A comprehensive profile of the sociodemographic, psychosocial and health characteristics of Ontario home care clients with dementia - CDIC: Vol 34, No 2-3, July 2014

Volume 34 · Number 2-3 · July 2014

A comprehensive profile of the sociodemographic, psychosocial and health characteristics of Ontario home care clients with dementia

M. Vu, MSc (1, 2); D. B. Hogan, MD (3, 4); S. B. Patten, MD, PhD (4); N. Jetté, MD, MSc (4, 5); S. E. Bronskill, PhD (6, 7); G. Heckman, MD (2); M. J. Kergoat, MD (8); J. P. Hirdes, PhD (2); X. Chen, MMSc (2); M. M. Zehr, BSc (1); C. J. Maxwell, PhD (1, 2, 4, 6)

https://doi.org/10.24095/hpcdp.34.2/3.08

This article has been peer reviewed.

Author references:

1. School of Pharmacy, University of Waterloo, Waterloo, Ontario, Canada
2. School of Public Health & Health Systems, University of Waterloo, Waterloo, Ontario, Canada
3. Division of Geriatric Medicine, Department of Medicine, University of Calgary, Calgary, Alberta, Canada
4. Department of Community Health Sciences and Institute for Public Health, University of Calgary, Calgary, Alberta, Canada
5. Department of Clinical Neurosciences and Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada
6. Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
7. Institute of Health Policy, Management & Evaluation, University of Toronto, Toronto, Ontario, Canada
8. Centre de recherche, Institut universitaire de gériatrie de Montréal, Département de médecine, Université de Montréal, Montréal, Quebec, Canada

Correspondence: Colleen J. Maxwell, School of Pharmacy, University of Waterloo, 200 University Avenue West, Waterloo, ON N2L 3G1; Tel.: 519-888-4567 ext. 21396; Fax: 519-883-7580; Email: colleen.maxwell@uwaterloo.ca

Abstract

Introduction: This study provides a comprehensive summary of the sociodemographic, psychosocial and health characteristics of a large population-based cohort of Ontario home care clients (aged 50 years and over) with dementia and examines the variation in these characteristics in those with co-existing neurological conditions.

Methods: Clients were assessed with the Resident Assessment Instrument-Home Care (RAI-HC) between January 2003 and December 2010. Descriptive analyses examined the distribution of these characteristics among clients with dementia relative to several comparison groups, as well as clients with other recorded neurological conditions.

Results: Approximately 22% of clients (n = 104 802) had a diagnosis of dementia (average age 83 years, 64% female) and about one in four within this group had a co-existing neurological condition (most commonly stroke or Parkinson disease). About 43% of those with dementia did not live with their primary caregiver. Relative to several comparison groups, clients with dementia showed considerably higher levels of cognitive and functional impairment, aggression, anxiety, wandering, hallucinations/delusions, caregiver distress and a greater risk for institutionalization. Conversely, they showed a lower prevalence of several chronic conditions and lower levels of recent health service use. Depressive symptoms were relatively common in the dementia and other neurological groups.

Conclusion: Clients with co-existing neurological conditions exhibited unique clinical profiles illustrating the need for tailored and flexible home care services and enhanced caregiver assistance programs.

Keywords: dementia, Alzheimer disease, neurological disorders, mental health, home care

Introduction

Current global estimates suggest that approximately 35.6 million people have a form of dementia, including Alzheimer disease.^{Footnote 1} Within Canada, approximately half a million people have dementia with prevalence estimates increasing exponentially beyond the age of 65 years.^{Footnote 2} Aside from its personal cost, the ongoing care of those with dementia poses a significant societal and economic burden both in terms of care provided by family as well as formal care services and costs.^{Footnote 3,Footnote 4,Footnote 5,Footnote 6} Although relatively few seniors will require costly institutional care as they age,^{Footnote 7} the risk increases significantly for older adults with dementia.^{Footnote 8} The provision of timely, appropriate and co-ordinated home care services to older Canadians with dementia may help mitigate institutional risk and costs while supporting seniors' preferences to remain at home surrounded by familiar settings and social networks for a longer period of time.^{Footnote 9}

