Introduction
The preference of people with dementia is to remain living at home and maintain normality for as long as possible (Brijnath, Reference Brijnath2012; Low et al., Reference Low, White, Jeon, Gresham and Brodaty2012; von Kutzleben et al., Reference von Kutzleben, Schmid, Halek, Holle and Bartholomeyczik2012). In developed countries, an estimated 60%–70% of persons with dementia live in the community (Knapp et al., Reference Knapp and Society2007; Australian Institute of Health and Welfare, 2012) and quality of life and quality of care have been rated at equal levels in home care and residential care (Beerens et al., Reference Beerens2014). Caring for persons with dementia in residential aged care is costly, and given the predicted doubling in worldwide prevalence by 2030 and tripling by 2050 (World Health Organization and Alzheimer's Disease International, 2012), there will be economic and workforce pressures to care for the majority of persons with dementia in their own homes rather than in nursing homes. It has been estimated that overall home care is cheaper than residential care although costs of informal care increase from mild through to severe dementia (Knapp et al., Reference Knapp and Society2007; Wübker et al., Reference Wübker2014). Other economic estimates suggest that when informal care costs are included, home care costs more than residential care and non-economic consequences such as carer burden are increased (König et al., Reference König2014).
The types and availability of home care services differ by country. A review of American studies reporting service use estimated that 46.7% of community-dwelling people with dementia used in-home health aide services during a one-year period and homemaker assistance was used by around 23%–36.8% (Weber et al., Reference Weber, Pirraglia and Kunik2011). Whereas, a UK. study of home care service use estimated that around 85% of community-dwelling people with dementia use some type of in-home care services (Alzheimer's Society, 2011).
Home care encompasses a broad range of services including needs assessment, personal care provision, leisure activities and rehabilitation at home. Home care has been found in the short term to increase time to hospitalization and death (You et al., Reference You, Dunt, White, Vander Hoorn and Doyle2014). Criticisms of home care services raised by researchers in Australia, the Netherlands, UK, and Canada are that they are fragmented and uncoordinated, not understood by potential users, and access is difficult because of complex aged care entry systems and inconsistent and sometimes poorly defined eligibility criteria (Low and Brodaty, Reference Low and Brodaty2008; Jansen et al., Reference Jansen2009; van der Roest et al., Reference van der Roest2009; Woolrych and Sixsmith, Reference Woolrych and Sixsmith2013). For this review we define home care as non-medical support to people in their own homes, also known as home care, community care, domiciliary care, social care, or in-home care. We exclude home healthcare which is provided by medical, nursing and allied health professionals as these services differ from home (non-health) care, typically focusing on improving or maintaining health rather than supporting function, and provided based on clients’ medical needs.
The aim of this paper is to review the effects of different models of delivery of home care services on outcomes for persons with dementia. The review will also discuss elements that contribute to making each model successful. The four models of home care delivery reviewed are: case management, integrated care, consumer directed care, and restorative care. Our previous systematic review identified the first three models as being the dominant ones in home care, and that restorative care was an emerging model (Low et al., Reference Low, Yap and Brodaty2011). These models are not mutually exclusive; integrated and restorative care usually include case management (Bernabei et al., Reference Bernabei1998; King et al., Reference King, Parsons, Robinson and Jörgensen2012), and a consumer directed approach could be used in combination with an integrated system (Kodner, Reference Kodner2003).
Case management
Case management (also known as care management or case coordination) involves a case manager or case management team making an assessment of the client's needs, providing information about possible care options, developing and implementing a care plan, and coordinating services (Case Management Society of America, 2008). Case management also involves ongoing monitoring, reassessment and care plan revision. Given that many aged care systems are fragmented and difficult to navigate, the case management model provides a central point of access, and an expert to provide information, advice and advocacy. Consultation with Dutch experts using an iterative methodology identified the essential components of case management for people with dementia as being: a patient-centered approach, information, support and counseling, coordination of care and, to a lesser extent, practical help (Verkade et al., Reference Verkade, Van Meijel, Brink, Van Os-Medendorp, Koekkoek and Francke2010). For this review we limited our definition of case management to services provided by a case manager or team which conducted an assessment, developed a care plan, and coordinated services as appropriate.
