A Prospective Longitudinal Study of Caregivers of Community Dwelling Persons with Severe Dementia (PISCES): Study Protocol

Abstract

Although many persons with severe dementia (PWSDs) are cared for at home by their family caregivers, few studies have assessed end of life (EOL) care experiences of PWSDs. We present the protocol for the PISCES study (Panel study Investigating Status of Cognitively impaired Elderly in Singapore) which aims to describe the clinical course, health care utilization, and expenditures for community-dwelling PWSDs; and perceived burden, coping, resilience, anticipatory and prolonged grief among their caregivers. This ongoing multi-center prospective longitudinal study is recruiting primary informal caregivers of 250 PWSDs from major restructured public hospitals, community hospitals, home care foundations, and hospices in Singapore. Caregivers are surveyed every four months for two years or until the PWSD passes away and then at eight weeks and six months post-death to assess the bereavement of the caregiver. Survey questionnaires included validated tools to assess PWSDs’ quality of life, suffering, behaviors, functional status, resource utilization; and caregiver’s satisfaction with care, awareness of prognosis, care preferences, resilience, coping, perceived burden, distress, positive aspects of caregiving, anticipatory grief, and bereavement adjustment. We also conduct qualitative in-depth interviews with a sub-sample of caregivers. The survey data is being linked with medical and billing records of PWSDs. The study has been approved by an ethics board. Results from the study will be disseminated through publications and presentations targeting researchers, policy makers and clinicians interested in understanding and improving EOL care for PWSDs and their caregivers.

Keywords

Dementia end of life health care utilization palliative care Singapore http://www.clinicaltrials.gov (NCT03382223)

INTRODUCTION

Dementia is a public health priority, increasing in prevalence with an aging population. It currently affects over 46 million people worldwide and is expected to affect 132 million by 2050 [1]. Those in its severe stages usually live for one to three years and experience profound deficits in memory, orientation, function, decision-making abilities, speech, and attention along with severe behavior or personality changes and loss of autonomy [2, 3]. Despite this, dementia is often not perceived to be a life-limiting or terminal condition resulting in less than optimal EOL care [4 –6]. Providing quality end-of-life (EOL) care to persons with severe dementia (PWSDs) and their caregivers thus constitutes an integral and essential part of dementia and palliative care services.

In Singapore (the setting for this study), dementia affects 10% of the elderly population [2]. Many PWSDs in Singapore, and in other Asian countries, are cared for at home by their family caregivers [7 –9]. However, most studies describing EOL experiences of PWSDs have been conducted in Western countries and have focused solely on those residing in nursing homes [10 –14]. Literature is sparse regarding experiences of community-dwelling PWSDs and their caregivers in Singapore and other Asian countries [15]. PWSDs residing in the community differ markedly from those in nursing homes in terms of their disease trajectory, management, and EOL care decision-making [16 –18]. EOL experiences of PWSDs and their caregivers are also likely to vary between countries [19]. As a result, findings from Western countries and from nursing homes offer limited value and direction to optimize care for community-dwelling PWSDs and their caregivers in Singapore and other Asian countries. Providing systematic information regarding the clinical course of community-dwelling PWSDs in Singapore can help identify gaps in current care in order to improve well-being of these PWSDs and potentially to prevent their institutionalization.

Several factors can determine care at EOL for PWSDs. One of these is caregivers’ understanding of the progressive and terminal nature of dementia [20 –22]. Studies among cognitively intact patients with advanced cancer show that most patients lack an accurate understanding of their prognosis and believe that the chemotherapies they are receiving are intended to cure their condition [23]. Such patients are also more likely to choose aggressive treatments at EOL [24]. Less is known about the extent to which caregivers of PWSDs understand the terminal nature of dementia and how this influences EOL care for PWSD, especially in Singapore and in other Asian countries.

Research over the past three decades has also extensively documented the negative psychological impact of caring for people with dementia [25], including a high prevalence of depressive symptoms among caregivers [26 –28]. Studies also show that many caregivers of PWSDs experience complicated grief during bereavement [29 –32]. The relationship between providing high-intensity care to PWSDs, caregiver burden and development of complicated grief symptoms during bereavement is, however, not clearly elucidated.

Several cost-of-illness studies have been conducted among persons with dementia [33 –37]. The worldwide cost of dementia care is estimated to be over US$818 billion and is expected to increase to a trillion dollar budget by 2018 [2]. According to Alzheimer Disease International (2014), the estimated annual cost of dementia in 2015 for Singapore was US$1,664 million [38]. Yet, less is known about annual and EOL direct and indirect costs for PWSDs and the extent to which they vary by factors such as type of health insurance for patients, patient’s comorbidities, presence of an advance care plan, family’s socioeconomic status, presence of a paid domestic helper (again, a factor relatively unique to Singapore and some other Asian countries), and caregiver burden. This data is essential to plan interventions that can potentially reduce the social cost of caring for PWSDs.

Another challenge is physicians’ difficulty in estimating prognosis for community-dwelling PWSDs [39]. Existing guidelines including Functional Assessment Staging (FAST) criteria [40, 41], Advanced Dementia Prognostic Tool [42], Alzheimer’s-Hospice Placement Evaluation Scale [43], Mitchell Novel Risk Score [44], and Mini Suffering State Examination [45 –47] have primarily been used to assess nursing home residents rather than community-dwelling PWSDs [41]. Assessing factors that predict mortality risk among community-dwelling PWSDs will allow physicians to more appropriately and timely refer PWSDs to palliative or hospice care services and their caregivers to make informed decisions regarding PWSD’s EOL care.

