Lifetime socioeconomic determinants of health trajectories among older adults

https://doi.org/10.1016/j.alcr.2021.100415Get rights and content

Highlights

  • We used a life course approach to examine health trajectories of adults ≥50 years old.

  • We used the Mexican Health and Aging panel survey from 2001 to 2015.

  • We identified four clusters of health trajectories for men and eight for women.

  • Childhood and adult socioeconomic factors shape health trajectories at older ages.

  • There is no simple monotonic relation between life-course circumstances and health.

Abstract

Drawing on life course theory and research, we explored how socioeconomic circumstances during childhood and adulthood shape self-reported health trajectories among older Mexican adults. We used data from the Mexican Health and Aging Study panel survey (2001–2015) and used sequence analysis to estimate types of self-reported health trajectories in older adulthood. We then explored the association between those health trajectories and socioeconomic determinants at different life stages, including education, occupation, employment, economic status, parental education, and adverse living conditions and illnesses during childhood. Our contributions are threefold. First, we identified four types of health trajectories for men and eight for women, representing a more nuanced longitudinal health status profile than previously shown. Second, we found that childhood and adult socioeconomic circumstances influence self-reported health trajectories at older age. Third, our results suggest there is no simple monotonic relationship between life course circumstances and self-reported health trajectories.

Introduction

The relationship between health and socioeconomic position has been the subject of a great deal of research during the last decades. There is consensus that individuals with disadvantaged socioeconomic position (SEP) (Galobardes, Shaw, Lawlor, Lynch, & Smith, 2006; Krieger, Williams, & Moss, 1997), such as those living in poverty or deprivation, are more likely to have poor health and are at higher risk of disease than their peers with higher SEP (Brunner, 1997; Deaton, 2013; Marmot, 2005; Stringhini et al., 2017). Despite the overwhelming amount of research pointing out social health inequalities, how health and socioeconomic position relate remains poorly understood, partially because most research has examined this association as a static phenomenon using a cross-sectional approach, considering a single point in time (Deaton, 2013). However, neither SEP nor health are stable phenomena throughout life.

Lives are ongoing processes and not just single states or events that can be adequately described using snapshots (Levy et al., 2005). For example, an individual can change her SEP throughout life due to childhood experiences, education attainments, or working opportunities, among other events. Likewise, health trajectories follow dynamic variations, with individuals experiencing periods of health, sickness, or disability throughout time. These changes occur due to biological factors such as aging, illnesses, and genes, and contextual factors such as retirement age, ethnicity, gender, behavior, environment, and economic status (Aisenbrey & Fasang, 2010; Bernardi, Huinink, & Settersten, 2019; Ferraro, Farmer, & Wybraniec, 1997; Haas, 2008; Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003; McDonough & Berglund, 2003; Sacker, Clarke, Wiggins, & Bartley, 2005; Settersten, 2003). Socioeconomic position and health should not be understood as single events but as longitudinal phenomena. The dynamic interrelationship between health trajectories and socioeconomic at different life stages can help us better understand the underlying mechanisms of social health inequalities.

Thus, a life-course perspective allows a more dynamic and nuanced understanding of the association between SEP and health by modeling exposure to socioeconomic factors throughout the life-course and its effects on the health trajectories of older adults (van der Linden et al., 2019, 2018; World Health Organization, 2000; Yu, 2006). The life-course theory uses two main approaches to study how changing SEP over the life-course affects an individual’s health. First, the critical periods approach posits that adverse events related to SEP, such as undernutrition or exposure to pollution, during critical life periods (e.g., early childhood) may alter an individual’s health trajectory or illness risk over the life course. Second, the cumulative model approach underscores the length and intensity of exposure to adverse events as having long-term effects on health status through the accumulation of relative advantages and disadvantages over specific domains, such as education or labor, that may affect health (Cunningham et al., 2018; Dannefer, 2003; Madero-Cabib, Azar, & Pérez-Cruz, 2019; Singh-Manoux, Marmot, & Adler, 2005). The number, duration, and severity of exposures to adverse events across different periods of life may reduce or even eliminate the reversibility of their negative health effects.

