SciELO - Scientific Electronic Library Online

vol.17 suppl.2Social factors associated with mental disorders with risk situations in the primary health careThe Porto Alegre paradox: social determinants and tuberculosis incidence author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Epidemiologia

Print version ISSN 1415-790X

Rev. bras. epidemiol. vol.17  supl.2 São Paulo  2014 

Original Articles

Race and perceived racism, education, and hypertension among Brazilian civil servants: the Pró-Saúde Study

Raça e racismo percebido, escolaridade e hipertensão em funcionários públicos brasileiros: estudo Pró-Saúde

Eduardo Faerstein I  

Dóra Chor I  

Guilherme Loureiro Werneck I  

Claudia de Souza Lopes I  

George Kaplan I  

IUniversidade do Estado do Rio de Janeiro - Rio de Janeiro (RJ), Brazil



Brazil has the largest population of African descendants outside Africa.


Mindful of the imprint of slavery on their contemporary social position, we investigated the relationship of perceived racism to hypertension.


We analyzed data (1999 - 2001) from 3,056 civil servants (mean age 42 years; 56% females) at university campuses in Rio participating in the longitudinal Pró-Saúde Study.


Cases of prevalent hypertension had measured blood pressure equal to or greater than 140/90 mmHg or used antihypertensive medication. Self-administered questionnaires assessed participants' perceived history of lifetime discrimination (due to race, gender, socioeconomic position, and other attributes) at work and school, neighborhood, public places, and in contact with the police. Participants used 41 terms as responses to an open-ended question on racial self-identification; for these analyses, 48% were classified as afrodescendants. Racial discrimination in at least one setting was reported by 14% of afrodescendants. Compared to whites, the age- and gender-adjusted prevalence of hypertension was higher for afrodescendants with history of self-perceived racism (prevalence ratio - PR = 2.1; 95%CI 1.5 - 3.0) than for those with no such history (PR = 1.5; 95%CI 1.2 - 1.8). Comparing the former to whites, the adjusted association with hypertension was stronger for those with elementary education (PR = 3.0; 95%CI 1.3 - 6.7) than for those with a college degree (PR = 1.7; 95%CI 1.0 - 3.1).


Racism may increase the risk of hypertension of afrodescendants in Brazil, and socioeconomic disadvantage - also influenced by societal racism - may further potentiate this increased risk.

Key words: Social determinants of health; Racism; Hypertension; Adult health; Black population health; Epidemiologic methods



O Brasil tem a maior população afrodescendente fora da África.


Em razão das sequelas da escravidão sobre sua posição social contemporânea, foi investigada a relação entre racismo percebido e hipertensão.


Analisamos dados (1999 - 2001) de 3056 funcionários públicos (idade média: 42 anos; mulheres: 56%) de campi universitários fluminenses, participantes do Estudo Pró-Saúde.


Casos de hipertensão prevalente tinham pressão arterial aferida igual ou superior a 140/90 mmHg ou uso de medicação antihipertensiva. Via questionários autopreenchíveis, aferiu-se história percebida de discriminação ao longo da vida (devido a raça, gênero, posição socioeconômica, outros atributos) nos domínios trabalho, moradia, escola, locais públicos e com a polícia. Participantes utilizaram 41 termos como respostas a pergunta aberta sobre autoidentificação racial; para estas análises, 48% foram classificados como afrodescendentes. Discriminação racial em ao menos um domínio foi relatado por 14% dos afrodescendentes. Em comparação com participantes brancos, as prevalências (ajustadas para idade e sexo) de hipertensão foram mais elevadas entre afrodescendentes com história percebida de racismo (RP = 2,1; IC95% 1,5 - 3,0) do que entre aqueles sem essa história (RP = 1,5; IC95% 1,2 - 1,8). Comparando-se afrodescendentes com história percebida de racismo com brancos, a associação com hipertensão foi mais forte entre participantes com instrução fundamental (RP = 3,0; IC95% 1,3 - 6,7) do que entre aqueles com nível universitário (RP = 1,7; IC95% 1,0 - 3,1).


É possível que o racismo aumente o risco de hipertensão entre afrodescendentes no Brasil; a adversidade socioeconômica - também influenciada pelo racismo - pode potencializar esse aumento de risco.

Palavras-Chave: Determinantes sociais da saúde; Racismo; Hipertensão; Saúde do adulto; Saúde da população negra; Métodos epidemiológicos


Hypertension (HT), a major cardiovascular risk factor, may be on the rise worldwide1 , 2 , and a similar trend has been observed in Brazil3. The excess risk of HT among afrodescendants, compared to whites, has been consistently observed in many populations4, including Brazil's5. In addition to a number of aspects of dietary intake and excess weight, among other possible causes, exposure to racial discrimination has been increasingly explored as underlying the higher risk of HT among populations of African descent, through a variety of mediating mechanisms, e.g. socioeconomic, behavioral and pschyconeuroendocrine. Overall, the findings suggest that racism may increase HT risk, but evidence is still mixed, and almost entirely restricted to more industrialized societies6 , 7.