To address these gaps in the literature, we initiated data collection for a 3-year prospective longitudinal study of community-dwelling PWSDs and their caregivers in May 2018. The study titled “Panel study Investigating Status of Cognitively impaired Elderly in Singapore (PISCES)” is being conducted in Singapore, a rapidly aging South-East Asian country with a multi-ethnic population [48]. Through PISCES, we aim to 1) describe the clinical course of community-dwelling PWSDs during their EOL period including their physical symptoms, behavioral problems, suffering, clinical complications experienced, use of burdensome interventions, physical restraints, and feeding tubes; 2) quantify their health care use and expenditures; 3) assess factors that predict their mortality; and 4) assess perceived burden, coping, resilience, anticipatory and prolonged grief among their caregivers. Through in-depth qualitative interviews with caregivers of PWSDs, we also aim to understand how caregivers make EOL care decisions for PWSDs.

This study with a mixed-methods design is the first-of-its-kind. It incorporates a prospective longitudinal study to assess the EOL period comprehensively in community-dwelling PWSD and their caregivers. This paper presents the full protocol of this study.

METHODS

Study participants and setting

Primary informal caregivers of 250 community-dwelling PWSDs are surveyed every four months for at least two years or until the PWSD passes away and then at eight weeks and six months post-death to assess bereavement among their caregivers. We also link the survey data with PWSDs’ medical and billing records.

Eligibility criteria for PWSDs include (a) being a Singapore citizen/permanent resident, (b) having a primary/secondary diagnosis of dementia with severity FAST stage 6 C or higher, (c) currently residing and expected to reside in Singapore for the next two years, and (d) having a family member residing in Singapore. PWSDs staying in long-term care facilities, such as nursing homes, are excluded. Inclusion criteria for caregivers are (a) being Singapore citizen/permanent resident, (b) aged ≥21 years, (c) having intact cognition as determined through the Abbreviated Mental Test (AMT) [49] for those aged ≥65 years, (d) family member of the PWSD, (e) currently residing and expected to reside in Singapore for the next 2 years, (f) main person or one of the main persons involved in making decisions regarding PWSD’s treatment or responsible for ensuring the well-being of the PWSD, and (g) meets the PWSD at least one day in a week (regardless of the amount of time spent seeing the PWSD in a day).

After the caregiver of PWSD is recruited in the study, the primary treating physician of that PWSD is also enrolled in the study and interviewed once during the course of the study.

The study protocol and all study-related documents were approved and are monitored by Institutional Review Boards at SingHealth (2017/2989) and National University of Singapore (H-18-024).

Recruitment and follow-ups

Eligible caregivers are recruited from memory/geriatrics clinics and general medicine wards of seven major public restructured hospitals in Singapore, six home care foundations and two hospices. The study workflow is described in Fig. 1. Health care providers at each site identify and approach eligible participants to take part in the study. This helps to improve response rate. Following verbal consent, health care providers send the details of eligible caregivers to the research team.

Fig.1

Study workflow.

As Singapore grants do not provide protected research time for our clinical co-investigators, a significant amount of time and effort is subsequently spent by the research staff hired on this project. A trained research coordinator (RC) from the team contacts the caregiver over the phone to schedule a face-to-face interview. Interviews are held at a time and place convenient to the caregivers. For working caregivers, interviews are also scheduled over weekends and after work hours.

Before initiating the interview, the caregiver provides written informed consent. Caregivers aged ≥65 years answer the Abbreviated Mental Test (AMT). Those without cognitive impairment (threshold score based on age and years of education) [50] subsequently answer the baseline survey questionnaire. Surveys are conducted in the caregivers’ preferred language (English/Mandarin/Malay). As PWSDs cannot be interviewed, all information about them is also collected from the caregivers. Survey data is captured electronically using tablets on Qualtrics. At the end of the survey, caregivers receive $80 as a token of appreciation for their participation.

Table 1 summarizes several ways (Categories A – F) caregivers can participate in the study, depending on their eligibility and interest and the type of data that is ultimately captured.

Table 1

Type of data retrieved from caregiver

	Caregiver Cognitive Status	Records review	Type of data consented to
Category A	Caregiver passed AMT¹	Patient and caregiver	1. Caregiver self-report survey2. Patient records review²3. Caregiver records review²
Category B	Caregiver passed AMT¹	Caregiver declined for both patient and caregiver	1. Caregiver self-report survey
Category C	Caregiver passed AMT¹	Patient and caregiver	1. Patient records review²2. Caregiver records review²
Category D	Caregiver passed AMT¹	Patient (Caregiver declined)	1. Patient records review²
Category E	Caregiver passed AMT¹	Patient (Caregiver declined)	1. Caregiver self-report survey2. Patient records review²
Category F	Caregiver passed AMT¹	Caregiver (Caregiver declined for patient)	1. Caregiver self-report survey2. Caregiver records review²
Scores required to pass AMT (by age group):
	60–74 years old	75+ years old
0 to 6 years of education	8 or more correct	6 or more correct
More than 6 years of education	9 or more correct	9 or more correct

AMT, Abbreviated Mental Test, a cognitive assessment for caregivers above 65 years old; Records review, review of medical and billing data.

Follow-up surveys with caregivers are conducted every four months for two years or until the PWSD’s death. A follow-up time of four months was scheduled to allow us to capture adequately the health care use, acute medical complications, changes in EOL care preferences and other dimensions of EOL care with minimal recall bias. We are using a similar approach for our other ongoing cohort studies with advanced cancer patients and heart failure patients [51, 52].

Interviews with caregivers assessing bereavement adjustment are conducted at eight weeks and six months after PWSD’s death. Medical and billing records of PWSDs are assessed for the treatments received and related costs and linked with survey records. A one-time survey is also administered to the primary treating physician of individuals at the time of recruitment.

Each caregiver is identified through a unique study code in the dataset. The project manager of the study checks survey responses every week for data quality. These include checking for consistency of responses across similar survey questions, reasonable range of responses and responses suggesting that a participant may not be engaging completely with the survey, e.g., always selects the same survey response option. A monthly progress report is sent to all investigators to update them about the study progress.

