Health inequalities in older populations: the association between housing tenure and health

Despite suffering the highest levels of morbidity and mortality, there has been relatively little research into health inequalities in older populations, compared with those of working age.Reference Connolly, Rosato and O'Reilly⁵ This may be partly explained by the difficulty in measuring socio-economic status in older populations, where commonly used indicators of socio-economic status may no longer be relevant. Indicators based on occupation, for example, may not be easily applied to older people, the majority of whom are no longer attached to the formal labour market; while an indicator based on education, for example, may not be appropriate for the current cohort of older people, many of whom have relatively little formal education.

Many of the studies examining inequalities in health at older ages have used housing tenure as an indicator of socio-economic status.Reference Grundy and Holt^6–Reference Huisman, Kunst, Andersen, Bopp, Borgan, Borrell, Costa, Deboosere, Desplanques, Donkin, Gadeyne, Minder, Regidor, Spadea, Valkonen and Mackenbach⁸ Tenure has a number of advantages including being readily available, applying to almost all people and potentially reflecting wealth accumulated through the life course. In general, owner-occupiers have been found to have better health than renters. Connolly et al.,Reference Connolly, Rosato and O'Reilly⁵ for example, looking at over 191,000 people aged 65 and older in Northern Ireland, found that owner-occupiers were significantly less likely to report poor general health and to die in a 5-year follow-up period compared with public sector renters. However, the association between tenure and health was found to be stronger for the younger old (65–74 years) compared with the older old (75 years and older). Huisman et al.Reference Huisman, Kunst, Andersen, Bopp, Borgan, Borrell, Costa, Deboosere, Desplanques, Donkin, Gadeyne, Minder, Regidor, Spadea, Valkonen and Mackenbach⁸ examined inequalities in mortality among older people across 11 European populations, and found that housing tenure was strongly related to mortality among the middle aged and early old age groups. Similarly to Connolly et al.,Reference Connolly, Rosato and O'Reilly⁵ they noted an age-related attenuation in mortality differences by housing tenure, especially for those aged 80 and over. They hypothesized that this may be due to the exclusion of institutionalized populations, in that the older population, who still rent a house, are a selection of healthy people because older tenants move more easily to institutions when they are faced with problems living on their own.

Additionally, possibly reflecting the different perceptions and rates of owner-occupation across countries mentioned above, Huisman et al.Reference Huisman, Kunst, Andersen, Bopp, Borgan, Borrell, Costa, Deboosere, Desplanques, Donkin, Gadeyne, Minder, Regidor, Spadea, Valkonen and Mackenbach⁸ found that the strength of association between tenure and mortality risk differed across countries; for example, among men aged 60–69 years, the strongest effects were found in Finland and England/Wales, and the smallest effects in France and Turin.

Somewhat similarly, Dalstra et al.Reference Dalstra, Kunst and Mackenbach⁷ looked at the strength of various socio-economic indicators for predicting less than good health among people aged 60–79 years across 10 European countries. There were substantial health differences among the study population according to education and income in each country, while health differences according to tenure were generally somewhat smaller. However, in Great Britain and the Netherlands housing tenure demonstrated large health differences even after adjustment for education and income.

A potential limitation of housing tenure as an indicator of socio-economic status in assessing health inequalities in older populations is that the majority of the population are included in the most advantaged group – owner-occupiers.Reference Grundy and Holt⁶ Overcoming this limitation, Connolly et al.Reference Connolly, Rosato and O'Reilly⁵ examined whether house value in conjunction with housing tenure was a predictor of ill-health in older populations. Owner-occupiers were divided into five groups based on the value of their home. The expected association between tenure and health was recorded; however, a step-wise gradient was also observed within the owner-occupier group, with those living in the highest valued properties reporting better health and lower mortality than those in lower valued properties. Similarly, Costa-FontReference Costa-Font⁹ looked at housing assets and health in older age using a representative sample of the Spanish population aged 55 and over, and found that while income did not exert appreciable effects on health and disability in old age, housing equity did. The study found that doubling housing equity resulted in an increase of about 3% in self-perceived health and a 4% reduction in disability.

