Associations between Socioeconomic Status and Adverse Childhood Experiences with Multidimensional Healthy Aging: Findings from the ELSI-Brazil.
FabianaSilvaRibeiro
Ph.D
1,3✉Email
GiuliaBuscicchio
Ph.D
1Email
HeloísaFerreira
Ph.D
2Email
AnjaK.Leist
Ph.D
1Email
1
A
A
Department of Social SciencesUniversity of Luxembourg, Esch-sur-AlzetteBelval Campus ,11, Porte des SciencesL-4366Luxembourg
2Department of Cognition and Human DevelopmentRio de Janeiro State UniversityRio de JaneiroBrazil
3Department of Social SciencesUniversity of Luxembourg, Esch-sur AlzetteLuxembourg
Fabiana Ribeiro, Ph.Da
Fabiana.ribeiro@uni.lu
Giulia Buscicchio, Ph.Da
giulia.buscicchio@uni.lu
Heloísa Ferreira, Ph.Db
helogf@gmail.com
Anja K. Leist, Ph.D a
anja.leist@uni.lu
aDepartment of Social Sciences, University of Luxembourg.
Belval Campus ,11, Porte des Sciences, L-4366, Esch-sur-Alzette, Luxembourg.
b Department of Cognition and Human Development, Rio de Janeiro State University, Rio de Janeiro, Brazil
Corresponding author: Fabiana Silva Ribeiro, PhD. fabiana.ribeiro@uni.lu. University of Luxembourg, Department of Social Sciences, Esch-sur Alzette, Luxembourg.
Associations between Socioeconomic Status and Adverse Childhood Experiences with Multidimensional Healthy Aging: Findings from the ELSI-Brazil.
Abstract
Background
By 2050, the global population aged 65 years and older is projected to double, reaching 1.5 billion, with the most rapid growth occurring in Latin America and the Caribbean. In Brazil, this demographic shift is advancing quickly within a context marked by profound social inequalities and insufficient preparation to address the challenges of an aging society. For this reason, this study aimed to estimate the prevalence of healthy aging in a representative sample of the Brazilian population and explore the role of socioeconomic conditions and adverse childhood experiences.
Methods
A
We analyzed data from 9,908 participants aged 50 and older from the 2019–2021 follow-up of the Brazilian Longitudinal Study of Aging (ELSI-Brazil). Participants were classified into healthy and less healthy aging based on the World Health Organization’s multidimensional definition of healthy aging. Logistic regression models were employed to identify sociodemographic and early-life predictors of healthy aging.
Results
Only 19.69% of the sample met the employed criteria for healthy aging. Women, older individuals, those self-identifying as mixed race (compared to those self-identifying as White), participants with no schooling or fewer than four years of education, and widowed individuals were more likely to not meet the criteria for healthy aging than their counterparts. Logistic regressions revealed that men are more likely to show healthy aging, as well as younger respondents, those with five years or more of education, and participants receiving two or more minimum wages. With respect to childhood experiences, those reporting poor health and who reported famine during childhood were less likely to meet the criteria for healthy aging.
Conclusions
The findings suggest that not only socioeconomic factors but also childhood experiences contribute to disparities in healthy aging in Brazil. These results underscore the importance of implementing early-life health and nutrition programs, advancing gender equity, and improving access to education and economic resources throughout the life course to support healthier aging for future generations.
A
Clinical trial number:
not applicable
Keywords:
healthy aging
socioeconomic factors
inequalities
childhood experiences
A
A
Background
The United Nations predicts that by 2050, the global population aged 65 or older will double to 1.5 billion, with the fastest growth occurring in Latin America and the Caribbean. In Latin America, the complexities of aging are manifested in varied life expectancies across cities and countries because of different social environments (1), such as economic inequalities and access to health care, education, and others. In Brazil, the increase in the number of older people is occurring rapidly, yet in a society that is not well prepared for such a transition (2).
Nevertheless, it is important to mention that this demographic transformation is occurring in a country marked by profound socioeconomic inequalities, limited healthcare infrastructure (3, 4), and historically low levels of education among older adults (5). As such, Brazil offers a valuable context in which to investigate and understand how socioeconomic status and early-life adversity shape aging trajectories.
Healthy Aging: Evolving Perspectives
The concept of healthy aging has evolved significantly over the last few decades, challenging traditional views that associated aging solely with losses and declines. Rowe and Kahn’s model of usual and successful aging was introduced three decades ago, with successful aging being characterized by three factors: a low likelihood of illness and disability related to diseases, strong cognitive and physical capabilities, and active participation in social life (6, 7).
Nevertheless, an important criticism from the World Health Organization (8) regarding Rowe and Kahn’s model is that identifying individuals on the basis of only health, that is, the absence of diseases, is extremely problematic, since some diseases may be treated and not limit individual capacities. Therefore, the WHO (8) defined healthy aging as the process of cultivating and sustaining the ability to function effectively, ensuring a sense of well-being during old age.
In accordance with the WHO, recent studies have further broadened the understanding of healthy aging, recognizing it as a complex construct with multiple dimensions (9–11). In a recent umbrella review, the authors showed that studies related to healthy aging, which are mostly carried out in developed countries, used heterogeneous conceptualization and operationalization. In addition, they highlight the importance of assuming context-specific conceptual guidance to fill gaps in the operationalization of healthy aging, notably with respect to gender, disabilities, and ethnicity (9).
According to the WHO (8), a multidimensional model must consider intrinsic capacity, social and political environments, and the interaction of older adults with their surroundings. This multidimensional perspective aligns with evolving concepts of healthy aging and highlights the interplay of individual, social, and environmental factors in the aging process‎.
Building on this approach, recent systematic review studies have proposed operational models that reflect the complexity of aging (12). For example, Rivadeneira et al. (11) proposed a multidimensional framework that includes three components: (1) intrinsic capacity, comprising geriatric syndromes, physiological health, risk factors, i.e., alcohol consumption, smoking, lack of physical activity, cognitive functioning, well-being, i.e., depression, and physical capacity; (2) social and political environments, such as participation in the community; and (3) interactions with the environment, including assistance offered by them to others. This model was developed and tested on nationally representative data from Ecuador. The WHO defines healthy aging as intrinsic capacity combined with functional ability in the given environment of an individual, which Rivadaneira et al. (11) combined into one multidimensional score to indicate the presence or absence of healthy aging. Their results revealed that women and individuals with lower incomes were less likely to meet the criteria for healthy aging. Owing to similar levels of socioeconomic inequalities and similar social and economic profiles, this model is suitable for application in other countries, such as Brazil.
