Author
BRIDGE Catherine; et al. ;
Title
The costs and benefits of using private housing as the 'home base' for care for older people: a systematic literature review.
Journal citation/publication details
Melbourne : Australian Housing and Urban Research Institute, November 2006. 69p.
Summary
This review of 51 diverse studies from around the industrialised world, including systematic reviews as well as primary research and secondary data analysis, provides very mixed findings and demonstrates many methodological, conceptual and other problems in cost-benefit research in this area. However, subject to these limitations, the authors argue that there is some evidence to suggest that home-based care can be more cost-effective that institutional care when targeted to meet individual need, and when it includes a preventive element. However, a true comparison of costs needs to encompass those incurred by informal care providers, and those related to accommodation setting.
Context
'Understanding the relationship of housing to the costs and benefits of home care for older people is a pressing matter for government' in light of an ageing population in Australia and elsewhere. It is generally believed that home-based care is cheaper than institutional alternatives, but 'the validity of that belief remains unclear'. The aim of this review is thus to establish a reliable evidence base on this key policy and practice issue.
Methods
The review was conducted on the basis of referenced guidelines developed by the first two authors for use in reviews of home modification studies.
What sources were used?
The following databases were searched: AgeLine; AMED (Allied and Complementary Medicine database); APAIS - Health [a subset of Australian Public Affairs Information Service]; ARCH [Australian architecture database]; CINAHL (Cumulative Index to Nursing and Allied Health Literature); Current Contents [Thomson Scientific service covering all disciplines]; 'EBM Review' [possibly Evidence Based Medicine Reviews]; EconLit; Expanded Academic Index ASAP [multidisciplinary database published by Gale]; FAMILY [Australian Family and Society Abstracts Database]; Health and Society [a subset of FAMILY]; Medline; Proquest [a database host rather than a database]; PsycINFO; Social Work Abstracts; and Web of Science. Google searches were also conducted, and grey literature was identified (one source - the Home Modification Information Clearinghouse library in Sydney , directed by the lead author - is mentioned).
What search terms/strategies were used?
Table two lists terms used in the five component categories of the search strategy, which were: problem; intervention; comparison; outcome; and target population. A sample search strategy used in AgeLine is included as Appendix 2 and suggests that an information professional carried out or advised on searching. Although the text states that Table 2 covers 'the full list of search terms', the AgeLine search includes many more synonyms and alternatives.
What criteria were used to decide on which studies to include?
Eligible studies were published in English between 1980 and 2006; included an abstract; were accessible online or via university sources; and reported cost estimates.
Who decided on their relevance and quality?
Figure 3 summarises the relevance assessment process. The searches delivered 614 studies, which were first screened by title. Abstracts of potentially relevant titles were then screened independently by two authors. After these processes 468 studies had been eliminated. The remainder were acquired in full text and the following information coded, apparently by one author: reference; author's country of origin; process and issues addressed; whether housing was mentioned; cost-benefit factors addressed; and methodology. A 10% sample was coded by a second author. The coding sheets were then scrutinised against the inclusion criteria, resulting in the elimination of more studies.
How many studies were included and where were they from?
Fifty-one studies were selected, and the coding sheets for all are reproduced as Appendix 1. Geographical origins of authors were the USA (23), UK (9), Canada (7), Europe (5), Australia (3), New Zealand (1), Hong Kong (1), Brazil (1) and Israel (1).
How were the study findings combined?
The selected studies were diverse, including systematic reviews (7), randomised controlled trials (3), quasi-experimental studies (18), secondary data analyses (15), expert opinion pieces (5) and case studies (3). The findings are presented according to these methodologies.
Findings of the review
The authors note that none of the studies considered all 18 cost-benefit variables included on their coding sheets, and that in some cases it was not entirely clear which variables were costed. Only 14 (25%) of the studies included accommodation as a variable, and they treated housing very differently: the variables actually measured ranged from land and building costs to tenure.
Systematic reviews
The findings of the seven reviews were varied, and suggest that there is limited and sometimes contradictory evidence about the cost-effectiveness of home-based care for older people. For example, one review showed that home-based care was not cost-effective and raised overall health care costs, while two claimed that it reduced mortality and improved well-being.
Randomised controlled trials
The three RCTs focused on hospital admissions, prescribed drug use and the effectiveness of managed care. All claimed that home-based care reduced hospital admissions by between a quarter and a third, while one also reported better health and functional outcomes, and less drug use.
Quasi-experimental studies
The results of these 18 studies were inconsistent. Three reported no difference in costs between home-based and institutional care, or greater costs for the former. Five suggested cost neutrality, but greater client satisfaction for home-based care, while four reported significant cost savings for home-based care, especially in relation to reduced bed-days. Findings on the impact of particular variables on the cost of home-based care were also inconsistent, and there was no consensus on the best independent predictors of either the cost or use of such care.
Secondary data analysis
Inconsistency was also the pattern among these 15 studies, both on cost-effectiveness of home-based compared with institutional care, and on the key determinants of home-based care costs, which were identified variously as housing fitness and location, client characteristics and case mix. Two studies highlighted the importance of including the cost of informal care in any calculation of the costs of home-based care, and four emphasised the role of 'case management and optimal or rationed service allocation' in keeping costs down. Another, however, questioned the cost-reducing powers of case management.
Expert opinion
The five expert opinion pieces mainly consisted of cost projections based on demographics and subsidies, and in most cases savings were achieved through modelling more targeted delivery and the direct substitution of formal for informal care. All these studies 'were limited in that they failed to consider the intangible and indirect effects of quality of care and the impact of informal care.'
Case studies
These three studies were an individual cost survey, a narrative review and a statistical study, and generally found home-based care to be less costly. Careful targeting of home-based services was advocated, but one study recognised that this would be politically difficult.
Authors' conclusions
'This review has revealed a number of significant limitations of previous cost-benefit research.' These include disparity in the terms used to describes economic evaluations (cost effectiveness, cost minimisation, cost-benefit etc.); inconsistency in the perspective from which costs are considered (societal, provider, carer, client etc.); variation in the cost models used; lack of blinding in quasi-experimental and randomised studies; poor consideration of informal care and travel time costs; and failure to make clear which model of home-based care is being considered (acute care substitution, long term care substitution etc.). Moreover, 'many current methods of economic analysis may have a bias against the elderly', including quality of life year scores that are dependent on the number of years of life that are left.
'The most surprising outcome of this review was that so few studies explored accommodation settings' such as tenure and dwelling type. In addition, the studies demonstrate 'very poor understanding and attention given to capital and maintenance costs of housing'. Issues requiring further research are identified, and the authors also spell out their own plans to improve the Australian evidence base (only two studies in the review were conducted here) by developing a model on the basis of data from the 2003 Survey of Disability, Ageing and Care.
Implications for policy or practice
'Subject to the limitations of existing work, there is some evidence that home care can be cost effective.' Home care has been found to be cost effective when services are targeted to meet individual needs and when they include a preventive element. However, formal home care costs may appear lower than those of institutional care because of a failure to take into account the informal care associated with the former: 'true cost-benefit analysis must appropriately account for the dollar value of informal care providers' time'.