Of the estimated 1 million Canadians receiving home care services at any given time,^{Footnote 10} over three-quarters (82%) are 65 years or older,^{Footnote 11} and about 20% have Alzheimer disease or other dementias.^{Footnote 12} Comprehensive understanding of the social, mental and physical health needs of older Canadians with dementia receiving community-based care is required to ensure responsive care planning and the optimal management of this growing and vulnerable population. A thorough examination of client characteristics and care needs may further facilitate the identification of supportive strategies for overwhelmed family caregivers.^{Footnote 13} Previous studies have primarily examined the care needs and service use of older adults with dementia living in residential or long-term care facilities in the United States.^{Footnote 14,Footnote 15,Footnote 16} Recent population-based studies of community-dwelling seniors with dementia across Canada are scarce. Earlier work (largely derived from the 1991–2001 Canadian Study of Health and Aging^{Footnote 17,Footnote 18}) may not reflect changes in the complexity of care or service needs facing people with dementia and their caregivers. There is also a paucity of research characterizing those who have dementia along with a comorbid neurological illness. This is an important sub-population given the probable rise in caregiver stress and health service use due to the increasing severity of symptoms related to co-occurring neurological conditions.^{Footnote 19,Footnote 20}

To address current knowledge and policy gaps relevant to the quality of life and care of older Canadians with dementia, our objectives were to

1. provide a comprehensive summary of the sociodemographic, psychosocial and health characteristics of a large population-based cohort of home care clients with dementia relative to several comparison groups; and
2. explore the variation in these characteristics in clients with dementia alone compared with those with co-existing neurological conditions (e.g. dementia with stroke or Parkinson disease).

Methods

Design and Sample

This cross-sectional study is part of a larger research program (Innovations in Data, Evidence, and Applications for Persons with Neurological Conditions, or ideas PNC)^{Footnote 21} designed to provide prevalence estimates and clinical profiles of people with one or more of 10 priority neurological conditions receiving continuing care services.

Our sample included all home care clients in Ontario aged 50 years or older assessed with the Resident Assessment Instrument-Home Care (RAI-HC) between January 2003 and December 2010. The RAI-HC provides a standardized comprehensive assessment of a client's sociodemographic characteristics, physical and cognitive status, health conditions and selected diagnoses, behavioural problems, medication use and receipt of specific services. Since 2002, the RAI-HC has been mandated for all long-stay (i.e. expected to receive services for more than 60 days) home care clients with assessment data captured in the Ontario Association of Community Care Access Centres (OACCAC) database.

We first excluded RAI-HC assessments completed in an inpatient acute care setting for the purpose of placement (7.6% of all assessments) and then selected the most recent assessment available for clients (n = 520 479). This sample was reduced to 488 374 following our age restriction (50–115 years). We excluded those assessed prior to 2003 (0.02%) due to concerns about data completeness during this initial implementation phase. The final analytical sample included 488 290 unique clients.

The University of Waterloo houses de-identified copies of OACCAC data as part of a license agreement between interRAI and the Canadian Institute for Health Information.^{Footnote 22} These holdings are governed by regulations to protect personal privacy but do not require individual client consent (beyond that already obtained by contributing organizations during assessment).

Our study received research ethics approval from the University of Waterloo's Office of Research Ethics (project #17045).

Measures

Trained case managers, usually nurses or social workers, perform routine RAI-HC assessments using the best available information (e.g. clinical judgement; case discussions with attending physicians, other formal care providers and family members; health record review). The reliability and validity of the instrument has been established across a range of populations and settings.Footnote 23,Footnote 24,Footnote 25,Footnote 26

We examined the following RAI-HC items: clients' sociodemographic status (age, sex, marital status, whether trade-offs in purchasing needed treatment were made due to limited funds); psychosocial characteristics (availability of a caregiver, living arrangements, presence of caregiver distress); health status (cognitive and functional impairment, health instability, depressive and other neuropsychiatric symptoms, behavioural problems, select disease diagnoses); recent hospitalization and emergency department visits; and medication use in the previous week (i.e. 9+ medications, 1+ medications from selected classes [antipsychotic, anxiolytic, antidepressant, hypnotic, cholinesterase inhibitor and/or memantine use]). Details regarding all medications used in the previous week are manually recorded from containers, verified with clients/caregivers and transcribed electronically.

