Quality of Care of Older People Living with Advanced Dementia in the Community in Israel Shelley Sternberg, MD,* Netta Bentur, PhD,† and Jennifer Shuldiner, MPH†
OBJECTIVES: To examine the quality of end-of-life (EOL) care of older people with advanced dementia (OPAD) living in the community. DESIGN: Retrospective cohort study. SETTING: The second largest preferred provider organization in Israel. PARTICIPANTS: Primary family caregivers of OPAD (N = 117; 52 bereaved caregivers, 65 still caring for OPAD). MEASUREMENTS: Primary outcome measures were quality-of-EOL-care scales regarding symptom management at the EOL with dementia (SM-EOLD) and comfort assessment in dying at the EOL with dementia (CAD-EOLD). RESULTS: Caregivers reported a mean symptom management score of 29 9.8 points (range 0–45 points), indicating poor symptom management in the areas of pain, shortness of breath, fear, skin breakdown, and resistiveness to care. In a multiple regression model, less comorbidity, longer duration of dementia, greater caregiver education, and absence of caregiver depression predicted better SM-EOLD. The 52 bereaved caregivers reported a mean CAD-EOLD score of 31.3 6.7 points (range 14–42 points), indicating poor comfort around dying, especially difficulty swallowing and discomfort. CONCLUSION: Because of the high rates of suffering, there is a need to address the complexity and severity of symptoms and problems faced by OPAD in the community and their caregivers. The priorities should include education of healthcare providers on palliative care, focusing especially on symptom-management needs of OPAD and comfort around dying. J Am Geriatr Soc 62:269–275, 2014.
Key words: advanced dementia; quality of care; end of life; symptom management; palliative care
From the *Maccabi Healthcare Services, Modiin; and †Myers-JDCBrookdale Institute, Jerusalem, Israel. Address correspondence to Netta Bentur, Myers-JDC-Brookdale Institute, JDC Hill, P.O.B. 3886, Jerusalem 91037, Israel. E-mail: [email protected] DOI: 10.1111/jgs.12655
JAGS 62:269–275, 2014 © 2014, Copyright the Authors Journal compilation © 2014, The American Geriatrics Society
W
ith the aging of the population, there has been a significant rise in the number of older people with dementia. By 2025 in the United States, 7.1 million will be affected—a 40% increase from the 5 million individuals aged 65 and older currently affected. Dementia is the sixth leading cause of death in the United States, reduces 10-year life expectancy at age 70 by half,1 and is a significant financial burden to society, similar to heart disease and cancer.2 In Israel, the frequency of dementia in people aged 65 and older was estimated at 19.8% in 2010 and is expected to increase by 88% by 2030.3 Because dementia is a progressive terminal illness in which it is often difficult to clinically define the beginning of the terminal stage,2,4 health and social care services may not be optimally coordinated to meet the multifaceted needs of older people with advanced dementia (OPAD).5 Studies of OPAD have concluded that this population often receives poor end-of-life (EOL) care, with inadequate pain and other symptom control, and lacks access to the type of palliative care services available to individuals with cancer.6,7 In Israel, a substantial proportion of EOL care for OPAD is provided in the community with home health services.8–10 According to a 2002 Israeli national survey, 84% of people with dementia were living in the community,3 compared with two-thirds in the United Kingdom11 and 60–70% in the United States.1 There is limited knowledge about the EOL experiences of OPAD,12,13 especially in the community.14 Several studies have examined EOL care of OPAD in nursing homes6,15–20 and a few in hospitals.21–24 Studies in nursing homes have shown that OPAD do not receive optimal care and that many have distressing symptoms such as dyspnea, pain, pressure ulcers, agitation, and aspiration,12 but little research has been done on EOL care of OPAD in a community setting.4,9 Assessment of EOL experiences is challenging in individuals with end-stage dementia, who cannot communicate their physical needs, understand treatment options, articulate their emotional needs, or express suffering at the EOL.25 Volicer and colleagues have developed standardized accepted tools to measure EOL care in advanced dementia. These scales have been used in several retrospective and prospective studies,9,16,26–29 but to the best of the
0002-8614/14/$15.00
270
STERNBERG ET AL.
knowledge of the authors of the current study, only one study included a group of OPAD living in the community.9 The aim of the current study was to examine the quality of EOL care of OPAD living in the community. This information can guide the development of health policy to improve care for individuals with advanced dementia in the community and organize health services to meet their complex needs.