Healthy aging and inequalities in different health outcomes in Latin America
Despite the growing interest in healthy aging, research employing a multidimensional framework remains scarce in Latin America. While multidimensional healthy aging as a concept has rarely been used in the Latin American context, with exceptions such as Rivadeneira et al.(11), several studies point to low prevalence and socioeconomic gradients in older-age health outcomes in Latin America and the Caribbean (LAC) (13, 14).
A recent systematic review exploring gradients and inequalities in dementia prevalence in 15 LAC countries on the basis of age, sex, rurality, and education revealed an overall pooled prevalence of all-cause dementia of 10.66%. Women, individuals with lower educational levels, and rural residents were identified as particularly vulnerable groups with higher dementia prevalence. The study emphasized the need for targeted public health efforts to address the unequal burden of dementia in these populations (15). Furthermore, a recent study (16) comprising data from a Brazilian representative sample revealed a significant increase in health risk factors among the older adult population from 2000 to 2015, specifically, overweight/obesity, as well as diabetes and hypertension.
The review by Nitrini et al. (17) stressed the modifiable nature of factors contributing to dementia, offering hope for preventive actions, and highlighted the importance of education and socioeconomic status. Educational attainment emerged as a positive factor, aligning with previous studies highlighting education's role in accessing health services and improving overall quality of life. Conversely, economic status plays a pivotal role, with those in the worst economic situation being less likely to experience healthy aging, underscoring the importance of addressing socioeconomic inequalities (11). Likewise, living as a rural resident was found to be positively associated with a higher level of dementia, especially among women (18). Notably, childhood adversity, including hunger and poor access to education, has also been linked to worse aging outcomes (19). Fair or poor childhood health is also considered an adverse childhood experience, but there is little research on how this and other adverse childhood experiences impact healthy aging over the long term in contexts such as Brazil.
Gender, Social Roles, and the Aging Process
Gender represents another important, yet often underexplored, dimension in aging research in Latin America, particularly given that many LAC countries are characterized by high levels of gender inequality. Importantly, studies have shown that gender inequalities might shape the aging experience through social roles, cultural expectations, and structural disadvantages across the whole life course (20).
In Brazil, older women born in the 1940s and 1950s faced profound constraints on formal education, limited formal labor market participation (21), and restricted financial autonomy (22) due to prevailing patriarchal norms. In addition, labor policies and cultural messages have reinforced the role of women as caregivers and homemakers (23, 24).
As a consequence, these gendered social roles have had lasting implications, with women predominantly in low-paid, insecure work or unpaid domestic labor, entering retirement with insufficient pension benefits and financial reserves (25, 26). Nevertheless, the biomedical literature further highlights gender differences in morbidity, showing women's greater susceptibility to various health issues, even when reproductive conditions are excluded (27). This disadvantage persists up to older age, with men maintaining better functioning and lower disability rates than women do. Gender differences also manifest in the experience of specific diseases and psychological well-being, with women reporting higher levels of anxiety and stress (27) and higher levels of depression and loneliness (28). The association between age and health-related quality of life remains debated, with complex intersections of race, ethnicity, and gender influencing these dynamics (29).
Regarding early life adversity, greater disparity in educational opportunities during schooling was consistently linked to diminished cognitive performance scores in older individuals, particularly among women, in a European sample (30). Moreover, adverse childhood experiences, such as famine, can lead to chronic diseases later in life; for instance, Félix-Beltrán and Seixas (19) reported that childhood hunger was associated with diabetes and osteoporosis later in life on the basis of ELSI-Brazil data. However, they did not explore other childhood experiences on a multidimensional model of healthy aging. Since adversities can influence biological aging, it is crucial to understand how these early experiences contribute to health disparities.
A
Early life adversity together with gender stereotypes across the life course might lead to cumulative social and biological disadvantages that unfold over the life course. Importantly, these gendered pathways intersect with socioeconomic status and childhood experiences in unique ways in high-inequality contexts such as Brazil (20). From the social environment influencing mortality in Latin American cities to the unequal burden of dementia based on education and gender, understanding the gender differences in healthy aging requires a multidimensional and context-sensitive approach. Additionally, recognizing the role of socio-economic status and early life experiences in shaping the aging trajectory further underscores the need for holistic interventions to promote healthy aging and reduce disparities.
Current Research
Our objective was to explore the prevalence of healthy aging via a multidimensional framework in a nationally representative sample of older Brazilian adults. Furthermore, we also aim to examine the role of environmental elements such as socioeconomic status, education, and childhood experiences. Finally, we also extend our investigation by incorporating the sex/gender dimension, enriching our understanding of the interplay between these factors and healthy aging.
Methods
Participants
In this study, we used data from the population-representative Brazilian Longitudinal Study of Aging (ELSI-Brazil) (31). ELSI-Brazil has been carried out with adults aged 50 years and older residing in diverse communities across different regions of Brazil. The first wave of data collection occurred from 2015–2016, and the second wave was conducted between 2019 and 2021. For the present analysis, we used data from the most recent wave to estimate the current prevalence of healthy aging. This sample consisted of 9,949 adults aged 50 years or more, of whom 9,908 participants (Mage = 66; SD = 10,06; max. = 109 years) had complete information on the relevant variables and were included in the analyses.
