Impact of home care versus alternative locations of care on elder health outcomes: an overview of systematic reviews

Background Many elders struggle with the decision to remain at home or to move to an alternative location of care. A person’s location of care can influence health and wellbeing. Healthcare organizations and policy makers are increasingly challenged to better support elders’ dwelling and health care needs. A summary of the evidence that examines home care compared to other care locations can inform decision making. We surveyed and summarized the evidence evaluating the impact of home care versus alternative locations of care on elder health outcomes. Methods We conducted an overview of systematic reviews. Data sources included MEDLINE, the Cochrane Library, EMBASE, and CINAHL. Eligible reviews included adults 65+ years, elder home care, alternative care locations, and elder health outcomes. Two independent reviewers screened citations. We extracted data and appraised review quality using the Assessing the Methodological Quality of Systematic Reviews (AMSTAR) checklist. Results were synthesized narratively. Results The search yielded 2575 citations, of which 19 systematic reviews were eligible. Three hundred and forty studies with 271,660 participants were synthesized across the systematic reviews. The categories of comparisons included: home with support versus independent living at home (n = 11 reviews), home care versus institutional care (n = 3 reviews), and rehabilitation at home versus conventional rehabilitation services (n = 7 reviews). Two reviews had data relevant to two categories. Most reviews favoured home with support to independent living at home. Findings comparing home care to institutional care were mixed. Most reviews found no differences in health outcomes between rehabilitation at home versus conventional rehabilitation services. Systematic review quality was moderate, with a median AMSTAR score of 6 (range 4 - 10 out of 11). Conclusions The evidence on the impact of home care compared to alternative care locations on elder health outcomes is heterogeneous. Our findings support positive health impacts of home support interventions for community dwelling elders compared to independent living at home. There is insufficient evidence to determine the impact of alternative care locations on elders’ health. Additional research targeting housing and care options for the elderly is needed. Electronic supplementary material The online version of this article (doi:10.1186/s12877-016-0395-y) contains supplementary material, which is available to authorized users.


Background
Location of care (LOC) for the elderly has become an increasingly important societal issue [1]. With a demographic shift towards an aging population with complex healthcare needs, healthcare systems face challenges in providing long-term care for elders [2,3]. One such challenge is matching the elders' health and wellbeing needs to the environment. The ecological theory of aging suggests that the environment influences elders' functional status. Promoting optimal outcomes requires a goodness of fit between the elders and the environment [4]. For example, the person-environment fit can positively or negatively influence health outcomes if personal competencies are well-suited, or alternatively, poorly matched to environmental demands. The World Health Organization's report on Aging and Health highlighted the importance of environmental influence on functioning and research has shown that elders' opportunity to build and maintain functional ability is enhanced when they live in an environment that addresses their needs [5]. As such, decision makers (e.g., policy makers, patients, family members, and healthcare providers) are challenged to better consider the impact of the environment, or LOC, on elders' health and wellbeing outcomes.
To make a well-informed decision, stakeholders need to know their options (i.e., available LOCs), the benefits and harms of each option, and how these benefits and harms relate to their personal situation [6]. Common long-term LOC options include home care and institutional care [7]. Home care typically includes independent living at home or living at home with supports and/or modifications to enhance health and independence. Institutional LOCs typically refer to nursing home care or skilled nursing care facilities. Elders and caregivers have identified deciding about moving from home to an alternative LOC as one of the most difficult decisions they face [8,9]. This decision is complicated by the evolving contextual factors relating to the care situation such as elder health status, characteristics of the caregivers, and physical environment [8]. Access to evidence describing the impact of home care versus an alternative LOC on elder health and wellbeing could help inform decision making.
Several systematic reviews have examined the impact of LOC on elder health outcomes. However, these reviews tend to narrowly focus on the health impact of a specific intervention (e.g., fall reduction interventions, palliative care, case management) delivered at different care locations. Although individual reviews provide important contributions, a single document that summarizes a broader range of findings will improve access to the literature and help inform stakeholders' decisions about elder LOC. Additionally, it will highlight important knowledge gaps and help prioritise future research questions. Therefore, we summarized available research findings that examined the impact of home care versus alternative LOCs on elder health and well-being.

