Racial inequities and biopsychosocial indicators in older adults

Abstract Objective to analyze the association of self-reported skin color/race with biopsychosocial indicators in older adults. Method cross-sectional study conducted with a total of 941 older adults from a health micro-region in Brazil. Data were collected at home with instruments validated for the country. Descriptive analysis and binary, multinomial and linear logistic regression (p<0.05) were performed. Results Most older adults were self-declared white color/race (63.8%). Black color/race was a protective factor for negative (OR=0.40) and regular (OR=0.44) self-rated health perception and for the indicative of depressive symptoms (OR=0.43); and it was associated with the highest social support score (β=3.60) and the lowest number of morbidities (β=-0.78). Conclusion regardless of sociodemographic and economic characteristics, older adults of black color/race had the best outcomes of biopsychosocial indicators.


Introduction
In Brazil, based on the 1991 demographic census, the Brazilian Institute of Geography and Statistics (IBGE) adopted the criterion of self-classification according to skin color/race into five categories: white, black, brown, indigenous and yellow (1) . The racial composition of the older adult population in the country has been changing over the years. In the year 2000, 61.7% of Brazilian older adults declared themselves white, 29.5% brown and 6.9% black (2) and, in the year 2015, there was a decrease in the proportion of white color/race older adults (52. 2 %) and an increase in browns (37.4%) and blacks (9.2%) (1) . However, the changes that have taken place in the country in the economic, political and health sphere have not yet mitigated the inequalities in the health conditions of racial groups of Brazilian older adults (3)(4) .
Racial inequities in health result from inappropriate living habits and access to social and health resources (3)(4)(5) .
In this context, color/race is seen as a marker of social position (4) which reflects on the distinct distribution of risk, protection and health hazards that accumulate throughout life (3) .
In national (3)(4) and international (6)(7)(8) studies, it was observed that health and social conditions differed among white, brown and black older adults. However, in other studies it was found, regardless of sociodemographic and economic differences and the context in which individuals are inserted, that health inequities cannot be directly attributed to color/race (9) . In this perspective, we sought to investigate this theme, with the aim of expanding knowledge and subsidizing clinical practice in the care of older adults.
It should be noted that ethnic or racial health disparities in older adults have been widely studied in developed countries, especially in the United States of America (USA) (11) . Although Brazil is a nation with ethnoracial diversity, scientific knowledge about biopsychosocial indicators among older adults with a racial background is still incipient (3)(4) , which makes it difficult to understand the influence of racial inequities at this stage of life.
Thus, the objective was to analyze the association of self-reported skin color/race with the biopsychosocial indicators of older adults.

Study design
Household, analytical and cross-sectional survey carried out in the urban area of a micro-health region in the state of Minas Gerais located in southwestern Brazil.
This study was developed in accordance with the Checklist for Reporting Results of Internet E-Surveys guidelines and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) for cross-sectional studies (12) .

Population and sample
The population consisted of individuals aged 60 years old or more living in the urban area. The multiple-stage cluster sampling technique was used to define the sample.

Explanatory and adjustment variables
Sociodemographic and economic data were obtained through the application of a structured questionnaire, developed and widely used by the researchers of this study, which includes the following information: gender (male and female); age group, in years old (60 to 69, 70 to 79 and 80 or more) and age (numerical variable); marital status (never married, married, widowed and divorced/separated); level of education, in years of study (no education, 1 to 3, 4 to 7 years and 8 or more) and education, in years of study (numerical variable); and monthly individual income, in minimum wages (without income, up to 1, 1 to 3, 4 and more).

Independent variable -color/race
The self-reported classification of skin color/race (white, black, brown and yellow) was used as defined in the country's demographic census (1) . Data were obtained through the question: How do you classify your skin color/race?