Outcome measures

We used standardized translations for all validated scales in the survey when available from developers, or translated and back-translated these scales into local languages (Mandarin and Malay). Bilingual study team members further reviewed the translations for equivalence and cultural appropriateness. We then conducted cognitive interviews with ten eligible caregivers to assess comprehension and relevance of our survey to the local context. We revised the survey to improve comprehension based on the feedback from the interviewer and respondents. The final survey assesses a broad range of topics that include caregivers’ and PWSDs’ quality of life, caregivers’ perception of quality of care for PWSDs, use of burdensome interventions for PWSDs, PWSDs’ health care utilization and the caregivers’ understanding of the PWSDs’ prognosis. If caregivers consent, medical and billing records are accessed for PWSDs’ and caregivers’ diagnostic and treatment information and costs. Data from medical records are linked to information from surveys. The following outcomes are measured in the study (details of scales are in Table 2):

Table 2

Summary of scales used in the survey

Domain	Sub-domain	Instrument	Subscales	Items	Response categories	Whether previously used in Singapore/other Asian countries
Clinical course of PWSDs	Quality of life	Quality of Life in Late-Stage Dementia Scale	None	11	5 response categories with increasing/decreasing frequency of the behavior	Used in Singapore [71]
	Patient’s suffering	Modified Version of Mini Suffering State Examination	None	8	3 response categories:•“Yes”,•“No” &•“Not sure”	Used in Singapore, Taiwan and Japan [72 –74]
	Agitated Behaviors	Cohen-Mansfield Agitation Inventory (Short form)	4 subscales:•Physical Non-Aggression,•Physical Aggression,•Verbal Non-Aggression,•Verbal Aggression	14	5 response categories:•1: Never•2: Less than once a week•3: Once or several times a week•4: Once or several times a day•5: A few times an hour or continuous for half an hour or more	Validated in Hong Kong [75]
	Functional status	Bedford Alzheimer Nursing Severity Scale	None	7	4 response categories that scale the frequency of the behaviors	–
Satisfaction with care received by PWSDs	Caregiver’s satisfaction with care received	Satisfaction with Care – End of Life in Dementia	None	10	4 response categories:•1: Strongly Disagree•2: Disagree•3: Agree•4: Strongly Agree	–
Resource utilization for PWSDs	Assessment of caregivers’ time spent on caregiving tasks	1. Resource Utilization in Dementia Lite2. Adapted questions from existing studies	2 subscales:•Caregiver’s healthcare resources and time spent on patient,•Patient’s utilization of healthcare resources	48	Mixed	Used in Singapore [76]
Caregiver resilience and coping	Resilience	Connor Davidson Resilience Scale	None	2	5 response categories:•1: Not true at all•2: Rarely true•3: Sometimes true•4: Often true•5: True nearly all of the time	Validated in Singapore [77]
	Coping	BriefCOPE	13 subscales:•Venting•Emotional support•Instrumental support•Religion•Active coping•Planning•Behavioral disengagement•Self-distraction•Denial•Self-blame•Humor•Positive reframing•Acceptance	26	•4 response categories:1: I haven’t been doing this at all•2: I’ve been doing this a little bit•3: I’ve been doing this a medium amount•4: I’ve been doing this a lot”	Validated in China [78]
Positive aspects of caregiving	Positive aspects of caregiving	Gain in Alzheimer Care Instrument	3 subscales:•Personal growth,•Gains in relationships,•Higher level gains	10	•5 response categories:1: Disagree a lot•2: Disagree a little•3: Neither agree nor disagree•4: Agree a little•5: Agree a lot	Developed and validated in Singapore [60]
Caregiver burden and psychological distress	Negative caregiving reaction	Caregiver Reaction Assessment	3 subscales:•Impact on schedule and health•Impact on finances•Lack of family support	15	5 response categories:•1: Strongly disagree•2: Disagree•3: Somewhat agree•4: Agree•5: Strongly agree	Validated in Singapore [48]
	Psychological distress	Hospital Anxiety and Depression Scale	2 subscales:•Anxiety•Depression**	14	4 response categories of decreasing severity of anxiety/depression	Validated in Singapore [79]
Caregiver anticipatory grief	Caregivers’ anticipatory grief	Marwit-Meuser Caregiver Grief Inventory	3 subscales:•Personal sacrifice•Heartfelt sadness and longing•Worry and felt isolation	18	5 response categories:•1: Strongly agree•2: Agree•3: Neither disagree nor agree•4: Disagree•5: Strongly disagree	Validated in Singapore [80]
Caregiver Quality of life 8 weeks and 6 months after PWSD’s death.	Caregiver’s quality of life after patient’s death	European Organization for Research and Treatment of Cancer	2 subscales:•Overall health•Overall quality of life	2	Likert Scale ranging from 1 (Very poor) to 7 (Excellent)	Validated in Singapore [81]
	Caregiver’s spiritual well-being 6 months after patient’s death	FACIT – Sp - Non-Illness	2 subscales:•Meaning/peace•Faith	12	5 response categories:•0: Not at all•1: A little bit•2: Somewhat•3: Quite a bit•4: Very much	Used in Singapore [82]
Bereavement Adjustment 6 months after PWSD’s death.	Caregiver’s bereavement adjustment	Brief Grief Questionnaire	None	7	Mixed	Validated in Japan [83]

Clinical course of PWSDs. Caregivers are Administered several validated measures to assess the clinical course of PWSDs including their symptoms, functional status, and behaviors. Quality of Life in Patient with Dementia, an eleven-item measure, is used to assess PWSDs’ engagement in basic activities over the last week [53]. PWSDs’ suffering is assessed through the Modified version of the Mini Suffering State Examination [45 –47]. Their behaviors are assessed using the fourteen-item Cohen-Mansfield Agitation Inventory–Short Form [54]. Their functional status is assessed using the seven-item Bedford Alzheimer Nursing Severity Scale [55]. We also collect caregivers’ report on PWSDs’ comorbidities, acute medical conditions in the past four months, and use of medical interventions, feeding tube, and physical restraints.