The study sampling was conducted by conglomerates, and Brazilian Institute of Geography and Statistics (IBGE) data were used for stratification and selection of regions. To guarantee a comprehensive representation of urban and rural areas across municipalities of varying sizes, ELSI-Brazil employed a multistage sampling approach. This method involves stratifying primary sampling units (municipalities) and further selection stages, including census tracts and households. For more details regarding the sampling methodology, see (31). In-person assessments were conducted in Portuguese, the official language of Brazil, by trained interviewers.
Measures
Operationalization of Healthy Aging
A
According to the WHO (8), healthy aging is a multidimensional concept that includes an individual’s intrinsic capacity and functional ability to address given environmental factors. Given that there is no clear definition of which variables should comprise healthy aging components, we adapted the multidimensional model proposed by Rivadeneira et al. (11) on the basis of the variables available in the ELSI-Brazil database (see Table 01). The composite score of intrinsic capacity included domains such as physical health, geriatric syndromes, physical capacity, cognitive ability, psychological well-being, and environmental aspects, including social participation. Following the operationalization by Rivadeneira et al. (11), individuals were classified into a healthy aging group or a less healthy aging group, as described in Table 1. We summed the scores for each domain and then recoded the results into a dichotomous variable. The specific variables used for each component are described in detail below.
Physical health
To be considered healthy, participants were assessed on the basis of the absence of diabetes, hypertension, cardiovascular disease, stroke, lung disease, vision, and hearing problems or whether participants with the diagnosis received treatment, i.e., if the disease was managed.
However, owing to limitations in the dataset, we were unable to determine whether participants were receiving treatment or had limitations caused by arthritis, rheumatism, osteoporosis, renal insufficiency, Parkinson's disease, and Alzheimer's disease. For this reason, participants who reported any of these conditions were included in the less healthy aging group.
Physical health conditions were self-reported by the following question: “Has a doctor or nurse ever told you that you had...?”
Geriatric syndrome
This factor comprises three variables: 1) whether participants had polypharmacy (according to the World Health Organization, polypharmacy is the concomitant and routine use of 4 or more medications with or without a prescription); 2) whether participants self-reported urinary or fecal incontinence; and 3) whether the respondent self-reported a fall in the last 12 months.
Functional Capacity
Functional capacity was assessed by self-perceived inability to perform basic activities of daily living (BADL) via the Katz Index (32), adapted for Brazil by Lino et al. (33). Subjects who did not need help with any of the activities assessed were considered independent. Disabilities related to instrumental activities of daily living (IADL) were assessed via the Lawton Scale (34), adapted in Brazil by Lopes and Virtuoso-Júnior (35). The subjects were considered independent and had no difficulty except with heavy domestic activities.
Psychological well-being
To create the well-being variable, two scales were considered: 1) Life satisfaction, which is the following response instruction from the individual ELSI-Brazil questionnaire, which has, as response options, increasing measures from 1 to 10 in the form of a MacArthur Scale ladder: "Please think about your level of satisfaction with life and point to the corresponding rung". Answers of 6 or above were considered ‘high’ satisfaction, and answers below or equal to 5 were considered ‘low’ satisfaction.
Depressive symptoms were assessed via the CES-D8 scale (an eight-item version of the Center for Epidemiological Studies Depression Scale). The affirmative answers to the items describing depressive symptoms were added. The cutoff point for categorizing depression was ≥ 4, which was based on the criteria adopted by Sandy Junior (36).
Cognitive functioning
Cognitive functioning was measured by four tasks as follows:
A
1) In the first task, participants were required to recall a list of 10 words immediately after hearing them.
2) The second task involved recalling the same set of 10 words after a 5-minute delay, following the completion of other cognitive tests. In both tasks, participants received one point for each correct answer, with a maximum score of 10 points.
3) The third assessment evaluated participants' awareness of the date, including the day, month, and year, as well as the day of the week. Each accurate response earned one point, with a total possible score of four points.
4) The fourth measure, focusing on language and processing speed, required participants to recall the names of the animals within a 1-minute timeframe. The participants received one point for each correctly remembered animal name.
The four cognitive measures were z-standardized, averaged and restandardized to calculate a global z score with a mean of zero and a standard deviation of 1. Cognitive impairment was determined for a global z score lower than or equal to -1.5, which is equal to -1.5 SD from the overall mean.
Environment
The environment comprises many factors, including the extrinsic world that forms the context of an individual’s life. These include the microlevel (e.g., home) to the macrolevel (e.g., community and broader society). In this study, we included social participation and other activities (i.e., social, productive, and entertainment activities), as assessed by the Advanced Activities of Daily Living (AADL, (37)).
For the present study, the total number of activities that each participant reported doing from a total of 13 activities in the scale was calculated. The participants were classified into two groups according to their total score: more active, with greater participation in AADL, or less active. Those who performed four or more activities were considered more active, and those who reported performing three or fewer activities were considered less active.
Table 1
Criteria for Defining Healthy Aging
Dimension of healthy aging
Domain
Presence of Healthy Aging
Absence of healthy aging
Intrinsic capacity
Physical health
Absence of the following diagnoses or the presence of diseases that are managed (treated) and do not limit the functioning of the participant:
Diabetes, hypertension, cardiovascular disease, stroke, lung disease, vision and hearing problems. arthritis, rheumatism, osteoporosis, renal insufficiency, Parkinson's, and Alzheimer's.
Diagnoses with at least one of the mentioned diseases that are not controlled or are limiting the functioning of the participant.
 
