Elsevier

Archives of Gerontology and Geriatrics

Volume 51, Issue 3, November–December 2010, Pages e37-e40
Archives of Gerontology and Geriatrics

Review
Socioeconomic gradients in mortality in the oldest old: A review

https://doi.org/10.1016/j.archger.2009.12.009Get rights and content

Abstract

This review aims at identifying gaps in knowledge on socioeconomic gradients in mortality in the oldest old. The authors review literature on oldest old population with a focus on unanswered questions: do socioeconomic status (SES) gradients in mortality persist after 80; does the magnitude of the gradient change as compared with younger populations; which socioeconomic/socio-demographic determinants should be used in this population with specific characteristics (e.g., with respect to sex ratio and household type)? Results are often inconsistent while conclusions drawn by selected studies are generally limited by the difficulty of disentangling the effects of age and cohort, and of generalizing results observed in preponderantly small, selected samples (which typically exclude institutionalized persons). Future research should explore the effects of socio-demographic indicators other than education and social class (e.g., marital status, loss of the partner) and adequately differentiate the social position of oldest old women. The authors recommend that research applies a life-course perspective combined with an interdisciplinary perspective to improve our understanding of the SES gradients in later life. Research is needed to elucidate which causal pathways depending on SES in younger age impact on mortality in higher ages up to oldest old.

Introduction

Mortality is consistently higher in persons with low SES than in those with high SES (e.g., Marmot, 1984, Kawachi et al., 1991, Hayward et al., 1997, Mackenbach et al., 1997). In various countries, SES inequalities in mortality increased during the last decades (Crimmins and Saito, 2001, Doblhammer et al., 2005, Menvielle et al., 2007, Jaffe et al., 2008, Meara et al., 2008, Ramsay et al., 2008). SES mortality gradients before the age 80 are fairly studied. However, many questions remain to be resolved in the oldest old: do SES gradients in mortality persist after age 80; if yes, does the magnitude of the gradient change as compared with younger populations; which socioeconomic determinants should be used in this population with specific characteristics (e.g., with respect to sex ratio and household type)? Literature brings sometimes inconsistent responses, while conclusions drawn by selected studies is often limited by the difficulty to disentangle the effects of age and cohort, and by the difficulty to generalize results observed in most often small, selected samples (which typically exclude institutionalized persons).

Section snippets

The oldest old as a new field of research

In this review, the oldest old are defined as being 80-year old and over, an age from which frail individuals become a majority in developed countries (Lalive d’Epinay et al., 2008). Mortality in oldest old populations has declined in low-mortality countries since the 1950s (Robine and Paccaud, 2005), however with important international variations (Janssen et al., 2004). This has contributed to the increase in the absolute and relative number of the oldest old in these countries, with an

Do SES gradients in mortality persist in the oldest old?

In general, studies show that SES (as measured by education level or social status) gradients in mortality persist in the oldest old (e.g., Manor et al., 2000, Crimmins and Saito, 2001, Marang-van de Mheen et al., 2001, Bopp and Minder, 2003), albeit smaller on a relative scale when compared to younger populations, especially among men (e.g., Huisman et al., 2004, Doblhammer et al., 2005). Selective survival favoring those enjoying better genetic, social and/or environmental resources is often

Which socioeconomic determinants should be used in the oldest old population?

The oldest old population has specific characteristics, e.g., with respect to sex ratio (i.e., larger proportion of women) and household type (i.e., larger proportions of both single person households and institutionalized persons). This suggests two recommendations for future research. First, future research on SES gradients in oldest old populations should include the institutionalized persons. Apart the fact they represent a significant proportion of the population, institutionalized persons

Life-course perspective to better study SES gradients in mortality

Future research should apply a life-course perspective to better assess the influence of SES on mortality in the oldest old. The life-course perspective has been efficiently used in younger ages. It has been shown that mortality risk for all causes was higher among those who experienced poorer SES circumstances during childhood (Galobardes et al., 2008). Trajectories in SES circumstances from childhood to adulthood are even more informative than SES circumstances at a single stage of life for

From pathways to prevention

Additional research is needed to elucidate which causal pathways operating between SES and mortality in younger populations maintain their impact up to oldest old. For instance, does higher exposure to chronic stressors during life course (e.g., unsafe environment, unfavorable work conditions or job insecurity) among lower SES groups explain their over-mortality up to oldest old age? Do SES gradients in mortality in oldest old populations reflect the lack of resources among lower SES groups

Conflict of interest statement

None.

References (61)

  • G. Menvielle et al.

    Évolution temporelle des inégalités sociales de mortalité en France entre 1968 et 1996. Étude en fonction du niveau d’études par cause de décès

    Rev. Epidemiol. Santé Publique

    (2007)
  • C.A. Mustard et al.

    Age-specific education and income gradients in morbidity and mortality in a Canadian province

    Soc. Sci. Med.

    (1997)
  • M. Noale et al.

