Life course and intrinsic ability in Brazilian adults and older adults: a mediation analysisABSTRACT
OBJECTIVE To analyze the direct and indirect associations between adverse life course conditions and impairment of intrinsic capacity in Brazilians aged 50 years or older.
METHODS An analytical cross-sectional study using data from the second wave of the Estudo Longitudinal da Saúde dos Idosos Brasileiros (ELSI-Brasil) (n = 4,817). Intrinsic capacity was assessed across five domains (cognition, vitality, mobility, psychological aspects, and sensory aspects) and was considered impaired when three or more domains were affected. Childhood adversities included parental illiteracy, lack of running water at home, and food insecurity. Educational attainment and income in adulthood were considered mediating variables. Path analysis was used.
RESULTS Childhood adversities had significant direct and indirect effects on the impairment of intrinsic capacity. These effects were mediated by lower educational attainment and income in adulthood. Educational attainment stood out as a central mediator, exerting a direct protective effect on intrinsic capacity and an indirect effect through its influence on income.
CONCLUSIONS The findings reinforce the relevance of the life-course perspective for understanding inequalities in aging and indicate educational attainment as a priority target for interventions aimed at promoting healthy aging in the Brazilian context.
DESCRIPTORS:
Life Course; Intrinsic Capacity; Adults; Older Adults; Aging
RESUMO
OBJETIVO Analisar as associações diretas e indiretas entre condições adversas ao longo do curso de vida e o comprometimento da capacidade intrínseca em brasileiros com 50 anos ou mais.
MÉTODOS Estudo transversal analítico com dados da segunda onda do Estudo Longitudinal da Saúde dos Idosos Brasileiros (ELSI-Brasil) (n = 4.817). A capacidade intrínseca foi avaliada em cinco domínios (cognição, vitalidade, locomoção, aspectos psicológicos e sensoriais), considerando-se como comprometida quando três ou mais domínios foram afetados. As adversidades na infância incluíram analfabetismo parental, ausência de água encanada no domicílio e insegurança alimentar. Escolaridade e renda na vida adulta foram consideradas variáveis mediadoras. Utilizou-se a análise de caminhos (path analysis).
RESULTADOS As adversidades na infância apresentaram efeitos diretos e indiretos significativos sobre o comprometimento da capacidade intrínseca. Esses efeitos foram mediados por menor escolaridade e renda na vida adulta. A escolaridade destacou-se como mediadora central, exercendo efeito protetor direto sobre a capacidade intrínseca e efeito indireto por meio de sua influência sobre a renda.
CONCLUSÕES Os achados reforçam a relevância da perspectiva do curso de vida para a compreensão das desigualdades no envelhecimento e indicam a escolaridade como alvo prioritário de intervenções voltadas à promoção do envelhecimento saudável no contexto brasileiro.
DESCRITORES
Curso de Vida; Capacidade Intrínseca; Adultos; Idosos; Envelhecimento
INTRODUCTION
Intrinsic capacity (IC) is defined as the set of physical and mental attributes of a person that interact with the environment and that, when maintained and optimized throughout life, promote healthy aging. Proposed in 2019 based on the English Longitudinal Study on Ageing (ELSA), the concept assesses five domains: mobility, vitality, cognition, psychological aspects, and sensory functions. Although it is a pillar of the “Decade of Healthy Aging (2021–2030),” evidence on IC in the Brazilian context remains scarce, especially regarding an understanding of how life trajectories in Brazil, marked by historical inequalities, impact these domains in an integrated manner. This gap limits the development of specific strategies that consider the heterogeneity of aging in Brazil1.
This comprehensive approach to the health of older adults reflects an important conceptual advance considering the realization that conceptions of aging centered exclusively on diseases or disabilities are insufficient to capture the complexity of people’s lives and health as they age1. While functionality generally focuses on what a person is able to do (such as performing activities of daily living), IC focuses on what a person possesses in terms of biological and psychological reserves and potential. Previous studies have identified sociodemographic and health factors associated with IC, including sex2, age3, income level4, marital status4, educational level5, behaviorshealth-related5, and the burden of chronic diseases3,4.
Despite the relevance of these studies, analyses conducted at a single point in time may not fully capture the factors that contribute to the occurrence of IC, given that each person’s life course varies over time and is influenced by individual, social, and contextual factors.