There have been four recent systematic reviews of case management for persons with dementia (Pimouguet et al., Reference Pimouguet, Lavaud, Dartigues and Helmer2010; Koch et al., Reference Koch2012; Somme et al., Reference Somme, Trouve, Dramé, Gagnon, Couturier and Saint-Jean2012; Tam-Tham et al., Reference Tam-Tham, Cepoiu-Martin, Ronksley, Maxwell and Hemmelgarn2013). These reviews did not explicitly define their inclusion criteria for case management, such that there were marked differences in the number (12, 7, or 6 and 17 studies respectively) and content of studies included (e.g. one review included carer counselling studies, one included an integrated care study), which in turn affected the conclusions drawn. The reviews concluded that there were inconsistent effects of case management on clinical outcomes for persons with dementia and their carers (Pimouguet et al., Reference Pimouguet, Lavaud, Dartigues and Helmer2010; Koch et al., Reference Koch2012; Somme et al., Reference Somme, Trouve, Dramé, Gagnon, Couturier and Saint-Jean2012). One review suggested that studies with longer follow-up periods tended to show delayed nursing home admission of people with dementia (Koch et al., Reference Koch2012), however another only found short-term benefits to reducing nursing home admission risk (Tam-Tham et al., Reference Tam-Tham, Cepoiu-Martin, Ronksley, Maxwell and Hemmelgarn2013). Another review suggested that better outcomes are observed with greater case management intensity (Somme et al., Reference Somme, Trouve, Dramé, Gagnon, Couturier and Saint-Jean2012). Case management was found to not affect nursing home placement, hospitalization, and emergency department visits (Tam-Tham et al., Reference Tam-Tham, Cepoiu-Martin, Ronksley, Maxwell and Hemmelgarn2013), and consistent with this, the reviews agreed that case management produced small to no decreases in resource utilization when healthcare expenditure, hospitalization or institutionalization were considered (Pimouguet et al., Reference Pimouguet, Lavaud, Dartigues and Helmer2010; Somme et al., Reference Somme, Trouve, Dramé, Gagnon, Couturier and Saint-Jean2012).
We only identified five controlled studies of home care case management for persons with dementia. These studies measured a diversity of outcomes which makes it difficult to draw conclusions. A Finnish study found that case management was effective in deferring nursing home placement particularly for persons with severe dementia, however by the end of the two years, the numbers of participants admitted to care were similar for case management and usual care groups (Eloniemi-Sulkava et al., Reference Eloniemi-Sulkava, Notkola, Hentinen, Kivela, Sivenius and Sulkava2001). An American study over 12 months found that in comparison to usual care, case managed clients improved adherence and quality of care on dementia guidelines, increased use of community agency assistance, and patient health-related quality of life, overall quality of patient care, caregiving quality, social support, and level of unmet caregiving assistance needs (Vickrey et al., Reference Vickrey2006). The American Medicare Alzheimer's Disease Demonstration (MADDE) provided case management, however results are confounded as case management clients also received an 80% subsidy of care costs. MADDE found that over one year there was increased use of home care services and adult day care. Over three years there were no differences on nursing home entry rates, carer burden or depression. Over three years carers in MADDE had a lower likelihood than controls of any hospitalization, but not of emergency department use, length of hospital stay or number of hospitalizations (Miller et al., Reference Miller, Newcomer and Fox1999; Newcomer et al., Reference Newcomer, Spitalny, Fox and Yordi1999).
A Dutch study found that, over 12 months, case management for carers of persons with early dementia did not increase carer's sense of competence, quality of life, depressive symptoms and burden, or patient's quality of life in comparison to usual care (Jansen et al., Reference Jansen2011). Another American study over 18 months found that in comparison to augmented usual care (written recommendations to primary care physicians and a resource guide to services), participants who received home-based case management had improved self-reported quality of life, and an estimated mean delay of 51 days to transition out of their home but not differences in total unmet needs, quality of life, neuropsychiatric symptoms or depression (Samus et al., Reference Samus2014).