Design
We conducted an overview of systematic reviews. An overview design attempts to survey, summarize, and describe the literature allowing findings to be efficiently compared and contrasted [10]. The paper is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [11]. However, we did not register our protocol in PROS-PERO [12]. We initially piloted our protocol using a single database [13]. Changes from the original protocol to the current study included limiting the population from elders and caregivers, to elders only. Our protocol can be available to readers upon request to the corresponding author.

Review eligibility criteria
The PICOS framework guided systematic review eligibility criteria (Table 1). We included systematic reviews that examined: adults aged 65 or greater, home care, alternative LOCs, and health outcomes (e.g., physical and mental health, morbidity, mortality, functional status and dependence, activities of daily living, quality of life, falls, etc.). We defined systematic reviews as a review of the evidence that included a clear research question, used systematic methods to identify, select and appraise the primary research, and extracted and analysed/synthesized data from included studies [14]. We did not restrict systematic reviews types (e.g. metaanalysis, narrative), dates, or language of publication. Non-systematic reviews, individual studies, and abstracts were excluded. Participant's age had to be reported in the results section of the systematic review to be included. When systematic reviews reported a broad age range, we included the review if the mean participant age was 65 years or greater and/or if a sub-group analysis was conducted on participants with a mean age of 65 years or greater. Although our focus was specific to LOCs (i.e., home or alternative setting), we included reviews that examined interventions, care options, and service delivery models within the context of eligible LOCs. Ineligible LOCs were acute hospital stays and temporary or transitional placements (e.g., respite care or community/institutional placements for less than 3 months) [15]. If two reviews had 100% overlap of included studies (e.g., systematic review updates, systematic reviews with the same aims and outcomes), we included the review with the most complete and up-to-date dataset. All eligible LOC comparisons were considered.

Search strategy
An Information Specialist designed and conducted the search with input from the research team. Our search strategy aimed to find all systematic reviews that compared home care to alternative locations of care. The search included a mix of subject headings and keywords related to the participants (e.g., aged, senior, older, elder, geriatric) and home care (e.g., home care services, home hospitalization). The rationale for searching 'home care' alone was to find all reviews concerning home care and select only those that compare home care to at least one other LOC. Databases, searched from journal inception to June 2016, included: MEDLINE, the Cochrane Library, EMBASE, and Cumulative Index to Nursing and Allied Health Literature (CINAHL). The search included a mix of subject headings and keywords related to the participants (e.g., aged, senior, older, elder, geriatric) and home care (e.g., home care services, home hospitalization). Limits were applied to study designs (e.g. reviews only) and participant characteristics (e.g., not infant, child or adolescent) (Additional file 1). We also scanned the reference list of included systematic reviews for systematic review eligibility. When we were unable to obtain the full text, we emailed the first author to request the paper.

Review selection
After removing duplicates, two reviewers independently screened citations in three phases: title, abstract, and full text. Title screening was guided by the citations' overall relevance to this study. The eligibility criteria guided the abstract and full text screenings. Reviewers made one of three screening choices: include, exclude, or unsure. Citations were removed from further screening when both reviewers decided the article should be excluded. When at least one reviewer assigned an article to the 'include' or 'unsure' category, the article moved to the next screening stage. Inclusion and exclusion of full text articles required reviewer consensus.

Data collection
One reviewer extracted data using a standardized pre-piloted form. A second reviewer independently verified all extracted data points for accuracy and completeness. All inconsistencies were resolved through consensus. Extracted variables included: citation information, review objectives, review methodology, review critical appraisal methods, number of studies included, number and characteristics of participants, LOCs, comparator LOCs, interventions, elder health and wellbeing outcomes, synthesis/summary of results, conclusions, and strengths/ limitations. We only extracted information from the systematic reviews that were relevant to this overview [16]. In other words, for some systematic reviews, we only examined a subset of studies (e.g., those related to the elderly or that compared LOCs).