Housing tenure and health: is tenure acting as a proxy for other indicators of socio-economic status?

Despite being widely used in the health inequalities literature as a proxy for socio-economic status, relatively little attention has been given to assessing what housing tenure is actually measuring. The association may be explained by tenure acting as a proxy for other indicators of socio-economic status (such as income and wealth accumulated across the life course that facilitate home ownership), which directly influence health, or alternatively other characteristics associated with tenure, including housing quality or location, may explain the association.

Assessing the first of these competing explanations, a number of studies have examined whether housing tenure continues to be associated with health in later life after adjustment is made for other indicators of socio-economic status of the population under consideration. A minority of studies have found that after controlling for socio-economic circumstances such as social class, the relationship between tenure and health disappears for older people. Arber and Ginn,Reference Arber and Ginn¹⁰ for example, found that the relationship between tenure and disability disappeared for older men (though not older women) once adjustment was made for social class. However, a majority of studies have found that the association continues to hold even after controlling for various other indicators of socio-economic status.Reference Connolly, Rosato and O'Reilly^5, Reference Dalstra, Kunst and Mackenbach^7, Reference Macintyre, Ellaway, Der, Ford and Hunt¹¹ Macintyre et al.,Reference Macintyre, Ellaway, Der, Ford and Hunt¹¹ for example, using the West of Scotland Twenty-07 study noted that a range of health measures were significantly associated with housing tenure even after controlling for age, sex and income, suggesting therefore that tenure was not simply acting as a proxy for income. Dalstra et al.Reference Dalstra, Kunst and Mackenbach⁷ found that in Great Britain housing tenure groups showed large health differences even after controlling for education and income among a cohort of people aged 60–79 years. In Northern Ireland, Connolly et al.Reference Connolly, Rosato and O'Reilly⁵ found that tenure was still significantly associated with morbidity and mortality for those aged 65–74 years after adjustment for social class and educational attainment of the study population.

Housing tenure and health: do other characteristics associated with tenure explain the relationship?

A number of alternative explanations for the association between housing tenure and health have been postulated: these include the location of owner-occupier and rental properties, the psychological impact of home ownership and the quality of the housing stock.

Much has been written about the potential effect of poor quality housing on health.Reference Howden-Chapman^2, Reference Evans, Wells and Moch^12, Reference Shaw¹³ A number of features of housing can impact on health, including over-crowding, temperature, dampness, indoor air pollution, noise, lead and fires.Reference Mackenbach and Howden-Chapman¹⁴ Aylin et al.,Reference Aylin, Morris, Wakefield, Grossinho, Jarrup and Elliott¹⁵ for example, found that lack of central heating was associated with high excess winter mortality. Howden-Chapman et al.Reference Howden-Chapman, Singal and Cranes¹⁶ describe a number of studies that have found links between damp housing, the presence of mould and high rates of respiratory illness and asthma, especially among those who spend more time inside, including older people. Older people, in particular, are likely to be more vulnerable to the effects of poor housing quality given the large proportion of time they spend within the home – with one study in Wales noting that older people spent an average of 83% of the day at home.Reference Windle, Burholt and Edwards¹⁷

However, while the negative impact of poor quality housing on health is well established, what is less clear is whether rental properties are of poorer quality than owner-occupier properties. In England, the House Condition Survey found that socially rented homes were less likely to be regarded as ‘non-decent’ than privately owned homes.¹⁸ However, privately rented accommodation was most likely to be ‘non-decent’. Conversely, Windle et al.Reference Windle, Burholt and Edwards¹⁷ looked at housing-related difficulties, housing tenure and variations in health in older people in Wales, and found that those in rented accommodation tended to report more problems than those who owned their homes. In particular, those in publically and privately rented accommodation were significantly more likely than owner-occupiers to report problems with heating, damp/condensation, draughts and security. Similarly, Macintyre et al.Reference Macintyre, Ellaway, Hiscock, Kearns, Der and Mckay¹⁹ found that, with the exception of smoke alarms, those in social rented dwellings were more likely to be exposed to health-damaging features of the environment such as damp, noise, crime and vandalism, and less likely to have access to health-promoting features such as gardens and local amenities, suggesting that not just the house but also the wider environment may be important in determining health status.