Satisfaction with care received by PWSDs. Caregivers’ satisfaction with the overall care received by the PWSD in the past four months is assessed using the ten-item Satisfaction with Care at the End-of-Life in Dementia [56].

Resource utilization for PWSDs. Resources utilized for the care of PWSDs include caregivers’ time, healthcare services, and finances such as the inpatient bill, pharmacy bills, outpatient bills, emergency department bills, hospice/homecare/day care agency bills, and nursing home bills. We use items from Resource Utilization in Dementia-Lite to assess caregivers’ time spent on caregiving tasks [57]. These include Activities of Daily Living (ADLs; assisting the patient with toilet visits, eating, dressing, grooming, walking and bathing), Instrumental ADLs (shopping, food preparation, housekeeping, laundry, transportation, taking medication and managing financial matters), and supervising the patient. Patients’ living accommodation and caregivers’ work status are also assessed.

Awareness of prognosis and care preferences. Caregivers’ awareness and understanding of the PWSDs’ prognosis is assessed through questions explicitly developed for the study. Items include caregivers’ level of agreement with the statement that dementia is a disease that one can die from, how old the PWSD might be when he/she passes away, what healthcare providers have told the caregiver about how long the PWSD may live, the primary goal of the PWSDs’ treatment and the caregivers’ preferences for the PWSDs’ treatment.

Caregiver resilience and coping. Caregivers’ adaptation to their caregiving responsibilities and coping with stress is measured using the Connor-Davidson Resilience Scale [58] and BriefCOPE [59]. Based on pilot interviews with caregivers, questions on substance use were removed from the survey to reduce respondent burden.

Positive aspects of caregiving. These are assessed using Gain in Alzheimer care Instrument (GAIN), which was developed and validated in Singapore (ten items) [60].

Caregiver burden and psychological distress. We use four subscales from the modified Caregiver Reaction Assessment Scale (fifteen items on a five-point scale) [61, 62] to understand negative reactions of caregivers to their caregiving role [61, 62]. We assess caregivers’ psychological distress (defined as depression and anxiety) using the Hospital Anxiety and Depression Scale [63, 64].

Caregiver anticipatory grief. Caregivers’ anticipatory grief is measured using the Singapore version of the Marwit Meuser Caregiver Grief Inventory-Short Form (eighteen items) [65].

Caregiver quality of life 8 weeks and 6 months after PWSD’s death. At 8 weeks and 6 months after PWSD’s death, caregivers are asked to rate their overall health and overall quality of life during the past week on a seven-point Likert scale. Additionally, at 8 weeks, we measure caregivers’ reactions to PWSDs’ death through five items assessing their preparedness for PWSDs’ death, the regrets they have, their mood in the last week, and whether the PWSDs’ death was a relief for the caregivers. At 6 months we also assess their spiritual well-being through the Functional Assessment of Chronic Illness Therapy-Spiritual Well-Being, modified version for non-illness [66 –68].

Bereavement adjustment 6 months after PWSD’s death. Caregivers are asked questions on their adjustment to the death of the PWSD through seven items which include their trouble in accepting the death of the PWSD, their amount of grief, how the PWSD’s death bothers them, things they used to do when the PWSD was alive that they no longer engage in, feeling cut off or distant from other people since the PWSD’s death, their mood in the last week, and if they need any assistance with coping with the loss of the PWSD.

Sample size calculation

The study is not a randomized controlled trial, but primarily descriptive and exploratory for multiple outcomes. Therefore, we considered sample sizes for estimating main effects averaged over all follow-up times as well as sample sizes for testing hypotheses involving proportions and means at any given follow-up time. The goal was to enroll an appropriate number of subjects to ensure adequate precision in estimation and power in hypothesis testing among subgroups and at the same time ensuring feasibility. We calculated that for an initial sample size of N = 250, in an analysis of subgroups consisting of n₁ = 100 and n₂ = 150 subjects with repeated measures over all follow-up times, and intra-class correlations of 0.20–0.50, will achieve 80% power at α= 0.05 to detect differences in main effect proportions of 0.10–0.13 (OR 1.50–1.74) and effect size (Δ/σ) differences in main effect means of 0.20–0.27. With the anticipated reduction in cohort sample size owing to mortality and attrition, a sample size of N = 200 and subgroups of n₁ = 75 and n₂ = 125 will achieve 80% power to detect differences in main effect proportions of 0.11–0.15 (OR 1.59–1.87) and effect size differences in main effects means of 0.23–0.31. We also calculated that at any given follow-up time a sample size of N = 250 with subgroups of n₁ = 100 and n₂ = 150 will achieve 80% power to detect differences in proportions of 0.18 (OR = 2.15) and effect size differences between means of 0.36.

Statistical analysis

Logistic mixed-effects models will be used to investigate PWSD and caregiver factors associated with the following PWSD outcomes: 1) receiving burdensome interventions, 2) use of tube feeding, and 3) use of physical restraint. Similarly, we will use Poisson mixed effects models to investigate factors associated with the PWSD’s 1) number of transitions in healthcare settings, 2) number of physical symptoms, 3) number of behavioral problems, and 4) number of hospitalizations, adjusting for confounders.

We anticipate using linear mixed effects models to investigate association between the following outcomes 1) caregiver bereavement adjustment scores, 2) PWSD quality of life score, and 3) PWSD suffering score and the independent variables - number of physical symptoms, receiving aggressive life-extending treatments, physical restraints, numbers of transitions in healthcare settings. We will also use linear mixed effects models to investigate the associations of perceived caregiver burden score (outcome) with caregiver stressors, coping, resilience, and assistance from a paid domestic helper. The interaction effects between caregiver stressors and caregiver resilience will also be tested.

The mixed effects model can handle data that are missing at random (MAR), i.e., chance of being missing does not depend on unobserved outcome values but can depend on covariates and observed outcomes. Assuming that the missing data mechanism involved in the current study satisfies the MAR assumption, we will retain participants with missing outcomes in the analyses rather than delete them.

