Age and Ageing Advance Access originally published online on June 4, 2007
Age and Ageing 2007 36(4):424-430; doi:10.1093/ageing/afm067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Socio-demographic variations in moves to institutional care 1991–2001: a record linkage study from England and Wales
Centre for Population Studies, London School of Hygiene & Tropical Medicine, 49–51 Bedford Square, London WC1B 3DP, UK
Adress correspondence to: Emily Grundy. Tel: 020 7299 4668. Email: Emily.grundy{at}lshtm.ac.uk
 |
Abstract |
|---|
 Top Abstract BackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
Background: Only a minority of older people in England and Wales live in institutional care, but among the older of the old, this minority is large. Disability is the major driver of admissions, but socio-demographic factors are also relevant. Understanding more about the influence of these is important for planning by long-term care.
Objective: To investigate effects of socio-demographic factors, including housing tenure, household type, marital status, and number of children, on the proportions of elderly people who made a transition from living in the community in 1991 to living in institutional care in 2001.
Subjects and setting: Nationally representative record linkage study including 36,650 people aged 65 years and over, living in the community in England and Wales in 1991, who were still alive in 2001. Nineteen thousand women aged 75–89 years in 2001 were included in additional analyses of effects of parity (number of children borne).
Methods: Bi-variate and multivariate analyses of variations in sample proportons, who b 2001 were resident in institutional care.
Results: 4.3% of men and 9.3% of women in the surviving sample then aged 75 years and over, were in institutional care in 2001. Older age, living in rented accommodation, living alone in 1991 and being unmarried in 2001, as well as long-term illness, were associated with higher proportions making this transition. Women had higher risks than men. Childless women aged 64–79 years in 1991 had a 25% higher risk than women with children of being in institutional care in 2001.
Conclusion: Socio-demographic factors continue to influence risks of entry to institutional care in England and Wales.
Keywords: institutionalisation, oldest old, parity, longitudinal studies, family and household, elderly
|
Background |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
Most elderly people are free from serious disability and prefer, and are able, to live in their own homes with no or relatively little help. However, among the older old—the most rapidly growing segment of the population in many developed countries—needs for assistance are much higher [1]. Much of this help is provided by family, neighbours and paid caregivers, and government policy in the UK is to enable older people to remain in their own homes as long as possible. This policy was reiterated and strengthened with the implementation of the NHS and Community Care Act, and other reforms, during the 1990s [2]. Nevertheless, in 2001, 15% of men and nearly a quarter of women aged 85 years and over lived in institutional care, which was the second-most usual type of living arrangement for women of this age [1, 3]. Institutional care is costly and policies on its provision and financing are contentious. It is, therefore, important to understand as much as possible about factors that increase the risk of admission.
There is a strong association between morbidity and disability, especially cognitive impairment, and institutional admission, [4–6] and a national survey conducted in England in 2000 showed that 75% of care homes residents were severely disabled [7]. A more detailed 2003 survey of UK residential and nursing homes managed by the largest not-for-profit provider found that half of all residents had dementia, stroke or other neurodegenerative disease. Family or social factors were identified among reasons for admission for 9% of nursing home and 20% of residential home residents, but over 90% of residents had an identifiable clinical condition leading to admission [8]. However, not all seriously disabled elderly people enter institutional care and there may still be a small minority of residents (especially among self funders who are not required to undergo an assessment) who are not seriously disabled [9]. In short, although serious disability is the major driver of institutional admission, other factors, such as availability of social and socio-economic resources, are also relevant.