Geriatric syndromes
Absence of polypharmacy, urinary and fecal incontinence, and no falls syndrome
Presence of the cited factors
 
Functional capacity
Absence of disability according to the Instrumental activities of daily living and Basic Activities of Daily Living
Physical activity
Presence of disabilities according to the cited tests.
 
Cognitive ability
Absence of cognitive impairment assessed by global Z score.
Presence of cognitive impairment, values of Z scores < = -1.5.
 
Psychological well-being
Global satisfaction with life and Absence of depression by using the Center for Epidemiological Studies Depression Scale
Presence of low satisfaction or/and symptoms of depression.
Environment
Social Participation
Advanced Activities of Daily Living.
Involvement in less than three Advanced Activities of Daily Living
Note: Variables adapted from Rivadeneira et al. (2021)
Covariates (Independent variables)
Socioeconomic characteristics included gender/sex (women/men), age groups (50–59 years, 60–69 years, 70–79 years, and 80 years of age or older), rurality (urban/rural), race (white, black, mixed), education levels (no school, 1–4 years, 5–8 years, 9–12 years, and 12 years or more), marital status (single, married/stable union, divorced/separated, widowed), and income calculated as multiples of the minimum wage (wage < 1, 1–2 wages, 2–3 wages, 3–4 wages, and wages > 4). The individual income was computed on the basis of the national minimum wage (NMW) applicable in the year of the interview.
Childhood experiences included three variables: health in childhood (excellent/very good, good, fair, and poor); hunger during childhood (no/yes); and living in rural areas until the age of 15 (no/yes).
Race categories were determined on the basis of participants' self-identification from a list of options aligning with the official classification of self-reported skin color in Brazil (Silva, 1997): white, brown or mixed, black, yellow, and indigenous. However, in this study, we included only participants who self-reported as white, black, or mixed since there were not enough cases in the other categories to fit the model.
Statistical analysis
First, we carried out descriptive analyses to observe the percentage distributions of all the study variables. Rao Scott's chi-square tests were subsequently used to assess potential differences in socioeconomic and childhood experiences between the healthy and less healthy aging groups.
Second, we ran logistic regression analyses to assess which independent variables influence healthy aging, which was treated as a dichotomous dependent variable. The reference group was the healthy aging group.
In addition, we included sampling weights in all analyses to adjust the complex sampling design of the ELSI-Brazil.
STATA (release 17, Stata Corp.) software was used for the statistical analyses.
Results
A
As shown in Table 2, 59.33% of the study participants were women. In terms of age, 65.24% were between 50 and 69 years old. Most participants lived in urban regions (83.8%), had 1–4 years of education (40.51%), and had up to 2 wages as income (57.03%). For the prevalence of healthy aging according to the adapted criteria, only 19.69% of the respondents were categorized as healthy aging, whereas 80.31% were categorized as less healthy aging.
Table 2
Socioeconomic and childhood characteristics of the study sample (N = 9,908).
Variables
Number of Participants
%
Gender/Sex
  
Women
5,819
59.33
Men
3,989
40.67
Age Groups
  
50–59 years
2,984
30.42
60–69 years
3,415
34.82
70–79 years
2,252
22.96
80 years or more
1,157
11.80
Rurality
  
Urban
8,219
83.80
Rural
1,589
16.20
Race
  
White
4,587
46.77
Black
1,052
10.73
Mixed
4,169
42.51
Education
  
no schooling
1,552
16.02
1–4 years
3,924
40.51
5–8 years
1,854
19.14
9–12 years
1,705
17.6
12 years or more
651
6.72
Marital Status
  
Single
1,195
12.18
Married/Stable union
5,212
53.14
Divorced/Separated
1,229
12.53
Widowed
2,172
22.15
Income
  
Wage < 1
1,905
20.50
1–2 wages
3,394
36.53
2–3 wages
1,714
18.45
3–4 wages
874
9.41
Wages > 4
1,405
15.12
Brazilian Regions
  
North
709
7.23
Northeast
2,659
27.11
Southeast
4,060
41.39
South
1,331
13.57
Midwest
1,049
10.70
Childhood experiences
Health in childhood
Excellent/Very good
2,236
22.96
Good
5,758
59.12
Fair
1,240
12.73
Poor
506
5.20
Hunger during childhood
  
No
7,293
75.41
Yes
2,378
24.59
Lived in rural areas until the age of 15
 
No
4,200
42.82
Yes
5,577
56.86
Healthy classification
  
Healthy
1,931
19,69
Less healthy
7,877
80,31
Table 2
Table 3 shows that there was a greater prevalence of healthy aging among men (55.1%) than among women (42.8%). Participants who did not meet the criteria for healthy aging were more likely to be older and have no schooling or up to four years of education. Moreover, participants reporting being widowed were more likely to be less healthy. Those receiving more than three wages a month were more likely to meet the criteria for healthy aging. With respect to childhood experiences, those who self-reported poor health in childhood and who experienced hunger during childhood were more likely to be allocated to the less healthy aging group.
Table 3
Weighted Analyses of Socioeconomic and Childhood Characteristics across Aging Health Categories.
Variables
Healthy
Less healthy
p
Gender/Sex
   
Women
44.9 (41.9–47.9)
57.2 (55.7–58.7)
< 0 .001*
Men
55.1 (52.1–58.1)
42.8 (41.3–44.3)
Age Groups
   
50–59 years
58.4 (55.1–61.5)
44.2 (42.1–46.2)
< 0 .001*
60–69 years
28.6 (26.7–30.5)
29.2 (28.1–30.4)
70–79 years
10.7 (9.3–12.2)
17.6 (16.5–18.6)
80 years or more
2.4 (1.7–3.2)
9.0 (8.3–9.8)
Rurality
   
Urban
87.9 (79.8–93.0)
83.4 (75.7–89.0)
0.02*
Rural
12.1(7.0–20.2)
16.6(11.0–24.3)
Race
   
White
51.2 (46.6–55.9)
45.3 (40.4–50.2)
< 0 .001*
Black
8.8(6.9–11.2)
11.5(9.7–13.6)
Mixed
39.9(36.0–44.0)
43.2 (39.2–47.4)
Education
   
no school
6.4(5.0–8.2)
14.4 (10.4–19.8)
< 0 .001*
1–4 years
29.5(23.6–36.3)
41.1 (37.6–44.7)
5–8 years
22.7(19.5–26.4)
19.9 (17.1–23.0)
9–12 years
29.6 (25.9–33.7)
18.5 (15.0–22.5)
12 years or more
11.6 (9.7–13.9)
6.1(4.4–8.3)
Marital Status
   
Single
12.1(10.4–14.1)
12.8 (11.2–14.7)
< 0 .001*
Married/Stable union
66.6 (62.7–70.2)
59.0 (56.5–61.4)
Divorced/Separated
11.7 (9.2–14.7)
10.9 (9.7–12.3)
Widowed
9.7(7.8–11.9)
17.3 (16.0–18.6)
Income
   
Wage < 1
13.6 (11.1–16.5)
20.2 (17.4–23.3)
< 0 .001*
1–2 wages
26.0 (21.6–31.0)
36.7 (34.6–38.8)
2–3 wages
20.7 (17.8–23.8)
18.6 (16.6–20.8)
3–4 wages
12.5 (9.6–16.0)
9.7 (8.3–11.3)
Wages > 4
27.3 (24.0–30.8)
14.8 (12.3–17.7)
Childhood experiences
   
Health in childhood
   
Excellent/Very good
26.7 (21.4–32.8)
23.1 (20.8–25.6)
< 0 .001*
Good
59.1 (54.0–64.0)
56.7 (54.9–58.6)
Fair
11.6 (10.2–13.2)
13.9 (12.8–15.1)
Poor
2.6 (1.8–3.6)
6.3 (5.6–7.0)
Hunger during childhood
   