To quantify PWSDs’ health care use and expenditures, we will estimate the total annual direct medical care, direct social care and indirect costs through an opportunity cost approach (loss of wage due to caregiving) using an appropriate generalized linear model specification. Information on direct medical costs will be assessed through patient bills from all sources of care. Indirect costs will be estimated through surveys. We will also analyze costs by time from death (e.g., last year or last six months) and by type of service (in total and separately for inpatient, non-inpatient, physician’s office, prescription drugs, and alternative medicines). Indirect costs will be stratified by the financial value of both formal and informal care giving, and other indirect costs (e.g., transportation costs). Predictor variables will be attitudes of caregivers towards life extension or quality of life, presence of advance care plan at baseline, presence of a paid domestic helper, and receiving specialist/community palliative care.

We will use univariable and multivariable logistic regression analysis to identify and assess risk factors for PWSD mortality at selected time points.

All analyses will be adjusted for potential confounding between outcomes and main independent variables. Possible confounders may include length of diagnosis with dementia, type of dementia, patient demographics including age, gender, education and ethnic group, caregiver’s relationship with PWSD and caregiver demographics.

Qualitative interviews

Qualitative in-depth interviews are being conducted with at least 10% of the recruited sample of the caregivers to gain a deeper understanding of EOL care decision making processes and the factors that influence their decisions. Interviews, conducted in the caregivers’ language of preference, are semi-structured, based on an interview guide, take about 1 to 3 hours to be completed and are audio recorded. The interview guide consists of open-ended questions that allow the participants to engage in the interviews, accompanied by probing and follow-up questions by the interviewer. The questions are primarily focused on eliciting caregivers’ overall goals of care and treatment preferences for PWSDs, factors that influence their treatment preferences including their perception of PWSDs’ illness, prognosis and quality of life, pressure and support from health care professionals and other family members, religious/spiritual beliefs, caregiver burden and fear of regretting their decision. The interviews are transcribed verbatim, translated to English (when necessary), coded in QSR Nvivo 11 and analyzed using thematic analysis [69].

DISCUSSION

By using a mix of prospective surveys, qualitative in-depth interviews, and medical and billing record abstraction and linkage, this study will be the first-of-its-kind to present a comprehensive picture of the EOL experiences of community-dwelling PWSDs and their caregivers.

These results from the study will be used for policy analysis and for developing interventions to reduce personal and social costs of caring for PWSDs and in efforts to reduce their suffering and caregiver burden. Results will also inform the development of appropriate care pathways, which can foster a comprehensive palliative support system to continue to care for these PWSDs in the community. In addition, the data collected will also help assess factors that predict 6-month and 1-year mortality for community-dwelling PWSDs. Further validation of the risk score will be conducted using a larger dataset in a future study. Using this risk score will ultimately help caregivers make informed EOL decisions for PWSDs, and physicians identify PWSDs eligible for community palliative/hospice care services. Understanding the role that resilience and coping play in buffering the effect of caregiving stressors on negative aspects of caregiving will allow dementia care services to develop practical interventions that focus on improving caregiver resilience and cultivating positive coping mechanisms.

The study is challenging in that it aims to recruit a large number of PWSDs from several hospitals and home care services in a short period of time. In the absence of a registry/sampling frame of PWSD, it is difficult to capture a representative population of PWSD. Given this challenge, as of yet, there has not been any past study in Singapore, or Asia, investigating EOL care for PWSD and their caregivers. Therefore, we are collaborating with all major practitioners caring for PWSD in Singapore including those in public restructured hospitals and home care teams. Results may still not be generalizable as some caregivers do not seek care from any hospital or home care team due to financial concerns and because they perceive cognitive decline to be a part of normal ageing process [33].

Due to stigma related to dementia [70], the study also posed some ethical challenges. To reduce these challenges, we are conducting all interviews in privacy at caregivers’ homes. All interviews are scheduled based on caregiver’s convenience after discussing with them over phone. Caregivers are assured that their data will be kept confidential. Caregivers engage with the same interviewer, when possible, for all follow-up interviews. Interviewers are also trained by the research team to avoid the use of any stigmatizing language. Lastly, attrition (for reasons other than the death of PWSDs) may be a problem. We will try to minimize this through appropriate financial incentives and scheduling interviews at a time and place convenient for caregivers. Data collection for this study was initiated in May 2018. To date, we have recruited 150 caregivers. We expect to attain our target sample size by October 2020 and follow-up assessments to continue for two years.

In summary, the information from this first-of-its-kind longitudinal study will not only help facilitate a deeper understanding of the most critical issues facing community-dwelling PWSD and their families, but will allow for the formulation of programs and policy solutions to effectively address these issues which have thus far received limited attention. The recommendations that will ultimately be generated from the study results will recognize the diversity of PWSD as well as caregiver experiences and preferences likely to occur at the EOL. With that, an appropriate care delivery system can be built to embrace those differences and provide a customized and affordable EOL experience that respects the autonomy of both PWSDs and caregivers.

Footnotes

ACKNOWLEDGMENTS

We would like to thank the team involved at each site for their assistance in the recruitment and screening of study participants, and our team of data collectors (Lien Centre for Palliative Care) for their contributions to the study. The study is funded by Health Services Research Grant (HSRGEoL16Dec002), Ministry of Health, Singapore.

Authors’ disclosures available online ().

References

Prince

, Wimo

, Guerchet

, Ali

, Wu

, Prina

(2015) World Alzheimer Report 2015. The Global Impact of Dementia: An analysis of prevalence, incidence, cost and trends. Alzheimer’s Disease International, London, UK.

Subramaniam

, Chong

, Vaingankar

, Abdin

, Chua

, Eng

, Heng

, Hia

, Huang

, Jeyagurunathana

, Kua

, Lee

, Mahendran

, Magadi

, Malladi

, McCrone

, Pang

, Picco

, Sagayadevan

, Sambasivam

, Seng

, Seow

, Shafie

, Shahwan

, Tan

, Yap

, Zhang

, Ng

, Prince

(2015) Prevalence of dementia in people aged 60 years and above: results from the WiSE Study. J Alzheimers Dis 45, 1127–1138.