    Predictors of mortality: an international comparison of socio-demographic and health characteristics from six longitudinal studies on aging: the CLESA project

    Exp. Gerontol.

    (2005)
  • F. Armi et al.

    Retaining or recovering robustness in very old age: what is the role of social relationships?

  • M. Beckett

    Converging health inequalities in later life—an artefact of mortality selection?

    J. Health Soc. Behav.

    (2000)
  • M. Ben-Ezra et al.

    Predictors of mortality in the old–old in Israel: the Cross-sectional and Longitudinal Aging Study

    J. Am. Geriatr. Soc.

    (2006)
  • J.-F. Bickel et al.

    Remaining active: a boost for well-being?

  • M. Bopp et al.

    Mortality by education in German speaking Switzerland, 1990–1997: results from the Swiss National Cohort

    Int. J. Epidemiol.

    (2003)
  • A. Bowling et al.

    Death after widow(er)hood: an analysis of mortality rates up to 13 years after bereavement

    Omega: J. Death Dying

    (1995)
  • E. Breeze et al.

    Socioeconomic and demographic predictors of mortality and institutional residence among middle aged and older people: results from the longitudinal study

    J. Epidemiol. Commun. Health

    (1999)
  • J. Cornoni-Huntley et al.

    Established populations for epidemiologic studies of the elderly: study design and methodology

    Aging (Milano)

    (1993)
  • E.M. Crimmins

    Trends in the health of the elderly

    Annu. Rev. Public Health

    (2004)
  • D. Dannefer

    Cumulative advantage/disadvantage and the life course: cross-fertilizing age and social science theory

    J. Gerontol. B: Psychol. Sci. Soc. Sci.

    (2003)
  • M.A. Davis et al.

    Living arrangements, changes in living arrangements, and survival among community dwelling older adults

    Am. J. Public Health

    (1997)
  • G. Doblhammer et al.

    Trends in educational and occupational differentials in all-cause mortality in Austria between 1981/82 and 1991/92

    Wien. Klin. Wochenschr.

    (2005)
  • A.S. Dontas et al.

    Survival in the oldest old: death risk factors in old and very old subjects

    J. Aging Health

    (1996)
  • A. Erlangsen et al.

    Loss of partner and suicide risks among oldest old: a population-based register study

    Age Ageing

    (2004)
  • K.F. Ferraro et al.

    Double jeopardy, aging as leveler, or persistent health inequality? A longitudinal analysis of white and black Americans

    J. Gerontol. B: Psychol. Sci. Soc. Sci.

    (1996)
  • F. Formiga et al.

    Predictors of two-year mortality in nonagenarians with severe functional impairment at baseline: the NonaSantfeliu study

    Age Ageing

    (2008)
  • B. Galobardes et al.

    Is the association between childhood socioeconomic circumstances and cause-specific mortality established? Update of a systematic review

    J. Epidemiol. Commun. Health

    (2008)

    • Cited by (18)

    • Contrasting temporal trends in lung cancer incidence by socioeconomic status among women in New South Wales, Australia, 1985–2009

      2017, Lung Cancer
      Citation Excerpt :

      Cases diagnosed before the age of 25 were excluded, as diagnoses of lung cancer are rare in younger people. Cases diagnosed after age 69 years were also excluded from the primary analysis because socioeconomic inequality tends to diminish in relative terms for older age groups [13–15], and social gradients in health are generally less consistent for the older population than those for population groups of working age [13]. To define the upper bound of the age range for the primary analysis in this study we tried three alternative age thresholds, 65, 70, and 75 years, and found that the resulting patterns of socioeconomic variation were similar for the three age thresholds.

    • Housing conditions and mortality in older patients hospitalized for heart failure

      2011, American Heart Journal
      Citation Excerpt :

      Furthermore, perceived income adequacy is personal and subjective and, thus, is free of obvious gender biases in a country where most older women have never had a paid work. Following recent recommendations for studies in the oldest elderly persons,42 we also considered other social indicators, such as social network, which complements the traditional indicators of education and income. The finding that housing conditions are associated with mortality independently of socioeconomic position is important because it is generally easier to intervene on specific housing deficiencies than to act directly on educational level, income, or social network.

    • Educational Differences in Life Expectancies With and Without Pain

      2023, Journals of Gerontology - Series B Psychological Sciences and Social Sciences

    • Self-assessment of health: how socioeconomic, functional, and emotional dimensions influence self-rated health among Italian nonagenarians

      2023, Quality and Quantity

    • Latent class trajectories of socioeconomic position over four time points and mortality: the Uppsala Birth Cohort Study

      2022, European Journal of Public Health

    • Association Between Dynapenic Abdominal Obesity and Fall Risk in Older Adults

      2022, Clinical Interventions in Aging
    View all citing articles on Scopus
    View full text
    Copyright © 2009 Elsevier Ireland Ltd. All rights reserved.