METHODS Study Population Information was obtained by surveying 117 family caregivers of OPAD, 44% (n = 52) of whom were bereaved of an OPAD who had died in the 3–6 months before study initiation and 65 of whom were taking care of an OPAD living in the community. Advanced dementia was defined as Stage 6 or 7 on the Global Deterioration Scale, a 7-point scale used to indicate the severity of primary degenerative dementia.30 All the OPAD were members of Maccabi Healthcare Services (MHS), the second-largest preferred provider organization in Israel. MHS has a nationwide network of more than 3,000 physicians providing healthcare services to 1.8 million members (24% of the country’s population) in Israel. This study was confined to the two largest regions of MHS. Family caregiver was defined as the person who provided most of the care for the OPAD, without payment; lived with the OPAD; and was primarily responsible for care, coordinating services, or maintaining contact with the health system. The sample of OPAD and their family caregivers was identified from the 2012 list of deceased members of MHS and from the roster of the homecare program. Interviews were conducted with 75% (117) of the caregivers of the 156 people identified. Interviews were not conducted for the following reasons: refusal (13%), geographical distance (3%), and other reasons (9%).
Data Collection To confirm the diagnosis, a MHS nurse contacted the responsible physician to verify the diagnosis of advanced dementia and that the older person had lived in the community in the last 3 months of life. The MHS nurse then telephoned the family member of the deceased or living OPAD, explained the study objectives, and requested consent for an interview. Names of consenting family members were passed on to the study interviewers to arrange an interview. The face-to-face interview was conducted in the family member’s home, in Hebrew, Russian, or English, using a closed structured questionnaire. Before the interview began, the interviewer explained the study to the family member and obtained written informed consent. The institutional review board (Helsinki committee) of MHS approved the study.
The Study Variables Primary dependent outcome measures in this study were the quality of EOL care provided to the deceased or living
FEBRUARY 2014–VOL. 62, NO. 2
JAGS
OPAD. The most-comprehensive and -specific set of instruments to measure EOL care with dementia (EOLD) were developed based on the literature and expert opinions and were developed for family caregivers to evaluate EOL care retrospectively.15,26 The EOLD scales include symptom management (SM) and comfort assessment in dying (CAD). They are specifically designed to measure and assess the overall quality of EOL care of OPAD, and are not limited to a single concept such as suffering. The SM-EOLD measures the frequency of nine symptoms and signs that occurred in the last 90 days of the illness or dying process: pain, shortness of breath, depression, fear, anxiety, agitation, calm, skin breakdown, and resistance to care. The scale has six response options (0–5, daily to never); nine items are summed to calculate a total score ranging from 0 to 45 (best symptom management). Eight items have negative responses, and one item (calm) is a positive condition. When caregivers were unable to answer all questions, up to two missing SM-EOLD items were imputed with subject means. A study on EOLD in nursing homes also used imputed values,28 whereas other studies did not use this technique.9,15,26 The CAD-EOLD scale, which measures symptoms commonly observed during the dying process, was administered only if the OPAD had died. CAD-EOLD items assess the frequency of the following 14 symptoms and conditions during the last 7 days of life (discomfort, pain, restlessness, shortness of breath, choking, gurgling, difficulty swallowing, fear, anxiety, crying, moaning, serenity, peace, and calm), divided into four subscales (physical distress, dying symptoms, emotional symptoms, and wellbeing). The scale has three response options (not at all, somewhat, a lot; scored 1–3). Summed total scores range from 14 to 42, and the sum of each subscale ranges from 4 to 12, with higher scores representing greater comfort. For all scales, the coding was reversed when applicable so that the higher scores would represent better outcomes. When caregivers were unable to answer all questions, up to three missing CAD-EOLD items were imputed with subject means. Independent variables in this study included demographic characteristics of the OPAD and their family caregivers (age, sex, marital status, education, country of origin, and relation to the OPAD). Independent variables also included medical variables of the OPAD: number of comorbidities, problems swallowing, weight loss, falls, number of medications, use of antipsychotics, and method of feeding (oral feeding, nasogastric tube feeding, and percutaneous endoscopic gastrostomy (PEG)). Types of services that OPAD used (physicians, nurses, and hospice) were also measured. All information was elicited from the interviews with the caregivers. The Agency for Healthcare Research and Quality twoquestion validated screening tool for depression was posed to the caregivers: “During the past month, have you often been bothered by feeling down, depressed, or hopeless? During the past month, have you often been bothered by little interest or pleasure in doing things?”31 Caregiver burden was assessed by asking the caregiver: “In general, the burden you feel from acting as a caregiver can be described as very heavy, heavy, not so heavy, or not at all heavy?”