No
80.6 (76.9–83.9)
73.7 (71.0–76.3)
< 0.001*
Yes
19.4(16.1–23.1)
26.3 (23.7–29.0)
Lived in rural areas until the age of 15
   
No
50.0 (37.0–63.1)
43.1 (33.3–53.4)
0.007*
Yes
49.9 (36.9–62.9)
56.7 (46.3–66.6)
Note: Weighted P values were determined via the Rao‒Scott test. Differences among groups were confirmed by nonoverlapping confidence intervals.
* P < .05
Rao-Scott analyses were carried out whether the experience of hunger during childhood was distributed differently across Brazilian regions. Participants living in the North (31.4%; I -23.7-40.3) and Northeast (33.4%; CI -29.1-38.0) regions were more likely to report famine during childhood compared to participants with residence in the Southeast of the country (20.4%; CI -18.0-23.0), p < .001.
Table 3
Table 4 displays the results of the logistic regression analyses with healthy aging as a dependent variable that showed that men were more likely to belong to the healthy aging group than women. Furthermore, older participants, especially those aged 80 or more, were more likely to be in the less healthy aging group. Older adults having five years or more of education were more likely to be healthily aging. Regarding childhood experiences, we detected that respondents who self-reported fair or poor health during childhood and suffered from hunger were less likely to meet the criteria for healthy aging.
Table 4
Associations between Healthy Aging, Socioeconomic Status, and Childhood Factors
Variables
Odds Ratio
95% CI
p
Gender/Sex
   
Women (Reference)
   
Men
0.63
0.56–0.71
< 0.001*
Age Groups
   
50–59 years (Reference)
   
60–69 years
1.29
1.11–1.51
0.002*
70–79 years
1.92
1.62–2.26
< 0.001*
80 years or more
4.08
2.91–5.71
< 0.001*
Rurality
   
Urban (Reference)
   
Rural
1.08
0.77–1.47
0.68
Race
   
White (Reference)
   
Black
1.19
0.93–1.52
0.15
Mixed
1.12
0.97–1.29
0.12
Education
   
no school (Reference)
   
1–4 years
0.85
0.55–1.31
0.45
5–8 years
0.62
0.43–0.88
0.008*
9–12 years
0.52
0.33–0.82
0.006*
12 years or more
0.45
0.25–0.80
0.008*
Marital Status
   
Single (Reference)
   
Married/Stable union
0.91
0.75–1.08
0.35
Divorced/Separated
0.84
0.53–1.34
0.45
Widowed
0.94
0.77–1.13
0.50
Income
   
Wage < 1 (reference)
   
1–2 wages
0.98
0.78–1.23
0.85
2–3 wages
0.69
0.57–0.83
< 0.001*
3–4 wages
0.63
0.48–0.82
< 0.001*
Wages > 4
0.53
0.43–0.66
< 0.001*
Childhood experiences
   
Health in childhood
   
Excellent/Very good (Reference)
   
Good
0.92
0.78–1.09
0.34
Fair
1.31
1.01–1.69
0.04*
Poor
2.35
1.66–3.33
< 0.001*
Hunger during childhood
   
No (Reference)
   
Yes
1.25
1.07–1.46
0.007*
Lived in rural areas until the age of 15
   
No (Reference)
   