Longitudinal studies, predominantly from North America, have shown that lower socio-economic status, including not being a home owner, being unmarried, and living alone, are associated with higher risks of entry to institutional care [10–14]. Nationally, representative longitudinal data are sparser for the UK, but previous analyses of the Office for National Statistics Longitudinal Study (ONS LS) showed associations between housing tenure and marital status and transitions to institutional care in the decades 1971–1981 and 1981–1991: these analyses included no control for health status as information on this was not collected in the 1971 and 1981 censuses [15–17]. Local studies have also shown that these factors, together with income and education, are associated with admissions [18, 19]. The prior national studies predate the extensive changes in long-term care policy introduced during the 1990s [2] which may have led to changes in associations between socio-economic and socio-demographic factors and admission to institutional care. To our knowledge, the last national British study to examine the influence of number of children was undertaken 40 years ago [20] and was cross-sectional. There is, however, recent Swedish evidence which suggests that elderly people with children are less likely to enter institutional care than the childless [14]. Some US studies have also shown that availability of a daughter, in particular, influences entry to nursing home care, although some contrary results have also been reported [13, 21, 22].
There are several reasons why socio-economic and socio-demographic factors might be associated with admission rates including the well-established association between socio-economic status and disability [2, 23]. Additionally those with greater socio-economic resources are better placed to pay for additional home care, and so postpone or avoid institutional admission. Means testing, whereby those with resources, including housing wealth, are required to contribute to the costs of care may provide a further disincentive to institutional admission among home owners [18] and descriptive studies of admissions to publicly-funded institutional care show a preponderance of tenants [24]. Socio-demographic characteristics, particularly marital status and availability of children, are important as it is well established that most extra-institutional care for older people with disabilities is provided by close relatives [2, 25].
We use data from a large nationally representative study of England and Wales, the ONS LS, to analyse effects of housing tenure and family status, including for women number of children borne, on transitions from living in the community in 1991 to living in an institution in 2001, taking account of age and presence of long-term illness.
|
Data and methods |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
The ONS LS is a record linkage study of approximately 1% of the population initially based on those enumerated in the 1971 Census of England and Wales and now including linked information from subsequent censuses and vital registration. The data-set is continually updated with the addition of 1% of births and immigrants and so remains nationally representative. Strengths of the data include large sample size, low non-response and attrition bias, and crucially for this study, inclusion of the institutional population. Limitations include the long interval between census data capture points and the relatively limited range of information collected in the census. Full details of the study have been reported elsewhere [26].
|
Measures |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
Our outcome measure is residence in a communal establishment in 2001. These were defined in the census as ‘establishments providing managed residential accommodation’. These include nursing homes, residential homes and long stay hospital accommodation but not sheltered housing (Appendix 1). The 2001 Census collected information on the basis of usual residence, so those temporarily in establishments (for example, patients in acute hospital care) are not counted as institutional residents.
|
The 1991 variables we included were age (single years); housing tenure; family/household type; and presence or absence of a limiting long-term illness. The household/family type indicator was based on the definitions employed in the 1991 Census. These defined a ‘family’ as a co-resident grouping comprising a couple, with or without never-married children (of any age); or a lone parent and never-married child. We derived from this a 4-fold classification distinguishing people living alone; those living just with a spouse; those living in another type of family (with at least one never-married child); and a fourth grouping we term ‘complex’. This category includes all those living with people other than a spouse or never-married child and very largely comprises elderly people living with married children and their families. This classification has been used in earlier work on transitions to institutions 1971–81 and 1981–91 [15, 16].
We also included marital status and limiting long-term illness in 2001. This information post-dates admission to institutional care (which we can date only as occurring between the 1991 and 2001 censuses), but as durations of stay in institutional care are relatively short [27], is probably a good indicator of status at admission. Certainly, marital status in 2001 is likely to be a better indicator of marital status at time of admission than the equivalent 1991 measure.
The long-term illness measures come from a question in the 1991 and 2001 censuses which asked whether a person had ‘any long-term illness, health problem or handicap [1991]/disability [2001] which limits his/her daily activities or the work he/she can do’. Completion notes instructed that problems due to old age should be included.
