Health inequalities in the older population: the role of personal capital, social resources and socio-economic circumstances

https://doi.org/10.1016/S0277-9536(02)00093-XGet rights and content

Abstract

Older people now constitute the majority of those with health problems in developed countries so an understanding of health variations in later life is increasingly important. In this paper, we use data from three rounds of the Health Survey for England, a large nationally representative sample, to analyse variations in the health of adults aged 65–84 by indicators of attributes acquired in childhood and young adulthood, termed personal capital; and by current social resources and current socio-economic circumstances, while controlling for smoking behaviour and age. We used six indicators of health status in the analysis, four based on self-reports and two based on nurse collected data, which we hypothesised would identify different dimensions of health. Results showed that socio-economic indicators, particularly receipt of income support (a marker of poverty) were most consistently associated with raised odds of poor health outcomes. Associations between marital status and health were in some cases not in the expected direction. This may reflect bias arising from exclusion of the institutional population (although among those under 85 the proportion in institutions is very low) but merits further investigation, especially as the marital status composition of the older population is changing.

Analysis of deviance showed that social resources (marital status and social support) had the greatest effect on the indicator of psychological health (GHQ) and also contributed significantly to variation in self-rated health, but among women not to variation in taking three or more medicines and among men not to self-reported long-standing illnesses. Smoking, in contrast, was much more strongly associated with these indicators than with self-rated health. These results are consistent with the view that self-rated health may provide a holistic indicator of health in the sense of well-being, whereas measures such as taking prescribed medications may be more indicative of specific morbidities. The results emphasise again the need to consider both socio-economic and socio-psychological influences on later life health.

Introduction

The increased representation of older people in the population of Britain and other developed countries, coupled with epidemiological changes which mean that older people constitute a large majority of those in poor health, has led to a growing concern with identifying determinants of health, and inequalities in health, in later life (Acheson, 1998; Department of Health, 1999). Variations in the mortality and morbidity of the elderly population by indicators of socio-economic status based on past occupation, education, housing tenure, income and wealth have been reported in a wide range of studies with those in the most socio-economically disadvantaged groups also suffering the greatest health disadvantage (Fox, Goldblatt, & Jones, 1985; Arber & Ginn, 1993; Menchik, 1993; Martelin, 1994; Elo & Preston, 1996; Marmot & Shipley, 1996; Rogers, 1996; Sundquist & Johansson, 1997; Grundy & Glaser, 1999; Grundy & Holt, 2000). In general, associations between health and socio-economic characteristics seem less pronounced than in young or middle-aged groups, but because morbidity and mortality is concentrated in elderly age groups, attributable differences are greater.

A substantial and to some extent parallel literature, has shown that socio-demographic and social-psychological characteristics such as marital status, household composition and social support are also associated with differentials in health and mortality in older age groups. In general, married people have the best health, followed by the never married and then the formerly married. Hypothesised reasons for these associations include both selection factors—good health increases the chances of marrying (including remarrying) and remaining married for longer—and the protective effects of care and support (Verbrugge, 1979; Hu & Goldman, 1990; Umberson, 1992; Hahn, 1993; Gliksman, Lazarus, Wilson, & Leeder, 1975; Waite, 1995; Cheung, 2000). Although these latter effects might be supposed to be particularly important in older age groups, some studies suggest a weaker, or even reversed, relationship between health and marriage with increasing age. Goldman, Korenman and Weinstein (1995), for example, found that never-married older women had better health outcomes than their married counterparts, a result they attributed to more extensive social ties built up over the lifetime as an alternative to marriage. However, their analysis was based on a sample that excluded the institutional population, which, as the unmarried are over-represented in institutions, may have biased results. Analyses of British data on differentials in limiting long standing illness which included the whole population have shown a continuing, although weaker, advantage for the married, even in the oldest age groups (Murphy, Glaser, & Grundy, 1997).