Yes
0.89
0.77–1.03
0.12
* P < .05
Table 4
Discussion
This study investigated the prevalence of healthy aging in a nationally representative sample of older adults in Brazil and explored the effects of socioeconomic conditions and childhood experiences in explaining the health of older adults. With a more comprehensive and thus less narrow view of healthy aging, considering multiple dimensions and whether diseases were treated or controlled, our findings suggest that more than four-fifths (80.3%) of our sample did not meet the criteria for healthy aging, suggesting that a substantial proportion of older Brazilian people must cope with many health challenges as they age.
Our findings on a Brazilian population-representative sample contrast with the findings of Rivadeneira et al. (11) for a sample from Ecuador, roughly half of which were found to be aging healthily (53.15% of the sample), even though their sample was aged 65 years or older, in contrast with our data from ELSI-Brazil, which comprised adults aged 50 and older. A more favorable socioeconomic profile of the Ecuadorian sample may have contributed to the differences in prevalence estimates. Furthermore, period effects may be partly responsible for the differences in findings; the data of the SABE Ecuador study of Rivadeneira et al., 2021 were collected in 2010, whereas the present study used 2019–2021 data from ELSI-Brazil. Similarly, earlier research revealed sharp increases in health risk factors such as obesity in Brazil over the period 2000–2015 (16). Finally, differences in socioeconomic conditions and healthcare access may have played a role, as Brazil has historically exhibited greater inequality and regional disparities (21).
On the basis of our data, older individuals with four years of formal education or less and household incomes lower than two wages are much less likely to meet the criteria for healthy aging, suggesting that these groups are particularly vulnerable. These findings align with prior research in Latin America and globally, which suggests that socioeconomic disadvantages accumulate across the life course and manifest in worse health outcomes in old age (14–16, 38), similar to the findings of other studies in other regions of the world (39, 40). The strong links between formal education and healthy aging are suggested to come from increased exposure to various risk factors; lower resources in terms of money, knowledge, prestige, power, and beneficial social connections (41, 42); higher stress levels; other biological mechanisms of ‘embodiment’ (43); and possibly limited access to (preventive) healthcare services starting from a younger age.
As discussed by previous studies (15, 20), gender inequality might lead to pronounced sex/gender differences in healthy aging, as men tend to be highly educated, perform cognitively demanding jobs, and are offered more strategies and opportunities to cope with the adverse external environment. Additionally, men and women who were born in the 1940s and 1950s played very different social roles, which deferentially impacted their mental health during their lifespan, with women being more likely to experience negative outcomes in mental health (28). Women also live longer, meaning that they spend more time coping with stressful life transitions, such as losing a spouse, caring for other people, and living with their own chronic illnesses (44).
Moreover, our results reinforce the impact of famine during childhood on healthy aging shown by Félix-Beltrán and Seixas (19). Self-reported lower health during childhood is associated with less healthy aging, given that a lack of nutrients during childhood could lead to epigenetic changes leading to metabolic dysfunction. These outcomes may reflect the long-term biological consequences of early nutritional deprivation and stress, including epigenetic changes linked to chronic disease later in life (45). Moreover, such early-life adversity was more commonly reported by participants from Brazil’s North and Northeast Regions, areas historically marked by higher poverty levels and food insecurity, underscoring the geographical dimension of inequality in aging trajectories.
Interestingly, no differences between groups were observed in urban versus rural residences. This suggests that environmental factors associated with locality, such as pollution or infrastructure, may be less predictive of healthy aging than broader structural inequalities, such as income, education, and life-course exposures. Future research could explore more nuanced aspects of the built and social environment, such as neighborhood safety, healthcare accessibility, and pollution exposure.
Strengths and limitations
A major strength of this study was the operationalization of a multidimensional framework to assess healthy aging, which is consistent with recent recommendations from the WHO (8) and prior empirical work (11) and provides a more holistic view of the intrinsic capacity of older adults in Brazil. Another strength is the use of recent nationally representative data from ELSI-Brazil, with a multistage sampling process and the use of weights to arrive at population-representative estimates.
However, limitations should be acknowledged. First, owing to differences in data collection, we had to slightly adapt the operationalization of intrinsic capacity used by Rivadeneira et al. (11). While some dimensions of healthy aging are more standardized, e.g., the classification of polypharmacy or the presence of chronic conditions, other dimensions are based on conventions that are sometimes less stringently used in the literature, such as the classification as cognitively impaired, with a score of 1.5 SD below the sample mean. The prevalence of healthy aging consequently is sensitive to changes in the operationalization of the different dimensions, including possible underreporting of chronic conditions, due to a lack of diagnosis, which may have led to an overestimation of healthy aging. Additionally, the domains constituting healthy aging may have differential importance, and different operationalizations exist in the literature; however, other studies have used country-specific operationalizations of intrinsic capacity as well (46), and we defined the concept as congruent as possible with existing studies in comparable contexts (9). Second, not all determinants of healthy aging found in earlier research were available in the data used from the ELSI-Brazil. Third, our estimates of the socioeconomic determinants of healthy aging may be impacted by premature mortality, which is still rather high in the Brazilian context (47). This selective attrition may have biased our estimates toward the null.
Policy implications
Gender-responsive policies are essential, as older women in Brazil face a double burden of socioeconomic disadvantage and greater longevity, leading to long periods of managing chronic illness while still fulfilling important family functions such as caregiving responsibilities. In this sense, some policy recommendations based on our findings might include macrolevel interventions that prioritize initiatives to promote access to education for future generations and support those in less favorable economic situations, optimizing the social and psychological environment, fostering a supportive social atmosphere through media campaigns, and promoting an appropriate understanding of family development laws (48). Moreover, community care for older people is crucial, especially among women, who represent the larger proportion of older adults in Brazil, emphasizing preventive services for their physical and mental health and the establishment of an effective old-age security system (48).