Hodgson

, Gitlin

, Winter

, Hauck

(2014) Caregiver’s perceptions of the relationship of pain to behavioral and psychiatric symptoms in older community-residing adults with dementia. Clin J Pain 30, 421–427.

Sampson

, Burns

, Richards

(2011) Improving end-of-life care for people with dementia. Br J Psychiatry 199, 357–359.

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

(2017) Dementia prevention, intervention, and care. Lancet 390, 2673–2734.

Sampson

(2010) Palliative care for people with dementia. Br Med Bull 96, 159.

Hsiao

, Chao

, Wang

(2013) Features of problematic eating behaviors among community-dwelling older adults with dementia: family caregivers’ experience. Geriatr Nurs 34, 361–365.

Tew

, Tan

, Luo

, Ng

, Yap

(2010) Why family caregivers choose to institutionalize a loved one with dementia: a Singapore perspective. Dement Geriatr Cogn Disord 30, 509–516.

Vaingankar

, Subramaniam

, Picco

, Eng

, Shafie

, Sambasivam

, Zhang

, Sagayadevan

, Chong

(2013) Perceived unmet needs of informal caregivers of people with dementia in Singapore. Int Psychogeriatr 25, 1605–1619.

10.

Olsen

, Pedersen

, Bergland

, Enders-Slegers

, Joranson

, Calogiuri

, Ihlebaek

(2016) Differences in quality of life in home-dwelling persons and nursing home residents with dementia - a cross-sectional study. BMC Geriatr 16, 137.

11.

Robinson

, Dickinson

, Rousseau

, Beyer

, Clark

, Hughes

, Howel

, Exley

(2012) A systematic review of the effectiveness of advance care planning interventions for people with cognitive impairment and dementia. Age Ageing 41, 263–269.

12.

Mitchell

(2015) Advanced dementia. N Engl J Med 372, 2533–2540.

13.

Cohen-Mansfield

, Dakheel-Ali

, Marx

, Thein

, Regier

(2015) Which unmet needs contribute to behavior problems in persons with advanced dementia? Psychiatry Res 228, 59–64.

14.

Shega

, Hougham

, Stocking

, Cox-Hayley

, Sachs

(2008) Patients dying with dementia: experience at the end of life and impact of hospice care. J Pain Symptom Manage 35, 499–507.

15.

Sternberg

, Bentur

, Shuldiner

(2014) Quality of care of older people living with advanced dementia in the community in Israel. J Am Geriatr Soc 62, 269–275.

16.

Mitchell

, Kiely

, Hamel

(2004) Dying with advanced dementia in the nursing home. Arch Intern Med 164, 321–326.

17.

Hoffmann

, Kaduszkiewicz

, Glaeske

, van den Bussche

, Koller

(2014) Prevalence of dementia in nursing home and community-dwelling older adults in Germany. Aging Clin Exp Res 26, 555–559.

18.

Gonzalez-Colaco Harmand

, Meillon

, Rullier

, Avila-Funes

, Bergua

, Dartigues

, Amieva

(2014) Cognitive decline after entering a nursing home: a 22-year follow-up study of institutionalized and noninstitutionalized elderly people. J Am Med Dir Assoc 15, 504–508.

19.

Goodman

, Evans

, Wilcock

, Froggatt

, Drennan

, Sampson

, Blanchard

, Bissett

, Iliffe

(2010) End of life care for community dwelling older people with dementia: an integrated review. Int J Geriatr Psychiatry 25, 329–337.

20.

van der Steen

, Onwuteaka-Philipsen

, Knol

, Ribbe

, Deliens

(2013) Caregivers’ understanding of dementia predicts patients’ comfort at death: a prospective observational study. BMC Med 11, 105.

21.

Mitchell

, Teno

, Kiely

, Shaffer

, Jones

, Prigerson

, Volicer

, Givens

, Hamel

(2009) The clinical course of advanced dementia. N Engl J Med 361, 1529–1538.

22.

Thuné-Boyle

, Sampson

, Jones

, King

, Lee

, Blanchard

(2010) Challenges to improving end of life care of people with advanced dementia in the UK. Dementia 9, 259–284.

23.

Weeks

, Catalano

, Cronin

, Finkelman

, Mack

, Keating

, Schrag

(2012) Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med 367, 1616–1625.

24.

Tang

, Liu

, Chow

, Chiu

, Hsieh

, Chen

, Liu

, Feng

(2014) Associations between accurate prognostic understanding and end-of-life care preferences and its correlates among Taiwanese terminally ill cancer patients surveyed in 2011–2012. Psychooncology 23, 780–787.

25.

, Cooper

, Bradley

, Shulman

, Livingston

(2012) Coping strategies and psychological morbidity in family carers of people with dementia: A systematic review and meta-analysis. J Affect Disord 139, 1–11.

26.

Schoenmakers

, Buntinx

, Delepeleire

(2010) Factors determining the impact of care-giving on caregivers of elderly patients with dementia. A systematic literature review. Maturitas 66, 191–200.

27.

Cooper

, Katona

, Orrell

, Livingston

(2008) Coping strategies, anxiety and depression in caregivers of people with Alzheimer’s disease. Int J Geriatr Psychiatry 23, 929–936.

28.

Borsje

, Hems

MAP

, Lucassen

PLBJ

, Bor

, Koopmans

RTCM

, Pot

(2016) Psychological distress in informal caregivers of patients with dementia in primary care: course and determinants. Fam Pract 33, 374–381.

29.

Adams

, Sanders

(2004) Alzheimer’s caregiver differences in experience of loss, grief reactions and depressive symptoms across stage of disease: a mixed-method analysis. Dementia 3, 195–210.