Given this, studying the IC from a life-course perspective is essential to understanding how experiences, exposures, and contexts encountered throughout an individual’s life course influence aging and health in later life, considering that aging is a multidetermined, continuous, and dynamic process. Life-course models theorize that events during critical periods of human development and the accumulation of risks throughout life influence health, disease, and aging6,7. Adversities in childhood, such as low socioeconomic status8 and food insecurity9, are associated with the development of physical diseases10 and mental disorders in adulthood11.
The relationship between life course variables and IC has already been documented12,13. Childhood circumstances such as low parental education, poorer childhood health, and living in substandard housing were directly associated with lower levels of IC in later life, with part of this effect mediated by accumulated socioeconomic inequalities13. Furthermore, the areas of IC most affected by these conditions are primarily the cognitive, sensory, and psychological domains12.
Evidence on the association between life course trajectories and impaired IC remains scarce in the Brazilian context, particularly when considering all its domains. Furthermore, this approach facilitates advances in scientific knowledge by integrating biological, psychological, and social perspectives, thereby enriching our understanding of the aging process.
In light of this, the objective of the study was to analyze the direct and indirect association between adverse life course conditions and impaired IC in a Brazilian population of adults aged 50 years or older, using baseline data from the Estudo Longitudinal da Saúde dos Idosos Brasileiros (ELSI-Brasil – Longitudinal Study on the Health of Brazilian Older Adults).
METHODS
Study Design and Population
This was a cross-sectional, analytical study based on data from the second wave of ELSI-Brasil, conducted between 2019 and 2021 (Figure 1). ELSI-Brasil is a national, household-based cohort representative of the Brazilian population aged 50 years or older living in the community, whose objective is to investigate the aging process and its determinants, thereby supporting the formulation and evaluation of public policies in the country14.
Study Sample and Sampling
The ELSI-Brasil sampling plan is complex, probabilistic, and multistage, involving the selection of municipalities, census tracts, and households. The design ensures national representativeness, covering urban and rural areas in 70 municipalities distributed across Brazil’s five regions (Figure 1).
The baseline phase of the study was conducted between 2015 and 2016, with 9,412 participants. In the second wave, 6,172 individuals were re-interviewed; there were 2,270 follow-up losses and 970 deaths. To maintain the cohort’s age representativeness, 3,777 new participants were included, bringing the total number of individuals interviewed in the second wave to 9,94914 (Figure 1).
For this study, only cross-sectional data from the second wave were used. All residents of the selected households aged 50 years or older were eligible. The final analytical sample consisted of 4,817 individuals, including only those participants with complete information for all variables of interest (Figure 1).
Data Collection
Data collection was performed by previously trained interviewers using standardized questionnaires, in addition to physical tests and laboratory exams, following protocols previously described in the ELSI-Brasil study14.
Study Variables
Dependent Variable
The variable of interest was IC, assessed based on five domains: cognition, psychological health, sensory function, vitality, and mobility1.
The cognitive domain was measured using the Verbal Fluency Test (VFT), in which participants named as many animals as possible in one minute. The cutoff scores were nine words for individuals with up to eight years of schooling and 13 for the others, indicating preserved cognition15.
The psychological domain was assessed using the abbreviated CES-D scale, consisting of eight items on depressive symptoms. Scores ≤ 4 indicated no impairment16.
Vitality was measured by handgrip strength using a dynamometer, with the average of three trials. The cutoff points were > 27 kgf for men and > 16 kgf for women, according to the criteria of the European Working Group on Sarcopenia17.
Locomotion was assessed using the habitual walking test over a three-meter distance, measured twice. Mean values ≥ 0.8 m/s indicated preserved function18.
Sensory assessment was based on self-reported visual and auditory function. Responses of “very good,” “good,” or “fair” for distance vision, near vision, and hearing indicated preserved function3.
Each IC domain was classified as 0 (preserved) or 1 (impaired). The sum of these scores yielded an index ranging from 0 to 5 points; IC was considered impaired if three or more domains were impaired2.