We examined five validated scales derived from RAI-HC items: Cognitive Performance Scale (CPS) (range 0–6);Footnote 27 Activities of Daily Living (ADL) Self-Performance Hierarchy Scale (range 0–6);Footnote 24,Footnote 28 Changes in Health, End-stage Disease and Signs and Symptoms (CHESS) Scale (range 0–5);Footnote 29,Footnote 30 Method for Assigning Priority Levels (MAPLe) (range 1–5);Footnote 31 and Depression Rating Scale (DRS) (range 0–14).Footnote 32,Footnote 33 We also examined a modified Aggressive Behaviour Scale (ABS)Footnote 34 derived from the sum of any occurrence of four behaviours (verbal abuse, physical abuse, socially inappropriate behaviour or resisting care) in the previous three days, and a summary measure of impairment in four instrumental ADLs (some or greater difficulty with meal preparation, managing finances, managing medications and transportation). Higher scores on all these scales indicate more severe impairment.

The CPS reflects level of cognitive impairment and has been validated against the Mini-Mental State Examination.^{Footnote 35} It includes four items (short-term memory, cognitive skills for daily decision making, expressive communication and eating self-performance) and ranges from 0 (intact) to 6 (very severe impairment).^{Footnote 27,Footnote 35} The CHESS scale ranges from 0 (stable) to 5 (unstable health) and combines symptoms (vomiting, dehydration, decline in food/ fluid intake, weight loss, shortness of breath, edema) with items capturing recent decline (in cognition and ADL) and end-stage disease. Higher CHESS scores predict mortality, institutionalization and hospitalization in older adults across care settings.^{Footnote 29,Footnote 36,Footnote 37} The MAPLe differentiates clients into five priority levels (low to very high) based on level of cognitive and ADL impairment, behavioural issues, environmental concerns and self-reliance. Higher levels are predictive of institutionalization and caregiver stress.^{Footnote 31}

The RAI-HC contains a diagnostic checklist for commonly occurring conditions in an older population. Conditions were considered present if a doctor diagnosed them, a home care professional was required to treat or monitor them, or the disease was a reason for hospitalization in the previous 90 days. Neurological diagnoses captured on this checklist include dementia (Alzheimer disease and/or other dementias), multiple sclerosis (MS), Parkinson disease/Parkinsonism (PD), traumatic brain injury (TBI, referred to as ''head trauma'' on the instrument) and stroke. There are open-ended fields for free-text entry of diagnoses not on the checklist. Six neurological conditions were coded as present/absent based on a review of all free-text entries: epilepsy/ seizure disorder, Huntington disease (HD), muscular dystrophy (MD), cerebral palsy (CP), spinal cord injury (SCI) and amyotrophic lateral sclerosis (ALS). The free-text terms were defined by consensus of an expert review committee including neurologists, psychiatrists and geriatricians. The conditions listed above (excluding stroke) are the 10 priority neurological diagnoses identified by the Public Health Agency of Canada for the ideas PNC program. We included stroke in our analyses because it is a common and disabling condition in older people.

Data supporting the accuracy of diagnoses recorded on RAI instruments have been published elsewhere.^{Footnote 37,Footnote 38,Footnote 39} Wodchis et al.^{Footnote 38} showed sensitivities of 0.80 or greater for several common conditions in Ontario complex continuing care settings (e.g. stroke, diabetes, cancer, chronic obstructive pulmonary disease [COPD], heart failure). Comparable sensitivity estimates were observed for PD (0.87), Alzheimer disease (0.85, allowing for a check of either ''Alzheimer's'' and/or ''Dementia other than Alzheimer's Disease'' on the RAI), CP (0.84) and seizure disorder (0.75). Sensitivity estimates were low (< 0.50) for other neurological conditions, including TBI and MS.

