/meeting room with appropriate seating. To meet learning objectives, participants were to 24. Identify information about the stages of AD with a focus on the end-stage. 25. Participate in an exercise focused on the personhood of the resident with advanced AD. 26. Define the terms used in advance directive forms. 27. List the treatment choices frequently made in advanced/end-stage AD care. The support session component for this study provided two additional 50-minute class sessions. The goal of the support sessions was to reinforce the didactic material about advance directives and decision-making, which was presented in the education session. These structured support sessions included a review of terms and concepts from the Standing By You content. In order to maintain consistency, a generic case study was employed in the support sessions to illustrate the elements in the didactic presentation rather than using anecdotal stories in the sessions. Chairs set in the round were used in the support sessions to provide a supportive atmosphere. Also used were an overhead projector with transparencies to facilitate recall and a handout of the case study that was returned at the end of the session. A timeline for the intervention is given in Figure 3-2.


68 Table 3-1. Study timeline Subjects Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Pre-Test O1 All X Education Session X1 All X Post-Test O2 All X Support Sessions X2 ESG only X X Post-Test O3 All X Subjects The study subjects were a convenience sample of CNAs who worked in nonprofit SNFs, with bed number 100 or greater, and with an active hospice contract. Four study sites were originally selected and three became actual study sites. The established setting characteristics assured that the study sites were as equal as possible to each other in organizational structure. A bed size of 100 or greater ensured a staff pool sufficient to generate adequate subjects for this study. An active hospice contract was defined for this study as having at least one resident enrolled in a hospice program during the 60 days prior to study site selection, and having a formal relationship between a hospice program and the SNF for at least 1 year prior to study enrollment. The geographic location of all sites was a central Florida metropolitan area. One site was not included due to a staff medical emergency. One wing of SNF A, a 430-bed facility, was identified as the Phase 1 site (SNF-Ar). Different subjects were recruited for the Phase 1 reliability study (n=18). Subjects were then recruited for the Phase 2 or implementation phase of the study from the other three units of Site A, and from the other two study sites (N=60), (SNF-B and SNF-C). A CNA in a licensed SNF in the state of Florida must be certified by the Florida Agency for Health Care Administration, which requires 40 hours of training using a set


69 curriculum. In addition, the state of Florida requires four hours of dementia training for all care staff in SNFs. Dementia care has recently become part of required training for CNAs in Florida, but the training curriculu m includes only a few sentences on end-of-life care and ethical issues are barely mentioned. Subject inclusion criteria were: Certified Nursing Assistant, certified by the state of Florida. Currently working in direct care of SNF residents. Volunteer to be a study participant. Sign the approved Informed Consent Form. Subject Recruitment Directors of Nursing and Staff Developers of the identified study sites received a written invitation to have their CNA staff participate in the study. A follow-up personal telephone invitation was delivered, and a facility contact person was identified in each participating site. In SNF-B and SNF-C the Director of Nursing was the contact person, and in SNF-Ar and SNF-A the Staff Developm ent Director was the contact person for the study. Study subjects were recruited through the identified on-site contact person in the study sites using a recruitment letter directed to all eligible CNAs and distributed to eligible CNAs with the payroll envelope. Eligible CNAs were those who were certified in the state of Florida, and who worked in direct care of residents. Study subjects were recruited by an invitation and flyer using an eye-catching design to enhance subjectsÂ’ interest. Recruitment also included flyers posted on nursing units and in break rooms. The dates, times, and location for the Education program were included in the recruitment letter and flyer for the EG. The recruitment flyer had a tearoff portion with instructions for completing and placing in the Standing By You collection box placed on nursing units. The ESG dates and times were announced through a


70 personal flyer to those subjects who were randomly selected for that group. They received a reminder, delivered by the facility contact person, prior to each of the two support sessions. To encourage the retention of participants over the 6 weeks of the study was important. An attempt was made to over sample in all groups in order to reduce the effects of attrition during the study time frame. Attrition was anticipated due to personal circumstances, change in employment hours related to census, or job change. Subjects received incentives at several points in the study to encourage them to continue participation. Incentive steps included the education program, post-test, support sessions, final post-test. Incentives included movie coupons and telephone calling cards. A grocery or restaurant coupon was given to participants who completed the study. The total value of incentives did not exceed $50 per participant. In addition, refreshments were available at all sessions. Measures Knowledge about end-stage AD was measured by the 8-item AD Knowledge Section (Part I) of the Certified Nursing Assistant Questionnaire (CNA-Q) adapted for this study from the Ethics Questionnaire (Chichin et al., 2000). The answers were Yes, No, and DonÂ’t Know. Knowledge about ethics of treatment decisions in end-stage AD was measured by the 8-item Knowledge about Ethics of Treatment Decisions (Part II) of the CNA-Q. Parts I and II were scored with 1 point for each correct answer and no points for each incorrect answer for a maximum of 8 points on each part. A higher score represented greater knowledge of the topic. The parts were scored and analyzed separately.


71 Attitudes about treatment choices in end-stage AD were measured by the 18-item Attitudes about Treatment Choices in End-Stage AD Section (Part III) of the CNA-Q. The answers were agree, disagree or donÂ’t know. While attitudes remain a very personal aspect of thought and belief, for the purposes of measuring change in this study there were preferred answers to this section supported by current research and medical ethics principles. Preferred choices were 8 agree answers and 10 disagree answers. Part III was scored right or wrong according to the preferred answers as supported by current research in ethics and medicine, with a total of 18 points possible. The key outcomes measures were the frequency and direction of change among the three testing times, and the relationship of those changes to the key variables and the demographic variables. All donÂ’t know answers were recorded as incorrect. The CNA-Q instrument, which was the pre and posttest for the study, included a total of 34 questions. This instrument was provided to each participant in an easy-to-read format. Participants were instructed to circle answers directly on the instrument. Demographic variables were measured by the 7-item Demographics Questionnaire (DQ) that was developed for this study. The DQ included age, gender, race, ethnicity, number of years worked in long-term care of older adults, number of years in the US if not US-born, and whether the person believed that membership and active participation in an organized religious group was not important, somewhat important, or very important. The DQ was administered in a one-page check-the-box questionnaire format to all subjects. All ques tionnaires were coded by the participant with the subjectÂ’s same unique code number.


72 Operationalization of Variables The major variables were operationalized and measured as Knowledge about EndStage AD (Part I), Knowledge about Ethics of Treatment Decisions (Part II), and Attitudes about Treatment Choices in End-stage AD (Part III). Additional variables were developed to determine whether other factors affected the change over time, including number of years working as a CNA in the SNF setting, ethnicity, and religiosity as well as demographic variables. Years worked as a CNA with older adults was operationalized as a continuous variable. Ethnicity was operationalized in the DQ in two items as Ethnicity and Number of years living in the United States if not US -born. Ethnicity was operationalized as either Hispanic or Latino and coded as 1 or Not Hisp anic or Latino coded as 0. Number of years living in the US if not US born was a continuous variable. The literature on end-of-life care and ethics refers to the influence of cu lture in relation to end-of-life care. In the study setting, it was estimated that over 50% of the CNAs in long-term care were not born in the United States. In central Florida the number of CNAs with Hispanic or Latino backgrounds approaches 40%. Religiosity was operationalized in the DQ as a choice of whether the person saw membership and active participation in an organized religious group as very important, coded as 2, somewhat important, coded as 1,or not important, coded as 0. The research literature on spirituality and health care refers to the important influence of religion in relation to attitudes about end-of-life care. Additional demographic variables were age of the participating CNA, operationalized as a continuous variable; and gender of the participating CNA,


73 operationalized using a dichotomous variable with female coded as 1, and male coded as 0. Race of the participating CNA was operationalized as a categorical variable: Asian coded as 3, Black or African-American coded as 2, White coded as 1, other coded as 0. Validity and Reliability Face validity for the CNA-Q was determined by a review of the instrument by a panel of experts. Four subject matter experts, professionally experienced in either the area of biomedical ethics, end-stage AD, end-of -life care, or advance directives, were recruited. Experts reviewed CNA-Q for suitability in addressing the study topics and all agreed that the questionnaire was suitable for the topics and appropriate for the target population of CNAs. Comments included that of one expert who acknowledged the challenge inherent in measuring change in attitude but agreed that the instrument was suitable for this study. Eighteen subjects were recruited to establish reliability for the CNA-Q. CNAs who cared for persons with end-stage AD were recruited from one SNF selected for the study. The CNA-Q was administered to these CNAs twice, approximately 2 weeks apart. These subjects were recruited by personal invita tion with a payroll envelope to participate in a research study designed to establish a good questionnaire for use with CNAs. The CNA-Q was provided to volunteer CNAs by a volunteer member of the nursing staff at the SNF. The testing location was a classroom setting within the SNF. Questionnaires were coded in the same way as those of the study participants, using a number created by each individual and composed of the first digit of the birth month, first two digits of social security number, and last digit of the birth year. Test booklets were coded by the subjects using this number and were matched to corresponding re-tests using the code


74 number. The DQ was administered at this time and was coded using the same unique code. Subjects placed the completed and coded questionnaire and the coded DQ in an envelope, which was sealed and stored with the on-site assistant for the PI. The re-test occurred between 13 and 16 days after the first test to allow for CNA work schedules. Incentives were available to the subjects upon completion of the re-test. The instrument was found to have a reliabilty coefficient of 0.72. Protection of Human Subjects Expedited review was granted and approval of the study given by the Institutional Review Board (IRB), Health Sciences Center , University of Florida. The IRB approval assured that human subjects were protected during the implementation of this study. The Informed Consent procedure for study subjects was followed as outlined by the IRB. Procedures and Protocols Procedures and protocols for this study included reliability group protocol, subject assignment, education program protocol, edu cation/support protocol, post-test protocol, and data collection protocol. Reliability Protocol CNAs in Phase I served as the reliability group. These subjects answered the CNAQ and approximately 2 weeks later they repeated the same questionnaire. They also completed the DQ. All tests were conducted in a classroom setting under the direction of the PI. Subject Assignment Subjects volunteered by using a tear-off sec tion of the invitation or flyer. The signed tear-off was placed in thematically designed drop boxes located in several places at the study sites in order to maximize voluntary response. Standing By You was used as