Independent Variables
The independent variables included sociodemographic factors—sex (female or male, self-reported) and age group (categorized as 50–59, 60–69, 70–79, and ≥ 80 years)—and life course variables, namely: parental illiteracy (report of illiterate father and mother)19, food insecurity (report of having ever gone to bed hungry or witnessed a lack of food from birth through age 15)9, lack of treated water (report of no running water in the home where the individual lived at age 10)20—all dichotomized as “yes” or “no.” Among the mediating variables in adulthood are educational attainment (complete years of schooling, categorized as: none, 1–4, 5–8, or ≥ 9 years) and monthly income (expressed in minimum wages, based on the value of R$ 1,047.67, and categorized as: < 1, 1–3, or ≥ 4 minimum wages)13,20.
Data Analysis
A descriptive analysis was performed to characterize the sample, using measures of absolute and relative frequency for the variables. Percentage distributions and their respective 95% confidence intervals were estimated using the survey (svy) module, including sample weights and stratification. Structural equation modeling was performed using the SEM command with robust standard errors (vce robust).
To study the relationship between the variables of interest, based on the pre-established theoretical model (Figure 2), structural equation modeling was used via the path analysis technique. This approach serves as an extension of the regression model, allowing for the evaluation of relationships among multiple variables21. Using this technique, it was possible to investigate the direct and indirect connections between independent and dependent variables.
The direct effects of childhood conditions on IC are indicated by the arrows that directly link the independent variables to the dependent variable. The indirect effects, mediated by educational attainment and income, are represented by the arrows that pass through the mediating variables (Figure 2).
After adjusting the indicators and conducting significance tests, the final path analysis model was constructed, which retained or eliminated the relationships from the initial theoretical model. The following tests and reference values for model fit thresholds were adopted in this study: chi-square test (χ2) for model fit (p > 0.05), chi-square-to-degrees-of-freedom ratio (χ2/df) < 2, Standardized Root Mean Square Residual (SRMR) ≤ 0.10, Root Mean Square Error of Approximation (RMSEA) ≤ 0.08, Comparative Fit Index (CFI) ≥ 0.90, and Tucker–Lewis Index (TLI) ≥ 0.90.
To assess the fit of the proposed paths, significance tests were performed on the path coefficients. Absolute values of t > 1.96 were considered statistically significant. The analyses were conducted using Stata software, version 15.1.
Ethical Considerations
The study was approved by the Ethics Committee of the Fundação Oswaldo Cruz (Fiocruz Minas – Oswaldo Cruz Foundation) (CAAE: 34649814.3.0000.5091), and all participants signed an informed consent form.
RESULTS
The sample consisted of 4,817 participants. Of the total number of participants (n = 4,817), 54.7% were women, 50.4% were between 50 and 59 years old, and 37.9% had 1 to 4 years of schooling. Regarding life course variables, 43.7% of the older adults had illiterate parents, 64.0% experienced a lack of piped water during childhood, and 24% experienced food insecurity (Table 1).
Figure 3 shows that both early-life variables (parental illiteracy, lack of piped water, and food insecurity) and adult-life variables (income and education) were directly associated with impaired cognitive function. It was observed that early-life variables, such as parental illiteracy (β = 0.07), lack of piped water (β = 0.03), and food insecurity (β = 0.05), were directly and positively associated with impaired CI. On the other hand, higher educational attainment (β = -0.14) and higher income (β = -0.04) showed direct protective effects. Adverse early-life conditions were also negatively associated with educational attainment (parental illiteracy: β = -0.36; lack of piped water: β = -0.24; food insecurity: β = -0.06), which, in turn, was positively associated with income in adulthood (β = 0.19). Additionally, it was observed that income and educational attainment mediated the association between early-life conditions and IC. Notably, educational attainment also acted as a mediator of the association between income and IC, as evidenced by the significant indirect effect (β = -0.01; p = 0.009) (Figure 3).
Table 2 presents the indirect effects, showing that all childhood conditions (parental illiteracy: β = 0.06; lack of piped water: β = 0.04; food insecurity: β = 0.01) significantly influenced impaired IC in adulthood, primarily through lower levels of education and income. Regarding adult-life variables, it is noteworthy that only educational attainment showed a protective indirect effect on IC mediated by higher income in adulthood (Table 2).
The final structural equation model demonstrated an adequate fit to the data, as indicated by the parameters described in the data analysis (CFI = 1.00; TLI = 1.00; RMSEA < 0.001; SRMR < 0.001), suggesting a good fit of the proposed structure. Overall, the model explained 26.1% of the variation in the IC commitment.