Analyses

We conducted descriptive analyses to examine the distribution of sociodemographic, psychosocial and health characteristics by the following comparison groups: ''Dementia,'' ''Stroke,'' ''Other Neurological Condition'' (presence of 1+ of the priority neurological conditions – MS, PD, TBI, HD, MD, CP, SCI, ALS, epilepsy) and ''Cognitively Intact Controls'' (clients without any of the selected 11 neurological conditions and a CPS score of 0 or 1).

Descriptive analyses were also performed comparing the characteristics of those with dementia alone to those with dementia and other documented neurological conditions (i.e. dementia with stroke only, dementia with PD only, dementia with PD and stroke only, dementia with TBI only). These comparison groups excluded clients with any of the other selected neurological conditions.

Results

Clients with dementia vs. stroke, other neurological conditions and cognitively intact controls (Tables 1A & 1B)

Our analysis included 104 802 clients (21.5%) with a diagnosis of dementia, 85 579 (17.5%) with stroke and 23 007 (4.7%) with one or more of the other priority neurological conditions (20 972 (4.3%) clients had a recorded diagnosis of both dementia and stroke). Almost half (n = 236 763; 48.5%) were in the cognitively intact control group. Excluded from the analyses were 59 089 clients (12.1%) with meaningful cognitive impairment (CPS 2+) but no priority neurological diagnosis, and 22 clients with missing CPS values.

Compared with the stroke and other neurological groups, clients with dementia were more likely to be female (63.7%) and older, with a mean age (standard deviation) of 83.2 (7.6) years. Across all groups, women were significantly less likely to be married than were men. Relatively few clients reported making economic trade-offs, and this was less common for dementia clients than for those with other neurological conditions. Compared with controls, clients across all three diagnostic groups were more likely to co-reside with their primary caregiver. Among those with dementia or stroke this caregiver was most often a child or child-in-law. Clients with dementia were more likely to have reported conflicts with others, a distressed caregiver, moderate to severe cognitive impairment, significant difficulties with Instrumental Activities of Daily Living (IADLs), and some impairment in ADLs (Table 1A).

Dementia clients were more likely to have some level of health instability and considerably higher levels of aggression, wandering, anxiety and hallucinations/ delusions than all other groups (Table 1B). These findings help to explain the significantly greater proportion of dementia clients with high to very high MAPLe scores (Figure 1) and distressed caregivers. For all sub-groups, the proportion of clients with a distressed caregiver increased with increasing MAPLe scores (Figure 2). However, for all levels of MAPLe, the proportion of clients with a distressed caregiver was relatively higher for those with dementia than those in the comparison groups. Clinically important depressive symptoms were slightly more common in dementia clients than the other groups, though depression and anxiety were not uncommon in clients with other neurological conditions or stroke.

For all groups, the most common comorbid diagnoses were cardiovascular diseases, arthritis and diabetes. Most clinical diagnoses were less prevalent in clients with dementia or other neurological conditions relative to clients in the stroke or control groups. All three neurological diagnostic groups showed a lower prevalence of cancer. A recent fall, unsteady gait and pressure ulcers were more common in clients with other neurological conditions and then in those with stroke. Relative to cognitively intact clients, swallowing problems were more prevalent in the three neurological diagnostic groups, particularly in clients with dementia.

Dementia clients were less likely than the other groups to have experienced one or more ED visits or hospitalizations in the previous 90 days or to use nine or more medications. Conversely, they were more likely than the other groups to be taking an antipsychotic/neuroleptic. Other psychotropic drug classes were more commonly used by clients with other neurological conditions. Multiple medication use (9+) was most common in stroke clients, presumably due to their relatively higher levels of comorbid illnesses (e.g. diabetes, cardiovascular diseases [CVDs]). Approximately half of dementia clients used a dementia medication.