A case series of eight case management programs for persons with dementia in the Netherlands explored factors relating to success and failure (Minkman et al., Reference Minkman, Ligthart and Huijsman2009). Success factors were: (1) investment in a strong provider network or care chain, and good personal connections with professionals; (2) expert knowledge of the case managers; (3) embedding in a multidisciplinary team and direct connection with medical staff; (4) support and recognition of local providers for the program; and (5) a low threshold for accessing support and care for patients and carers. Another study found that, characteristics of case management, which negatively influence implementation, are low intensity of case management (e.g. infrequent follow-up), a large caseload (60 patients or more per case manager), and a reactive (rather than proactive) approach to care (Khanassov et al., Reference Khanassov, Vedel and Pluye2014). In the Netherlands, the model of case management in which case manager and client services are embedded within one organization is more successful in facilitating implementation of care in comparison to the model where the case manager coordinates services from a range of organizations.
In summary, case management for persons with dementia may increase use of community based services and delay nursing home admission.
Integrated care
Integrated care is described as “the management and delivery of health services so that clients receive a continuum of preventive and curative services, according to their needs over time and across different levels of the health system” (World Health Organisation, 2008). There is a continuum of integration within care models described as “integrated.” This ranges from establishing lines of communication between different services (linkage), having explicit structures to coordinate care across services (coordination), to reorganizing services so they fall under one funding source and organization (full integration) (Leutz, Reference Leutz1999; Beland and Hollander, Reference Beland and Hollander2011). Integrated models were proposed to produce more seamless and flexible care for clients and fully integrated systems are thought to lower costs through reduced administration (Reed et al., Reference Reed, Cook, Childs and McCormack2005).
There have been three controlled trials of integrated care models that included persons with cognitive impairment however none of these have published analyses specifically on the subsamples of persons with dementia.
A randomized controlled trial was conducted in Canada of the System of Integrated Care for Older Persons (SIPA) where just under 1/3 of participants had cognitive impairment. SIPA is an almost fully integrated model where publically funded community-based multidisciplinary teams fully provide community health and social services and coordinated hospital and nursing home care (Beland et al., Reference Beland2006). This trial found increased accessibility of health and social home care with increased intensification of home healthcare, 50% reduction in hospital alternate level inpatient stays (“bed blockers”) but no significant differences in utilization and costs of emergency department, hospital acute inpatient, and nursing home stays. Satisfaction was increased for SIPA carers with no change in carer burden or out-of-pocket costs or improvements in health outcomes (Beland et al., Reference Beland2006).
The Program of Research to Integrate Services for the Maintenance of Autonomy (PRISMA) was evaluated in a non-randomized controlled trial in Canada. The sample had mild levels of cognitive impairment (Mini-Mental State Examination average score of 24). PRISMA used a coordination-type model of integration where all the public, private, and voluntary health-and-social-services organizations involved in caring for older people cooperated to enable (1) co-ordination between decision-makers and managers, (2) a single entry point, (3) a case management process, (4) individualized service plans, (5) a single assessment instrument based on the clients’ functional autonomy, and (6) a computerized clinical chart for communicating between institutions (Hebert et al., Reference Hebert, Durand, Dubuc and Tourigny2003). After four years, the annual incidence of functional decline was lower by 137 cases per 1,000 in the SIPA group compared to the controls, whereas the prevalence of unmet needs in the control group was nearly double the prevalence observed in the SIPA group. Satisfaction and empowerment were significantly higher in the SIPA group. A lower number of visits to emergency rooms and hospitalizations than expected was observed in the SIPA group (Hébert et al., Reference Hébert, Raîche, Dubois, Gueye, Dubuc and Tousignant2010; Dubuc et al., Reference Dubuc, Dubois, Rache, Gueye and Hébert2011).