DISCUSSION
The results of this study showed that conditions indicative of low socioeconomic status in childhood—parental illiteracy, lack of piped water, and food insecurity—were significantly associated with impaired IC, an effect that occurred both directly and through the mediating role of lower educational attainment and income.
Although life-course evaluation is the subject of many studies, its relationship with IC has been assessed in a few recent studies12,13. In a general analysis of the effect of various childhood conditions on the domains of IC in adults and older adults, two studies show findings similar to ours. A cross-sectional study in China involving individuals aged ≥ 45 years showed that low parental education, poorer childhood health, and poor housing conditions were associated with lower IC, with part of this effect mediated by socioeconomic inequalities accumulated over the life course. The strongest associations were identified in the cognitive, sensory, and psychological domains of IC13. In an observational cohort study involving participants aged ≥ 50 years from the United States, the United Kingdom, Europe, China, and Japan, higher levels of parental education, socioeconomic status in childhood, and household income in adulthood were positively associated with IC in all analyzed cohorts12.
Regarding parental illiteracy and impaired IC, research with results similar to ours explains this relationship, specifically attributing it to the cognitive domain of IC20. Higher maternal and paternal educational attainment is associated with a slower rate of cognitive decline in their children’s middle age and old age, possibly mediated by more stimulating family environments and better childhood health care19,22.
Food insecurity in childhood, characterized by reduced or missed meals and the substitution of nutritious foods with cheaper, nutrient-poor alternatives, can lead to insufficient intake of essential nutrients, compromising physical, cognitive, and psychosocial development. Nutritional deprivation in childhood has been consistently linked to cognitive decline in later life23, to greater functional limitations, to poorer lung function24, and to reduced sensory capacity (impaired vision and hearing development)13. Evidence of food insecurity among older adults has shown that the constant search for food leads to ongoing worry, mental distress, and chronic stress, which can affect psychological well-being by increasing the risk of depression, depressive symptoms, and cognitive impairment25.
In Brazil, access to piped water during childhood is a significant marker of inequities in access to essential services throughout the life course. Despite expanded coverage, inequalities in water supply and limitations regarding water quality persist, particularly affecting low-income populations and rural areas, with more than 40 million Brazilians still lacking adequate access26. Although its importance as a health determinant is recognized, there is a scarcity of studies that assess its association with IC in isolation. International evidence indicates that early-life exposures related to water and sanitation influence outcomes in old age, including poorer cognitive performance and markers of accelerated cellular aging27.
Regarding variables from adulthood, both educational attainment and income had a direct effect on IC and mediated the effects of childhood conditions. Education, in addition to mediating adult income, emerged as the primary determinant of IC in older adults and the elderly, indicating that early exposure to education may mitigate adverse effects throughout life and promote better health in old age.
Previous studies show similar results: Huang et al.12 observed that upward social mobility—an improvement in socioeconomic status over the course of one’s life—is associated with better IC in old age. Complementarily, another study found that individuals who grew up in unfavorable socioeconomic contexts but attained a high level of education in adulthood had an active life expectancy similar to that of those from economically privileged homes, highlighting the protective role of education in mitigating early-life disadvantages28.
A study of Brazilian older adults showed that unfavorable socioeconomic conditions in childhood are linked to poorer cognitive performance in old age. However, socioeconomic status in adulthood was the factor that explained this relationship: it fully mediated the effect on memory and partially mediated the effect on verbal fluency20.
According to the cognitive reserve theory, cognitive abilities accumulated over the course of a lifetime—such as those resulting from the level of education attained by early adulthood—persist, at least in part, even as one ages and strengthen the neurobiological reserve that promotes greater resistance to cognitive and functional decline in later life29.
In this context, favorable socioeconomic conditions—such as higher levels of education—expand access to professions that require complex cognitive skills and offer better-paying job opportunities, which in turn ensure greater participation in leisure activities, social engagement, and regular physical activity. These factors act synergistically, which appear to strengthen the development and maintenance of cognitive reserve throughout life30, thereby optimizing IC. Considering these findings and given the growing elderly population and rising educational levels among more recent generations, education may positively influence IC.