75 the theme phrase to describe and identify the study. This was also the name of the education program. Subjects were randomized to EG or ESG using a sequential numbering of response forms with odd numbers assigned to EG and even numbers to ESG. A final sign-up date was identified, but exceptions for late sign-ups were made. A reminder flyer to all CNAs highlighting the Standing By You presentation was distributed after the initial invitation to boost interest. Eligible subjects who stopped by the class site at the time of the education program were randomized by the same process and were included in the study. Confidentiality was stressed in all materials and at the education and support sessions. Subjects were asked not to discuss the material among themselves outside of the sessions until the study conclusion. Each participant created a PIN composed of the number of the birth month, the first two digits of the social security number, and the last digit of the year of birth. This PIN was identifiable only by the principal investigator (PI). The key to the identification of subjects was created by the PI and was stored in a password-protected personal computer file with one copy in a locked file. Data collected were reviewed only by the research team and were reported only in the aggregate. The PI kept study data in a locked file. Reliability study data were stored and locked separately from intervention study data. Informed Consent was obtained from all study participants. Education Program Protocol The education program was presented in an appropriate classroom or meeting room in each study site. Refreshments were available for all sessions. The education program was offered up to twice in each SNF to assure an adequate sample in each group. Times were chosen to facilitate attendance by the highest possible number of CNAs. CNAs who


76 arrived on time to the session, but who had not responded using the tear-off response, were included and randomized to groups. The PI was the sole presenter in all sessions. The CNA-Q (34 items) was administered at the beginning of each education program as a pre-test. The DQ (7 items) was administered at the same time. The two-hour education program included at least 20 minutes for the pre-test using the CNA-Q. The education program itself was conducted in 90 minutes, covering seven modules of information about AD, ethics and treatment decisions in end-stage AD. The education program, presented by the PI, was didactic and included the use of overhead transparencies and posters. Education/Support Protocol For those CNAs randomized into the ESG, the support sessions followed the first post-test in Week 3 at approximately weekly intervals. One-half the study subjects were randomized into this group. These ESG subjects were notified in writing of their ESG group status and provided the dates and times of the support sessions. These CNAs were expected to participate in the ESG if they were randomized to that group. Both support sessions were required. Reminder flyers were distributed personally prior to each support session to highlight session dates and times. ESG subjects were asked not to discuss the support session content with other staff until the end of the study. The support sessions consisted of two follow-up structured discussion sessions, 1 week and 2 weeks after the first post-test, prepared and facilitated by the PI. Support session topics included a review of terms and concepts from the education program and a sample generic case study. Participation in the support session was limited to CNAs who completed the pre-test and participated in the Standing By You intervention . The support


77 session protocol included a sign-in form using the same coding system in order to document participation for comparison. Post-Test Protocol The post-test (02) was administered to all participants approximately 2 weeks after the intervention. A second post-test (O3) was administered to all participants approximately 1 week after the two Support Sessions were completed. Data Collection Protocol The pre-test and post-test were administered to individual study participants at the beginning of the education program as a pape r and pencil questionnaire test booklet. Each participant completed the test booklet that was coded using the PIN. The DQ was done at the time of the education program using a one-page form with check-off answer boxes. To facilitate data analysis using the SAS program, the PI entered information directly from the test booklet into an Excel spreadsheet. A research assistant randomly selected 15% (9 subjects) of the entered data to review for accuracy and one error was found and corrected. Statistical Analysis of Data Descriptive statistics such as frequency di stributions, means, and standard deviation were used to describe demographic characteristics of the sample and the questionnaire data. Repeated Measures Analysis of Variance was used to address each hypothesis in order to determine if there was a difference in the questionnaire scores between testing times and between education and education plus support groups. Chi-square tests were used to analyze the research question in order to determine if ethnicity, religiosity, or years working in long-term care affected the rate of change in knowledge about ethics of


78 treatment decisions and CNAs attitudes about treatment choices for SNF residents with end-stage AD after the implementation of an education intervention. Table 3-2. Major Study Variables Variables Measure Validity Reliability Knowledge about end-stage AD Knowledge about AD Section (Part I) of the CNA-Q Face validity established in Phase I Test-retest reliability established in Phase I Knowledge about the ethics of decision-making about treatment choices in endstage AD Knowledge about Ethics of Treatment Decisions in EndStage (Part II) of CNA-Q Face validity established in Phase I Test-retest reliability established in Phase I Attitudes about treatment choices in AD at the end-oflife Attitudes about Treatment Choices in End-Stage AD (Part III) of CNA-Q Face validity established in Phase I Test-retest reliability established in Phase I Sample Size and Power Considerations The sample was a convenience sample selected within similar settings. The reliability phase included 18 subjects and the implementation phase included 60 subjects. The sample size of N=60 participants was determined to have 78% power for detecting a medium effect size (0.5) in the change betw een preand post-test using a significance level of .05. Table 3-3. Hypothesis testing Hypotheses Statistical Analyses CNAs will have greater knowledge about the ethics of treatment choices in the care of persons with end-stage AD after the implementation of an educational intervention than before the intervention. Repeated measures MANOVA were used to analyze the change in knowledge after the intervention. CNAs will have a greater number of preferred attitudes about treatment choices in the care of persons with end-stage AD after the implementation of an educational intervention than before the intervention. Repeated measures MANOVA were used to analyze the change in attitudes after the intervention.


79 Table 3-3. Continued Hypotheses Statistical Analyses CNAs who have participated in the educational intervention and additional support intervention will have a greater increase in knowledge about ethics of treatment choices in end-stage AD than before the support intervention and greater than those who did not receive additional support. Repeated measures MANOVA were used to determine if the change in knowledge related to the added support. CNAs who have participated in the educational intervention and additional support intervention will have a greater increase in preferred attitudes about the treatment choices in end-stage AD than before the support intervention and greater than those who did not receive additional support. Repeated measures MANOVA were used to determine if the change in preferred attitudes related to the added support.


80 CHAPTER 4 RESULTS The purposes of this research were to measure and analyze the change in CNAsÂ’ knowledge about ethics of treatment deci sions and CNAsÂ’ attitudes about treatment choices for SNF residents with end-stage AD after the implementation of an education intervention. An additional purpose was to determine if added support sessions made a difference in the degree of change in knowledge and attitudes compared to those who received only the education intervention. From these purposes four research hypotheses were developed and tested. Additional research questions were addressed. This chapter contains a description of the SNF setting, demographics of the sample, questionnaire data, and a report of the results from analyses of these data. Description of the Setting The selected settings for this study implementation were three nonprofit SNFs located in metropolitan central Florida, SNF-A, SNF-B and SNF-C (Table 4-1). A subgroup of Facility A (SNF-Ar) was designated as the reliability study (N = 18), which was completed prior to the intervention phase of the study. Table 4-1. Description of the study settings Facility settings Study sample (n) Support sessions (n) Status Number of beds Active hospice SNF-A 26 13 Nonprofit 420 Yes SNF-B 25 12 Nonprofit 182 Yes SNF-C 9 5 Nonprofit 100 Yes


81 Demographic Characteristics of the Sample The sample was a convenience sample of CNAs currently working in three SNFs in metropolitan central Florida (N = 60). The sample was randomly assigned to one of two groups: Education Group (EG) which received the Education Session only (n = 30), and the Education/Support Group (ESG) which received the Education Session plus two structured Support Sessions (n = 30). Four subjects that originally completed informed consent did not participate in the full study and were not part of the sixty subjects. Three did not continue due to work schedules, and one did not continue in the job. Table 4-2. Frequencies and percentages of the sample characteristics (N = 60) Characteristics Frequency Percentage Sex Female 52 86.7 Male 8 13.3 Ethnicity Not Hispanic 56 93.3 Hispanic 4 6.7 Race Black/African American 44 73.3 White 5 8.3 Asian 3 5.0 Other 8 13.4 Race (collapsed to two categories) Black 44 73.3 Non-Blackb 16 26.7 Religiosity Religion very importanta 50 83.3 Religion somewhat importanta 6 10.0 Religion not importanta 4 6.7 Not born in the United States 36 60 Born in the United States 24 40 NOTE: aReligion defined as self-report of regular participation in a faith community bCombined White, Asian, Other The sample of CNAs was predominantly non-Hispanic and female (Table 4-2). Black or African-American subjects comprised almost three-fourths of the sample. The


82 collected data on race were collapsed to 73.3% black or African-American and 26.7% not black or African-American. The majority (83.3%) of the CNAs rated themselves as believing that regular participation in a faith community was very important, while the remainder believed faith community participation to be somewhat important or not important. The age of the CNA sample was older than expected, a range of 20 to 70 years with a mean of a little more than 45 years. Sixty percent of the CNAs were not born in the United States. Of this group, the years lived in the United States ranged from 4 to 40. The number of years each CNA spent working in long-term care was also very broad, ranging from 1 year to 35.. Ethnicity data were different from exp ectations. Hispanic persons make up about 30% of the CNA population throughout central Florida but only 6.7% of the sample was Hispanic. That the program was offered only in English may have been a barrier. It is also possible that the reference to end-of-life care on the flyer may have been a barrier to more religiously fundamentalist Hispanic persons. Data on ethnicity and race for the three SNFs in the study shows that the sample characteristics are not representative of the study sites in the category of ethnicity, which was defined for this study as either Hispanic/Latino or non-Hispanic/Latino. In th e demographic categories of Black/African American and Asian the sample is only slightly lower than the study sites (Table 4-3). Table 4-3. Comparison of ethnicity and race by study sites Study site % Hispanic %Black/African American % Asian SNF-A 16 73 10 SNF-B 10 86 5 SNF-C 21 78 3 Sample 7 73 5