Associations between other indicators of social connectedness, including social network size and characteristics and church or club membership, and both mortality and other health indicators have also been reported (Blazer, 1982; Seeman, Kaplan, Knudsen, Cohen, & Guralnik, 1987; House & Landis, 1988; Sugisawa, Liang, & Liu (1996), Grundy, Bowling, & Farquhar (1996); Sugisawa, Liang, & Liu, 1994). However, in some studies effects appear much weaker, or non-existent, in older than in younger groups, possibly because of inconsistencies in the measures of health and social support used and the relatively small numbers of older people included in some analyses (Orth-Gomer & Unden, 1987; O’Reilly, 1988; Bowling & Grundy, 1998).

A wide range of studies thus show variations in the health of elderly people according to differences in socio-economic circumstances and socio-demographic or socio-psychological characteristics, although in both cases variations seem less marked than in younger age groups. These domains, although often considered separately (Preston & Taubman, 1994), are clearly intertwined in several ways. It is known, for example, that there are social class differences in patterns of social interaction and in marriage and divorce patterns (House & Landis, 1988; Ben-Shlomo, Smith, Shipley, & Marmot, 1993; Schoeni, 1995; Stansfield, 1999). Among elderly women, in particular, being married is associated with indicators of economic advantage such as income and housing tenure (Hahn, 1993; Murphy et al., 1997).

Apart from these interrelationships, there may be factors which influence both socio-economic and socio-demographic circumstances at older ages and exert an influence on health, such as legacies from earlier life. Attributes present or acquired in childhood may have an important and lasting effect on life chances, health behaviours and coping mechanisms, and so on health, throughout the life course, as well as exerting a strong influence on adult socio-economic and socio-demographic experiences and later life circumstances (Barker, 1992; Bartley, Blane, & Montgomery, 1999; Bosma, van de Mheen, & Mackenbach, 1999; Brunner, Shipley, Blane, Smith, & Marmot, 1999; Davey Smith, Hart, Blane, Gillis, & Hawthorne, 1997; Wadsworth, 1997). Education, for example, may increase feelings of personal control and promote better health behaviours as well as providing a route to higher status well paid occupations and so to accumulated wealth and better pensions in later life (Bosma, Schrijvers, & Mackenbach, 1999). Height, often used as an indicator of childhood circumstances and development, is associated with both adult social class and with the marriage chances of men, as well as with health (Blane et al., 1996; Kuh & Wadsworth, 1989; Murray, 2000; Phillips et al., 2001).

As well as common, or overlapping, pathways to particular health, socio-economic and socio-demographic statuses in adult life, there may also be common mechanisms whereby socio-economic and socio-demographic or psychological characteristics influence health. Most obviously, social class, education and housing tenure are all strongly associated with smoking (Blaxter, 1990; Bennett, Dodd, Flatley, Freeth, & Bolling, 1995); so too is marital status and, among elderly men, household type (Umberson, 1992). Smoking is clearly not the only explanation for variations in health as persistent socio-economic inequalities in mortality and morbidity in the whole adult population are found even when smoking is allowed for (Blaxter, 1990; Suadicani, Hein, & Gyntelberg, 1994). Similarly, studies have reported associations between social and blood pressure after adjustment for smoking, exercise, alcohol use and body mass index (Hanson, Isacsson, Janzon, Lindell, & Rastam, 1988).

Synergistic or offsetting interactions between economic and socio-psychological domains may also be important (Roberts, Dunkle, & Haug, 1994). Material advantages not only enable the purchase of better food and housing, but also the purchase of services that may preserve feelings of control and autonomy and enable social participation, all factors hypothesised to have important influences on health behaviours and physiological functions (Seeman, 2000). Supportive networks may buffer the effects of stress, including socio-economic stress, and enhance the operation of immunological functions, as well as being a potential source of practical help and advice and so an alternative to purchased assistance (Berkman, Leo-Summers, & Horowitz, 1992; Uchino, Cacioppo, & Kiecolt-Glaser, 1996). Conversely, the combination of poor socio-economic and poor socio-psychological circumstances may be particularly harmful (Ben-Shlomo et al., 1993; Martikainen & Valkonen, 1998).