Overall, the results contribute to the gerontological literature by demonstrating that adverse socioeconomic conditions in childhood remain associated with impaired IC six to seven decades later, reinforcing the relevance of the life-course perspective in the study of aging. Education also stands out as a central axis in these relationships, acting both as a direct protective factor and as a mediator of the effects of early disadvantages. By quantifying these mediational pathways, the study advances our understanding of the mechanisms through which early inequities accumulate and are perpetuated throughout life.
Taken together, these findings broaden our understanding of the determinants of healthy aging and reinforce the potential of interventions that act early and continuously throughout the life course6,7. In this regard, the results point to the need for public policies that integrate actions aimed at improving conditions in childhood, strengthening education, and expanding socioeconomic opportunities in adulthood. Such an approach promotes a comprehensive perspective on aging, aligned with the needs of older adults and the organization of health care systems based on integrated, person-centered care.
This study has some limitations. First, the cross-sectional design of the study prevents the establishment of causal relationships between the variables. Although the associations found are statistically significant, data collection at a single point in time does not allow for determining the temporal precedence of events. Additionally, the study relied on retrospective measures of life course, which introduces potential memory bias. Reliance on self-reports for events that occurred several decades ago may be subject to recall bias or a subjective reinterpretation of past experiences, influenced by the participants’ current health status. Although the use of validated life course instruments and questions aims to minimize such distortions, the subjective nature of retrospective reporting remains a limiting factor in the accuracy of the magnitude of these events. Regarding data analysis, it should be noted that the models were estimated without applying sample weights. As a strategy to reduce potential violations of assumptions (particularly those related to heteroscedasticity), robust standard errors (vce robust) were used. It should also be noted that, although ELSI-Brasil is a representative survey of Brazilian adults and older adults, the variables used in the analyses had approximately 40% missing data; however, there were no statistically significant differences between the included and excluded groups for the life course variables. Despite these limitations, path analysis is a robust strategy, as it allows for the simultaneous identification of direct and indirect effects between variables, representing an advantage over traditional regression analyses.
In conclusion, it was evident that adverse childhood conditions related to a low socioeconomic status (parental illiteracy, lack of treated water, and food insecurity) are associated with impaired IC. This association manifests itself both through a direct effect of these adversities and through an indirect effect, mediated by life trajectories characterized by lower educational attainment and income in adulthood.
REFERENCES
-
1 Beard JR, Jotheeswaran AT, Cesari M, Carvalho IA. The structure and predictive value of intrinsic capacity in a longitudinal study of ageing. BMJ Open. 2019;9(11):e026119. https://doi.org/10.1136/bmjopen-2018-026119
» https://doi.org/10.1136/bmjopen-2018-026119 -
2 Plácido J, Marinho V, Ferreira JV, Teixeira IA, Costa EC, Deslandes AC. Association among race/color, gender, and intrinsic capacity: results from the ELSI-Brazil study. Rev Saúde Pública. 2023;57:29. https://doi.org/10.11606/s1518-8787.2023057004548
» https://doi.org/10.11606/s1518-8787.2023057004548 -