83 The difference in the percentage of Hispanic/Latino subjects in the study may be related to the fact that the invitation to the education program was provided in English. Spanish-speaking CNAs may be reluctant to volunteer for a program that could be difficult to understand or during which their language challenges may be highlighted. Almost no CNA training is provided in Spanish in the long-term care system. The inferred requirement is that applicants to basic CNA training programs speak and write English despite the fact that their preferred language for conversation and reading is Spanish. In central Florida there is a robust Spanish language community including radio stations, print media, neighborhoods, faith communities and retail establishments. In exploring the Hispanic CNA experience for this research it was found that the language issue is also a barrier to both supervisory roles as well as to further study, perhaps in LPN or RN programs. Subjects not Born in the US Demographic analysis revealed that 60% of the sample was not born in the US. These subjects were predominantly Black/African American (86%) and it is likely these subjects were from a variety of countries th at immigrate to central Florida, such as Jamaica, Haiti, Bahamas, and the Dominican Republic. The culture, customs, and religiosity of each country are unique. Since over 13% of the sample (8 subjects) characterized themselves as other in the category of race, the true number in each racial category may be higher. Description of the Questionnaire Data The 34-item CNA-Q was divided into three subscales: Knowledge about end-stage AD, Knowledge about ethics of treatment deci sions in end-stage AD, and Attitudes about


84 treatment choices in end-stage AD. The C NA-Q means table (Table 4-4) and subscale means (Figure 4-1) show the overall improvement in scores across the three testing times for 60 subjects. Internal consistency was established for the CNA-Q by computing alpha coefficients for the entire questionnaire for each testing time (Time1 = 0.60, Time2 = 0. 73, Time3 = 0.68). The Knowledge about End-Stage AD subscale showed improvement in each of the 8 questions from Time1 to Time3, with information about the length of AD illness course, options for care of persons with end-stage AD, and the availability of physical therapy for persons with end-stage AD having the most improvement. The Knowledge about Ethics of Treatme nt Decisions subscale showed improved mean scores on each item, with greatest increases on the items that referred to the various powers of the health care agent in making treatment decisions for the resident with AD. Items showing a smaller gain in mean scores were at a 0.750 mean or higher at Time1. In the Attitudes about Treatment Choices in End-Stage AD subscale, every item showed an increased mean score in preferred answers over the three testing times. The most improvement was shown in the statements about the choice to stop a treatment, the choice to always feed even if that requires a feeding tube placement, belief that some treatments are too uncomfortable for the dying, and the fact that appropriate pain management may contribute to the individualÂ’s death even though that is not the intention. Subjects also showed increasing agreement in recognizing that it is not true that people want to be protected from knowledge of their own dying. There was less improvement in four statements that had a Time1 mean score of over 0.700. The main theme in these statements was autonomy; the statements focused on refusing life saving


85 treatment, letting others make decisions, pr eventing pain, and the belief that stopping a feeding tube is the same as killing someone. Table 4-4. CNA-Q means table (N = 60) Means Item Time1 Time2 Time3 Knowledge about end-stage AD (8 items) 1 0.667 0.833 0.917 2 0.933 0.900 0.933 3 0.783 0.833 0.867 4 0.617 0.683 0.800 5 0.650 0.783 0.817 6 0.700 0.850 0.850 7 0.483 0.700 0.683 8 0.933 0.983 0.950 Knowledge about ethics of treatment decisions in end-stage AD (8 items) 9 0.333 0.350 0.483 10 0.267 0.383 0.500 11 0.750 0.750 0.850 12 0.333 0.567 0.583 13 0.367 0.633 0.617 14 0.417 0.667 0.683 15 0.883 0.983 0.967 16 0.817 0.933 0.900 Attitudes about treatment choices in end-stage AD (18 items) 17 0.667 0.867 0.833 18 0.567 0.767 0.833 19 0.400 0.417 0.517 20 0.150 0.283 0.367 21 0.383 0.533 0.633 22 0.600 0.850 0.817 23 0.500 0.750 0.850 24 0.300 0.717 0.717 25 0.583 0.617 0.683 26 0.933 0.950 0.967 27 0.717 0.783 0.850 28 0.567 0.533 0.733 29 0.783 0.850 0.850 30 0.700 0.717 0.817 31 0.700 0.867 0.917 32 0.333 0.350 0.550 33 0.400 0.583 0.600 34 0.400 0.433 0.467 NOTE: The higher mean reflects preferred scores.


86 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Time 1Time 2Time 3 Knowledge about End-Stage AD Knowledge about Ethics of Decisions Attitudes about Treatment Choices Figure 4-1. Mean subscale scores The attitudes subscale contained two statements that were fundamental to an understanding of palliative care for persons with end-stage AD. These had little increase in mean score over the time of the study. The first statement addressed the issue of whether talking to people about a DNR would make them feel their condition is hopeless (Time1 – 0.583; Time3 – 0.683), and the second addressed whether letting patients die by not giving a treatment was the same as helping them to commit suicide (Time1,0.567; Time3, 0.733). Standing By You The education program Standing By You was a 90-minute didactic program presented in a classroom setting within the SNF study sites. The PI was the sole interventionist. The presentation text and overheads were the same in each education group and the support


87 session presentations were the same. CNAs were discouraged from talking about the education and/or support session until after the last day of the study. . The response of the groups of CNAs to the program was one of observable keen interest. Some of the facts presented were probably unfamiliar to the CNAs, particularly the various elements of treatment choices and the terms used in advance directives. The program may also have been personally challe nging in that the topic of death and dying is often difficult. Despite these issues, the attendance was excellent and participation was positive and enthusiastic. Hypothesis 1 Hypothesis 1 was that the CNAs would have greater knowledge about the ethics of treatment decisions in the care of persons with end-stage AD after the implementation of an education intervention than before the education. Data from the Knowledge about Ethics of Treatment Decisions in End-Stag e AD subscale of the CNA-Q were analyzed using Repeated Measures Analysis of Variance. Results indicated that there was a significant difference in level of knowledge about the ethics of treatment decisions in the care of persons with end-stage dementia after the education intervention compared to before the education (F = 18.05, p = <0.0001) (Table 4-5). Hypothesis 2 Hypothesis 2 stated that the CNAs would have a greater number of preferred attitudes about treatment choices in the care of persons with end-stage AD after the implementation of an educational intervention than before the education. Data from the Attitudes about Treatment Choices in E nd-Stage AD subscale of the CNA-Q were analyzed using Repeated Measures Analysis of Variance. Results indicated that there was a significant difference in number of preferred attitudes about treatment choices in

PAGE 100

88 the care of persons with end-stage AD after the implementation of the educational intervention compared with before the education (F = 29.16, p = <0.0001). Hypothesis 3 Hypothesis 3 further stated that CNAs who had participated in the educational intervention plus additional support sessions would have a greater increase in knowledge about ethics of treatment decisions in endstage AD than before the support sessions, and greater than those who did not receive the additional support sessions. Data from the Knowledge about Ethics of Treatment Deci sions in End-Stage AD subscale of the CNAQ were analyzed using Repeated Measures An alysis of Variance. Results indicated that there was no interaction between the groups related to additional support (F = 1.40, p = 0.2410). . There was, however, a sustained increase in mean scores between Time2 and Table 4-5. Impact of the Education Intervention and Education plus Support Sessions on Level of CNA Knowledge about Ethics of Treatment Decisions and the Number of Preferred Attitudes Among CNAs about Treatment Choices for Persons with End-Stage AD Change within subjects Effect of support sessions Time1 to Time 2 Time1 to Time3 Time2 to Time3 CNA knowledge about ethics of treatment decisions in end-stage AD F = 18.05 p = <0.0001 F = 0.17 p = 0.8439 F = 21.84 p = <0.0001 F = 29.79 p = <0.0001 F = 1.40 p = 0.2419 CNA number of preferred attitudes about treatment choices F = 29.16 p = <0.0001 F = 1.99 p = 0.1459 F = 24.50 p = <0.0001 F = 53.98 p = <0.0001 F = 5.22 p = 0.0260 Time3 in five out of the eight items in the subscale. The knowledge items that were not sustained through Time3 related to specific health care agent powers to not give

PAGE 101

89 permission for a feeding tube and the nursing home or physicianÂ’s power to override the residentÂ’s written wishes. Hypothesis 4 Hypothesis 4 went on to state that CNAs who had participated in the educational intervention and additional support sessions would have a greater increase in preferred attitudes about the treatment choices in end-stage AD than before the support sessions and greater than those who did not receive the additional support sessions. Data from the Attitudes about Treatment Choices in E nd-Stage AD subscale of the CNA-Q were analyzed using Repeated Measures Analysis of Variance. In this area of preferred attitudes there was a significant difference between Times2 and 3, indicating interaction between the education intervention alone and the education plus support sessions (F = 5.22, p = 0.0260). Other Findings Knowledge about End-Stage AD was the subject area of the first subscale of the CNA-Q, an instrument adapted for this study. Using Repeated Measures Analysis of Variance, results of analysis showed that there was a significant change in knowledge about end-stage AD after the education session (F = 16.03, p = <0.0001). There was no interaction showing that the pattern of change was the same in both the education only group and the group that participated in the support sessions (F = 0.44, p = 0.6487). Research Questions Research questions asked whether other factors affected the study results. Variables examined were ethnicity (Hispanic or not Hispanic), religiosity (self-described

PAGE 102

90 as regular participation in an organized faith group is important, somewhat important, or not important) and number of years working as a CNA in the long-term care setting. Group Comparison Independent Samples Chi-Square tests were used to compare the two groups (education only and education plus support) at baseline on the categories of ethnicity and religiosity as well as the demographic categories of gender and race. The groups were determined to be homogeneous (Table 4-6). There was no significant association between religiosity and group, ethnicity and group, gender and group, or the four categories of race and group. A t-test procedure was performed to analyze any difference between the education group and the education plus support group at baseline with regard to years worked as a CNA in long-term care. There was no significant difference between the groups (p = 0.7520). The group means were 13.6 and 12.8 years. Table 4-6. Group comparison Comparison DF p values Religiosity by group 2 p = 0.2844 Ethnicity by group 1 p = 0.2844 Gender by group 1 p = 0.6120 Race by group 3 p = 0.5327 Table 4-7. Means table for knowledge about ethics Variable N Mean (SD) Min – Max Time 1 Time 3 Time 1 Time 3 Group Education only 30 4.23 (1.30) 5.60 (1.65) 2-6 3-8 Education+support 30 4.10 (1.24) 5.57 (1.85) 2-7 2-8 Ethnicity Non-Hispanic 56 5.13 (1.25) 5.55 (1.73) 2-7 2-8 Hispanic 4 4.75 (1.50) 6.00 (2.16) 3-6 3-8 Religiosity (Rel) Not important 4 3.75 (0.50) 5.50 (1.73) 3-4 4-7 Somewhat imp 6 4.50 (1.76) 5.83 92.32) 2-6 3-8 Very imp 50 4.16(1.250 5.56 (1.70) 2-7 2-8