For a subset of women who had been present in the sample in 1971 we analyse associations between parity (number of children ever borne) and institutional admission 1991–2001. Parity information comes from fertility histories collected in the 1971 Census from women who were then ever-married and aged 16–59 years, together with information on subsequent births from linked birth registration data (very few for the women we consider who were aged 45–59 years in 1971); post-1971 immigrants are necessarily excluded from this analysis. Information on non-marital births was not collected in 1971 and we have assumed women who were then never-married were childless. This is only a slight potential source of bias as prior to 1971 non-marital births were relatively few. Full details of the derivation and comparisons with other official sources, which show a very good match, have been reported elsewhere [28]. Parity relates to children ever borne, rather than living children. However, there is a strong correlation between the two and demographic modelling has shown that for women born in the early twentieth century the vast majority (over 95%) of parous women have or will have a child alive when they themselves reach the age of 80 [29].
All analyses refer to LS members who were in the sample in 1991 and still alive and in the sample in 2001. Loss to follow-up was very low (<4%) but there was considerable loss through mortality. This is not a source of bias as our outcome is residence or not in an institution in 2001 and the sample is representative of the population aged 75 and over in 2001 (immigrants between 1991 and 2001 are not included, but rates of immigration in these age groups are very low). However, it should be recognised that the outcome indicator—proportion of people in the community in 1991 who were in institutional care in 2001—is not a measure of incidence of institutional admission as it excludes those who entered institutions after 1991 but died before 2001.
We excluded those not enumerated at their usual residence or already in institutional care in 1991. Two hundred and thirty people were excluded because of inconsistent information on age and sex in the 1991 and 2001 censuses, and 49 because of missing 2001 data on marital status. A larger group had missing information on limiting long-standing illness in 2001; these were separately identified and included. After these exclusions, the main sample comprised 36,647 people aged 65 years and over in 1991 who survived to 2001. In similar analyses of transitions to institutions 1971–81 and 1981–91 people identified as ‘staff’ rather than residents were also excluded [15, 16]. However, detailed investigation by the Office of National Statistics showed that in 2001 changes to the design of the census form for institutional residents led to erroneous classification of elderly residents of institutions as staff. We therefore here follow official ONS guidance on this issue and have not excluded residents (all of whom are aged 75 years or over) recorded as staff [30].
|
Results |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
Table 1 shows sample size and characteristics and also numbers and proportions in institutional care in 2001. Five hundred and eighty-three (4.3%) of the 13,543 men and 2,143 (9.3%) of the 23,104 women in the study lived in institutions in 2001. Of these, 55.4% were in residential homes, 37.5% in nursing homes and the remaining 7.2% in other types of establishment, predominantly homes described as ‘other medical and care’ or psychiatric hospitals (Appendix). There was a strong association between age and transition to an institution. Twenty-nine percent of those aged 80 years and over in 1991 (20% of men and 32% of women in this age group) were in institutional care by 2001 (when aged 90 years and over). The proportions in communal establishments by 2001 were higher for tenants than for owner occupiers; higher for those who had lived alone or in a complex household in 1991 than for those living with a spouse or other family; and much lower among those married in 2001 than for the unmarried. Differentials by long-term illness, especially 2001 long-term illness, show higher rates of transition among those with a long-term illness or with missing information for 2001.
|
Results from multivariate analysis (logistic regression) are shown in Table 2. Results from two fully adjusted models are presented; model 1 includes only variables measured in 1991; model 2 additionally includes the 2001 indicators. Results from model 1 show that older age, being a social housing tenant (and among men being tenants in privately rented accommodation), and long-term illness were all independently associated with higher risks of institutional residence in 2001. Those who in 1991 lived in a family with a spouse or a never-married child had significantly lower risks than either those living alone (the reference category) or those living in complex households; there was no significant difference between the latter group—mostly people living with married children—and those living alone. When 2001 marital status and long-term illness were also included (model 2), age, housing tenure and long-term illness in 1991 remained significant. Living in the ‘other family’ type of household in 1991 was significantly negatively associated with institutional residence in 2001, even allowing for 2001 marital status, but for women having lived with a spouse in 1991 no longer appeared protective. 2001 marital status was strongly associated with institutional residence, particularly for men. In the whole sample single people were five and a half times as likely, and widowed or divorced people three and a half times as likely as the married to be in an institution. Presence of limiting long-term illness conferred a 9-fold risk; but the odds of having made a transition to an institution were even higher for those with missing information for this variable. After adjustment for all these factors, risks for women were higher than for men (OR 1.13, 1.02–1.26) and were strongly associated with age (OR 1.15, 1.14–1.16).