The Program for All-Inclusive Care of the Elderly (PACE) is an American Medicare/Medicaid funded program which has been evaluated in a non-controlled trial. In 2001, 49% of PACE clients had a dementia diagnosis (National Pace Association, 2012). In PACE an interdisciplinary team is responsible for developing and implementing a comprehensive care plan within a fully integrated program of health and social support. Compared with community dwelling people who declined to enrol, PACE participants had lower rates of nursing home utilization, higher use of ambulatory services, better health status and quality of life, less physical function deterioration, and lower mortality rates (Chatterji et al., Reference Chatterji, Burstein, Kidder and White1998). Lower hospital utilization of an average 0.2 days per month compared to community dwelling non-enrollees’ average of 0.8 days per month has also been reported (Meret-Hanke, Reference Meret-Hanke2011). PACE does not appeal to many older adults because they are required to attend a day care center frequently and to switch to the primary care physician at the health center (Lynch et al., Reference Lynch, Hernandez and Estes2008). There are inconsistencies between PACE sites on achieving service utilization, care quality and place of death outcomes (Pacala et al., Reference Pacala, Kane, Atherly and Smith2000; Mukamel et al., Reference Mukamel, Bajorska and Temkin-Greener2002; Temkin-Greener and Mukamel, Reference Temkin-Greener and Mukamel2002).
A review of North American integrated models identified four elements of successful programs: (1) umbrella organizational structures to guide integration on the strategic, managerial and service delivery levels, encourage and support effective joint working, ensure efficient operations, and maintain overall accountability for service, quality and cost outcomes; (2) case-managed, multidisciplinary team care that allows for the effective evaluation and planning of client needs, provides a single contact or entry point into the healthcare and social service systems, packages and coordinates services, and triages or allocates clinical responsibility; (3) organized provider networks joined together by standardized referral procedures, service agreements, joint training, shared information systems, and even common ownership of resources enhance access to services, provide seamless care and maintain quality; (4) financial incentives to promote prevention, rehabilitation and the downward substitution of services, as well as enable service integration and efficiency (Kodner, Reference Kodner2006).
Overall the research suggests that integrated care is associated with greater client satisfaction, increased use of community based services, and reduced hospital days however the specific impacts on persons with dementia and their carers are not known.
Consumer directed care
The aim of consumer directed care programs is to provide home care recipients (consumers) with more choice and control than they would have with the traditional or agency directed programs in order to maximize independence and autonomy (Department of Health, UK, 2012). Consumer directed programs offer a continuum of choice and responsibility to care recipients. Consumers may be paid a cash amount or allocated a budget which they manage or direct the spending of. Consumers choose what they purchase, either from a menu of services or are given carte blanche choice. One of the most popular choices in comparison to traditional agency directed care is the use of their budget to hire family or friends. An advisor (counselor) or professional care manager or fiscal manager may or may not be involved in care. Typically persons with dementia require a proxy or advocate to make decisions on their behalf. Consumer directed care is the most popular form of home care in Germany (Wiener, Reference Wiener2007). Both the UK (Department of Health, Reference Health2012) and Australia (Department of Health and Ageing, Reference Ageing2012) are planning to make all their home care services consumer directed. Consumer directed care is also known as Personal Budgets (the Netherlands), Individual budgets, or Direct Payments (UK).
To date, there has been little focus on how to most effectively deliver consumer directed care for persons with dementia. Concerns have been raised that persons with dementia are not being offered consumer directed care (Lakey and Saunders, Reference Lakey and Saunders2011). Four randomized controlled trials of consumer directed care have all included persons with cognitive impairment – 28% of the Medicare Primary and Consumer-Directed Care Demonstration were cognitively impaired (Meng et al., Reference Meng, Friedman, Wamsley, Mukamel and Eggert2005), an unreported proportion of the Cash and Counseling program and Individual Budgets evaluation had cognitive impairment (Carlson et al., Reference Carlson, Foster, Dale and Brown2007; Glendinning et al., Reference Glendinning2008) and a small Italian trial included frail elderly with cognitive impairment (Giannini et al., Reference Giannini2007). None of these trials published analyses focusing particularly on the cognitively impaired subsample.