PAGE 103

91 Table 4-8. Means table for attitudes about treatment choices Variable N Mean (SD) Min – Max Time 1 Time 3 Time 1 Time 3 Group Education only 30 9.96 (2.50) 12.53 (3.40) 5-14 5-18 Education+support 30 9.40 (3.18) 13.47 (2.52) 4-18 7-18 Ethnicity Non-Hispanic 56 9.50 (2.84) 12.84 (3.02) 4-18 5-18 Hispanic 4 12.25 (1.50) 15.25 (1.71) 11-14 13-17 Religiosity (Rel) Not important 4 9.25 (2.99) 11.50 (3.70) 5-12 7-15 Somewhat imp 6 10.83 (2.14) 13.17 (4.26) 6-14 5-17 Very imp 50 9.58 (2.93) 13.10 (2.82) 4-18 5-18 Means were calculated for the Knowledge about Ethics of Treatment Decisions in End-Stage AD subscale (Table 4-7) and the Attitudes about Treatment Choices in EndStage AD subscale (Table 4-8). While the trend is up for all group means it is interesting to note that there remain some areas that were resistant to change. The data was also analyzed using analysis of variance: the groups were the education only group (n=30), the education plus support group (n=30), and the reliability group (N=18), acting as a no-contact cont rol group(Tables 4-9 and 4-10). In the Knowledge about Ethics of Treatment Decisions subscale there was no significant difference in the means between the three groups at pretest, and no change in the control group at posttest. In the Attitudes about Treatment Choices subscale there was no significant difference in the means between the three groups at pretest, and no change in the control group at posttest. This analysis supports the finding of significant change in both subscales resulting from the education intervention.

PAGE 104

92 Table 4-9. Means of three groups at Time 1 and Time 3 N Mean Std. Deviation Knowledge of Ethics Time 1 Education only 30 4.2333 1.30472 Educ + support 30 4.1000 1.24152 Control 18 4.2778 1.12749 Total 78 4.1923 1.22800 Knowledge of Ethics Time 3 Education only 30 5.6000 1.65258 Educ + support 30 5.5667 1.85106 Control 18 4.1667 1.54349 Total 78 5.2564 1.79057 Attitude about Treatment Time 1 Education only 30 9.9667 2.49805 Educ + support 30 9.4000 3.17968 Control 18 9.8889 2.65438 Total 78 9.7308 2.78993 Attitude about Treatment Time 3 Education only 30 12.5333 3.40115 Educ + support 30 13.4667 2.51524 Control 18 9.0556 2.04284 Total 78 12.0897 3.26009 Table 4-10. One-way ANOVAs to compare 3 groups at time 1 and time 3 Sum of Squares df F Sig . Knowledge of Ethics Time 1 Between Groups .438 2 .142 .86 8 Within Groups 115.678 75 Total 116.115 77 Knowledge of Ethics Time 3 Between Groups 27.805 2 4.760 .01 1 Within Groups 219.067 75 Total 246.872 77 Attitude about Treatment Time 1 Between Groups 5.402 2 .341 .71 2 Within Groups 593.944 75 Total 599.346 77 Attitude about Treatment Time 3 Between Groups 228.494 2 14.526 .00 0 Within Groups 589.878 75 Total 818.372 77

PAGE 105

93 CHAPTER 5 DISCUSSION AND RECOMMENDATIONS Standing By You , an ethics education program developed for this study, and the CNA-Q were used for this research. Data were collected from a convenience sample of CNAs in non-profit SNFs in central Florida. The purposes of this research were to measure and analyze the change in CNAsÂ’ knowledge about ethics of treatment decisions and CNAsÂ’ attitudes about treatment choices for SNF residents with end-stage AD after the implementation of an education intervention. An additional purpose was to determine if added support sessions made a difference in the change in knowledge and attitudes compared to those who received only the education intervention. Research questions addressed the interaction of ethnicity, religiosity and years worked in long-term care. Implications More Americans can anticipate time spent in a SNF, especially if they live past 85 years and are cognitively impaired. CNAs provide more than 90% of the hands-on care for persons in end-stage AD, often over months and even years of residential care. CNAs observe profound physical and cognitive decline in persons in their care, regularly interact with physicians and families, and are expected to provide care for these dying persons. It is widely accepted that the work of CNAs is physically difficult and emotionally demanding, but little research has been conducted to examine the possible burden of ethical or spiritual conflicts within the individual CNA. CNAs are not generally included in the end-of-life care discussions with families, or in the care planning process that occurs for each resident which may directly affect

PAGE 106

94 their care practices . Even the new, nationwide interest in improving end-of-life care has not significantly influenced the structure of the SNF sufficiently to highlight the knowledge needs and daily care challenges of the CNA staff. The data showed that there was a significant increase in knowledge about the ethics of treatment decisions in end-stage AD and a significant increase in the number of preferred attitudes about treatment choices in end-stage AD after the Standing By You educational intervention. However, this positive result, in both knowledge and attitudes, was not significantly increased after additional support sessions, although the initial increase was sustained through the 6-week time frame of the study. One important implication is that the subject of ethics, traditionally seen as subject matter for management and supervisory staff levels and those more academically prepared, can be presented to and received by staff members that have less formal educational preparation. CNAs, as moral agents, are responsible for their own actions and have a stake in the care provided to persons with end-stage AD. This moral agency of the individual CNA, along with the principles of everyday ethical care, provided the framework for this study. This same moral agency is required for the truly compassionate palliative or comfort care that is advocated for persons with end-stage AD. Change in Knowledge about Ethics of Treatment Decisions in End-Stage AD The Standing By You education program provided a didactic presentation that addressed both the formal ethical principles and the ethics of everyday care in the SNF setting. The program addressed concepts such as the health care agent and what that individual can and cannot do, identifying and defining terms like substituted judgment, ethics, values, advance directives, autonomy, beneficence, nonmaleficence, and everyday ethical care. Data analysis following the intervention showed significant increase in

PAGE 107

95 knowledge about ethics of treatment decisions in end-stage AD compared with scores at Time1 prior to the intervention. Eight of the 34 CNA-Q items were in the section labeled Knowledge about Ethics of Treatment Decisions in End-Stage AD. Time3 scores in this section were significantly higher than at Time1. It is important to recognize that CNAs, whose training does not include the ethical dimensions of treatment decisions, had significant positive change in knowledge around these issues. Further study can examine whether the change was in specific areas, and whether the knowledge can be maintained over time. Applied research can address the use of this new knowledge in the care setting. Change in Attitude about Treatment Choices in End-Stage AD The Standing By You Program provided an educational presentation about preferred treatment choices in end-stage AD such as stopping a feeding tube, treating pain, and who makes those decisions, as well as Do Not Resuscitate (DNR) orders and the concept of a person’s right to refuse treatment and how that is complicated by AD. Answers were identified as “preferred” answers based on current research and practice in end-of-life care. A case study of a typical long-term resident with end-stage AD was woven throughout the educational program. Data analysis showed a significant increase in number of preferred attitudes about treatment choices in end-stage AD compared with scores at Time1. The Attitudes about Treatment Choices in End-Stage AD section of the CNA-Q was comprised of 18 items. The Standing By You presentation provided a careful discussion of treatment choices without emphasizing an agenda even though there were preferred answers. For example, information was presented about the mechanism and efficacy of gastrostomy tube feedings in such a way that the CNA could see the role of his own values, the range

PAGE 108

96 of values a person with end-stage AD might have had, and the potential competing values of family members. An education intervention developed to contribute to a sustained positive change in preferred attitudes is very important for future work with CNAs if we are to bring evidence-based practice to the SNF setting. Additional Support Sessions CNAs that were randomly selected in each SNF study site participated in two onehour support sessions. These structured support sessions did not significantly enhance the scores for level of knowledge although significance was sustained through the 6 weeks of the study, from Time1 through Time3. Results may be due to the fact that the significant increase in both knowledge and attitudes reached a maximum point after the educational intervention for the time allotted to this study. Sessions were timed within tw o and three weeks after the initial education session, and significance was maintained over that time frame. In addition, the nature of this study required a structured support session. A more responsive, coaching model allowing for instructor interaction and questions about specific issues may actually have elicited additional knowledge or attitude change. This active coaching model is currently a key component of the culture change movement in long term care in central Florida and elsewhere in the US. This approach is far diffe rent from the didactic training model of the past and prepares CNAs as team leaders who are fully invested in the planning and decision making process for resident care. Level of Knowledge of AD The first eight questions (Section I) on the CNA-Q were about AD and reflected the focus of the study on care for persons with end-stage AD. A hypothesis was not developed specifically for this section since a more robust presentation of AD

PAGE 109

97 information would be preferred in order to draw conclusions about level of knowledge of this subject. However, scores on this section, Knowledge about End-Stage AD, significantly improved after the Standing By You Program. Although there was no significant change related to the additional support sessions, the positive change was sustained over the study time frame from Time1 through Time3. The Certified Nursing Assistants The sample was 87% female which was representative of the whole population of CNAs in each facility. However, the convenience sample in this study presented other unexpected elements for this research. An informal telephone survey of six large central Florida SNFs, including the study sites, indicated that an average of 30 – 35% of CNAs are Hispanic, yet only 7% of the convenience sample in the study described themselves as Hispanic. It is possible that a study invitation and program presented in English and not also presented in Spanish would discourage CNAs who are Hispanic from participating. It is unknown whether other voluntary opportunities for learning related to their job responsibilities are bypassed by CNAs due to the language difference. It is clear that this study did not include the anticipated balance of Hispanic subjects and therefore the results cannot be generalized to that population. In addition to Hispanics from Puerto Rico, Cuba, and Central America, central Florida is also home to people from Haiti, Jamaica, the Bahamas and other Caribbean countries. The study sample was 73.3% black or African American and 60% of the subjects were not born in the United States. The racial and ethnic breakdown did provide an anticipated rich diversity of cultural backgrounds and values to this research. The significant findings in both the area of knowledge and attitudes in this sample implies that