30.

Chan

, Livingston

, Jones

, Sampson

(2013) Grief reactions in dementia carers: a systematic review. Int J Geriatr Psychiatry 28, 1–17.

31.

Meuser

, Marwit

(2001) A comprehensive, stage-sensitive model of grief in dementia caregiving. Gerontologist 41, 658–670.

32.

Ott

, Sanders

, Kelber

(2007) Grief and personal growth experience of spouses and adult-child caregivers of individuals with Alzheimer’s disease and related dementias. Gerontologist 47, 798–809.

33.

Rice

, Fox

, Max

, Webber

, Lindeman

, Hauck

, Segura

(1993) The economic burden of Alzheimer’s disease care. Health Aff (Millwood) 12, 164–176.

34.

Kelley

, McGarry

, Gorges

, Skinner

(2015) The burden of health care costs for patients with dementia in the last 5 years of life. Ann Intern Med 163, 729–736.

35.

Joling

, Schöpe

, van Hout

HPJ

, van Marwijk

HWJ

, van der Horst

, Bosmans

(2015) Predictors of societal costs in dementia patients and their informal caregivers: a two-year prospective cohort study. Am J Geriatr Psychiatry 23, 1193–1203.

36.

Jönsson

, Wimo

(2009) The cost of dementia in Europe. Pharmacoeconomics 27, 391–403.

37.

Gustavsson

, Brinck

, Bergvall

, Kolasa

, Wimo

, Winblad

, Jönsson

(2011) Predictors of costs of care in Alzheimer’s disease: A multinational sample of 1222 patients. Alzheimers Dement 7, 318–327.

38.

Alzheimer’s Disease International, Alzheimer’s Australia (2014) Dementia in the Asia Pacific Region. Alzheimer’s Disease International, London, UK.

39.

Sachs

, Shega

, Cox-Hayley

(2004) Barriers to excellent end-of-life care for patients with dementia. J Gen Intern Med 19, 1057–1063.

40.

Reisberg

(1988) Functional Assessment Staging (FAST). Psychopharmacol Bull 24, 653–659.

41.

Sclan

, Reisberg

(1992) Functional assessment staging (FAST) in Alzheimer’s disease: reliability, validity, and ordinality. Int Psychogeriatr 4 Suppl 1, 55–69.

42.

Mitchell

, Miller

, Teno

, Davis

, Shaffer

(2010) The advanced dementia prognostic tool: a risk score to estimate survival in nursing home residents with advanced dementia. J Pain Symptom Manage 40, 639–651.

43.

Marsh

, Prochoda

, Pritchett

, Vojir

(2000) Predicting hospice appropriateness for patients with dementia of the Alzheimer’s type. Appl Nurs Res 13, 187–196.

44.

Mitchell

, Kiely

, Hamel

, Park

, Morris

, Fries

(2004) Estimating prognosis for nursing home residents with advanced dementia. JAMA 291, 2734–2740.

45.

Aminoff

(2004) Mini Suffering State Examination (MSSE) Scale: first 5 years. Isr Med Assoc J 6, 645–646.

46.

Aminoff

, Purits

, Noy

, Adunsky

(2004) Measuring the suffering of end-stage dementia: reliability and validity of the Mini-Suffering State Examination. Arch Gerontol Geriatr 38, 123–130.

47.

Aminoff

, Adunsky

(2006) Their last 6 months: suffering and survival of end-stage dementia patients. Age Ageing 35, 597–601.

48.

Malhotra

, Bautista

MAC

, Müller

, Aw

, Koh

GCH

, Theng

, Hoskins

, Wong

, Miao

, Lim

, Malhotra

, Chan

(2018) The aging of a young nation: population aging in Singapore. Gerontologist 59, 401–410.

49.

Sahadevan

, Tan

(1997) Cognitive testing of elderly Chinese people in Singapore: influence of education and age on normative scores. Age Ageing 26, 481–486.

50.

Sahadevan

, Lim

, Tan

, Chan

(2000) Diagnostic performance of two mental status tests in the older chinese: influence of education and age on cut-off values. Int J Geriatr Psychiatry 15, 234–241.

51.

Teo

, Singh

, Malhotra

, Ozdemir

, Dent

, Kumarakulasinghe

, Yeo

, Cheung

, Malhotra

, Kanesvaran

, Yee

ACP

, Chan

, Wu

, Chin

, Allyn

HYM

, Yang

, Neo

PSH

, Nadkarni

, Harding

, Finkelstein

(2018) Cost of Medical Care of Patients with Advanced Serious Illness in Singapore (COMPASS): prospective cohort study protocol. BMC Cancer 18, 459–459.

52.

Malhotra

, Foo

, Singh

, Ozdemir

, Teo

, Sim

, Jaufeerally

, Aung

, Keong

, Nadkarni

, Finkelstein

(2018) Study protocol for a cohort study of patients with advanced heart failure in Singapore. BMJ Open 8, e022248.

53.

Weiner

, Martin-Cook

, Svetlik

, Saine

, Foster

, Fontaine

(2000) The quality of life in late-stage dementia (QUALID) scale. J Am Med Dir Assoc 1, 114–116.

54.

Cohen-Mansfield

(1991), Instruction manual for the Cohen-Mansfield Agitation Inventory (CMAI).

55.

Volicer

, Hurley

, Lathi

, Kowall

(1994) Measurement of severity in advanced Alzheimer’s disease. J Gerontol 49, M223–226.

56.

Volicer

, Hurley

, Blasi

(2001) Scales for evaluation of end-of-life care in dementia. Alzheimer Dis Assoc Disord 15, 194–200.

57.

Wimo

, Winblad

(2003), Resource utilisation in dementia: RUD Lite.

58.

Vaishnavi

, Connor

, Davidson

JRT

(2007) An abbreviated version of the Connor-Davidson Resilience Scale (CD-RISC), the CD-RISC2: psychometric properties and applications in psychopharmacological trials. Psychiatry Res 152, 293–297.