Data synthesis
We synthesised the data narratively using groupings of LOC comparisons that emerged from the included reviews. Pooling the data was not appropriate due to the heterogeneity in study design, population characteristics, LOCs, interventions, measures, and outcomes, within and across included systematic reviews.

Critical appraisal of included reviews
Two raters assessed the methodological quality of included systematic reviews using the 11-item Assessing the Methodological Quality of Systematic Reviews (AMSTAR) instrument [17]. Disagreements were resolved through consensus and third person arbitration. A sensitivity analysis was conducted at the systematic review level (i.e., not primary studies). We re-examined higher quality systematic reviews that scored at or above the mean AMSTAR score.

Review selection
The search yielded 2,575 citations ( Fig. 1). After removing duplicates and screening titles and abstracts, we examined 207 full texts, of which 17 were eligible for inclusion. Screening the reference lists of included systematic reviews revealed 16 potential titles, of which 2 were included. In total, 19 systematic reviews met our inclusion criteria.

Characteristics of reviews
Included systematic reviews were published in English between 2000 and 2016 and originated from seven different countries: United Kingdom (n = 6), Canada (n = 5), Australia (n = 2), the Netherlands (n = 2), United States (n = 2), Switzerland (n = 1), and Sweden (n = 1) ( Table 2). Three hundred and forty studies (range 1 to 110) with a reported 271,660 participants (range 96 to 108,838) were synthesized across the systematic reviews. LOC comparisons were: home with support versus independent living at home (n = 11 reviews), home care versus institutional care (n = 3 reviews), and rehabilitation at home versus other conventional rehabilitation services (n = 7). Data from the Mehta [18] and Fens [19] systematic reviews were included in two categories.

Summary of findings
Home with support versus independent living at home (n = 11 reviews) Compared to usual care, typically defined as independent living at home, seven reviews favoured home with support, three found no difference, and one reported insufficient evidence (Tables 3 and 4). Home support by an interdisciplinary team reduced nursing home and hospital admissions, decreased falls, and improved physical function [20]. Two reviews found that preventative home visits for community dwelling elders improved health and functional status, mortality rates, and delayed hospitalization and nursing home use [21,22]. However, another review found no difference in health outcomes due to preventative home visits [23]. Transitional support provided by clinical nurse specialists reduced re-hospitalizations [24]. Two reviews examined case management interventions (i.e., collaborative patient care to meet patients' holistic needs). One found delayed onset of nursing home placement and decreased nursing home admissions, length of stay, and community care service use [25]. The other found that case management approaches for people with dementia improved cognitive status and decreased institutionalization and caregiver outcomes at certain time points [26]. One review concluded that home health monitoring technologies reduced undesired health outcomes for the elderly [27]. Another review found limited evidence for the effectiveness of multidisciplinary care for stroke patients at home compared to usual care [19]. Mehta [18] found no difference between receiving home physiotherapy for hip fractures compared to no physiotherapy. Finally, a Cochrane review found little high-grade evidence that physical modifications to the home environment affected the likelihood of sustaining an injury in the home [28].        Home care versus institutional care (n = 3 reviews) Three reviews compared home care to institutional care. Although each review differed in their purpose and scope, two found insufficient evidence, and one review favoured home care (Tables 3 and 4). A Cochrane review that examined the health outcomes of elders receiving home care compared to institutional care found only one randomized controlled trial and concluded that there was insufficient evidence to make recommendations about alternatives to institutional care for the elderly [7]. Similarly, a review that examined various home care options compared to nursing homes for caredependent elders found no significant differences for most health outcomes [29]. Another Cochrane review found that elders who received home palliative care compared to institutional palliative care were more likely to die at home and experience less symptom burden, particularly for patients with cancer [30].
Home rehabilitation versus conventional rehabilitation care (n = 7 reviews) Within this category, five reviews reported no differences between home and conventional rehabilitation care, one concluded insufficient evidence to make a recommendation, and one reported that home rehabilitation was superior to conventional options in the short-term (Tables 3 and 4). Specifically, one Cochrane review included seven studies that compared homebased rehabilitation to medical day hospital rehabilitation for community dwelling elders [31]. Findings showed no difference between these settings on health outcomes. Three reviews examined post-stroke rehabilitation. In one review, there were no significant differences between home and conventional rehabilitation for activities of daily living, depression, quality of life, and social activities [32]. Pooled evidence from the other review suggested that home-based rehabilitation, provided between 6 and 12 weeks after discharge from inpatient care, might yield superior functional benefit and satisfaction; however, benefits were less clear after 6 months [33].
The third review showed no difference between home and conventional rehabilitation for outcomes related to activities of daily living and quality of life [19]. Three reviews examined rehabilitation after hip fractures. One review found no difference between home and outpatient physiotherapy for patient-reported healthrelated quality of life, but suggested that performancebased outcomes might be marginally better following outpatient services [18]. However, due to the low quality of the primary studies the authors concluded that there was insufficient evidence to recommend one service delivery model over the other. A similar review found that home-based rehabilitation programs involving physical therapy for hip fractures were as beneficial as hospital rehabilitation for patients who had not lost many functional abilities prior to the fracture [34]. The other review examined elders with dementia and hip fractures and found no differences in health outcomes between the settings [35].