|
Of the 18,951 slightly younger women included in the analysis of the effects of parity, 6.9% were resident in a communal establishment in 2001 (Table 1). Women who had had one or more children were significantly less likely than childless women to be in institutional care in 2001, even when the range of other characteristics already discussed was controlled (OR 0.74, 0.63–0.87) (Table 3). However, although the odds ratio was lowest for women who had had four or more births, there were no statistically significant differences between women of higher parities. As in the main analysis, housing tenure, family/household type, marital status and long-term illness were also associated with differential risks.
|
|
Discussion |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
The 1990s were a time of considerable change in policy on long-term care [2] but these results show that socio-economic and socio-demographic factors remain important influences on institutional residence. To some extent, differentials may reflect variations in disability not captured by the indicator available in this study—limiting long-term illness—which is inadequate to distinguish, for example, people with chronic conditions not involving serious ADL limitations from people with advanced cognitive impairment. However, the size of the effects also suggests that socio-economic and family resources have a further influence. Among women aged 75–89 years, the childless were 25% more likely to be in an institution in 2001 than those who had children. Being married, as shown elsewhere [10–22], conferred a strong protective advantage. Even allowing for marital status in 2001 (and in the sub-analysis, for parity), those who were living with a never-married child in 1991 (and for men, those then living with a spouse) had a reduced chance of being in an institution in 2001 when compared with those then living alone. However, those who in 1991 lived in complex households, predominantly with married children, had similar risks as those living alone. This might reflect a higher level of disability among those who had already made a move to living with a married child; it may also be the case that spouses and never-married children are able or willing to carry on caring for a disabled spouse or parent for longer than married children with competing responsibilities [11, 13, 31]. Targeting of home care services on those with the perceived highest risks of institutional admission (such as those living alone) may mean that insufficient support is provided for these carers, an issue worthy of further attention.
Differentials by housing tenure in 1991 were marked with significantly higher risks of institutional residence in 2001 among those who had been social housing tenants in 1991. Owner occupiers (and their children) are both more likely than tenants to have the resources to pay for additional care at home and a financial disincentive to enter institutional care.
Our results refer to people born in or before 1926. These early twentieth century cohorts were characterised by high proportions of the unmarried and childless. Their successors born in the 1930s and 1940s, the parents of the post-war baby boom had, by contrast, unusually high rates of marriage and low rates of childlessness. Additionally, a recent narrowing of sex differentials in mortality has led to some postponement of widowhood [3, 29]. These trends suggest decreasing propensities to enter institutional care in the short-term future, but only if families continue to provide current levels of support. The large growth in the numbers of older old people projected for the coming decades will, however, mean an increasing need for long-term care even if levels of disability fall to some extent [32] providing a further imperative for policies and resources to support seriously disabled older people and their carers. Finally it should be recognised that moves to residential settings are sometimes the only way of meeting the needs of older people with complex disabilities. High-standard individualised residential care may be associated with improvements in the quality of life of some older people, ensuring that good practice in this regard is the norm will also require additional resources.
|
Key points |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
- Among women aged 65–79 years in England and Wales in 1991 the childless were 25% more likely to be in institutional care ten years later than women with children.
- Being married in 2001, having lived with a never-married child in 1991 and then being in owner occupied housing, were associated with lower risks of a move to institutional care.
- Allowing for these factors, and for age and long-term illness, women were more likely than men to move to institutional care.
|
Conflicts of interest |
|---|
None.
|
Acknowledgements |
|---|
The research reported here was funded by the Joseph Rowntree Foundation.