Research in high-income countries suggests that SEP gradients are associated with health status (Case & Deaton, 2005; Deaton & Paxson, 2001). This association persists into old age for a broad range of health measures, such as mortality (Demakakos, Biddulph, Bobak, & Marmot, 2016; Marmot & Shipley, 1996; Mortensen et al., 2016) and self-reported health (Hu et al., 2016; Huisman, Kunst, & Mackenbach, 2003; Semyonov, Lewin-Epstein, & Maskileyson, 2013; van Kippersluis, O’Donnell, van Doorslaer, & Van Ourti, 2010). Mortality is the most commonly used outcome to measure the relationship between SEP inequalities and health (Lobmayer & Wilkinson, 2002; Lynch, Smith, Kaplan, & House, 2000; Pickett & Wilkinson, 2015; Wilkinson & Pickett, 2006). However, mortality is of limited use when examining the way health changes throughout the life course. Self-reported health is more appropriate than mortality to describe dynamic variations over time (Case & Deaton, 2005). Self-reported health provides a simple, direct way of capturing health perceptions. It is a valid, reliable, low cost predictor of morbidity (Chandola & Jenkinson, 2000) and mortality (Kaplan & Camacho, 1983; Lee, 2000).

Despite referring to current circumstances, self-reported health reflects the accumulation of health over the life course. Research suggests that self-rated health adequately represents morbidity and disability (Idler & Benyamini, 1997). For example, Idler and Kasl (1995) found a significant association between self-rated and disability in a sample of American adults. In Mexico, self-rated health and mortality have been strongly associated with infectious and chronic diseases (Markides, Salinas, & Sheffield, 2008; Samper-Ternent, Michaels-Obregon, Wong, & Palloni, 2012). Similarly, using a national representative sample of the older Mexican population, Bustos-Vázquez, Fernández-Niño, and Astudillo-Garcia (2017) reported direct and indirect associations between self-rated health with disability and morbidity.

Previous studies show women perceive their health less favourably than men at older ages (Anson, Paran, Neumann, & Chernichovsky, 1993; Hunt, McEwen, & McKenna, 1984). Explanations of sex differences in self-reported health can reflect actual differences in health as a consequence of biological and social factors and reflect a difference in the tendency to report health problems with women having a greater awareness of the socioemotional effect of health (Ross & Bird, 1994). Therefore, self-reported health trajectories might differ by gender. Researchers have also found gender differences in the relationship between self-rated health and mortality. A systematic review conducted by (Idler & Benyamini, 1997) reported that poor self-rated health poses a greater risk of mortality for men than women. There are also documented gender differences in the association between self-reported health and SEP, especially when including occupation and income. For example, a Mexican study including individuals over 50 years old showed a stronger association between lifetime occupation and self-rated health on men compared to women (Torres, Rizzo, & Wong, 2018). Similarly, an analysis of urban adults 50 years and over in Latin America found that women were more likely to report poor health than men. Socioeconomic disadvantage over the life course did not fully explain gender differences (Zunzunegui, Alvarado, Béland, & Vissandjee, 2009).

Researchers have identified an association between low SEP and health indicators among the elder, including low educational attainment, income, self-esteem, and life satisfaction (Idler & Benyamini, 1997). This association is often smaller at older ages, possibly related to how studies measure SEP at different life stages (Sacker et al., 2005). There is also some evidence supporting the age as-leveler hypothesis. This hypothesis posits that health inequalities may diminish as a consequence of older age if health declines surpass cumulative life-course disadvantages (Dupre, 2007; Siegel, Luengen, & Stock, 2013; van Kippersluis et al., 2010). The higher mortality of poorer individuals at younger ages or the reduction in inequality with older age may explain this effect (Dupre, 2007; Siegel et al., 2013; van Kippersluis et al., 2010).