PAGE 110

98 educational programs in ethics can be successful in a culturally diverse group, even though it may be important to provide discussion and information in Spanish. Religiosity was defined as the level of importance placed on regular participation in an organzed religious group. The level of very important was self-identified by 83.3% of the study sample. The topics of personal values, ethical care, and end-of-life issues were likely to be important to a group with high religiosity, since these topics are within the realm of spirituality and faith experience. Homogeneity in religiosity within this group may have influenced the lack of significant findings on this variable, or changes may have occurred independent of religiosity. A larger sample with more variability in religiosity would increase knowledge in this area. It is also true that believing in the importance, as stated in the DQ, may not be the same as participating in a religious group or organization. Further study on this definition of religiosity and its possible influence in the end-of-life care practices of CNAs is recommended. The data included information on the number of years subjects worked in the longterm care setting in order to explore the informally held notion that the longer people work in a setting the harder it is for them to adopt new ideas. The sample ranged from one year to 35 years working in the long-term setting. Findings described significant change only for attitude, and only at Time3, indicating that longer time workers can indeed adopt new attitudes. This is an important finding in that it challenges the educational apathy that sometimes exists at the long-term care management level. Although there was a significant change in knowledge and attitudes related to race, there was not a specific group interaction, indicating that there may not be a need to

PAGE 111

99 target ethics education to specific racial groups. Again, larger, diverse samples are needed for further study in this area. The Experience of Implementing Standing By You The Standing By You educational intervention was didactic and structured, as were the two one-hour support sessions. The presentation delivery was simple and straightforward using a case study-type example to illustrate the concepts and definitions. The manner in which CNAs responded to the program was a profound experience for the researcher. The learning atmosphere was marked by a very high level of interest among the CNAs who participated. Even though there were incentives provided, these appeared not to be the most important features of th e research. It was apparent by observed facial expression and head nodding that the subjects actively recognized the ethics elements presented and very definitely connected with the case example. In fact, the subjectsÂ’ energy around the educational program was also conveyed by facility contacts in each setting. This positive interpersonal response from CNAs supports the development of Standing By You into a formal program offered to SNFs as part of their continuing education for direct care staff and may actually be the basis for similar presentations to professional staff in SNFs. A large amount of CNA education and training is mandated by state statutes and often described by CNAs as boring and routine. Such training may also hold low status in the minds of professional staff perpetuating this culture. Discussion about ethical and values-cente red care at the end-of-life can provide personal growth and discovery for the individual learner as well as directly affect care for residents. This study did not address the ways in which changes in knowledge and attitudes around ethics and treatment decisions could affect care but it is clear from the

PAGE 112

100 movement toward culture change in long-term care that these changes are important to many stakeholders (Fagan, Williams, & Burger, 1997). The New Culture of Residential Long-Term Care In daily practice, the CNA must participate in the provision of care with or without an understanding of the origin or the rationale around treatment decisions. It is not reasonable to ascribe to the CNA a level of professional judgment required of physicians and nurses who make care decisions. However, it is important to recognize the role of CNAs in the care of SNF residents who are dying, and most especially to focus on the many residents who will die with AD or a re lated disorder. Highly trained professional staff need critical and timely information from direct-care workers on the front line, and the front line workers need to be involved in decisions and understand the care rationale (Bowers, 2001), most especially at the end-of -life when those decisions are so fraught with meaning and consequences. The new culture of long-term care emerging across our country was an underlying energy in this research because so much depends on the CNA in the everyday care setting. The philosophy of personhood (Kitwood, 1997) informs and shapes appropriate care for persons with AD. Inviting the everyday care practitioner to become engaged with the person, and then relate this philosophy to his or her own values, provides a true connection between values and practice. This connection is what will ultimately change by forming our beliefs out of sound principles and then calling us to practice what we believe. It is certainly within the realm of professional nursing to work toward long-term care settings in which principled care is empowered at all levels. Alzheimer care is fundamentally relational care, hard to measure and evaluate but required in order to

PAGE 113

101 address the essential needs of the person who cannot direct his own care (Kitwood, 1997). Unloving care was a descriptor of nursing homes coined in the 1980s and as we continue to try to create the loving care we envision for the future in our long-term residential care settings, the emerging models are very adaptable to the principles of palliative care that are part of the end-of-life care movement (Eaton, 2000; Thomas, 1992). Dr. Thomas’ Eden Alternative (1996) has invited a whole new approach to an environment of living. Relationships precede tasks in this new culture of care (Misiorski, 2001), and it is only when we educate CNAs at a level where their values and ethics are recognized and incorporated can they become invested in the future vision of change. Recommendations To improve care for SNF residents who have end-stage AD it will take more than a few sessions of education for direct care sta ff. Simply mandating a new approach is not enough as noted by the experiences with mandating advance directives by the PSDA. That action has failed to adequately convert aging individuals or their families to be proactive planners of their own future care. We also know that the system is bowed by regulations that actually interfere with care, discouraging innovation and worker autonomy. Do we really “lack the moral, legal and ethical consensus as to when someone with dementia has begun the dying process, or how to manage symptoms related to pain and nutritional status in persons with dementia?” (Mezey et al., 2002, p. 11). By embracing the relational ethic of care and focusing on the true principles of palliative care, nursing home care for persons with end-stage AD can be improved. Cassell and Foley (1999) outlined the Core Principles of Palliative Care which were adopted by most major healthcare specialties, including the American Medical

PAGE 114

102 Association, signaling a step forward in this previously neglected or avoided area of medicine. Key elements of these palliative care principles that apply to those with endstage dementia include: 28. A focus on dignity and the individual’s wishes 29. Pain relief and comfort, 30. Assessment and alleviation of new problems 31. Continuity of care 32. Alternative therapies as appropriate 33. Access to hospice care 34. Awareness of the burden-benefit ratio in treatment choices 35. Use of evidence-based care These elements can and should be accomplished as the goals of care are identified for residents with end-stage AD. Furthermore, the findings of this study indicate that CNAs, whose everyday responsibility is this very important care, grasped and retained the ideas of palliative care when provided with a focused learning experience. The research has found that a major barrier to a good death is a lack of training and poorly defined comfort care (Hanson et al., 2002). Most SNFs do not have ethics committees or any mechanism for ongoing, formal discussion of ethical issues, some of which are a daily occurrence (Kelly, 2000). In a central Florida survey study (N = 132), only 9% of SNFs had an ethics committee and several of those were in their first year of operation. We should strive for education at every staff level, facilities that engage in Ethics Committee discussion, and open dialogue with families about the goals of care in light of the resident’s wishes – all these working together will enhance the new culture of care in nursing homes. Future Research End-of-life care in the nursing home setting is a rich area for nursing research. Standing By You is a basic educational intervention that can be incorporated into a SNF

PAGE 115

103 training calendar, and easily adapted to the learning needs of other staff levels. Future research objectives for this program include expanding discussion around the questions that arise for participants in the educational session, translation and presentation in Spanish, support sessions that have an agenda set by the participants, adaptation to the assisted living facility (ALF) setting, and development of a module for family participation. Item analysis of the CNA-Q was not the focus of this study but is a future goal. Analysis of each question in order to better understand the data from this study may lead to a more robust conclusion about the knowledge and attitude changes for these CNA participants. Another key area for future research is that of incorporating palliative care principles into the every day life of a nursing home. One goal is to eliminate the prevalent denial aspects and assure appropriate comfort care for all residents, especially those who are dying. By educating staff and then measuring key outcomes observed in the daily life in the facility, a research agenda could become an ongoing commitment. By linking this relationship with an academic setting a substantive benefit could be realized for residents, staff, students, and the community. Specific critical indicators in nursing home surveys can also be incorporated into this future research conversation. Falls, nutrition issues, and behavior concerns are some areas where a newly energized and committed direct care staff could make a positive difference. Staff retention, community volunt eer participation and survey status are organizational areas that could benefit from a focus on relational care.

PAGE 116

104 APPENDIX A CERTIFIED NURSING ASSISTANT QUESTIONNAIRE Part 1: End-Stage AlzheimerÂ’s Disease (1-8) 1. The person with AD in the nursing home may have had the disease for more than 8 years already. 1. Yes 2. No 3. DonÂ’t Know 2. In the early stage, AD causes memory loss, personality changes, and poor judgment. 1. Yes 2. No 3. DonÂ’t Know 3. In the middle stage, persons with AD usually cannot walk. 1. Yes 2. No 3. DonÂ’t Know 4. In the final or end-stage of AD, the only place for care is the nursing home. 1. Yes 2. No 3. DonÂ’t Know 5. Hospice care may be a choice for persons with end-stage AD, since it is a terminal disease. 1. Yes 2. No 3. DonÂ’t Know 6. Since persons with end-stage AD cannot talk much, it must be true that they donÂ’t hear or understand what we say. 1. Yes 2. No 3. DonÂ’t Know 7. End-stage AD care may include physical therapy. 1. Yes 2. No 3. DonÂ’t Know

PAGE 117

105 8. Families often feel sad and guilty that they cannot provide care at home. 1. Yes 2. No 3. DonÂ’t Know Part 2: Knowledge about Ethics of Treatment Decisions (9-16) 9. Can the health care agent (the person named in the health care proxy form) make financial decisions for the resident? 1. Yes 2. No 3. DonÂ’t Know 10. Can the health care agent tell the doctor not to give the resident antibiotics when the resident has pneumonia? 1. Yes 2. No 3. DonÂ’t Know 11. Can the health care agent sell the residentÂ’s home or apartment? 1. Yes 2. No 3. DonÂ’t Know 12. Can a health care agent ask the doctor to stop a feeding tube once it has already begun? 1. Yes 2. No 3. DonÂ’t Know 13. Can a health care agent ask the doctor not to start a tube feeding if the resident can no longer eat by mouth? 1. Yes 2. No 3. DonÂ’t Know 14. Can health care agent sign permission for an amputation if the resident has gangrene? 1. Yes 2. No 3. DonÂ’t Know 15. If a resident left instructions (for exampl e, a living will) that he would not want to be fed by feeding tube if he could no longer swallow food and would never regain the ability to swallow food could the nursing home give him a feeding tube anyway?