Compared with cognitively intact clients, those with dementia, stroke or other neurological conditions were more likely to have received care from home health care aides (61%–66% vs. 48%) and homemaking services (42% vs. 31%) in the previous seven days but were less likely to have received care from a visiting registered nurse (25%–28% vs. 40%). Clients with dementia were also less likely than all other groups to have received physical therapy (7% vs. 13%–15%) or occupational therapy (8% vs. 10%–16%) in the previous week (data not shown; details available on request).

Clients with dementia alone vs. dementia with other neurological conditions (Tables 2A & 2B)

The dementia cohort included 77 670 (74.1%) clients with dementia alone, 19 061 (18.2%) with co-existing stroke, 4480 (4.3%) with PD, 1182 (1.1%) with both PD and stroke, and 763 (0.7%) with TBI (Table 2A). There were 1646 (1.6%) clients with dementia and some other combination(s) with selected neurological conditions that were rare and thus not presented.

Dementia clients with co-existing PD (with or without stroke) or TBI were generally younger and more likely to be male compared with the other two groups. This age/sex distribution likely explains the relatively higher proportion of married clients (with an available co-residing spousal caregiver) in the groups with dementia and PD (with or without stroke). Dementia clients with TBI were more likely than the other groups to have reported conflicts with others, and along with those with PD (with or without stroke), were more likely to have a distressed caregiver. Dementia clients with co-existing PD and stroke were more likely to exhibit higher levels of cognitive impairment and ADL dependence than the other groups.

Moderate to high health instability was slightly more common in dementia clients with stroke (including stroke/PD) or TBI (Table 2B). All groups showed similar proportions at high/very high MAPLe levels and with clinically important depressive symptoms. Aggressive behaviours were less prevalent in dementia clients with PD (with or without stroke) and more common in those with coexisting TBI. Anxiety symptoms were slightly more common in dementia clients with TBI, whereas hallucinations/delusions were more prevalent in dementia clients with PD.

Generally, various comorbid illnesses (e.g. arthritis, diabetes, CVDs) were more common in dementia clients with co-existing stroke and less common in those with coexisting PD only. A recent fall, unsteady gait and pressure ulcers were more common in dementia clients with co-existing PD. Overall, compared with dementia-only clients, all four groups with coexisting neurological conditions showed a higher prevalence of recent falls, unsteady gait and problems with swallowing (the latter were especially common in those with dementia, PD and stroke). A recent ED visit or hospitalization was also more common in the four groups with a co-existing neurological condition relative to the dementia-only group. A recent hospitalization was especially common in dementia clients with stroke (including stroke/PD) or TBI. The use of nine or more medications was less common in those with dementia alone or with TBI, and more common in those with coexisting stroke or PD (particularly stroke with PD).

Dementia clients with PD (with or without stroke) generally showed higher use of antipsychotic/neuroleptic and antidepressant medications compared with the other groups. Clients with PD (no stroke) and with dementia alone were more likely than the other groups to be using a cholinesterase inhibitor and/or memantine, whereas those with TBI or stroke (no PD) showed the lowest use.

Discussion

Findings from this population-based study of home care clients in Ontario highlight the substantial psychosocial, functional and mental health needs of people with dementia who live in the community. Our work expands on previous literature by providing a recent and comprehensive profile of the key domains relevant to the care, quality of life and health outcomes of this growing population. As a further contribution, we provide estimates of the prevalence of common co-existing neurological conditions and the associated complexity of health and care planning needs imposed by this comorbidity.

Clients with dementia vs. stroke, other neurological conditions and cognitively intact controls

Approximately 22% of Ontario long-stay home care clients (n = 104 802) had been diagnosed as having dementia. A common profile was that of an older (>75 years) widowed woman supported by a child (or child-in-law) as her primary caregiver. However, in about one-third of dementia clients, the primary caregiver was a spouse who was likely of the same age or older and likely soon facing challenges to his/her own health and social well-being. Approximately 43% of dementia clients (and 50% of those with cognitive impairment but no diagnosis) did not co-reside with their primary caregiver. The lack of a close or well-informed advocate available to monitor and communicate their needs in a timely manner may lead to an increased risk of fragmented or sub-optimal care and more rapid disease progression.^{Footnote 40,Footnote 41}