JAGS
2014–VOL. 62, NO. 2
Statistical Analysis Descriptive statistics (means, standard deviations, and medians for continuous variables and frequency for categorical variables) were used to describe the demographic characteristics of the OPAD and their caregivers. Bivariate analyses were performed using chi-square tests and t-tests to estimate the relationship between the quality of EOL care (SM-EOLD, CAD-EOLD) and OPAD and caregivers characteristics, health status of the OPAD, and their health service utilization. For variables that were normally distributed, a Spearman rank-order correlation was run for continuous variables and a Kruskal-Wallis test for categorical variables. Multivariate analysis was performed using multiple regression analysis to describe the relationship between the dependent variable (SM-EOLD) and the characteristics of the OPAD, their caregivers, disease characteristics, and healthcare utilization. Data were analyzed using SPSS 20 software for Windows (SPSS, Inc., Chicago, IL).
RESULTS Characteristics of OPAD Sixty-five percent of the OPAD were female, and at the time of death or interview, their mean age was 86.6 7.6 (range 64–101; 34% aged
OBJECTIVES: To examine the quality of end-of-life (EOL) care of older people with advanced dementia (OPAD) living in the community. DESIGN: Retrospective cohort study. SETTING: The second largest preferred provider organization in Israel. PARTICIPANTS: Primary family caregivers of OPAD (N = 117; 52 bereaved caregivers, 65 still caring for OPAD). MEASUREMENTS: Primary outcome measures were quality-of-EOL-care scales regarding symptom management at the EOL with dementia (SM-EOLD) and comfort assessment in dying at the EOL with dementia (CAD-EOLD). RESULTS: Caregivers reported a mean symptom management score of 29 9.8 points (range 0–45 points), indicating poor symptom management in the areas of pain, shortness of breath, fear, skin breakdown, and resistiveness to care. In a multiple regression model, less comorbidity, longer duration of dementia, greater caregiver education, and absence of caregiver depression predicted better SM-EOLD. The 52 bereaved caregivers reported a mean CAD-EOLD score of 31.3 6.7 points (range 14–42 points), indicating poor comfort around dying, especially difficulty swallowing and discomfort. CONCLUSION: Because of the high rates of suffering, there is a need to address the complexity and severity of symptoms and problems faced by OPAD in the community and their caregivers. The priorities should include education of healthcare providers on palliative care, focusing especially on symptom-management needs of OPAD and comfort around dying. J Am Geriatr Soc 62:269–275, 2014.