Macintyre and colleagues have explored the meaning of housing tenure in Scotland, and in particular its association with health. In one study they found a significant association between housing tenure and a range of health measures; however, controlling for various housing features (including overcrowding, access to a garden, and housing fixtures) significantly reduced the explanatory power of housing tenure.Reference Macintyre, Ellaway, Hiscock, Kearns, Der and Mckay¹⁹ Howden-Chapman et al.Reference Howden-Chapman, Chandola, Stafford and Marmot³ found that after controlling for a range of variables including indicators of socio-economic status and housing quality, housing tenure was no longer a significant explanation of mental health in retirement among the Whitehall II cohort. This evidence suggests that housing quality may be on the causal pathway between housing tenure and health outcomes.

An alternative, but related, explanation for the association between housing tenure and health relates to the location of properties for different tenure groups. It may be that the health advantage of owner-occupiers relative to renters is explained, or at least partly explained, by the location of owner-occupier properties in more affluent areas and the concentration of rental properties (especially socially rented accommodation) in more deprived areas. There are a number of potential ways in which area of residence may influence health status including norms of behaviours, access to services and the potential stress associated with living in deprived areas that often experience higher levels of degradation and crime.

While the literature on the potential effects of area of residence on health tend to show modest effects of area characteristics compared with individual characteristics,Reference Pickett and Pearl²⁰ it is possible that area effects are important for some particular causes of ill-health and for particular groups of people. Ellaway and MacintyreReference Ellaway and Macintyre²¹ found, for example, that owner-occupiers had significantly better assessments of their areas – they reported more local amenities, more friendly neighbours, more satisfaction with their areas, fewer local incivilities or problems and lower fear of crime. In addition, they found that tenure was no longer associated with a number of health measures once housing stressors and type and neighbourhood conditions were considered simultaneously. They suggested that rather than just acting as a marker of material well-being, housing tenure may be providing some information on exposures to health-promoting or health-damaging features of the dwelling and its immediate environment.

Home-ownership too may confer psychological advantages to home-owners. Psychologically owning a home rather than renting seems to confer greater autonomy and social status.Reference Kearns²² Rental housing is more insecure. Leases do not give the same security to tenants as a house title gives to an owner. Stability may reduce stress and anxiety.Reference Costa-Font⁹ However, in times of economic recession and falling house prices, home-owners, who bought in a boom, may be left with negative equity in their houses, making home-ownership potentially less secure than rental housing.Reference Howden-Chapman, Chandola, Stafford and Marmot³ This is, however, likely to be less relevant for the current cohort of older people, the majority of whom are mortgage free. Perhaps a more relevant consideration of the potential negative impact of home ownership for older people is the issue of the repair and upkeep of an owned property at a time of generally declining income. Bull and PooleReference Bull and Poole²³ discussed the issue of house rich, income poor older people, who have valuable assets in their homes, but living standards constrained by low incomes. Despite this, home-owners consistently report better health than renters.

Housing tenure and the risk of admission to long-stay residential care

The preference of most older people is to remain in their own home as they age,Reference Windle, Burholt and Edwards^17, Reference Gitlin²⁴ and the concept of ageing in place has become increasingly important in relation to health and social care policy for older people. The idea behind ageing in place is that helping older people to remain living at home contributes to well-being and independence, as well as healthy ageing.Reference Sixsmith and Sixsmith²⁵