The Office for National Statistics provided access to the data and the Centre for Longitudinal Study Information and User Support (CeLSIUS) service at the London School of Hygiene & Tropical Medicine facilitated and assisted with data extraction. CeLSIUS is funded by the UK Economic and Social Research Council Grant Reference RES-348-25-0004. The study was approved by the ONS Longitudinal Study Research Board.
|
References |
|---|
|
TopAbstractBackgroundData and methodsMeasuresResultsDiscussionKey pointsReferences |
|---|
- Tomassini C. The oldest old in Great Britain: change over the past twenty years. Popul Trends (2006) 23:32–9.
- Wanless D. Securing Good Care for Older People: Taking a Long-Term View (2006) London: King's Fund.
- Grundy E, Murphy M. Human Longevity, Individual Life Duration, and the Growth of the Oldest-Old Population—Crimmins EM, Horiuchi S, Zeng Y, Robine J-M, eds. (2006) Dordrecht: Springer. Marital status and family support for the oldest old in Great Britain.
- Aguero-Torres H, von Strauss E, Viitanen M, Winblad B, Fratiglioni L. Insitutionalization in the elderly: The role of chronic diseases and dementia. Cross-sectional and longitudinal data from a population-based study. J Clin Epidemiol (2001) 54:795–801.[CrossRef][ISI][Medline]
- Bharucha AJ, Pandav R, Shen C, Dodge HH, Ganguli M. Predictors of nursing facility admission: a 12-year epidemiological study in the United States. J Am Geriatr Soc (2004) 52:434–9.[CrossRef][ISI][Medline]
- Tomiak M, Bertholet J-M, Guimond E, Mustard CA. Factors associated with nursing-home entry for elders in Manitoba, Canada. J Gerontol A Biol Sci Med Sci (2000) 55A:M279–87.
[Abstract/Free Full Text] - Bajekal M. Health Survey for England 2000: Characteristics of Care Homes and their Residents (2002) London: The Stationery Office.
- Bowman C, Whistler J, Ellerby M. A national census of care home residents. Age Ageing (2004) 33:561–6.
[Abstract/Free Full Text] - Challis D, Mozley CG, Sutcliffe C, et al. Dependency in older people admitted to care homes. Age Ageing (2000) 29:255–60.
[Abstract/Free Full Text] - Hebert R, Dubois MF, Wolfson C, Chambers L, Cohen C. Factors associated with long-term institutionalization of older people with dementia: data from the Canadian Study of Health and Aging. J Gerontol A Biol Sci Med Sci (2001) 56:M693–9.
[Abstract/Free Full Text] - Wang JJ, Mitchell P, Smith W, Cumming RG, Leeder SR. Incidence of nursing home placement in a defined community. Med J Aust (2001) 174:271–5.[ISI][Medline]
- Soto ME, Andrieu S, Gillette-Guyonnet S, Cantet C, Nourhashemi F, Vellas B. Risk factors for functional decline and institutionalisation among community-dwelling older adults with mild to severe Alzheimer's disease: one year of follow-up. Age Ageing (2006) 35:308–10.
[Free Full Text] - Boaz RF, Muller CF. Predicting the risk of permanent nursing-home residence- the role of community help as indicated by family helpers and prior living arrangements. Health Serv Res (1994) 29:391–414.[ISI][Medline]
- Larsson K, Silverstein M. The effects of marital and parental status on informal support and service utilization: a study old older Swedes living alone. J Aging Stud (2005) 18:231–44.[CrossRef][ISI]
- Grundy EMD. Socio-demographic variations in rates of movement into institutions among elderly people in England and Wales: an analysis of linked census and mortality data 1971–1985. Popul Stud (1992) 46:65–84.[CrossRef]
- Grundy E, Glaser K. Trends in, and transitions to, institutional residence among older people in England and Wales, 1971 to 1991. J Epidemiol Community Health (1997) 51:531–40.[Abstract]
- Breeze E, Sloggett A, Fletcher A. Socioeconomic and demographic predictors of mortality and institutional residence among middle aged and older people: results from the Longitudinal Study. J Epidemiol Community Health (1999) 53:765–74.[Abstract]
- Hancock R, Arthur A, Jagger C, Matthews R. The effect of older people's economic resources on care home entry under the United Kingdom's long-term care financing system. J Gerontol B Psychol Sci Soc Sci (2002) 57:S285–93.
