|Since the provision of suitable housing is an essential part of community care, the aggregate housing needs of different community care groups have to be estimated. A method of doing so was developed by Smart and Titterton, based on prevalence rates of different conditions among the general population. This new study was commissioned to critically review the prevalence rate method and Smart and Titterton's estimates before they become widely accepted as targets. It addresses 2 issues: are national prevalence rates useful in community care planning? and are reliable prevalence rates available for all community care groups?|
- Prevalence based estimates are robust for the larger, better defined and more fully researched community care groups (older people, people with physical and learning disabilities, and people with mental health problems) but not for other groups. For these larger groups, the estimates provide a "headline" estimate of the approximate number of people who suffer such problems and some indication of the distribution of severity of their condition.
- Prevalence based estimates should be used with caution: they do not show that people within a community care group who have a certain level of dependency should always be matched to a specific type of housing or support. They can, however, indicate the range of housing or support that may be suitable.
- Local planning should not rely solely on prevalence based estimates; they should be one of a battery of tools. There is no single well-established method of estimating housing needs for community care groups at the local level; the pathways approach has the greatest potential.
|The prevalence based method of estimating the housing needs of community care groups was developed by Smart and Titterton, first for Glasgow District Council, 1 and then for Scotland as a whole in a report for Scottish Homes. 2 The method involves the following 4 steps:|
|1. estimating the prevalence rate for a particular condition or problem, using past research evidence;|
|2. estimating the distribution of severity of the condition within the group in terms of high, moderate and low needs groups;|
|3. making a household adjustment to convert individual needs into household needs;|
|4. allocating the estimated number of households with housing needs into preferred housing and support solutions depending on their characteristics.|
|This study scrutinises each of these 4 steps in the Smart-Titterton method, and considers what role prevalence rate based estimates should have in the planning of housing for community care users. The study included consultation with potentially interested groups, and 4 local case studies in contrasting areas of Scotland.|
|Estimating prevalence rates|
|The review found that the estimates produced by Smart and Titterton were not always the best available: |
- calculating the prevalence rate as the average from the findings of a number of different studies is questionable unless all the studies are believed to be of equal validity and relevance; this is seldom the case; some studies contributing to the averages were rather dated, some were from overseas, and some used different definitions.
- there are ambiguities in some use of the original source material.
- there are some arithmetical errors and inaccuracies in Smart and Titterton's estimates.
|The review was, however, able to estimate relatively robust prevalence rates for the larger, better defined and more fully researched community care groups: older people, people with physical and learning disabilities, and people with mental health problems.|
|Estimating the distribution of severity|
|The studies on which the Smart-Titterton estimates were based are particularly inconsistent in their treatment of institutional populations. These inconsistencies have a significant effect on the estimates of those with high needs. In the studies there were many other differences of detail in the consideration of severity.|
|Nevertheless, broad estimates of the numbers with moderate or high needs are possible within the major community care groups.|
|The household adjustment|
|In many of the estimates produced, the translation of individual needs into household needs is problematic: assumptions have to be made about the extent to which people with the same condition live together and about the proportion of people who get all the support they need from within their families. Assumptions that those sharing accommodation are adequately provided for are liable to under-estimate physical housing needs (particularly needs for aids and adaptations) and the support required.|
|All the estimates are in terms of adults; children's disability is little recognised but, clearly, appropriate housing can assist the care of disabled children.|
|There is a danger of double counting: one individual can belong to several community care groups. Conversely, the standard community care groups may not cover all needs.|
|Fitting households to preferred housing|
|Many ad hoc assumptions are made to produce estimates of the distribution of severity. The estimates should be used only to indicate the range of housing or support which may be suitable for a particular need group. Using them to make exact recommendations for specific housing solutions is not appropriate. In addition, an assumption needs to be made about the proportion of people who provide solutions to their needs privately.|
|Potential community care users are likely to be concentrated in those groups designated as having high or moderate needs. In general those estimated to have low needs appear to be outwith the population requiring specialist housing or significant formal support.|
|Prevalence rate based estimates of housing need assume an ideal situation of adequate services being available to support the great majority of people within the community; they do not make explicit what pace of long stay hospital closure is assumed, and they do not state the timescale for achieving the targets of housing provision.|
|Using national estimates for local planning|
|Use of national prevalence rates may not be appropriate in all local areas because: |
- analysis of the Census shows that poor health is not equally distributed across Scotland.
- some needs (eg mental illness) may be systematically related to socio-economic conditions and hence prevalence rates may need to be adjusted.
- some needs groups are so small that in small local areas there would be very large margins of error if national prevalence rates were used.
|If used with caution, national estimates are of some use in local planning. In the best researched community care groups they can provide a relatively reliable guide to the numbers of a client group, by needs category, who need housing and support in a local area. But for most groups they cannot provide firm targets for local housing provision: they can only provide broad brush targets which should be used in conjunction with other measures and estimates.|
|The review concluded that evidence in relation to many important aspects of the needs of community care groups is limited but, as local planning develops, more and better information may become available; it concludes that prevalence based estimates provide an easy, cost effective way of providing a "headline" view of likely needs, and a way of looking consistently at needs across different groups.|
|What are the alternatives?|
|This study briefly reviewed alternative methods of estimating needs at the local level, including the pathways approach. 3 This approach consists of 3 basic steps:|
|1. All the people in a local area in a community care client group are estimated and their current living circumstances described.|
|2. The factors that will cause current living circumstances to become unsuitable are described, and the proportion of cases to which these factors will apply is estimated, thus providing an estimate of gross housing need.|
|3. Net outstanding needs are estimated by estimating the extent to which needs will be met from existing supply.|
|The study found that, of the alternatives examined, the pathways approach has the greatest potential but recognised that the approach requires detailed work at the local level and is therefore more expensive than using prevalence rates. As yet, relatively few local authorities have implemented it for even some of the community care groups.|
|1 Smart H and Titterton M (1994) "Housing and community care in the City of Glasgow", Edinburgh: TREC.|
|2 Titterton M (1994). "Housing for special needs", Scottish Homes Working Paper, Edinburgh: Scottish Homes.|
|3 Watson L (1996) "Housing needs and Community Care: the housing pathways pilot programme", London/Coventry, NFHA/CIH.|
|'Estimating the Housing Needs of Community Care Groups', the research report which is summarised in this Research Findings, is available priced £5.|
|Cheques should be made payable to The Stationery Office and addressed to: |
The Stationery Office,
Mail Order Department,
21 South Gyle Crescent,
Edinburgh EH12 9EB
Telephone: 0131-479 3141, or Fax: 0131-479 3142.
The report can also be ordered online from:www.thestationeryoffice.co.uk
|Further copies of this Research Findings may be obtained from: |
The Scottish Office Central Research Unit
Room 2J CRU
Telephone: 0131-244 7562.