Key words: advanced dementia; quality of care; end of life; symptom management; palliative care
From the *Maccabi Healthcare Services, Modiin; and †Myers-JDCBrookdale Institute, Jerusalem, Israel. Address correspondence to Netta Bentur, Myers-JDC-Brookdale Institute, JDC Hill, P.O.B. 3886, Jerusalem 91037, Israel. E-mail: [email protected] DOI: 10.1111/jgs.12655
JAGS 62:269–275, 2014 © 2014, Copyright the Authors Journal compilation © 2014, The American Geriatrics Society
W
ith the aging of the population, there has been a significant rise in the number of older people with dementia. By 2025 in the United States, 7.1 million will be affected—a 40% increase from the 5 million individuals aged 65 and older currently affected. Dementia is the sixth leading cause of death in the United States, reduces 10-year life expectancy at age 70 by half,1 and is a significant financial burden to society, similar to heart disease and cancer.2 In Israel, the frequency of dementia in people aged 65 and older was estimated at 19.8% in 2010 and is expected to increase by 88% by 2030.3 Because dementia is a progressive terminal illness in which it is often difficult to clinically define the beginning of the terminal stage,2,4 health and social care services may not be optimally coordinated to meet the multifaceted needs of older people with advanced dementia (OPAD).5 Studies of OPAD have concluded that this population often receives poor end-of-life (EOL) care, with inadequate pain and other symptom control, and lacks access to the type of palliative care services available to individuals with cancer.6,7 In Israel, a substantial proportion of EOL care for OPAD is provided in the community with home health services.8–10 According to a 2002 Israeli national survey, 84% of people with dementia were living in the community,3 compared with two-thirds in the United Kingdom11 and 60–70% in the United States.1 There is limited knowledge about the EOL experiences of OPAD,12,13 especially in the community.14 Several studies have examined EOL care of OPAD in nursing homes6,15–20 and a few in hospitals.21–24 Studies in nursing homes have shown that OPAD do not receive optimal care and that many have distressing symptoms such as dyspnea, pain, pressure ulcers, agitation, and aspiration,12 but little research has been done on EOL care of OPAD in a community setting.4,9 Assessment of EOL experiences is challenging in individuals with end-stage dementia, who cannot communicate their physical needs, understand treatment options, articulate their emotional needs, or express suffering at the EOL.25 Volicer and colleagues have developed standardized accepted tools to measure EOL care in advanced dementia. These scales have been used in several retrospective and prospective studies,9,16,26–29 but to the best of the
0002-8614/14/$15.00
270
STERNBERG ET AL.
knowledge of the authors of the current study, only one study included a group of OPAD living in the community.9 The aim of the current study was to examine the quality of EOL care of OPAD living in the community. This information can guide the development of health policy to improve care for individuals with advanced dementia in the community and organize health services to meet their complex needs.
METHODS Study Population Information was obtained by surveying 117 family caregivers of OPAD, 44% (n = 52) of whom were bereaved of an OPAD who had died in the 3–6 months before study initiation and 65 of whom were taking care of an OPAD living in the community. Advanced dementia was defined as Stage 6 or 7 on the Global Deterioration Scale, a 7-point scale used to indicate the severity of primary degenerative dementia.30 All the OPAD were members of Maccabi Healthcare Services (MHS), the second-largest preferred provider organization in Israel. MHS has a nationwide network of more than 3,000 physicians providing healthcare services to 1.8 million members (24% of the country’s population) in Israel. This study was confined to the two largest regions of MHS. Family caregiver was defined as the person who provided most of the care for the OPAD, without payment; lived with the OPAD; and was primarily responsible for care, coordinating services, or maintaining contact with the health system. The sample of OPAD and their family caregivers was identified from the 2012 list of deceased members of MHS and from the roster of the homecare program. Interviews were conducted with 75% (117) of the caregivers of the 156 people identified. Interviews were not conducted for the following reasons: refusal (13%), geographical distance (3%), and other reasons (9%).
Data Collection To confirm the diagnosis, a MHS nurse contacted the responsible physician to verify the diagnosis of advanced dementia and that the older person had lived in the community in the last 3 months of life. The MHS nurse then telephoned the family member of the deceased or living OPAD, explained the study objectives, and requested consent for an interview. Names of consenting family members were passed on to the study interviewers to arrange an interview. The face-to-face interview was conducted in the family member’s home, in Hebrew, Russian, or English, using a closed structured questionnaire. Before the interview began, the interviewer explained the study to the family member and obtained written informed consent. The institutional review board (Helsinki committee) of MHS approved the study.