Almost half of dementia clients had moderate to severe cognitive impairment (CPS score 3+) and almost all experienced some or great difficulty with multiple IADLs. As informal and formal care costs increase with dementia severity,^{Footnote 3,Footnote 6,Footnote 42} this finding has important implications for family caregivers, health care providers and policy makers. Consistent with their level of cognitive impairment, dementia clients showed a significantly higher prevalence of aggression, anxiety, wandering and hallucinations/delusions than other diagnostic groups. They were also more likely to exhibit clinically important depressive symptoms. In their examination of 2005 Canadian Community Health Survey participants aged 55 years and over, Nabalamba and Patten^{Footnote 43} also observed higher levels of mood (19.5%) and anxiety (16.3%) disorders in people with dementia. The clustering of cognitive, behavioural and psychiatric issues evident in dementia clients helps to explain the greater likelihood of caregiver distress^{Footnote 17} (approximately 35% of family caregivers in our study) as well as clients' increased risk of institutionalization^{Footnote 17,Footnote 44} and higher care costs.^{Footnote 42,Footnote 45} Specifically, 82% of clients with dementia displayed high to very high MAPLe scores indicating an imminent risk for transition to a higher level of care.

Clients with dementia (and those with other neurological conditions) showed a lower prevalence of several chronic conditions (including cardiovascular diseases, arthritis, diabetes, COPD and cancer) and lower levels of recent health service use (e.g. emergency room visits or hospitalizations in the previous 3 months and use of 9+ medications). While earlier research reported people with dementia (particularly those with Alzheimer disease) as being relatively healthier,^{Footnote 46,Footnote 47} recent findings have been inconsistent.^{Footnote 45,Footnote 48} The one exception is the lower prevalence of cancer consistently noted for those with dementia and other neurological conditions.^{Footnote 49} These inconsistencies likely reflect variations across investigations in study design and samples (e.g. sociodemographic characteristics, dementia severity and sub-types examined) and in the diagnostic and clinical health measures employed. Several studies have reported higher rates of comorbid health conditions, medication and health service use for those with vascular dementia (as compared with Alzheimer disease).^{Footnote 46,Footnote 48} Our findings for dementia clients with co-existing stroke (Table 2B) are consistent with these reports. For some conditions, a lower prevalence may be the consequence of poorer detection and under-diagnosis in people with a dementia disorder.^{Footnote 48} Factors underlying this poorer recognition may include the atypical presentation of some conditions and the under-reporting of symptoms in patients with dementia as well as the stigma associated with the diagnosis of dementia. Additional efforts to investigate this possibility and potential strategies for improved detection of existing comorbidities in patients with dementia are warranted.^{Footnote 50} It should also be noted that dementia clients were significantly more likely to experience swallowing difficulties and to use antipsychotic/neuroleptic medications, both of which represent risk factors for decline and hospitalization.^{Footnote 45,Footnote 51}

Clients with dementia alone vs. dementia with other neurological conditions

Approximately one in four dementia clients had a co-existing neurological condition (specific to our targeted conditions), most often stroke and then PD and TBI. Others have documented the relatively common co-occurrence of dementia in people with stroke or with PD.^{Footnote 20,Footnote 52} Clients documented as having all three conditions (dementia, PD and stroke) showed the greatest burden in terms of more severe cognitive and ADL impairment. Compared with dementia-only clients, all four groups with co-existing neurological conditions showed a higher prevalence of recent falls, unsteady gait, swallowing problems (with the latter present in almost 50% of those with dementia, PD and stroke) and recent health service use.

Dementia clients with selected co-existing neurological conditions were also found to exhibit unique sociodemographic and health profiles. Those with dementia and PD were more likely to be younger and male and consequently more likely to co-reside with a spousal caregiver. In dementia clients with PD or with TBI, approximately 40% were noted to have a distressed caregiver (compared with about 35% for the other groups). Yet the underlying factors possibly contributing to caregiver burden varied in these groups. For example, dementia clients with TBI were more likely than other groups to experience conflicts with others, aggressive behaviours, wandering and recent worsening of mood and/or behaviours. Conversely, those with PD were less likely to have behavioural issues or conflicts but more likely to exhibit hallucinations and/or delusions.