PAGE 118

106 1. Yes 2. No 3. Don’t Know 16. If a resident left instruction (for exampl e, a living will) that she would not want to have any amputations – could a doctor perform an amputation is the resident gets gangrene? 1. Yes 2. No 3. Don’t Know Part 3: Attitudes about Treatment Choices (17-34) 17. Sometimes there are acceptable reasons not to start a treatment, such as a feeding tube, even if it means a resident may die sooner. 1. Agree 2. Disagree 3. Don’t Know 18. Sometimes there are acceptable reasons to stop a treatment, such as a feeding tube, even if it means a resident may die sooner. 1. Agree 2. Disagree 3. Don’t Know 19. A dying person should always be offered food. 1. Agree 2. Disagree 3. Don’t Know 20. A dying person should always be offered water. 1. Agree 2. Disagree 3. Don’t Know 21. A dying person should always receive food and water, even if by feeding tube if necessary. 1. Agree 2. Disagree 3. Don’t Know 22. Sometimes we make dying residents uncomfortable by continuing to give them food and water by mouth. 1. Agree 2. Disagree

PAGE 119

107 3. DonÂ’t Know 23. Sometimes we make dying residents uncomfortable by giving them tube feedings. 1. Agree 2. Disagree 3. DonÂ’t Know 24. Sometimes we make dying residents uncomfortable by giving them IVs. 1. Agree 2. Disagree 3. DonÂ’t Know 25. If you talk to residents about a DNR (Do Not Resuscitate) order they will think their condition is hopeless. 1. Agree 2. Disagree 3. DonÂ’t Know 26. All residents who are able to make decisions for themselves have the right to refuse life-saving treatment, even if refusing such treatment may result in death. 1. Agree 2. Disagree 3. DonÂ’t Know 27. Residents usually prefer to let other people make decisions for them. 1. Agree 2. Disagree 3. DonÂ’t Know 28. Letting patients die by not giving them a certain treatment is the same thing as helping them commit suicide. 1. Agree 2. Disagree 3. DonÂ’t Know 29. It is possible to prevent dying residents from feeling much pain. 1. Agree 2. Disagree 3. DonÂ’t Know 30. Stopping a feeding tube is the same as killing a resident. 1. Agree 2. Disagree 3. DonÂ’t Know

PAGE 120

108 31. Sometimes it is right to give pain medication to relive suffering, even if it may make a person die sooner. 1. Agree 2. Disagree 3. DonÂ’t Know 32. Most patients prefer not to know they are dying. 1. Agree 2. Disagree 3. DonÂ’t Know 33. If a resident is not able to make decisions (for example, if the resident has AlzheimerÂ’s disease), the doctor is the best person to make treatment decisions for the resident. 1. Agree 2. Disagree 3. DonÂ’t Know 34. If a resident is not able to make decisions (for example, if the resident has AlzheimerÂ’s disease), a family member is the best person to make treatment decisions for the resident. 1. Agree 2. Disagree 3. DonÂ’t Know

PAGE 121

109 APPENDIX B DEMOGRAPHIC QUESTIONNAIRE Demographic Questionnaire for CNAs Personal Identification Number (PIN) _______________ 1. __________ Female ________ Male 2. __________ Asian ________ White _________ Black or African-American ________ Other 3. _________ Number of years I have worked as a CNA in nursing homes. 4. _________ Hispanic or Latino _________ Not Hispanic or Latino 5. _________ Number of years living in USA if not born in the USA. (If you were born in USA, leave a blank space.) 6. __________ Age ( or Birth Date month ______ year _______) 7. I believe that active participation in an organized religious group is: ________ Very Important ________ Somewhat important ________ Not important

PAGE 122

110 LIST OF REFERENCES Ahronheim, J.C., Morrison, R.S., Morris, J., Baskin, S., & Meier, D.E. (2000). Palliative care in advanced dementia: A randomized controlled trial and descriptive analysis. Journal of Palliative Medicine, 3 (3), 265-273. Allen, R.S., DeLaine, S.R., Chaplin, W.F., Marson, D.C., Bourgeois, M.S., Dijkstra, K., & Burgio, L.D. (2003). Advance care planning in nursing homes: Correlates of capacity and possession of advance directives. Gerontologist, 43 (3), 309-317. AlzheimerÂ’s Association. (2004). Fact Sheet. Retrieved May 16, 2004, from Baer, W.M., & Hanson, L.C. (2000). FamiliesÂ’ perception of the added value of hospice in the nursing home. Journal of the American Geriatrics Society, 48 (8), 879-882. Bargmann, P.L. (2001). Alzheimer care training program for staff in assisted living facilities in Florida . Orlando, FL: AlzheimerÂ’s Association Central and North Florida Chapter. Beauchamp, T.L., & Childress, J.F. (2001). Principles of medical ethics . Oxford, UK: Oxford University Press. Blasi, Z.V., Hurley, A.C., & Volicer, L. (2002) . End-of-life care in dementia: A review of the problems, prospects, and solutions in practice. Journal of the American Medical Directors Association, 3 (2), 57-65. Bosek, M.S.D., Lowry, E., Lindeman, D.A., Burck, J.R., & Gwyther, L.P. (2003). Promoting a good death for persons with dementia in facilities. JONAÂ’S Healthcare Law, Ethics, and Regulation, 5 (2), 34-41. Bottrell, M.M., OÂ’Sullivan, J.E., Robbins, M.A., Mitty, E.L., & Mezey, M.D. (2001). Transferring dying nursing home residents to the hospital: DON perspectives on the nurseÂ’s role in transfer decisions. Geriatric Nursing, 22 (6), 313-317. Bowers, B. (2001). Organizational change and workforce development in long-term care. Paper presented for technical Expert Panel Meeting, November 30, 2001. Washington, DC. Bradley, E.H., Blechner, B.B., Walker, L.C., & Wetle, T.T. (1997). Institutional efforts to promote advance care planning in nursing homes: Challenges and opportunities. Journal of Law and Medical Ethics, 25 , 150-159.

PAGE 123

111 Bradley, E.H., Peiris, V., & Wetle, T. ( 1998). Discussions about end-of-life care. Journal of the American Geriatrics Society, 46 , 1235-1241. Brenner, P.R. (1998). The experience of Jac ob Perlow Hospice: Hospice care of patients with AlzheimerÂ’s disease. In L. Volicer & A. Hurley, (Eds.), Hospice care for patients with advanced progressive dementia . New York: Springer Publishing Company. Brenner, P.R. (1999). Issues and themes in the care of persons with AlzheimerÂ’s disease. The American Journal of Hospice & Palliative Care, 1 , 391-393. Brock, D.B., & Foley, D.J. (1998). Demography and epidemiology of dying in the US with emphasis on deaths of older persons. Hospice Journal, 13 , 49-60. Brod, M., Stewart, A.L., Sands, L., & Walton, P. (1999). Conceptualization and measurement of quality of life in dementia: The dementia quality of life instrument (DQoL). The Gerontologist, 39 (1), 25-35. Brooks, S., Warshaw, G., Hasse, L., & Kues, J. (1999). The physician decision-making process in transferring nursing home patients to the hospital. Archives of Internal Medicine, 154 , 902-908. Buntin, M.B., & Huskamp, H. (2002). What is known about the economics of end-of-life care for Medicare beneficiaries? The Gerontologist, 42 (Special Issue III), 40-48. Burack, O.R., & Chichin, E.R. (2001). A s upport group for nursing assistants: Caring for nursing home residents at the end of life. Geriatric Nursing, 22 (6), 299-305. Byock, I.R. (1996). The nature of suffering and the nature of opportunity at the end of life. Clinics in Geriatric Medicine, 12 (2), 237-252. Campbell, S.L. (2003). Empowering nursing st aff and residents in long-term care. Geriatric Nursing, 24 (3), 170-175. Caplan, A.L., & Kane, R.A. (1990). Everyday ethics: Resolving dilemmas in nursing home life . New York: Springer Publishing Co. Carter, M.W. (2002). Advancing an et hical framework for long-term care. Journal of Aging Studies, 16 (1), 57-71. Casarett, D., Takasaka, J., Karlawish, J., Hirschman, K.B., & Clark, C.M. (2002). How should clinicians discuss hospice for patients with dementia? Anticipating caregiversÂ’ preconceptions and meeting their information needs. AlzheimerÂ’s Disease and Associated Disorders, 1 6(2), 116-122. Centers for Medicare and Medicaid Services (2002). Program information. Retrieved July 18, 2004 from

PAGE 124

112 Center for Medicare Education. (2001). Medicare hospice benefit. Issue Brief 2 (9): Washington, DC. Chichin, E.R., Burack, O.R., Olson, E., & Likourezos, A. (2000). End of life ethics and the nursing assistant . New York, NY: Springer, Publishing Company, Inc. Christakis, N.A., & Escarce, J.J. (1996). Surviv al of Medicare patients after enrollment in hospice programs. New England Journal of Medicine, 335 (3), 172-178. Cohen-Mansfield, J., Libin, A., & Lipson, S. (2003). Differences in presenting advance directives in the chart, in the Minimum Data Set, and through the staffÂ’s perceptions. The Gerontologist, 43 (3), 302-308. Eaton, S,C. (2000). Beyond unloving care. International Journal of Human Resource Management, 1 1(3), 591-616. Ersek. M., Kraybill, B.M., & Hansberry, J. (2000). Assessing the educational needs and concerns of nursing home staff regarding end-of-life care. Journal of Gerontological Nursing, 26 (10), 16-26. Evers, M.M., Purohit, D., Perl, D., Khan, K., & Marin, D.B. (2002). Palliative and aggressive end-of-life care for patients with dementia. Psychiatric Services, 53 (5), 609-613. Fagan. R.M., Williams, C.C., & Burger, S.G. (1997). Meeting of pioneers in nursing home culture change . Rochester, NY: LIFESPAN of Greater Rochester. Ferrera-Reid, R. (2004). Access barriers to hospice care for non-cancer conditions. Journal of Hospice & Palliative Nursing, 6 (2), 103-107. Field, M.J., & Cassell, C. (1997). Approaching death: Improving care at the end of life . Institute of Medicine, Washington, DC: National Academy Press. Finucane, T.E. (1999). How gravely ill becomes dying: A key to end-of-life care. Journal of the American Medical Association ,282(17), 1670-1672. Finucane, T.E., Christmas, C., & Travis, K. (1999). Tube feeding in patients with advanced dementia: A review of the evidence. Journal of the American Medical Association, 282 (14), 1365-1370. Forbes, S., Bern-Klug, M., & Gessert, C. ( 2000). End-of-life decision making for nursing home residents with dementia. Journal of Nursing Scholarship, 3 , 251-258. Fried, T.R., & Mor, V. (1998). Frailty and hos pitalization of long-term stay nursing home residents. Journal of the American Geriatrics Society, 45 , 265-269.