Critical appraisal with sensitivity analysis
Based on the AMSTAR ratings, the average quality of included systematic reviews was moderate, with a mean AMSTAR score of 7 of 11 (median = 6; range = 4 to 10) ( Table 5). No systematic reviews met the criteria of disclosure of financial interest (Item 11) because none reported funding/support of each included study. We re-examined systematic reviews scoring at or above the mean. In the home with support versus independent living at home category, 5 of 11 reviews were of higher quality. Four reviews, including one Cochrane, which examined preventative home visits, transitional support, case management, and home health technologies, concluded that home with support was superior to independent living at home [22,24,26,27]. The other, a Cochrane review, concluded that inseparable or insignificant results for home modification to reduce falls [28].
Two of three reviews in the home care versus institutional care category were high quality (AMSTAR ratings were 9 and 10 of 11, both Cochrane reviews). The findings, however, were mixed: one review concluded that there was insufficient evidence to determine whether elders fare better with home care compared to institutional care [7] and the other supported home palliative care over institutional palliative care to improve the likelihood of dying at home and decreasing symptom burden [30]. Only the Brown et al., [31] Cochrane review was of high enough quality (AMSTAR score of 10/11) to be included in the home versus conventional rehabilitation category. This review found no significant differences in elder health outcomes between those who received home rehabilitation compared to those who attended medical day hospitals for rehabilitation services.