There was evidence of greater health instability (e.g. higher CHESS scores and recent hospital use) in dementia clients with co-existing stroke or TBI. Further, those with dementia and co-existing stroke showed a higher prevalence of common comorbid health conditions (including cardiovascular illness, diabetes and arthritis) and polypharmacy (9+ medications). Although less likely than others to exhibit these comorbid health conditions, dementia clients with PD were more likely to have a recent fall, unsteady gait and pressure ulcers. The variation in cholinesterase inhibitor and/or memantine use observed in dementia clients with co-existing neurological conditions is intriguing and has been more fully examined in a separate publication.^{Footnote 53}

The unique care needs observed for particular dementia sub-groups illustrate the importance of tailored and coordinated home care services.^{Footnote 13} For example, further educational resources and behavioural management strategies may be a priority for dementia clients with TBI (and their caregivers) whereas dementia clients with PD may have a greater need for fall prevention strategies and rehabilitation services and dementia clients with coexisting stroke will need enhanced chronic disease management.

Important strengths of our study include the examination of a large population-based sample of older home care clients (allowing for greater precision in estimates, stratification by diagnostic subgroups and generalizability) and the complete and comprehensive nature of the RAI-HC assessment data. However, some limitations should be noted. Despite evidence supporting the validity of diagnostic data on the RAI-HC (including dementia),^{Footnote 37,Footnote 38,Footnote 39} further validation work is required. In addition, the diagnostic and cognitive data captured on the RAI-HC does not permit a differentiation of dementia sub-type (an important predictor of care needs and service use). Approximately 12.1% of clients (without a recorded neurological diagnosis) had moderate to significant cognitive impairment, and a proportion in this group (particularly those with a CPS score of 4+) are likely to have had a dementia disorder. The potential for diagnostic misclassification may have resulted in a reduced ability to detect relevant differences in client characteristics across some of our comparison groups. The cross-sectional nature of our data and the absence of prospective data on actual health system and home care use also limits our ability to comment on the differential burden and unmet care needs associated with selected co-existing neurological conditions in dementia clients.

Conclusions

Our findings suggest that a significant proportion of clients currently living with dementia in the community may be close to the tipping point in terms of their continued ability to remain in their own homes. These data support the argument that more flexible and enhanced community-based and caregiver assistance programs may be needed to ensure continued client and caregiver well-being and quality of care.^{Footnote 13,Footnote 54} Of critical importance for all dementia clients (given the burden of mood and anxiety disorders) is the immediate need for improved, coordinated and integrated psychiatric and mental health services (with intensive case management).^{Footnote 43} Care providers (including case managers, primary care physicians and family caregivers) may face numerous structural barriers in obtaining access to appropriate mental health specialists and services,^{Footnote 55} leading to an increased likelihood for delayed or inappropriate treatment and poor outcomes for community-dwelling seniors with dementia. Further work detailing the extent and consequences of unmet needs associated with co-existing mental health and neurological conditions in dementia is clearly warranted.

Acknowledgments

This study is part of the National Population Health Study of Neurological Conditions. We wish to acknowledge the membership of Neurological Health Charities Canada and the Public Health Agency of Canada for their contribution to the success of this initiative.

Funding for the study was provided by the Public Health Agency of Canada. The opinions expressed in this publication are those of the authors/researchers, and do not necessarily reflect the official views of the Public Health Agency of Canada.

Dr. Hogan holds the Brenda Strafford Foundation Chair in Geriatric Medicine. Dr. Jetté holds an Alberta Innovates Health Solutions Population Health Investigator Award and a Canada Research Chair Tier 2 in Neurological Health Services Research. Dr. Bronskill is supported by a New Investigator Award in the Area of Aging from the Canadian Institute of Health Research. Dr. Hirdes holds the Ontario Home Care Research and Knowledge Exchange Chair funded by the Ontario Ministry of Health and Long-Term Care.

Conflict of interest

None.

References