Encouraging the development of human resources among older people involves promoting their participation in social activities, amplifying their network of social support, and leveraging their knowledge in community initiatives (48). In addition, childhood health and nutrition must be core components of aging policy, given the long-term impact of early adversity on later-life outcomes. Investments in maternal and child health, nutrition programs, and poverty alleviation can have intergenerational effects on healthy aging (20).
On a micro level, mastering skills to understand the psychological functioning of older women is crucial, emphasizing respectful communication and active listening to promote happiness and confidence (48). Additionally, addressing the prevalence of dementia is imperative, considering its impact on dependence among older adults and the substantial medical and care costs involved (49–52). Finally, while this study did not directly assess loneliness or social participation, our findings still suggest that psychosocial conditions, such as widowhood and poor mental health, are important domains for policy intervention. However, the central policy message should not be about promoting social participation per se, as structural influences shape downstream conditions and opportunities for social participation. Thus, addressing the structural inequalities that shape healthy aging trajectories from birth to old age is essential.
Conclusion
This study operationalized healthy aging with a comprehensive and multidimensional score. A large majority (80%) of respondents in a population-representative sample of the older Brazilian population did not meet the criteria for healthy aging. Our findings highlight substantial gaps in health equity by gender, socioeconomic status, and childhood adversity across aging individuals. As Brazil continues to age, ensuring that longer lives are accompanied by better health will require integrated, equity-focused public policies. Without addressing the root causes of inequality across the life course, gains in life expectancy may not translate into gains in health and quality of life for most of the population.
References
1.
Bilal U, Hessel P, Perez-Ferrer C, Michael YL, Alfaro T, Tenorio-Mucha J, et al. Life expectancy and mortality in 363 cities of Latin America. Nat Med. 2021;27(3):463–70.
2.
Kirby T. Brazil facing ageing population challenges. Lancet. 2023;402(10415):1821.
3.
de Barros RP, de Carvalho M, Franco S, de Mendonça RSP. Determinantes imediatos da queda da desigualdade de renda brasileira. 2007 Jan [cited 2025 Oct 1]; Available from: http://repositorio.ipea.gov.br/handle/11058/2161
4.
Pimentel FG, Buchweitz C, Onocko Campos RT, Hallal PC, Massuda A, Kieling C. Realising the future: Health challenges and achievements in Brazil. SSM - Ment Health. 2023;4:100250.
5.
Travassos GF, Coelho AB, Arends-Kuenning MP. The elderly in Brazil: demographic transition, profile, and socioeconomic condition. Rev Bras Estud Popul. 2020;37:e0129.
6.
Rowe JW, Kahn RL. Human Aging: Usual and Successful. Science [Internet]. 1987 July 10 [cited 2025 Oct 1]; Available from: https://www.science.org/doi/10.1126/science.3299702
7.
Rowe JW, Kahn RL. Successful Aging1. Gerontologist. 1997;37(4):433–40.
8.
WHO. World report on ageing and health [Internet]. 2015 [cited 2025 Oct 1]. Available from: https://iris.who.int/items/5db2c88f-941a-472d-ac6e-f08a7e9ae20b
9.
Menassa M, Stronks K, Khatami F, Díaz ZMR, Espinola OP, Gamba M et al. Concepts and definitions of healthy ageing: a systematic review and synthesis of theoretical models. eClinicalMedicine [Internet]. 2023 Feb 1 [cited 2025 Oct 1];56. Available from: https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(22)00550-8/fulltext
10.
Michel JP, Sadana R. Healthy Aging Concepts and Measures. J Am Med Dir Assoc. 2017 June;1(6):460–4.
11.
Rivadeneira MF, Mendieta MJ, Villavicencio J, Caicedo-Gallardo J, Buendía P. A multidimensional model of healthy ageing: proposal and evaluation of determinants based on a population survey in Ecuador. BMC Geriatr. 2021;21(1):615.
12.
Piriu AA, Bufali MV, Cappellaro G, Compagni A, Torbica A. Conceptualisation and measurement of healthy ageing: Insights from a systematic literature review. Soc Sci Med. 2025 June;1:374:117933.
13.
Castillo-Riquelme M, Yamada G, Diez Roux AV, Alfaro T, Flores-Alvarado S, Barrientos T, et al. Aging and self-reported health in 114 Latin American cities: gender and socio-economic inequalities. BMC Public Health. 2022;22(1):1499.
14.
Santamaria-Garcia H, Sainz-Ballesteros A, Hernandez H, Moguilner S, Maito M, Ochoa-Rosales C, et al. Factors associated with healthy aging in Latin American populations. Nat Med. 2023 Sept;29(9):2248–58.
15.
Ribeiro FS, Teixeira-Santos AC, Leist AK. The prevalence of mild cognitive impairment in Latin America and the Caribbean: a systematic review and meta-analysis. Aging Ment Health [Internet]. 2022 Aug 24 [cited 2025 Oct 1]; Available from: https://www.tandfonline.com/doi/abs/10.1080/13607863.2021.2003297
16.
Ribeiro FS, de Oliveira Duarte YA, Santos JLF, Leist AK. Changes in prevalence of cognitive impairment and associated risk factors 2000–2015 in São Paulo, Brazil. BMC Geriatr. 2021;21(1):609.
17.
Nitrini R, Barbosa MT, Dozzi Brucki SM, Yassuda MS, Caramelli P. Current trends and challenges on dementia management and research in Latin America. J Glob Health. 2020;10(1):010362.
18.
Prina AM, Stubbs B, Veronese N, Guerra M, Kralj C, Llibre Rodriguez JJ, et al. Depression and Incidence of Frailty in Older People From Six Latin American Countries. Am J Geriatr Psychiatry Off J Am Assoc Geriatr Psychiatry. 2019;27(10):1072–9.
19.
Félix-Beltrán L, Seixas BV. Childhood hunger experiences and chronic health conditions later in life among Brazilian older adults. Rev Panam Salud Publica Pan Am J Public Health. 2021;45:e39.
20.
Ribeiro FS, Crivelli L, Leist AK. Gender inequalities as contributors to dementia in Latin America and the Caribbean: what factors are missing from research? Lancet Healthy Longev. 2023 June 1;4(6):e284–91.
21.
World Bank. Inequality and Economic Development in Brazil [Internet]. Ferreira FHG, Velez CE, Paes. De Barros R, editors. The World Bank; 2004 [cited 2025 Oct 1]. (Country Studies). Available from: https://documents.worldbank.org/en/publication/documents-reports/documentdetail/456611468744062519/Inequality-and-economic-development-in-Brazil
22.
Gacitua-Mario E, Woolcock M, editors. Social Exclusion and Mobility in Brazil [Internet]. The World Bank; 2008 [cited 2025 Oct 1]. Available from: https://documents.worldbank.org/en/publication/documents-reports/documentdetail/551501468227638533/social-exclusion-and-mobility-in-brazil
23.
Berquó E. Considerações sobre o envelhecimento da população no Brasil. In: Velhice e Sociedade. Papirus; 1999.
24.
Nicodemo D, Godoi MP. Youth in the 1960’s-70’s and elderly: case study on feminization and rights of elderly women. Rev Ciênc Em Ext. 2010;6(1):40–53.
25.
OECD. Towards Improved Retirement Savings Outcomes for Women [Internet]. 2021 Mar [cited 2025 Oct 1]. Available from: https://www.oecd.org/en/publications/towards-improved-retirement-savings-outcomes-for-women_f7b48808-en/full-report.html
26.