3 Aliberti MJR, Bertola L, Szlejf C, Oliveira D, Piovezan RD, Cesari M et al. Validating intrinsic capacity to measure healthy aging in an upper middle-income country: findings from the ELSI-Brazil. Lancet Reg Health Am. 2022;12:100284. https://doi.org/10.1016/j.lana.2022.100284
» https://doi.org/10.1016/j.lana.2022.100284 -
4 Ma L, Chhetri JK, Zhang L, Sun F, Li Y, Tang Z. Cross-sectional study examining the status of intrinsic capacity decline in community-dwelling older adults in China: prevalence, associated factors and implications for clinical care. BMJ Open. 2021;11(1):e043062. https://doi.org/10.1136/bmjopen-2020-043062
» https://doi.org/10.1136/bmjopen-2020-043062 -
5 Jiang X, Chen F, Yang X, Yang M, Zhang X, Ma X et al. Effects of personal and health characteristics on the intrinsic capacity of older adults in the community: a cross-sectional study using the healthy aging framework. BMC Geriatr. 2023;23:643. https://doi.org/10.1186/s12877-023-04354-0
» https://doi.org/10.1186/s12877-023-04354-0 -
6 Cao X, Ma C, Zheng Z, He L, Hao M, Chen X et al. Contribution of life course circumstances to the acceleration of phenotypic and functional aging: a retrospective study. eClinicalMedicine. 2022;51:101578. https://doi.org/10.1016/j.eclinm.2022.101548
» https://doi.org/10.1016/j.eclinm.2022.101548 -
7 Kemoun P, Ader I, Planat-Benard V, Dray C, Fazilleau N, Monsarrat P, et al. A gerophysiology perspective on healthy ageing. Ageing Res Rev. 2022;73:101537. https://doi.org/10.1016/j.arr.2021.101537
» https://doi.org/10.1016/j.arr.2021.101537 -
8 Cheval B, Boisgontier MP, Orsholits D, Sieber S, Guessous I, Gabriel R et al. Association of early- and adult-life socioeconomic circumstances with muscle strength in older age. Age Ageing. 2018;47(3):398-407. https://doi.org/10.1093/ageing/afy003
» https://doi.org/10.1093/ageing/afy003 -
9 Royer MF, Ojinnaka CO, Zhang X, Thornton AG, Blackhorse K, Bruening M. Food insecurity and adverse childhood experiences: a systematic review. Nutr Rev. 2022;80(10):2089-99. https://doi.org/10.1093/nutrit/nuac029
» https://doi.org/10.1093/nutrit/nuac029 -
10 Hu Z, Kaminga AC, Yang J, Liu J, Xu H. Adverse childhood experiences and cancer risk. Child Abuse Negl. 2021;117:105088. https://doi.org/10.1016/j.chiabu.2021.105088
» https://doi.org/10.1016/j.chiabu.2021.105088 -
11 Park S, Kwon E, Lee H. Life Course Trajectories of later-life cognitive functions: does social engagement in old age matter? Int J Environ Res Public Health. 2017;14(4):393. https://doi.org/10.3390/ijerph14040393
» https://doi.org/10.3390/ijerph14040393 -
12 Huang Z, Lai ETC, Hashimoto H, Marmot M, Woo J. Life-course socioeconomic inequalities and healthy aging. Arch Gerontol Geriatr. 2025;133:105829. https://doi.org/10.1016/j.archger.2025.105829
» https://doi.org/10.1016/j.archger.2025.105829 -
13 Si Y, Hanewald K, Chen S, Li B, Bateman H, Beard JR. Life-course inequalities in intrinsic capacity and healthy ageing, China. Bull World Health Organ. 2023;101(5):307-316C. https://doi.org/10.2471/BLT.22.288682
» https://doi.org/10.2471/BLT.22.288682 -
14 Lima-Costa MF, Mambrini JVM, Andrade FB, Souza Júnior PRB, Vasconcellos MTL, Neri AL et al. Cohort Profile: The Brazilian Longitudinal Study of Ageing (ELSI-Brazil). Int J Epidemiol. 2023;52(1):e57-65. https://doi.org/10.1093/ije/dyac132
» https://doi.org/10.1093/ije/dyac132 -
15 Brucki SMD, Malheiros SMF, Okamoto IH, Bertolucci PHF. Dados normativos para o teste de fluência verbal categoria animais em nosso meio. Arq Neuropsiquiatr. 1997;55(1):56-61. Available from: https://www.scielo.br/j/anp/a/LVLVRv9fG3zgszsFhvfrv7p/?lang=pt
» https://www.scielo.br/j/anp/a/LVLVRv9fG3zgszsFhvfrv7p/?lang=pt -
16 Gallagher D, Kiss A, Lanctot K, Herrmann N. Depressive symptoms and cognitive decline: a longitudinal analysis of potentially modifiable risk factors in community dwelling older adults. J Affect Disord. 2016;190:235-40. https://doi.org/10.1016/j.jad.2015.09.046
» https://doi.org/10.1016/j.jad.2015.09.046 -