59.

Carver

(1997) You want to measure coping but your protocol’s too long: consider the brief COPE. Int J Behav Med 4, 92–100.

60.

Yap

, Luo

, Ng

, Chionh

, Lim

, Goh

(2010) Gain in Alzheimer care INstrument—a new scale to measure caregiving gains in dementia. Am J Geriatr Psychiatry 18, 68–76.

61.

Given

, Given

, Stommel

, Collins

, King

, Franklin

(1992) The caregiver reaction assessment (CRA) for caregivers to persons with chronic physical and mental impairments. Res Nurs Health 15, 271–283.

62.

Malhotra

, Chan

, Malhotra

, Ostbye

(2012) Validity and reliability of the Caregiver Reaction Assessment scale among primary informal caregivers for older persons in Singapore. Aging Ment Health 16, 1004–1015.

63.

Zigmond

, Snaith

(1983) The hospital anxiety and depression scale. Acta Psychiatr Scand 67, 361–370.

64.

Yang

, Ding

, Hu

, Zhang

, Sheng

(2014) Reliability and validity of a Chinese version of the HADS for screening depression and anxiety in psycho-cardiological outpatients. Compr Psychiatry 55, 215–220.

65.

Marwit

, Chibnall

, Dougherty

, Jenkins

, Shawgo

(2008) Assessing pre-death grief in cancer caregivers using the Marwit-Meuser Caregiver Grief Inventory (MM-CGI). Psychooncology 17, 300–303.

66.

FACIT Questionnaires, https://www.facit.org/FACITOrg/Questionnaires.

67.

Peterman

, Fitchett

, Brady

, Hernandez

, Cella

(2002) Measuring spiritual well-being in people with cancer: The functional assessment of chronic illness therapy—spiritual well-being scale (FACIT-Sp). Ann Behav Med 24, 49–58.

68.

Bredle

, Salsman

, Debb

, Arnold

, Cella

(2011) Spiritual well-being as a component of health-related quality of life: The Functional Assessment of Chronic Illness Therapy—Spiritual Well-Being Scale (FACIT-Sp). Religions 2, 77–94.

69.

Braun

, Clarke

(2006) Using thematic analysis in psychology. Qual Res Psychol 3, 77–101.

70.

Setia

, Islam

, Thompson

, Matchar

(2011) Stakeholders’ perspective on issues and challenges associated with care and treatment of aging-related cognitive impairment disorders in Singapore. Int Psychogeriatr 23, 1421–1432.

71.

Hum

, Tay

, Wong

YKY

, Ali

, Leong

IYO

, Wu

, Chin

, Lee

AOK

, Koh

MYH

(2019) Advanced dementia: an integrated homecare programme. BMJ Support Palliat Care, doi: 10.1136/bmjspcare-2019-001798.

72.

Tan

, Seng

(2012) First presentation psychosis among the elderly in Singapore. Singapore Med J 53, 463–467.

73.

Shigemori

, Ohgi

, Okuyama

, Shimura

, Schneider

(2010) The factorial structure of the Mini-Mental State Examination (MMSE) in Japanese dementia patients. BMC Geriatr 10, 36.

74.

Liang

, Chou

, Chen

, Wang

, Lin

, Chen

, Lin

, Liu

, Loh

(2017) Delaying cognitive and physical decline through multidomain interventions for residents with mild-to-moderate dementia in dementia care units in Taiwan: A prospective cohort study. Geriatr Gerontol Int 17 Suppl 1, 36–43.

75.

Choy

, Lam

, Chan

, Li

, Chiu

(2001) Agitation in Chinese elderly: validation of the Chinese version of the Cohen-Mansfield Agitation Inventory. Int Psychogeriatr 13, 325–335.

76.

Chong

, Tan

, Chan

, Lim

, Ali

, Ang

, Chua

(2013) Cost of informal care for community-dwelling mild-moderate dementia patients in a developed Southeast Asian country. Int Psychogeriatr 25, 1475–1483.

77.

Kwan

, Ng

, Lim

, Fong

, Phang

, Chew

, Lui

, Tan

, Thumboo

, Ostbye

, Leung

(2019) Validity and reliability of the ten-item Connor-Davidson Resilience Scale (CD-RISC10) instrument in patients with axial spondyloarthritis (axSpA) in Singapore. Rheumatol Int 39, 105–110.

78.

, Lau

, Mak

, Choi

, Feng

, Chen

, Liu

, Chen

, Song

, Zhang

, Zhao

, Zhu

, Cheng

(2015) A preliminary validation of the Brief COPE instrument for assessing coping strategies among people living with HIV in China. Infect Dis Poverty 4, 41.

79.

Beck

, Tan

, Lum

, Lim

, Krishna

(2016) Validation of the emotion thermometers and hospital anxiety and depression scales in Singapore: Screening cancer patients for distress, anxiety and depression. Asia Pac J Clin Oncol 12, e241–249.

80.

Liew

, Yeap

, Koh

GC-H

, Gandhi

, Tan

, Luo

, Yap

(2018) Detecting predeath grief in family caregivers of persons with dementia: validity and utility of the Marwit-Meuser Caregiver Grief Inventory in a multiethnic Asian population. Gerontologist 58, e150–e159.

81.

Luo

, Fones

CSL

, Lim

, Xie

, Thumboo

, Li

SCJQoLR

(2005) The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30): Validation of English version in Singapore. Qual Life Res 14, 1181–1186.

82.

Yang

, Tan

, Cheung

, Lye

, Lim

SHA

, Ng

, Puchalski

, Neo

PSH

(2017) Effect of a spiritual care training program for staff on patient outcomes. Palliat Support Care 15, 434–443.

83.

Ito

, Nakajima

, Fujisawa

, Miyashita

, Kim

, Shear

, Ghesquiere

, Wall

(2012) Brief measure for screening complicated grief: reliability and discriminant validity. PLoS One 7, e31209.