Overall, these trials showed that consumer directed care improved satisfaction with care and community service use, but had little effect on clinical outcomes (Low et al., Reference Low, Yap and Brodaty2011). Since consumer directed care increases service usage, it also is associated with higher costs than traditional care models (Grabowski, Reference Grabowski2006).
Restorative care
Restorative home care aims to maximize the functional ability and independence of service clients and may also aim to maximize self-esteem and health related quality of life. Restorative home care programs usually include case management, and often are goal focused. Restorative programs are also known as reablement or active service models.
All of the published evaluations of restorative models of home care to date excluded persons with dementia (SPRU University of York and PSSRU University of Kent, 2009; Lewin and Vandermeulen, Reference Lewin and Vandermeulen2010; King et al., Reference King, Parsons, Robinson and Jörgensen2012; Parsons et al., Reference Parsons, Sheridan, Rouse, Robinson and Connolly2013). These randomized controlled trials showed that goal focused restorative home care improved physical function and health related quality of life (King et al., Reference King, Parsons, Robinson and Jörgensen2012; Parsons et al., Reference Parsons, Sheridan, Rouse, Robinson and Connolly2013), functional dependency, confidence in performing everyday activities without falling and functional mobility (Lewin and Vandermeulen, Reference Lewin and Vandermeulen2010), improved social care, health-related quality of life, and perceived health (SPRU University of York and PSSRU University of Kent, 2009). Initially clients of restorative models used more services, but after 3–12 months there was decreased service use in comparison to usual models of care (SPRU University of York and PSSRU University of Kent, 2009; Lewin and Vandermeulen, Reference Lewin and Vandermeulen2010). For example, one study reported that compared with 57% of the control group after 12 months, 85.8% of the restorative care group no longer required a service (Lewin and Vandermeulen, Reference Lewin and Vandermeulen2010). The break-even point for costs in one study for restorative care in comparison to usual services, was approximately 30 weeks, with restorative care clients initially using more services than usual care (SPRU University of York and PSSRU University of Kent, 2009).
An Australian pilot (n = 11) restorative care program demonstrated the feasibility of this approach for people with dementia, with the authors suggesting that restorative care may be most beneficial to people with early dementia (Lewin et al., Reference Lewin2013). A pilot restorative care program for persons with dementia is also being conducted in the United Kingdom (Moray Council, 2012). Restorative interventions in residential care exclusively for or including persons with cognitive impairment have shown improvements in function, mood, and behavior, so the approach has potential to improve outcomes for persons with dementia living at home (Galik et al., Reference Galik, Resnick, Gruber-Baldini, Nahm, Pearson and Pretzer-Aboff2008).
Conclusions
There has been a little research into models of home care for people with dementia, and no head-to-head comparison of the different models. The only trials which published data on people with dementia were for the case management model which was shown to increase use of community services and delay nursing home admission. Integrated care and consumer directed care can be delivered to people with dementia, however it is not clear whether the benefits of these models, in terms of improving satisfaction and service use, apply equally to older people with and without cognitive impairment. Restorative care is just starting to be trialed for people with dementia.
Research into models of care are typically pragmatic trials that require collaboration with service providers, funders and policy makers; these studies can be challenging logistically and costly. A mixed methods approach may maximize the learnings from future research by collecting quantitative data on the effects on client outcomes, and qualitative information about factors that make models more successful in delivering those outcomes. Future research needs to address how persons with dementia can be assisted to operate within consumer directed and restorative models which require greater active participation of the care recipient, and investigate the role of family carers in this participation. This is required to inform evidence-based policy for dementia.
Conflict of interest
None.
Description of author's roles
L.-F. Low conducted literature searches, reviewed the papers, drafted the manuscript. J. Fletcher conducted grey literature searches, checked reviewed papers, and critically revised the manuscript.