17 Lauretani F, Russo CR, Bandinelli S, Bartali B, Cavazzini C, Di Iorio A et al. Age-associated changes in skeletal muscles and their effect on mobility: an operational diagnosis of sarcopenia. J Appl Physiol. 2003;95(5):1851-60. https://doi.org/10.1152/japplphysiol.00246.2003
» https://doi.org/10.1152/japplphysiol.00246.2003 -
18 Montero-Odasso M, Pieruccini-Faria F, Ismail Z, Li K, Lim A, Phillips N et al. CCCDTD5 recommendations. Alzheimers Dement (N Y). 2020;6(1):e12068. https://doi.org/10.1002/trc2.12068
» https://doi.org/10.1002/trc2.12068 -
19 Frangos E, Barriguete-Mélendez A, Debré P, Gutiérrez Robledo LM, Parodi AL, Michel JP. Life course education, health, and ageing well: a short inter-academic report. Gerontology. 2023;69(7):799-808. https://doi.org/10.1159/000529869
» https://doi.org/10.1159/000529869 -
20 Seixas BV, Macinko J. Distinct domains of childhood disadvantage and cognitive performance among older brazilians: evidence from ELSI-Brazil. SSM Popul Health. 2023;22:101416. https://doi.org/10.1016/j.ssmph.2023.101416
» https://doi.org/10.1016/j.ssmph.2023.101416 -
21 Harris JE, Gleason PM. Application of path analysis and structural equation modeling in nutrition and dietetics. J Acad Nutr Diet. 2022;122(11):2023-35. https://doi.org/10.1016/j.jand.2022.07.007
» https://doi.org/10.1016/j.jand.2022.07.007 -
22 Luo S, Chen W, Hu W, Wang HHX, Li J, Guo VY et al. Parental education, own education, and cognitive function in middle-aged and older adults. JAMA Netw Open. 2025;8(5):e2513036. https://doi.org/10.1001/jamanetworkopen.2025.13036
» https://doi.org/10.1001/jamanetworkopen.2025.13036 -
23 Saenz J, Avila JC. Late-life food insecurity and cognition: exploring timing, duration, and mechanisms among older Mexican adults. BMC Geriatr. 2023;23(1):788. https://doi.org/10.1186/s12877-023-04497-7
» https://doi.org/10.1186/s12877-023-04497-7 -
24 Peele ME. Childhood conditions predict chronic diseases and functional limitations among older adults: the case of Indonesia. J Aging Health. 2019;31(10):1892-916. https://doi.org/10.1177/0898264318799550
» https://doi.org/10.1177/0898264318799550 -
25 Lu P, Kezios K, Yaffe K, Kim S, Zhang A, Milazzo FH et al. Depressive symptoms mediate the relationship between sustained food insecurity and cognition: a causal mediation analysis. Ann Epidemiol. 2023;81:6-13.e1. https://doi.org/10.1016/j.annepidem.2023.02.009
» https://doi.org/10.1016/j.annepidem.2023.02.009 -
26 Paz MGA, Fracalanza AP, Alves EM, Silva FJR. Os conflitos das políticas da água e do esgotamento sanitário: que universalização buscamos? Estud Av. 2021;35(102):193-208. https://doi.org/10.1590/s0103-4014.2021.35102.012
» https://doi.org/10.1590/s0103-4014.2021.35102.012 -
27 Lee H, Lee MW, Warren JR, Ferrie J. Childhood lead exposure is associated with lower cognitive functioning at older ages. Sci Adv. 2022;8(45):eabn5164. https://doi.org/10.1126/sciadv.abn516
» https://doi.org/10.1126/sciadv.abn516 -
28 Montez JK, Hayward MD. Cumulative childhood adversity, educational attainment, and active life expectancy among U.S. Adults. Demography. 2014;51(2):413-35. https://doi.org/10.1007/s13524-013-0261-1
» https://doi.org/10.1007/s13524-013-0261-1 -
29 Lövdén M, Fratiglioni L, Glymour MM, Lindenberger U, Tucker-Drob EM. Education and cognitive functioning across the life span. Psychol Sci Public Interest. 2020;21(1):6-41. https://doi.org/10.1177/1529100620920576
» https://doi.org/10.1177/1529100620920576 -
30 Arenaza-Urquijo EM, Bartrés-Faz D, Belleville S, Cantilon M, Chetelat G et al. Whitepaper: defining and investigating cognitive reserve, brain reserve, and brain maintenance. Alzheim Dement. 2020;16(9):1305-11. https://doi.org/10.1016/j.jalz.2018.07.219
» https://doi.org/10.1016/j.jalz.2018.07.219
Note: The arrows represent directional associations estimated in the model, and the values correspond to the standardized coefficients (β). Thicker arrows indicate direct associations.