For all these reasons it would seem essential to consider both socio-economic and socio-demographic factors in analyses of health differentials in later life and preferably also indicators of childhood legacies and health behaviours. This type of analysis is, however, relatively unusual, partly due to data limitations (Preston & Taubman, 1994). Here we use data from a large nationally representative sample of the older population of England to investigate the effect of attributes acquired in childhood and young adulthood, which we term personal capital, current social resources and current socio-economic circumstances on health variation in later life, while also controlling for smoking behaviour, one of the mechanisms whereby the other domains may influence health. The personal capital variables used were height and educational qualifications. Educational qualifications may be gained in adulthood, but in the cohorts with which we are concerned this was relatively unusual and highest qualification obtained is a good indicator of educational experiences and outcomes in childhood and early adulthood, themselves strongly influenced by social class of origin (White, Blane, Morris, & Mourouga, 1999).

As already noted there is an extensive literature on links between marital status and health, although still some uncertainty as to whether associations are attenuated, or even reversed in older age groups. We used this, together with a variable measuring perceived social support, as indicators of social resources.

Housing tenure has been shown in numerous studies to be associated with other indicators of socio-economic status, such as income and social class, and to be strongly associated with differentials in health (Fox et al., 1985). It has the advantage, in comparison with occupationally based measures, of relating to current material circumstances and applying equally to men and women (Arber & Ginn, 1993; Grundy & Holt, 2001). We used this as an indicator of socio-economic circumstances together with receipt of income support, a means tested benefit paid to those on low incomes.

Our aims were to quantify the extent of inequalities in the health of older people, see which of the broad domains considered was most strongly associated with health in later life, and investigate possible interactive effects of socio-economic and socio-demographic disadvantage on health in later life. Health, whether conceptualised negatively as the absence of disease, positively as a complete state of well-being, or normatively as the average, is a multidimensional concept that is difficult to measure (Ware, Allyson, & Robert, 1980). It is well recognised, for example, that self-reported and observational measures produce different results, although which gives a better indicator of ‘true’ health status remains a matter of debate. Several analysts have concluded that self-reported and directly measured indicators of physical function represent different dimensions of health (Guralnik, Branch, Cummings, & Curb, 1989; Merrill, Seeman, Kasl, & Berkman, 1997). In this study we use a range of six indicators of health status, four based on self-reports and two based on nurse collected data, which we hypothesised would identify different dimensions of health. We aimed to investigate both associations between the explanatory domains and these outcome indicators and whether the explanatory variables had consistent or varying effects on different indicators of health status.

Section snippets

Data and methods

The data we use come from the 1993–95 rounds of the Health Survey for England (HSfE) (Bennett et al., 1995). We chose this study because it is nationally representative, includes a range of indicators of health status, together with information on health-related behaviours and socio-demographic and socio-economic characteristics, and has a large enough sample size, when, as here, 3 years data are combined to allow detailed analysis. Data in the HSfE are collected through a questionnaire, nearly

Results

Fig. 1 shows the age-adjusted prevalence of self-reported bad or very bad health according to the pairs of variables representing the three domains of interest. Among both men and women a higher proportion of those with no educational qualifications rated their health bad or very bad and among men and, to a lesser extent women, there was a gradient in prevalence with height (Fig. 1, Panel a). Short men and women with no qualifications were twice as likely to report poor health as their tall

Discussion

In this paper we have used data from a large nationally representative sample to analyse differentials in the health of older adults according to variables selected to represent personal capital, social resources and socio-economic resources. Unlike many analyses which have considered only one outcome, we analysed variations in six indicators of health status including both self-reported and nurse-measured variables. We also controlled for smoking status as well as for age. Our initial analyses

Acknowledgements

This research was supported by the Economic and Social Research council, grant reference number L128 251040, as part of its Health Variations (Phase II) Programme. Gemma Holt provided valuable assistance with data management and preparation in the early stages of the work reported here.

References (66)

  • D Acheson

    Independent inquiry into inequalities in health report

    (1998)
  • D.J.P Barker

    Fetal and infant origins of adult diseases

    (1992)
  • M Bartley et al.