Furthermore, there is little evidence about the association between SEP and health trajectories before and after retirement. A recent systematic review by Cullati, Rousseaux, Gabadinho, Courvoisier, and Burton-Jeangros (2014) found that self-reported health trajectories of disadvantaged populations decline faster rate than health trajectories of relatively advantaged populations. Those authors also found that young age, high SEP, and entering a partnership are protective factors for health. Using hierarchical models for repeated measures, a recent study in Mexico found childhood SEP was associated with later-life health trajectories, suggesting persistent disparities into old age (Torres et al., 2018). These findings highlight the relevance of early-life conditions in shaping late-life health and life-time health trajectories. While there is some evidence supporting the association between SEP and self-reported health for older adults in Latin America (Smith & Goldman, 2007; Subramanian, Delgado, Jadue, Vega, & Kawachi, 2003; Torres et al., 2018; Wong, Michaels-Obregon, & Palloni, 2015), most studies have focused on western populations (i.e., conducted mostly in Europe and North America). This may limit the generalizability of results (Henrich, Heine, & Norenzayan, 2010). Further, most studies have not considered childhood SEP (see Shuey & Willson, 2014 for an exception), raising questions about how much of the association between SEP and health trajectories is explained by early life circumstances.

In sum, little is known about the association between lifetime SEP and health trajectories. Most previous research have used cross-sectional methods, modeled the association on a single point in time, focused on Western populations, and have not considered childhood SEP, baseline health status, and heterogeneity in time trends. The aims of this study are to (i) describe health trajectories using sequence analysis, (ii) to describe changes in SEP throughout the life course, and (iii) to estimate the association between SEP throughout life course with the health trajectories identified. We use data from a nationally representative panel survey of Mexican men and women between 50 and 65 years of age to address these questions and to capture the transition to retirement. Many biological and contextual factors can affect older adults’ self-reported health status during this period (Bernardi et al., 2019). At least some of these negative impacts could be prevented if understood better. We explicitly draw from life-course theory to recognize the relevance of individual biographies shaped by social circumstances and their long-term effects (Bernardi et al., 2019). Accounting for diversity at the individual level is the first step toward better understanding health over the life course and its dynamic relationship with social exposures (McDonough & Berglund, 2003). The sequence analysis is particularly useful for this purpose as it allows us to understand long-term patterns and differences between individuals through illustrative visualizations and a holistic perspective (Madero-Cabib & Fasang, 2016).

We believe Mexico provides an apt setting to examine the association between lifetime SEP and health trajectories among older adults for several reasons. First, Mexico is a middle-income country going through a fast demographic transition in the context of high income inequality (Consejo Nacional de Población, 2000; OECD, 2016). Among OECD countries, Mexico has the highest income inequality and lowest life expectancy at age 60 (20.8 years for men and 23.0 years for women; Dicker et al., 2018). Demographic projections suggest that by 2050 about 30 % of Mexicans will be older than 60 years of age. Longer lives are associated with increased healthcare utilization and costs (de Meijer, Wouterse, Polder, & Koopmanschap, 2013; Payne, Laporte, Deber, & Coyte, 2007; Spillman & Lubitz, 2000). However, Mexico’s fast aging process occurs under conditions of limited social security and a fragmented health system (Aguila, 2014; Angel, Vega, & López-Ortega, 2017; De Souza, Queiroz, & Skirbekk, 2019). There is substantial inequality in healthcare access and quality (Atun et al., 2015; Lozano et al., 2020), and the health system has one of the highest out-of-pocket expenditures among OECD countries, imposing significant financial risks to the population from catastrophic expenditures (Atun et al., 2015; OECD, 2016). The public health security system is based primarily on private contributions schemes linked to formal employment, although healthcare has been substantially extended in the past two decades to households that were not previously covered by the Seguro Popular (Angel et al., 2017; Atun et al., 2015). Access to social security is more limited in rural areas, where 20 % of the population lives (World Bank, 2020). A large proportion of the employed population (53.4 %) remains in informal employment (International Labour Office (ILO), 2020), and a large majority of older adults (∼80 %) do not have a retirement pension. Many of those who can access a pension delay their retirement until after 65 years of age (median retirement age is 69.4 years), further enhancing income inequalities and leaving many older adults financially vulnerable (Consejo Nacional de Población, 2000; OECD, 2016).