Approximately 4% of people aged 65 and older are in receipt of long-term residential care across OECD (Organisation for Economic Co-operation and Development) counties (http://www.oecd.org/health/healthataglance). While for many increased disability and declining health (in particular the progression of dementia-related symptoms) largely explain an admission to long-stay residential care, a range of demographic and socio-economic characteristics (including marital status and living arrangements) have also been found to be important determinants of long-stay residential care admission risk. A number of studies have shown housing tenure to be associated with long-stay residential care admission risk, with owner-occupiers less likely to be admitted, a finding that holds even after controlling for the health advantage of owner-occupiers.Reference Grundy and Jitlal^26–Reference Hancock, Arthur, Jagger and Matthews²⁸ Grundy and Jitlal,Reference Grundy and Jitlal²⁶ for example, found that those in socially rented properties were 36% more likely to enter a care home than owner-occupiers after adjusting for a range of other variables including health status and living arrangements, while in Australia, Wang et al.Reference Wang, Mitchell, Smith, Cumming and Leeder²⁷ noted that those who owned their homes had a decreased likelihood of nursing home placement in their 6-year follow-up study of a population in Sydney.

Grundy and JitlalReference Grundy and Jitlal²⁶ suggest that the higher risk of institutional residence among social renters may be explained by the fact that owner-occupiers (and their children) are more likely than tenants to have the resources to pay for additional care within the home. Testing this assertion, McCann et al.Reference McCann, Grundy and O'Reilly²⁹ looked at the probability of admission to residential care by housing tenure and house value on the basis that if the association between housing tenure and risk of long-stay residential care admission reflected differences in access to financial resources, then there would be a graded relationship between housing value and admission, with owner-occupiers with higher valued properties less likely to be admitted than those with lower valued properties. They found, however, that those with higher value homes (and therefore probably more likely to afford community-based care) were as likely to enter a long-stay residential care facility as those with the lowest valued homes, and therefore discounted increased wealth as the reason for reduced admission among owner-occupiers. Instead, they argued that the current method used to assess eligibility for financial assistance with care home fees within the UK as being an important explanatory factor. In some cases, means testing may result in older home owners selling their home if admitted to a residential or nursing home. They argue that families may choose to significantly increase the amount of informal care they provide rather than see the asset sold. While such informal care giving may allow an older person to remain within their own home, it also raises potential concerns about the burden of care on informal caregivers and the potential negative effects on caregivers and care recipients.

The bulk of care received by older people is provided within their own home by family members and friends. Generally this care is provided without payment, and caregivers often remain unknown and unsupported by the authorities until a crisis occurs (often involving the declining health of the caregiver or care recipient). A number of studies have highlighted how effective supports for caregivers can lead to a delay in nursing home placement,Reference Mittleman, Haley, Clay and Roth^30–Reference Broadaty, Green and Koscher³² yet in many countries informal caregivers remain unacknowledged and unsupported.

In addition to informal care provided by family and friends, timely and appropriate community-based care has the potential to facilitate older people to remain within their own home. Whereas in the past, older people with complex physical and psychosocial health needs were generally cared for in nursing and residential homes, a change in policy in many countries has facilitated such older people to remain within the home.Reference Ryan, McCann and McKenna³³ Government policy, which recognizes and supports informal caregivers and provides adequate community-based care, will help ensure that the greatest possible number of older people remain within their own home and that the risk of admission is not determined by demographic and socio-economic characteristics including housing tenure.

Increasingly, the potential use of assistive technologies, and to a lesser extent equity release schemes, as a means of enabling older people to remain within their own home, have been acknowledged. However, the extent to which both can be utilized is likely to be influenced by housing tenure and may cause further discrepancies in long-stay residential care use by tenure grouping. Appropriately selected assistive technologies have been found to potentially lead to important improvements in independence and potential savings in formal care services;Reference Tinker and Lansley³⁴ however, concerns about who is responsible for the cost and maintenance of such technologies will probably result in differences in the use of such technologies for different tenure groups. Equity release – which allows the continual use of the home while also receiving a lump sum or regular income, which is paid back at a later date – is only relevant for those that own their own home, and not those renting a property. More work is required to ensure that whatever policies are implemented to facilitate people to remain within their own home as they age, they do not discriminate between owner-occupiers and renters.

Conflict of interest

The author declared no conflict of interest.