    Socio-economic determinants of health: Health and the life courseWhy safety nets matter

    British Medical Journal

    (1997)
  • Y Ben-Shlomo et al.

    Magnitude and cause of mortality differences between married and unmarried men

    Journal of Epidemiology and Community Health

    (1993)
  • N Bennett et al.

    Health survey for England 1993

    (1995)
  • L.F Berkman et al.

    Emotional support and survival after myocardial infarction

    Annals of Internal Medicine

    (1992)
  • D Blane

    The life course, the social gradient, and health

  • D Blane et al.

    Association of cardiovascular disease risk factors with socioeconomic position during childhood and during adulthood

    British Medical Journal

    (1996)
  • M Blaxter

    Health and lifestyles

    (1990)
  • D.G Blazer

    Social support and mortality in an elderly community population

    American Journal of Epidemiology

    (1982)
  • H Bosma et al.

    Social class in childhood and general health in adulthoodQuestionnaire study of contribution of psychological attributes

    British Medical Journal

    (1999)
  • H Bosma et al.

    Socio-economic inequalities in mortality and importance of perceived controlCohort study

    British Medical Journal

    (1999)
  • A Bowling et al.

    The association between social networks and mortality in later life

    Reviews in Clinical Gerontology

    (1998)
  • E Brunner et al.

    When does cardiovascular risk start? Past and present socio economic circumstances and risk factors in adulthood

    Journal of Epidemiology and Community Health

    (1999)
  • Y.B Cheung

    Marital status and mortality in British womenA longitudinal study

    International Journal of Epidemiology

    (2000)
  • G Davey Smith et al.

    Life-course approaches to socio-economic differentials in cause-specific adult mortality

  • G Davey Smith et al.

    Lifetime socioeconomic position and mortalityProspective observational study

    British Medical Journal

    (1997)
  • Department of Health, 1999. NHS R&D Strategic Review, Ageing and age associated diseases and disability. Department of...
  • A.J Fox et al.

    Social class mortality differentialsArtefact, selection or life circumstances?

    Journal of Epidemiology and Community Health

    (1985)
  • D.P Goldberg et al.

    A user's guide to the general health questionnaire

    (1988)
  • E Grundy

    Population reviewThe population aged 60 and over

    Population Trends

    (1996)
  • E Grundy et al.

    Trends in, and transitions to, institutional residence among older people in England and Wales, 1971 to 1991

    Journal of Epidemiology and Community Health

    (1997)
  • E Grundy et al.

    Differentials in the onset and progression of disability in early old age in Great Britain

    Age and Ageing

    (1999)
  • Cited by (220)

    • The racial and cultural ecology of home and community-based services for diverse older adults

      2022, Journal of Aging Studies
      Citation Excerpt :

      Moreover, while foreign-born Latinos have the highest life expectancy, this dubbed the “Latino Paradox,” increasing acculturation among Latino immigrants is associated with a higher risk of mortality (Carter-Pokras et al., 2008; Garcia, Angel, Angel, Chiu, & Melvin, 2015; Markides & Coreil, 1986). Social determinants of health are regarded as substantial considerations for practitioners addressing the care of older adults who are, or are close to becoming, homebound (Grundy & Sloggett, 2003; White, Philogene, Fine, & Sinha, 2009). In this regard, research has consistently demonstrated the importance of HCBS in facilitating holistic, person-centered care for older adults and allowing them to age in place.

    • Financial hardship and suicide ideation: Age and gender difference in a Korean panel study

      2021, Journal of Affective Disorders
      Citation Excerpt :

      Though the explanation is tentative and needs future studies, we hypothesize that direct measurements of economic situation such as financial hardship are more closely linked to actual inability due to economic reasons in the elderly population compared to conventional SES. This provides further confirmation of a previous evidence that has shown advantages of hardship indicators such as car ownership and housing tenure over conventional SES indicators in measuring health inequalities in the elderly population (Grundy and Sloggett, 2003; Lorant et al., 2005). Alternatively, this finding can be viewed within the context of specific countries like Korea where progressive income redistribution including public pension is not favourable to the post-working population.

    View all citing articles on Scopus
    View full text