The Study Variables Primary dependent outcome measures in this study were the quality of EOL care provided to the deceased or living
FEBRUARY 2014–VOL. 62, NO. 2
JAGS
OPAD. The most-comprehensive and -specific set of instruments to measure EOL care with dementia (EOLD) were developed based on the literature and expert opinions and were developed for family caregivers to evaluate EOL care retrospectively.15,26 The EOLD scales include symptom management (SM) and comfort assessment in dying (CAD). They are specifically designed to measure and assess the overall quality of EOL care of OPAD, and are not limited to a single concept such as suffering. The SM-EOLD measures the frequency of nine symptoms and signs that occurred in the last 90 days of the illness or dying process: pain, shortness of breath, depression, fear, anxiety, agitation, calm, skin breakdown, and resistance to care. The scale has six response options (0–5, daily to never); nine items are summed to calculate a total score ranging from 0 to 45 (best symptom management). Eight items have negative responses, and one item (calm) is a positive condition. When caregivers were unable to answer all questions, up to two missing SM-EOLD items were imputed with subject means. A study on EOLD in nursing homes also used imputed values,28 whereas other studies did not use this technique.9,15,26 The CAD-EOLD scale, which measures symptoms commonly observed during the dying process, was administered only if the OPAD had died. CAD-EOLD items assess the frequency of the following 14 symptoms and conditions during the last 7 days of life (discomfort, pain, restlessness, shortness of breath, choking, gurgling, difficulty swallowing, fear, anxiety, crying, moaning, serenity, peace, and calm), divided into four subscales (physical distress, dying symptoms, emotional symptoms, and wellbeing). The scale has three response options (not at all, somewhat, a lot; scored 1–3). Summed total scores range from 14 to 42, and the sum of each subscale ranges from 4 to 12, with higher scores representing greater comfort. For all scales, the coding was reversed when applicable so that the higher scores would represent better outcomes. When caregivers were unable to answer all questions, up to three missing CAD-EOLD items were imputed with subject means. Independent variables in this study included demographic characteristics of the OPAD and their family caregivers (age, sex, marital status, education, country of origin, and relation to the OPAD). Independent variables also included medical variables of the OPAD: number of comorbidities, problems swallowing, weight loss, falls, number of medications, use of antipsychotics, and method of feeding (oral feeding, nasogastric tube feeding, and percutaneous endoscopic gastrostomy (PEG)). Types of services that OPAD used (physicians, nurses, and hospice) were also measured. All information was elicited from the interviews with the caregivers. The Agency for Healthcare Research and Quality twoquestion validated screening tool for depression was posed to the caregivers: “During the past month, have you often been bothered by feeling down, depressed, or hopeless? During the past month, have you often been bothered by little interest or pleasure in doing things?”31 Caregiver burden was assessed by asking the caregiver: “In general, the burden you feel from acting as a caregiver can be described as very heavy, heavy, not so heavy, or not at all heavy?”
JAGS
2014–VOL. 62, NO. 2
Statistical Analysis Descriptive statistics (means, standard deviations, and medians for continuous variables and frequency for categorical variables) were used to describe the demographic characteristics of the OPAD and their caregivers. Bivariate analyses were performed using chi-square tests and t-tests to estimate the relationship between the quality of EOL care (SM-EOLD, CAD-EOLD) and OPAD and caregivers characteristics, health status of the OPAD, and their health service utilization. For variables that were normally distributed, a Spearman rank-order correlation was run for continuous variables and a Kruskal-Wallis test for categorical variables. Multivariate analysis was performed using multiple regression analysis to describe the relationship between the dependent variable (SM-EOLD) and the characteristics of the OPAD, their caregivers, disease characteristics, and healthcare utilization. Data were analyzed using SPSS 20 software for Windows (SPSS, Inc., Chicago, IL).
RESULTS Characteristics of OPAD Sixty-five percent of the OPAD were female, and at the time of death or interview, their mean age was 86.6 7.6 (range 64–101; 34% aged