PAGE 125

113 Gessert, C.E., Mosier, M.C., Brown, E.F., & Frey, B. (2000). Tube feeding in nursing home residents with severe and irreversible cognitive impairment. Journal of the American Geriatrics Society, 48 , 1593-1600. Gillick, M.R. (2000). Rethinking the role of tube feeding in patients with advanced dementia. New England Journal of Medicine, 342 (3), 206-210. Gillick, M.R. (2002). Do we need to create geriatric hospitals? Journal of the American Geriatric Society, 50 (1), 174-177. Gregory, S.R. (2001). The nursing home workfo rce: Certified nursing assistants. AARP Public Policy Institute. Retrieved June 2, 2004 from Gwyther, L. (2001). Caring together. Alzheimer Care Quarterly, 2 (1), 64-72. Hanrahan, P., & Luchins, D.J. (1995). Access to hospice programs in end-stage dementia: A national survey of hospice programs. Journal of the American Geriatrics Society, 43 (1), 56-59. Hanrahan, P., Raymond, M., McGowan, E., & Luchins, D.J. (1999). Criteria for enrolling dementia patients in hospice: A replication. American Journal of Hospice & Palliative Care, 1 , 395-400. Hanson L.C., Danis, M., & Garrett, J. (1997). What is wrong with end-of-life care? Opinions of bereaved family members. Journal of the American Geriatrics Society, 45 , 1339-1344. Hanson, L.C., Henderson, M., & Menon, M. (2002) . As individual as death itself: A focus group study of terminal care in nursing homes. Journal of Palliative Medicine, 5 (1), 117-125. Hayley, D.C., Muir, J.C., Stocking, C., Hougham, G., & Sachs, G.A. (2001) Not ready for hospice: Characteristics of patients in a pre-hospice program. American Journal of Hospice & Palliative Care, 18 (6), 377-382. Head, B. (2003). Palliative care for persons with dementia. Home Healthcare Nurse, 21 (1), 53-60. Hebert, L.E., Scherr, P.A., Bienias, J.L., Bennett, D.A., & Evans, D.A. (2003). Alzheimer disease in the US population: Prevalence estimates using the 2000 census. Archives of Neurology, 60 (8), 1119-1122. Hofmann, J.C., Wenger, N.S., Davis, R.B., Teno, J., Connors, A.F., Desbiens, N., Lynn, J., & Phillips, R.S. (1997). Patient preferences for communication with physicians about end-of-life decisions. Annals of Internal Medicine, 127 (1), 1-12.

PAGE 126

114 Hogan, C., Lunney, J., Gabel, J., & Lynn, J. ( 2001). Medicare beneficiariesÂ’ costs of care in the last year of life. Health Affairs, 120 (4), 188-195. Holstein, M. (2001). A spiritual role for the elderly. Health Progress, 81 (2), 12-16. Hurley, A.C., Bottino, R., & Volicer, L. (1994). Nursing role in advance proxy planning for Alzheimer patients. Caring, (August) , 72-76. Hurley, A.C., Volicer, L., & Blasi, Z.V. (2000). End-of-life care for patients with advanced dementia. JAMA, 284 (19), 2449-2450. Hurley, A.C., Volicer, B.J., Hanrahan, P.A., Houde, S., & Volicer, L. (1992). Assessment of discomfort in advanced Alzheimer patients. Research in Nursing & Health, 15 , 369-377. Hurley, A.C., Volicer, L., Rempusheski, V.F ., & Fry, S. (1995). Reaching consensus: The process of recommending treatment decisions for Alzheimer patients. Advances in Nursing Science, 18 (2), 33-43. Juozapivicius, K.P., & Weber, J.A. (2001). A reflective study of AlzheimerÂ’s caregivers. American Journal of AlzheimerÂ’s Disease and Other Dementias, 16 (1), 11-20. Kane, R.A. (2003). Definition, measurement, and correlates of quality of life in nursing homes: Toward a reasonable practice, research and policy agenda. The Gerontologist, 43 (Spec No.2), 28-36. Katz, J.S., Sidell, M., & Komaromy, C. (2001). Dying in long-term care facilities: Support needs of other residents, relatives and staff. American Journal of Hospice and Palliative Care, 18 (5), 321-326. Kayser-Jones, J. (2002). The experience of dying: An ethnographic nursing home study. The Gerontologist, 42 (Special Issue III), 11-19. Kayser-Jones, J., Schell, E., Lyons, W., Kris, A.E., Chan, J., & Beard, R.L. (2003). Factors that influence end-of-life care in nursing homes: The physical environment, inadequate staffing, and lack of supervision. The Gerontologist, 43 (Special Issue II), 76-84. Keay, T.J. (1999). Palliative care in the nursing home. Generations, 23 (1), 96-98. Keay, T.J., Cranmer, K., & Hanson, L. (2002). Attitudes toward hospice care. Journal of the American Medical Directors Association, 4 (2), 121. Kelly, A.M. (2000). Ethics committees in nursing homes in central Florida . Unpublished survey, Orlando, FL: AlzheimerÂ’s Association Central and North Florida.

PAGE 127

115 Kinosian, B.P., Stallard, E., Lee, J.H., Woodbury, M.A., Zbrozek, A.S., & Glick, H.A. (2000). Predicting 10-year care requirements for older people with suspected AlzheimerÂ’s disease. Journal of the American Geriatrics Society, 48 (6), 631-638. Kovach, C, Wilson, S., & Noonan, P. (1996). The effects of hospice interventions on behaviors, discomfort, and physical comp lications of end-stage dementia nursing home residents. American Journal of Alzheimer Care, 4 , 7-15. Kuhn, D. (2001). Living with loss in AlzheimerÂ’s disease. Alzheimer Care Quarterly, 2 (1), 12-22. Lawton, M.P. (1994) Quality of life in AlzheimerÂ’s disease. AlzheimerÂ’s Disease and Associated Disorders, 8 (Supp. 3), 138-150. Luchins, D.J., & Hanrahan, P. (1993). What is appropriate health care for end-stage dementia? Journal of the American Geriatrics Society, 41 (1), 25-30. Luchins, D.J., Hanrahan, P., & Litzenberg, K. (1998). Acceptance of hospice care for dementia patients by health care professionals and family members. In L. Volicer & A. Hurley, (Eds). Hospice care for patients with advanced progressive dementia . New York: Springer Publishing Company. Luchins, D.J., Hanrahan, P., & Murphy, K. (1997). Criteria for enrolling dementia patients in hospice. Journal of the American Geriatrics Society, 45 , 1054-1059. Lynn, J. (2000). Learning to care for people w ith chronic illness facing the end of life. Journal of the American Medical Association, 284 (19), 2508-2510. Lynn, J., Wilkinson, A., Cohn, F., & Jones, S.B. (1998). Capitated risk-bearing managed care systems could improve end-of-life care. Journal of the American Geriatric Society, 46 (3), 322-330. Magaziner, J., German, P., Zimmerman, S.I ., Gebel, H.R., Burton, L., Gruber-Baldini, A.L., et al. (2000). The prevalence of dementia in a statewide sample of new nursing home admissions aged 65 and older: Diagnosis by expert panel. The Gerontologist, 40 , 663-672. Meier, D.E., Ahronheim, J.C., Morris, J., Baskin-Lyons, S., & Morrison, R.S. (2001). High short-term mortality in hospitalized patients with advanced dementia: Lack of benefit of tube feeding. Archives of Internal Medicine, 161 (19), 2385-2386. Meier, D.E., & Morrison, R.S. (1999). Introduction: Old age and care near the end of life. Generations, 23 (1), 1-11. Meisel, A. (1995). Barriers to forgoing nutrition and hydration in nursing homes. American Journal of Law and Medicine, 21 (4), 334-382.

PAGE 128

116 Mezey, M., Dubler, N.N., Mitty, E., & Brody, A.A. (2002). What impact do setting and transitions have on the quality of life at the end of life and the quality of the dying process? The Gerontologist, 42 , 54-67. Mezey, M., Miller, L.L., & Linton-Nelson, L. (1999). Caring for caregivers of frail elders at the end of life. Generations, 23 (1), 44-51. Mezey, M., Mitty, E., Rappaport, M., & Ramsey, G. (1997). Implementation of the PSDA in nursing homes in New York City. Journal of the American Geriatrics Society, 45 , 43-49. Mezey, M., Teresi, J., Ramsey, G., Mitty, E., & Bobrowitz, T. (2000). Decision-making capacity to execute a health care proxy: Development and testing of guidelines. Journal of the American Geriatrics Society, 48 (2), 179-187. Miller, S.C., Gozalo, P., & Mor, V. (2001). Hospice enrollment and hospitalization of dying nursing home patients. American Journal of Medicine, 111 (1), 38-44. Misiorski, S. (2001). Building a better life: Transforming institutional culture. Alzheimer’s Care Quarterly, 2 (3), 16-19. Mitchell, S.L., Kiely, D.K., & Hamel, S.B. (2004). Dying with advanced dementia in the nursing home. Archives of Internal Medicine, 162 (3), 321-326. Molloy, D.W., Guyatt, G.H., Russo, R., Goeree, R., O’Brien, B.J., Bedard, M., et al. (2000). Systemic implementation of an advance directive program in nursing homes. Journal of the American Medical Association, 283 (11), 1437-1444. Morrison, R.S. (2000). Planning and providing care at the end of life. Hospital Practice, 35 (10), 61-64. Morrison, R.S., & Siu, A.L. (2000). Survival in end-stage dementia following acute illness. Journal of the American Medical Association, 284 (1), 47-52. Moss, M.S., Braunschweig, M.A., & Rubinstein, R.L. (2002). Terminal care for nursing home residents with dementia. Alzheimer’s Care Quarterly, 3 (3), 233-246. Moss, M.S., Moss, S.Z., Rubenstein, R.L., & Black, H.K. (2003). The metaphor of “family” in staff communication about dying and death. Journal of Gerontology: Social Sciences, 583 (5), S290-296. National Hospice Organization (2000). NHO prognostication guide lines. Alexandria, VA: National Hospice and Palliative Care Organization. Ortigara, A. (2001). Families and care sta ff: Bridging the divide, creating caring relationships. Alzheimer Care Quarterly, 2 (1), 73-76.