Garnick D. Gender Retirement Gap [Internet]., Rochester NY. Social Science Research Network; 2016 [cited 2025 Oct 1]. Available from: https://papers.ssrn.com/abstract=2888911
27.
Carmel S. Health and Well-Being in Late Life: Gender Differences Worldwide. Front Med [Internet]. 2019 Oct 10 [cited 2025 Oct 1];6. Available from: https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2019.00218/full.
28.
Ferreira HG. Gender Differences in Mental Health and Beliefs about Covid-19 among Elderly Internet Users. Paid Ribeirão Preto. 2021;31:e3110.
29.
Lee KH, Xu H, Wu B. Gender differences in quality of life among community-dwelling older adults in low- and middle-income countries: results from the Study on global AGEing and adult health (SAGE). BMC Public Health. 2020;20(1):114.
30.
Leist AK, Bar-Haim E, Chauvel L. Inequality of educational opportunity at time of schooling predicts cognitive functioning in later adulthood. SSM - Popul Health. 2021 Sept 1;15:100837.
31.
Lima-Costa MF, de Andrade FB, de Souza PRB, Neri AL, Duarte YA, de O, Castro-Costa E et al. The Brazilian Longitudinal Study of Aging (ELSI-Brazil): Objectives and Design. Am J Epidemiol. 2018 July 1;187(7):1345–53.
32.
Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. STUDIES OF ILLNESS IN THE AGED. THE INDEX OF ADL: A STANDARDIZED MEASURE OF BIOLOGICAL AND PSYCHOSOCIAL FUNCTION. JAMA 1963 Sept 21;185:914–9.
33.
Lino VTS, Pereira SRM, Camacho LAB, Ribeiro Filho ST, Buksman S. Adaptação transcultural da Escala de Independência em Atividades da Vida Diária (Escala de Katz). Cad Saúde Pública. 2008;24:103–12.
34.
Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9(3):179–86.
35.
dos Santos RL, Júnior JSV. Confiabilidade da versão brasileira da escala de atividades instrumentais da vida diária. Rev Bras Em Promoção Saúde. 2008;21(4):290–6.
36.
Sandy Júnior PA, Borim FSA, Neri AL. Solidão e sua associação com indicadores sociodemográficos e de saúde em adultos e idosos brasileiros: ELSI-Brasil. Cad Saúde Pública. 2023;39:e00213222.
37.
Reuben DB, Laliberte L, Hiris J, Mor V. A hierarchical exercise scale to measure function at the Advanced Activities of Daily Living (AADL) level. J Am Geriatr Soc. 1990;38(8):855–61.
38.
Alvarado BE, Zunzunegui MV, Béland F, Bamvita JM. Life Course Social and Health Conditions Linked to Frailty in Latin American Older Men and Women. J Gerontol Ser A. 2008;63(12):1399–406.
39.
Kollia N, Caballero FF, Sánchez-Niubó A, Tyrovolas S, Ayuso-Mateos JL, Haro JM, et al. Social determinants, health status and 10-year mortality among 10,906 older adults from the English longitudinal study of aging: the ATHLOS project. BMC Public Health. 2018;18(1):1357.
40.
Zhao B, Liu Z, Fu Y, Zhang H, Wu J, Lai C, et al. Social Determinants of Intrinsic Capacity: A National Cohort Study. Am J Prev Med. 2024;66(3):559–67.
41.
Masters RK, Link BG, Phelan JC. Trends in education gradients of ‘preventable’ mortality: A test of fundamental cause theory. Soc Sci Med. 2015;127:19–28.
42.
Phelan JC, Link BG, Tehranifar P. Social Conditions as Fundamental Causes of Health Inequalities: Theory, Evidence, and Policy Implications. J Health Soc Behav [Internet]. 2010 Oct 8 [cited 2025 Oct 1]; Available from: https://journals.sagepub.com/doi/10.1177/0022146510383498
43.
Krieger N. Embodiment: a conceptual glossary for epidemiology. J Epidemiol Community Health. 2005;59(5):350–5.
44.
Languirand M. Who I was, who I am: Gender and generativity in the assessment of older adults. Handbook of gender and sexuality in psychological assessment. New York, NY, US: Routledge/Taylor & Francis Group; 2016. pp. 578–602.
45.
Abeliansky AL, Strulik H. Hungry children age faster - ScienceDirect. Economics & Human Biology [Internet]. 2018 [cited 2025 Oct 1]; Available from: https://www.sciencedirect.com/science/article/pii/S1570677X17302757?via%3Dihub
46.
Beard JR, Hanewald K, Si Y, Amuthavalli Thiyagarajan J, Moreno-Agostino D. Cohort trends in intrinsic capacity in England and China. Nat Aging. 2025;5(1):87–98.
47.
de Azeredo Passos VM, Champs APS, Teixeira R, Lima-Costa MFF, Kirkwood R, Veras R, et al. The burden of disease among Brazilian older adults and the challenge for health policies: results of the Global Burden of Disease Study 2017. Popul Health Metr. 2020 Sept;30(1):14.
48.
Sarkar SM, Dhar BK, Crowley SS, Ayittey FK, Gazi MAI. Psychological Adjustment and Guidance for Ageing Urban Women. Ageing Int. 2023;48(1):222–30.
49.
Custodio N, Duque L, Montesinos R, Alva-Diaz C, Mellado M, Slachevsky A. Systematic Review of the Diagnostic Validity of Brief Cognitive Screenings for Early Dementia Detection in Spanish-Speaking Adults in Latin America. Front Aging Neurosci. 2020;12:270.
50.
Ibáñez A, Pina-Escudero SD, Possin KL, Quiroz YT, Peres FA, Slachevsky A, et al. Dementia caregiving across Latin America and the Caribbean and brain health diplomacy. Lancet Healthy Longev. 2021;2(4):e222–31.
51.
Parra MA, Baez S, Allegri R, Nitrini R, Lopera F, Slachevsky A, et al. Dementia in Latin America. Neurology. 2018;90(5):222–31.
52.
Prince MJ, Wu F, Guo Y, Gutierrez Robledo LM, O’Donnell M, Sullivan R, et al. The burden of disease in older people and implications for health policy and practice. Lancet Lond Engl. 2015;385(9967):549–62.
List of abbreviations
LAC
Latin America and the Caribbean
WHO
World Health Organization
ELSI
Brazil-Brazilian Longitudinal Study of Aging
IBGE
Brazilian Institute of Geography and Statistics
BADL
Basic Activities of Daily Living
IADL
Disabilities related to instrumental activities of daily living
CES
D8-Center for Epidemiological Studies Depression Scale
NMW
National minimum wage
Ethics approval and consent to participate
The ELSI-Brazil study was approved by the Ethics Committee of the Oswaldo Cruz Foundation-Minas Gerais, and the process is registered on Plataforma Brasil (CAAE: 34649814.3.0000.5091).
A
The participants signed separate informed consent forms for each of the research procedures and authorized access to corresponding secondary databases.
A
This study received ethics approval from the Ethics Review Committee of the ERC in November 2018.
Consent for publication
Not applicable
A
Data Availability
The data supporting the findings of this study are available from the ELSI-Brazil repository [http://elsi.cpqrr.fiocruz.br]. Access can be obtained upon reasonable request and with permission from the ELSI-Brazil coordination.
Competing interests
The authors declare that they have no competing interests.
A
Funding
ELSI-Brazil was supported by the Brazilian Ministry of Health: DECIT/SCTIE (Grants: 404965/2012-1 and TED 28/2017) and COPID/DECIV/SAPS (Grants: 20836, 22566, 23700, 25560, 25552, and 27510).
A
Moreover, part of this work was supported by the National Council for Scientific and Technological Development (CNPq; Novation Process: 229520/2013-8), and it is part of the CRISP project funded by the European Research Council (ERC; grant agreement no. 803239).
A
Author Contribution
Research design and data analysis by FR; FR, GB, AL wrote the article and were responsible for the final content; and FR, GB, HF, and AL assisted in the interpretation of the results and critical revision of the manuscript. All the authors contributed substantially to the development of the manuscript and approved its final version.
Acknowledgments
The authors express their gratitude to all the researchers, interviewers, and participants of ELSI-Brazil.
Yes
Total words in MS: 5777
Total words in Title: 16
Total words in Abstract: 318
Total Keyword count: 4
Total Images in MS: 0
Total Tables in MS: 4
Total Reference count: 52