[Abstract/Free Full Text] - Crawford VLS, Beringer TRO, Stout RW. Comparison of residential and nursing home care before and after the 1991 community care policy. BMJ (1999) 318:366.
[Free Full Text] - Townsend P, Wedderburn D. Occasional Papers in Social Administration (1965) London: London School of Economics. The Aged in the Welfare State. No. 14.
- Wolf DA. Demography of Aging—Martin L, Preston SH, eds. (1994) Washington, DC: Academy Press. The elderly and their kin: patterns of availability and access.
- Freedman VA. Family structure and the risk of nursing home admission. J Gerontol B Soc Sci (1996) 51:S61–9.
- MRC CFAS. Socioeconomic status and the expectation of disability in old age: estimates for England. J Epidemiol Community Health (2000) 54:286–92.
[Abstract/Free Full Text] - Bebbington A, Darton R, Netten A. Care Homes for Older People, vol 2, Admissions, Needs and Outcomes, the 1995/96 National Longitudinal Survey of Publicly-Funded Admissions (2001) London: Personal Social Services Research Unit.
- Hirst M. Trends in informal care in Great Britain during the 1990s. Health Soc Care Community (2001) 9:348–57.[CrossRef][ISI][Medline]
- Blackwell L, Lynch K, Smith J, Goldblatt P. Longitudinal Study 1971–2001: Completeness of Census Linkage (2003) London: Office for National Statistics.
- Laing and Buisson Ltd. Care of Elderly People, UK Market Report (2005) 8th edition. London: Laing and Buisson.
- Grundy E, Tomassini C. Fertility history and health in later life: a record linkage study in England and Wales. Soc Sci Med (2005) 61:217–28.[CrossRef][ISI][Medline]
- Murphy M, Grundy E. Mothers with living children and children with living mothers: the role of fertility and mortality in the period 1911–2050. Popul Trends (2003) 112:36–45.[Medline]
- Bajekal M, Dix D, Whelloff L. Estimating residents and staff in communal establishments from the 2001 Census. Health Stat Q (2006) 31:42–50.[Medline]
- Pot AM, Deeg DJ, Knipscheer CP. Institutionalization of demented elderly: the role of caregiver characteristics. Int J Geriatr Psychiatry (2001) 16:273–80.[CrossRef][ISI][Medline]
- Comas-Herrera A, Wittenberg R, Costa-Fonti J, et al. Future long-term care expenditure in Germany, Spain, Italy and the United Kingdom. Ageing Soc (2006) 26:285–302.[CrossRef][ISI]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Nihtila and P. Martikainen Institutionalization of Older Adults After the Death of a Spouse Am J Public Health, July 1, 2008; 98(7): 1228 - 1234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Martikainen, E. Nihtila, and H. Moustgaard The Effects of Socioeconomic Status and Health on Transitions in Living Arrangements and Mortality: A Longitudinal Analysis of Elderly Finnish Men and Women From 1997 to 2002 J. Gerontol. B. Psychol. Sci. Soc. Sci., March 1, 2008; 63(2): S99 - S109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Nihtila and P. Martikainen Why older people living with a spouse are less likely to be institutionalized: The role of socioeconomic factors and health characteristics Scand J Public Health, January 1, 2008; 36(1): 35 - 43. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Bowman Tracking demographic footprints Age Ageing, July 1, 2007; 36(4): 360 - 361. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||