Dependent variables -biopsychosocial indicators
The frailty syndrome was assessed using the five components of the frailty phenotype (14)  Older adults with scores of two or three on any of the questions met the frailty criterion for this item (15)  Questionnaire (IPAQ), adapted for older adults (16) . The classification used for this component considered active those who spent 150 min or more of weekly physical activity; and inactive people who spent 0 to 149 min of weekly physical activity (17) . Older adults with three or more of the items described above are classified as frail, those with one or two items as pre-frail and those with all negative tests as robust or non-frail (14) .
The number of self-reported morbidities and selfrated health was measured by applying the instrument developed by the study researchers. The self-rated health variable was classified into three categories: very positive/ positive, regular and negative/very negative. The number of morbidities was considered a numerical variable.
The Basic Activities of Daily Living (BADL) were measured using the Katz Index, adapted to the Brazilian context and composed of six items that measure the individual's performance in self-care activities (18) . For each item, there are three possible answers, the first and second denoting independence and the third one, dependence (18) . For Instrumental Activities of Daily Living (IADL), we used the Lawton & Brody Scale (1969), adapted in Brazil (19) . This scale is composed of nine items that have three answer alternatives for each question: independence, need for partial help and need for full help/ cannot perform the activity. Based on these instruments, older adults were classified as independent or dependent for BADL and IADL.
To measure physical performance, we used the Brazilian version of the Short Physical Performance Battery (SPPB), consisting of the sum of the scores acquired in the tests of balance, gait speed and standing up from a chair five consecutive times and with a total score that varies from 0 (disability) to 12 (better performance), that is, the highest score represents a better physical performance (20) .
Depressive symptoms were assessed using the ranging from 0 to 15 points (21) . The total sum of points greater than 5 was considered indicative of depressive symptoms (21) .
To identify the network and social support, the Network and Social Support Scale was used, translated and validated in Brazil (22) . The social network was measured using two questions, including: "How many relatives do you feel comfortable with and can talk about almost anything?" and "How many friends do you feel comfortable with and can talk about almost anything?". Social support is measured by the frequency with which the person has material support, that is, the provision of practical and material resources, such as help at work or financial assistance; positive social interaction/affective support that reflect the possibility of having someone to perform leisure activities and offer physical demonstrations of love and affection; and emotional/information support, which consists of the social network's ability to meet individual needs in relation to emotional problems and the fact that it can count on people to advise, inform and guide (22) . The final score for each of the dimensions ranges from 20 to 100 points, and the higher the score, the better the level of social support (22) .
Communication independence was assessed using the Communication Skills Functional Assessment Scale (ASHA-FACS) (23) applied to the caregiver/family member and composed of four domains: Social Communication, related to social situations that require interaction with the speaker; Communication of Basic Needs, that is, the reaction to situations of need and emergency; Reading, Writing and Numerical Concepts, which consist of the ability of the person to take a message, identify food labels and/or fill in small forms; and the Daily Planning, which involves the notion of agenda to be fulfilled and appointments, use of the telephone and calendar (23) .
The ASHA-FACS is graded as a seven-point scale, which assesses the performance of communication along the "continuum" of independence, in terms of levels of assistance and/or readiness necessary for communication (24) . In this graduation, seven means that the individual has a proper performance in the item, without the need for any assistance; six -needs minimal assistance for proper performance; five -minimal to moderate assistance; four -moderate assistance; three -moderate to maximum assistance; two -maximum assistance; and one -not capable of certain behavior, even with maximum assistance for it. At the end, the weighted average is calculated, reaching the average value of communication independence (23) . the significance level (α) of 5% and the tests considered significant when p ≤ α were adopted.

Ethical aspects
The project was approved on

Results
Of the total number of study participants (n=941), it was found that most of them self-declared themselves as white (63.8%), followed by brown (25.3%) and black (10.9%).
As shown in Table 1 Table 2 shows the distribution of absolute and relative frequencies, mean and standard deviation of biopsychosocial indicators of older adults living in the health micro-region (MG) according to skin color/race self-reported.   (Table 3).