Section snippets

Materials and methods

We used data from the Mexican Health and Aging Study (MHAS), an ongoing longitudinal panel study with a nationally representative sample of ∼21,500 adults ≥50 years old in Mexico (Wong et al., 2015). MHAS, among other various life-course events, characterized socioeconomic status across life and health transitions in older adulthood. We used individual-level data from the four publicly available survey waves of MHAS (2001, 2003, 2012, and 2015).

Descriptive statistics

Table 1 shows weighted descriptive statistics separately for men and women in our sample (see Appendix C for unweighted descriptive statistics). The average age is 55.6 for men and women (median = 56). Women show worse indicators of self-reported health than men at baseline and in the last observation. However, more men than women died during the study period (15.6 versus 10.6, respectively). About half the adults in our sample had a mother with no formal education (men: 51.3 %, women: 54.5 %),

Discussion

Drawing on life course theory and research, we explored how socioeconomic circumstances during childhood and adulthood shape self-reported health trajectories among older Mexican adults. The analysis showed at least three relevant findings. First, we found substantial heterogeneity in the health trajectories of Mexican adults around the age of retirement. Second, disadvantaged childhood living conditions and SEP during adulthood are significantly associated with changes in self-reported health

Conclusion and policy implications

We examined the impact of lifetime SEP factors on self-reported health trajectories on a Mexican population. Understanding the link between childhood and adult disadvantage and later-life health is critical to help prevent negative impacts at an older age and inform policy design and implementation. Our results suggest that early and later life socioeconomic circumstances have lasting impacts on health trajectories. The dynamic approach to health supports that policies targeting transitions in

Funding

This work was supported by:

  • i

    ANID/Millennium Science Initiative/Grant NCS17_062 ‘Millennium Nucleus for the Study of the Life Course and Vulnerability (MLIV)’

  • ii

    ANID/FONDECYT/INICIACION/N°11180360

  • iii

    ANID/FONDAP/Nº15130009

Declaration of Competing Interest

No competing interests were reported by the authors of this paper.

Acknowledgments

Study sponsors did not have any role in study design; collection, analysis, and interpretation of data; writing the report; or in the decision to submit the report for publication.

References (124)

  • M. Huisman et al.

    Socioeconomic inequalities in morbidity among the elderly; a European overview

    Social Science & Medicine

    (2003)
  • R. Leinonen et al.

    Self-rated health and self-assessed change in health in elderly men and women—A five-year longitudinal study

    Social Science & Medicine

    (1998)
  • R. Levy et al.

    Why look at life courses in an interdisciplinary perspective?

    Toward an interdisciplinary perspective on the life course

    (2005)
  • I. Madero-Cabib et al.

    Gendered work–family life courses and financial well-being in retirement

    Advances in Life Course Research

    (2016)
  • M.G. Marmot

    Social determinants of health inequalities

    Lancet

    (2005)
  • P. McDonough et al.

    Time on my side? Life course trajectories of poverty and health

    Social Science & Medicine

    (2005)
  • P. McDonough et al.

    Socioeconomic inequalities in health dynamics: A comparison of Britain and the United States

    Social Science & Medicine

    (2010)
  • K.E. Pickett et al.

    Income inequality and health: A causal review

    Social Science & Medicine

    (2015)
  • P.J. Rousseeuw

    Silhouettes: A graphical aid to the interpretation and validation of cluster analysis

    Journal of Computational and Applied Mathematics

    (1987)
  • M. Semyonov et al.

    Where wealth matters more for health: The wealth–health gradient in 16 countries

    Social Science & Medicine

    (2013)
  • K.M. Shuey et al.

    Economic hardship in childhood and adult health trajectories: An alternative approach to investigating life-course processes

    Advances in Life Course Research

    (2014)
  • E. Aguila

    Male labor force participation and social security in Mexico

    Journal of Pension Economics & Finance

    (2014)
  • S. Aisenbrey et al.