Discussion
We synthesized systematic reviews evaluating the impact of home care versus alternative LOCs on elder health outcomes. Overall, we found 19 eligible systematic reviews covering 340 studies and 271,660 participants. Reviews originated from seven industrialised countries that are known to be experiencing rapid growth among their elder population. Categories of comparisons included: home with supports versus independent living at home, home care versus institutional care, and home rehabilitation versus conventional rehabilitation. Most reviews favoured home with support to independent living at home. Results for home care versus institutional care were mixed. Most reviews found no differences in health outcomes between rehabilitation at home versus conventional rehabilitation services. The quality of included systematic reviews was moderate. Our results lead us to make the following observations. First, as aging in a desirable LOC may contribute to overall health and wellbeing in the late-life period [36,37], we were not surprised to find several reviews that examined LOC as a function of elder health and wellbeing. Many elderly people prefer and choose to age at home [38][39][40][41][42]. Healthcare organizations and policy makers are increasingly challenged to better support a shift from institutionalized long-term care Table 5 Authors' judgements about each risk of bias item of included systematic reviews = yes = no = unsure to support elders to remain in their community [43]. The included reviews indicate a growing literature of interventions designed to promote health, function and independent living among community dwelling elders. Yet, many older adults anticipate moving, often to a more institutionalized location. Even those who do not anticipate moving might be compelled to move due to unforeseen life changes (e.g. loss of a spouse, illness) [40,44]. As life expectancy increases and more people in late life suffer from multimorbidity, their relocation will represent a significant market for new residential options beyond the limited choices of private nursing homes or public long-term care facilities [45]. Therefore, an improved understanding of the role of long-term LOCs on elders' health and wellbeing is central to any strategy aimed at fostering elders' quality of life [46].
Second, the impact of home care compared to institutional care on elder health outcomes was less clear. The Mottram et al. [7] and Wysocki et al. [29] reviews had similar PICOS questions, and despite the 13-year publication gap and new research, both concluded that there was insufficient evidence to draw firm conclusions. Similarly, most systematic reviews that examined location of rehabilitation services and elder health outcomes found no differences or insufficient evidence. Inconclusive findings suggest that informed patient preferences and individual needs should guide elders' decision to move from home to an alternative LOC. Examination of LOC options and preferences should also incorporate contextual factors, such as the quality of available services or other interventions. Consistent the ecological model of aging and person-environment fit, elders' specific needs (e.g., mobility, cognitive, social) must be evaluated and matched to available LOC options with appropriate interventions or services to address those needs.
Third, our results highlight that most home care research focuses on few LOC comparisons. Further, most systematic reviews did not compare across multiple potential LOCs. When examined individually, each comparison provided insight regarding potential outcomes of one LOC option compared to the other. However, it was not possible to integrate the findings to determine the best LOC based on elder health outcomes. Although we did not examine the primary studies, our overview identifies a need for better research targeting long-term LOC options for the late-life years. Improved research and additional investment will translate into more rigorous methods and generate a more complete understanding of the impact of LOCs on elders' health.
Our findings should be interpreted within the context of its limitations. First, our search strategy focused on home care, thus terms such as 'day hospital' and 'assisted living' were not included. Also, we did not search the grey literature. Therefore, we might have missed eligible systematic reviews. Second, several factors made this literature challenging to synthesize. We found considerable heterogeneity within and across included systematic reviews regarding population characteristics, locations considered, measures used, and outcomes reported, precluding firm conclusions. LOCs were also poorly and inconsistently described in the literature. Development of an accepted taxonomy for LOCs would help advancement in this field. Third, the study designs of the included reviews made direct comparisons for LOCs difficult and we noted a risk of indication bias (elders' health status correlates with the intervention and is a risk indicator for the outcome) within several of the primary studies. Causal relationships between LOCs and elder health outcomes could not be measured. Instead, this overview provides a broad summary of the state of knowledge regarding elder home care compared to alternative LOCs and elder health outcomes. Fourth, there was some primary study overlap across the systematic reviews, potentially over-representing certain findings. However, the purpose of an overview of systematic reviews is to provide a high level description of the state of the evidence [16]. To offset this limitation, we excluded studies with 100% of primary study overlap. Finally, new and relevant studies might have been excluded if they were not yet included in a systematic review.

Conclusions
Results from this overview of systematic reviews, which evaluated the impact of home care compared to alternative long-term LOCs on elder health outcomes, suggests that home interventions and/or supports that promote elder health and independence might be effective in helping elders age at home. However, we are unable to make recommendations regarding the impact of other LOCs compared to home on elders' health and wellbeing. Without a more robust evidence base, elders considering moving from home will be unable to make evidence informed decisions about which long-term alternative LOC to choose that fits best with their needs and preferences.