Biopsychosocial indicators
Color/race

Discussion
This study explored the association of self-reported skin color/race with biopsychosocial indicators in a representative sample of community older adults in a micro-health region in Minas Gerais, Brazil. The findings showed that: a) most of them self-declared as being of white color/race, followed by brown and black; b) black color/race was consolidated as a protective factor for negative and regular self-rated health and for the indication of depressive symptoms, regardless of gender, age, marital status, education and income; c) black color/ race was also associated with a higher social support score and a lower number of morbidities, even after adjustment.
According to the National Household Sample Survey (1.2%). However, when analyzing these data according to age group, it was found that, among Brazilian older adults, most self-reported as white (50.7%), followed by brown (39.2%) and black (8.8%) (25) , which corroborates the findings of this study and demonstrates a sample alignment in relation to the older adults' population in the country.
The highest rate of self-declared white older adults is in line with the country's mortality and life expectancy statistics, which show a higher proportion of early death among blacks and browns (3) . These data reinforce racial inequities observed in Brazilian society, causing many black people to not experience old age (26) , especially those in unfavorable living, health and socioeconomic conditions.
As a result, according to a study carried out in the USA, in Pittsburgh and Memphis, older adults who declared themselves black had lower survival rates when compared to white ones (7) , which would result in the survival bias of those who are part of the most vulnerable racial groups (27) , that is, a more select sample of surviving black people.
In line with this assumption, the study showed a lower percentage of black and brown older adults aged 80 years old and over compared to white ones. In order to minimize the confounding effect in this study, the association of selfreported skin color/race with biopsychosocial indicators was adjusted for the age variable.
In addition, another fact that drew attention was the higher percentage of brown and black older adults without education compared to white ones. It is noteworthy that health literacy can interfere with the self-perception of black people's health (11) . Considering that education is one of the causes of health inequity among older adults (3) , the analyzes were also adjusted for this variable.
In the world literature, contradictory findings were identified regarding the association of color/race with the biopsychosocial indicators of older adults (11,28) . If on one hand studies found that the racial inequities observed in health status remained independently of socioeconomic and contextual diversities (11) , on the other hand, evidence showed that this variable partially explained the analyzed outcomes, with socioeconomic inequalities and/or the social context had a more relevant effect on the health of the older adults (28)(29) .
Regardless of the direct effect of this variable or in conjunction with socioeconomic aspects, it is essential to consider color/race to understand the health inequities of the older adults population in Brazil (29) . As for the selfrated health indicators of this study, a similar result was evidenced in a survey conducted with 3594 older adults in the USA, which found that black people were more likely to have a positive self-rated health (excellent/ very good/good) compared to Hispanics and Chinese (p=0.015) (30) . In South Africa, a study carried out with 3284 older adults also found that white and brown color/ race participants had worse self-reported health status compared to black Africans (11) . It is noteworthy that South Africa is a developed country, with a great ethical-racial and cultural diversity and with a mostly black older adults population (11) .
In Brazilian literature, on the other hand, previous findings are divergent; while some showed that black older adults had worse self-rated health compared to the white ones (29,31) , another study found no association between these variables (3) . It is known that racial/ethnic and cultural differences may reflect in the self-reported subjective measures of older adults (27,32) such as self-rated health. Therefore, this measure may not represent the same aspects between different racial/ethnic and cultural groups, which makes the differences in the findings between the aforementioned studies consistent.
Some American scholars have promoted the theory that black individuals, due to material deprivation and/ or racism throughout life, develop healthier adjustments to deal with adversities in relation to white ones (33)(34) .
Therefore, it is possible that the black color/race as a protective factor for negative health self-assessment in this study is related to this group's greater coping capacity to deal with the challenges arising from senility and senescence, acquired in the past.
Additionally, older adults who make up racial/ethical minorities, such as African-Americans, tend to maximize resources and optimize their well-being (35) , with a positive impact on mental health. This data is consistent with this study, as it verified that black color/race was shown to be a protective factor for the indication of depressive In a cohort study in the USA, it was observed that black older adults had higher levels of depressive symptoms compared to white ones. However, in a situation of higher chronic exposure to stress, those were less likely to report these symptoms compared to these (36) .
It is possible that black individuals who have experienced situations of segregation, racial discrimination and/or economic deprivation throughout their lives may have developed resilience, which, consequently, represents a protective factor against depression during old age (36)(37) .
In this study, black color/race was associated with a higher social support score, which can be explained by the greater development of social cohesion attributes compared to the white population. This finding can also be understood as an adjustment or adaptive response to high exposure to stressors resulting from structural racism (33,37) .
Thus, it has been investigated whether social support influences black older adults' health. A study developed with North American black men showed that social support, when assessed individually, was predictive of better selfrated health. However, when evaluated collectively with other psychosocial resources (e.g., optimism, sense of mastery and religiosity), it did not show significantly protective effects (38) . It is worth mentioning that the nurse plays an important role in caring for older adults in the community. Therefore, the social support network is one of the aspects that must be included in the nursing consultation (39) .
In addition, an association of black color/race with the lowest number of morbidities was identified,  (3) .
Race contrasts in the presence of chronic morbidities may be related to the fact that groups of older adults evaluated and distributed by race/color are unequally exposed to several risk factors that influence the adoption of healthy behaviors or that pose a risk to lifelong health, favoring racial differences in chronic disease estimates.
Thus, it is possible that the person self-declared as black present a greater number of chronic morbidities as a result of the conditions being associated with the most vulnerable social groups (29) .
Considering the results of this research, related to the association of black color/race with the lowest number of morbidities, it is important to emphasize that black people have greater difficulty in accessing health services due to structural barriers, social and economic aspects, as well as the cultural, ethnic and racial prejudice, as demonstrated in a review study (41) . Therefore, these access limitations can lead to a lack of diagnosis for chronic conditions and, consequently, influence the results of research that obtain the number of morbidities through self-report.
It is noteworthy that the survival bias of black older adults may represent a potential limitation of this research and partially explain the findings by identifying that black color/race was associated with better outcomes

Conclusion
When analyzing the association of self-reported skin color/race with the biopsychosocial indicators of older adults, it was found that black color/race was considered a protective factor for worse self-rated health and for the indicative of depressive symptoms and was related to the highest social support score and the lowest number of morbidities, regardless of sociodemographic and economic characteristics. www.eerp.usp.br/rlae