    New life for old ideas: The" second wave" of sequence analysis bringing the" course" back into the life course

    Sociological Methods & Research

    (2010)
  • J.L. Angel et al.

    Aging in Mexico: Population trends and emerging issues

    The Gerontologist

    (2017)
  • S. Barquera et al.

    Obesidad en México, prevalencia y tendencias en adultos. Ensanut 2018-19

    Salud Publica de Mexico

    (2020)
  • J.R. Behrman

    Intrahousehold allocation of nutrients in rural India. Are boys favored? Do parents exhibit inequality aversion?

    Oxford Economic Papers-New Series

    (1988)
  • Y. Ben-Shlomo et al.

    A life course approach to chronic disease epidemiology: Conceptual models, empirical challenges and interdisciplinary perspectives

    International Journal of Epidemiology

    (2002)
  • Y. Ben-Shlomo et al.

    Life course epidemiology

    Handbook of epidemiology

    (2014)
  • E. Brunner

    Socioeconomic determinants of health. Stress and the biology of inequality

    British Medical Journal

    (1997)
  • C. Buchmann

    The debt crisis, structural adjustment and women’s education

    International Journal of Comparative Sociology

    (1996)
  • E. Bustos-Vázquez et al.

    Self-rated health, multimorbidity and depression in Mexican older adults: Proposal and evaluation of a simple conceptual model

    Biomedica

    (2017)
  • K.F. Butcher et al.

    The effect of sibling sex composition on women’s education and earnings

    The Quarterly Journal of Economics

    (1994)
  • S.v. Buuren et al.

    Mice: Multivariate imputation by chained equations in R

    Journal of Statistical Software

    (2010)
  • R. Caballero et al.

    El consumo de tabaco, alcohol y drogas ilegales, en los adolescentes de diferentes estratos socioeconómicos de Guadalajara

    Salud Mental

    (1999)
  • A. Case et al.

    Broken down by work and sex: How our health declines

    Analyses in the economics of aging

    (2005)
  • T. Chandola et al.

    Validating self-rated health in different ethnic groups

    Ethnicity & Health

    (2000)
  • Consejo Nacional de Población

    Envejecimento demográfico de México: retos y perspectivas. Por una sociedad para todas las edades

    (2000)
  • S.A. Cunningham et al.

    County-level contextual factors associated with diabetes incidence in the United States

    Annals of Epidemiology

    (2018)
  • D. Dannefer

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

    The Journals of Gerontology Series B, Psychological Sciences and Social Sciences

    (2003)
  • C. de Meijer et al.

    The effect of population aging on health expenditure growth: A critical review

    European Journal of Ageing

    (2013)
  • A. Deaton

    Health, inequality, and economic development

    Journal of Economic Literature

    (2003)
  • A. Deaton

    The great escape: Health, wealth, and the origins of inequality

    (2013)
  • A. Deaton et al.

    Mortality, education, income, and inequality among American cohorts

  • P. Demakakos et al.

    Wealth and mortality at older ages: A prospective cohort study

    Journal of Epidemiology and Community Health

    (2016)
  • D. Dicker et al.

    Global, regional, and national age-sex-specific mortality and life expectancy, 1950-2017: A systematic analysis for the Global Burden of Disease Study 2017

    Lancet

    (2018)
  • M.E. Dupre

    Educational differences in age-related patterns of disease: Reconsidering the cumulative disadvantage and age-as-leveler hypotheses

    Journal of Health and Social Behavior

    (2007)
  • L. Feinstein

    The relative economic importance of academic, psychological and behavioural attributes developed on childhood

    (2000)
  • K.F. Ferraro et al.

    Health trajectories: Long-term dynamics among black and white adults

    Journal of Health and Social Behavior

    (1997)
  • J. Fox et al.

    Effect displays in R for multinomial and proportional-odds logit models: Extensions to the effects package

    Journal of Statistical Software

    (2009)
  • A. Gabadinho et al.

    Analyzing and visualizing state sequences in R with TraMineR

    Journal of Statistical Software

    (2011)
  • View full text