PAGE 129

117 Parker-Oliver,D., & Bickel, D. (2002). Nursing home experience with hospice. Journal of the American Medical Directors Association, 3 (2), 46-50. Petrisek, A.C., & Mor, V. (1999). Hospice in nursing homes: A facility-level analysis of the distribution of hospice beneficiaries. The Gerontologist, 39 (30), 279-290. Pillemer, K., Hegeman, C.R., Albright, B ., & Henderson, C. (1998). Building bridges between families and nursing home staff: The Partners in Caregiving Program. The Gerontologist, 38 (4), 499-503. Pioneer Network (2002). Values, visi on and mission. Retrieved May 16, 2004 from Post, S.G. (2001). Tube feeding and advanced progressive dementia. Hastings Center Report, 31 (1), 36-42. Post, S.G., Volicer, L., & Ortigara, A (1999) . The real-world ethics of dementia. Annals of Long-term care, 7 , 6-11. Reisberg, B. (1982). The Global Deterioration Scale for assessment of primary degenerative dementia. American Journal of Psychiatry, 139 , 1136–1139. Reisberg, B. (1988). Functional Assessment Staging (FAST). Psychopharmacology Bulletin, 24 (4), 653-659. Sachs, G. (2000). Redefining end of life care for people with dementia. Last Acts Newsletter, Fall , 2-4. Schonwetter, R.S., Han, B., Small, B.J., Martin, B., Tope, K., & Haley, W.E. (2003). Predictors of six-month survival among patients with dementia: An evaluation of hospice Medicare guidelines. American Journal of Hospice & Palliative Care, 20 (2), 105-113. Shawler, C., Rowles, G.D., & High, D.M. (2001), Analysis of key decision-making in the life of a nursing home resident, The Gerontologist, 41 (5), 612-622. Smith, S.A. (2001). Treatment of end-stage non-cancer diagnoses . Hospice and Palliative Nurses Association: Pittsburgh, PA. Steinhauser, K.E., Christakis, N.A., Clipp, E.C., McNeilly, M., McIntyre, L., & Tulski, J.A. (2000). Factors considered important at the end of life by patients, families, physicians, and other care providers. JAMA, 284 (19), 2476-2482. Strahan, G.W. (1999). An overview of nursing homes and their current residents: Data from the 1995 National Nursing Home Survey . Hyattsville, MD: National Center for Health Statistics.

PAGE 130

118 Sullivan, M.D. (2002). The illusion of patient choice in end-of-life decisions. American Journal of Geriatric Psychiatry, 10 (4), 365-372. Tauber, A.L. (2003). Sick autonomy. Perspectives in Biological Medicine, 46 (4), 484495. Taylor, D.H., & Sloan, F.A. (2000). How much do persons with AlzheimerÂ’s disease cost Medicare? Journal of the American Geriatrics Society, 48 , 739-746. Teno, J. (1999). TIME: Toolkit of instruments to measure end-of-life care. Brown University. Retrieved July 18, 2004 from Teno, J. (2002) Now is the time to embrace nursing homes as a place of care for dying persons. Innovations in End-Of-Life Care, 4 (2) Retrieved July 10, 2004 from Teno, J., Licks S., Lynn, J., Wender, N., Connors, A.F., Phillips, R.S., et al. (1997a). Do advance directives provide instructions that direct care? Journal of the American Geriatrics Society, 45 , 508-512. Teno, J., & Lynn, J. (1996). Putting advance care-planning into action. Journal of Clinical Ethics, 7 , 205-213. Teno, J., Lynn, J., Weenger, N., Phillips, R.S., Murphy, D.P., Connors, A.F., et al. (1997b). Advance directives for seriously ill hospitalized patients: Effectiveness with PSDA and the SUPPORT intervention. Journal of the American Geriatrics Society, 45 , 500-507. Thomas, W. (1996). A life worth living: The Eden Alternative . Acton, MA: Vander Wyk & Burnham. Topinkova, E., & Callahan, D. (1999). Culture, economics, and AlzheimerÂ’s disease; Social determinants of resource allocation. Journal of Applied Gerontology, 18 (4), 411-422. Travis, S.S., Bernard, M., Dixon, S., McAuley, W.J., Loving, G., & McClanahan, I. (2002). Obstacles to palliation and end-of-life care in a long-term care facility. The Gerontologist, 42 (3), 342-349. Troxel, D., & Bell, V. (1997). The best friends approach to Alzheimer care . Baltimore, MD: Health Professions Press. Tuch, H., Parrish, P., & Romer, A.L. (2002). Integrating palliative care into nursing homes, Journal of Palliative Medicine, 6 (2), 297-309.

PAGE 131

119 US Department of Health and Human Services (2004a). National Home and Hospice Care Data . Retrieved on July 18, 2004 from US Department of Health and Human Services (2004b). Nursing Home Data . Retrieved on July 18, 2004 from Vladek, B. (1999). The problem isnÂ’t payment: Medicare and the reform of end-of-life care. Generations, 23 (1), 52-57. Volicer, L. (1997). Goals of care in advanced dementia: Comfort, dignity, and psychological well-being. American Journal of Alzheimer Disease, Sept/Oct , 196197. Volicer, L. (2001). Management of severe Al zheimerÂ’s disease and end-of-life issues. Clinics in Geriatric Medicine, 17 (2), 377-391. Volicer, L. (2002). Is percutaneous endoscopic gastrostomy really better? Journal of the American Geriatrics Society, 50 (1), 1907-1908. Volicer, L., & Bloom-Charette, L. (1999a). A ssessment of quality of life in advanced dementia. In L. Volicer, & L. Bloom-Charette, (Eds). Enhancing the qulaity of life in advanced dementia . Philadelphia, PA: Brunner/Mazel. Volicer, L., Camberg, L., Hurley, A.C., Askle y, J., Woods, P., Ooi, W.L., et al. (1999b). Dimensions of descreased psychological well-being in advanced dementia. AlzheimerÂ’s Disease & Associated Disorders, 13 (4), 192-201. Volicer, L., & Hurley, A. (Eds.). (1998). Hospice care for patients with advanced progressive dementia . New York, NY: Springer Publishing Company. Volicer, L., Hurley, A.C., & Blasi, Z.V. (2001) . Scales for evaluation of end-of-life care in dementia. AlzheimerÂ’s Disease and Associated Disorders, 15 (4), 194-200. Volicer, L., Hurley, A.C., & Blasi, Z.V. ( 2003). Characteristics of dementia end-of-life care across care settings. American Journal of Hospice & Palliative Care, 20 (3), 191-200. Volicer, B.J., Hurley, A.C., Fabiszewski, K.J., Montgomery, P., & Volicer, L. (1993). Predicting short-term survival for patients with advanced AlzheimerÂ’s disease. Journal of the American Geriatrics Society, 41 , 535-540. Volicer, L., Rheaume, Y., Riley, M.E., Karner, J., & Glennon, M. (1990). Discontinuation of tube feedings in patients with dementia of the Alzheimer type. American Journal of Alzheimer Care, 5 , 22-25.

PAGE 132

120 Walker, L., Blechman, B., & Gruman, C. (1998). Assessment of capacity to discuss advance care planning in nursing homes. Journal of the American Geriatrics Society, 46 (8), 1055-1056. Walker, L., Bradley, E., Blechner, B., & Tetle, T. (1998). Problems implementing the PSDA in nursing homes. Journal of Mental Health and Aging, 4 (1), 83-96. Wilmot, S., Legg, L., & Barratt, J. (2002). Ethical issues in the feeding of patients suffering from dementia: A focus group study of hospital staff responses to conflicting principles. Nursing Ethics, 9 (6), 599-610. Wilson, S.A., & Daley, B.J. (1999). Family perspectives on dying in long-term care settings. Journal of Gerontological Nursing, 25 (110), 19-25. Wilson, S.A., Kovach, C.R., & Stearns, S.A. (1996). Hospice concepts in the care for end-stage dementia. Geriatric Nursing, 17 (1), 6-10. World Health Organization. (1990). Technical report series 804, cancer pain and palliative car e. Geneva: World Health Organization. Zerzan, J., Stearns, S., & Hanson, L. (2000). Access to palliative care in nursing homes. JAMA, 284 (19),2489-2494. Zimmerman, S., Sloane, P.D., Hanson, L., Mitchell, C.M., & Shy, A. (2003). Staff perceptions of end-of-life care in long-term care. Journal of the American Medical Directors Association, 4 (1), 23-26.

PAGE 133

123 BIOGRAPHICAL SKETCH Annette Kelly has been the President/CEO of the AlzheimerÂ’s Association (Central and North Florida Chapter) since 1993. As a Geriatric Nurse Practitioner, and a nurse for over 40 years, Annette also has a clinical background and experience in the health care of older adults. Annette received her diploma in nursing from St. VincentÂ’s Hospital School of Nursing, in New York City in 1962, and a bachelorÂ’s degree in nursing from George Mason University in Fairfax, Virginia, in 1982. Annette received her masterÂ’s degree in nursing from the University of Florida in 1993. Nationally, Annette is part of the Ethics Panel of the National AlzheimerÂ’s Association, and contributes to several Advisory Boards on AlzheimerÂ’s and dementia care issues. Annette is a Florida department of Elder Affairs-certified dementia trainer. Annette has been married to Jerry Kelly for 42 years. They have four adult children and three grandchildren, and live in central Florida.