Brazil has the largest population of African descendants outside Africa, and their contemporary social position carries the intergenerational imprint of the enslavement of about 3 million individuals brought from West Africa over three centuries8, which became a major structural component of Brazilian society. Because little is known about the inter-relationships between racial and socioeconomic inequalities in this country, we report the possible combined effects of African descent/perceived racial discrimination and socioeconomic disadvantage on HT risk in Brazil, a multiracial society in which socioeconomic inequality is pervasive.


Our subjects participated in two stages of baseline data collection (1999 - 2001) of a cohort study (the Pró-Saúde Study) of non-faculty civil servants at university campuses in Rio de Janeiro, Brazil. All eligible employees (4,177) were invited to participate; the response rate was 78% (3,253). More details about the study are reported elsewhere9. The present analyses are based on data from 3,056 subjects after the exclusion of 197 individuals with missing data (blood pressure, age, sex, color/race, history of lifetime discrimination, or education) or who self-identified as Asian or Indigenous.

Blood pressure was measured twice with mercury sphygmomanometers, in compliance with standard techniques and international protocols, by trained and certified technicians. Cases of prevalent hypertension had (by averaging the two measurements) systolic blood pressure ≥ 140 mmHg, or diastolic blood pressure ≥ 90 mmHg, or reported use of anti-hypertensive medication. Quality control indicators (e.g. final digit preference, differences between subsequent measurements, proportion of missing data) monitored pari passu with data collection suggest that blood pressure measurements achieved high quality10.

We used self-administered questionnaires to assess the other variables. Participants used 41 terms as responses to an open-ended question on racial self-identification; for these analyses, 48% of respondents were classified as afrodescendants.

By means of an abridged adaptation of the Everyday Discrimination Scale11, the history of perceived lifetime discrimination was measured by responses to five questions with the same initial phrase: "Have you ever been unfairly treated, discriminated against at your workplace, for example, while getting fired, applying for a job or being denied a promotion?", "[...] while getting housing or interacting with neighbors?"; "[...] by the police, such as being accused, searched or arrested?"; "[...] in public places, such as banks, shops, hospitals, government offices, etc?"; "[...] in your school or college, for example, being discouraged to continue your studies?" Participants who reported experiences of discrimination were then asked: "The last time this happened to you, which was, in your opinion, the main reason for being discriminated against?" The following response options were offered: skin color or race; being a man or a woman; religion or creed; illness or disability; sexual orientation; socioeconomic condition, education or job; political activism; age; physical appearance; other (specify). In a test-retest reliability study (n = 92, 2 weeks apart), we estimated a kappa coefficient of 0.85 (95%CI 0.72 - 0.98) for responses related to overall discrimination.

For the purpose of the present analyses, we created a composite exposure variable combining dichotomous data on racial self-identification (white, afrodescendant) with dichotomous data on the history of skin color/racial discrimination (ever, never). Because no white participant reported racial discrimination, our exposure variable has 3 categories: whites (reference), afrodescendants with no history of perceived racism ("afro/no racism"), and afrodescendants with history of perceived racism ("afro/racism"). Age and sex-adjusted associations of the exposure categories with prevalent hypertension were estimated through Poisson regression models with robust variance, and subsequently analyzed according to three education categories (elementary or less, high school, college or more). Statistical analyses were performed using the SPSS for Windows version 11. The study was approved by the State University of Rio de Janeiro research ethics committee.


Our study population was within the age range of 22 - 69 years (mean = 42 years); 56% of the subjects were females, and 48% were afrodescendants. The age and sex-adjusted prevalence of hypertension was 21%, 26% and 33%, respectively, for white people, "afro/no racism", and "afro/racism". Compared to the whites, individuals in the "afro/racism" category had a 2.1 times increased prevalence of HT (95%CI 1.5 - 3.0), and those with no such history showed a prevalence ratio of 1.4 (95%CI 1.2 - 1.8) (Table 1)

Table 1 Race/perceived racism and prevalent hypertension. Pró-Saúde Study, Rio de Janeiro, Brazil, 1999 - 2001. 

n % Hypertension PR* 95%CI
%* 95%CI
White people 1,600 52.4 21.0 19.1 – 23.0 1.0
no perceived racism
1,258 41.2 26.3 24.1 – 28.5 1.4 1.2 – 1.8
perceived racism
198 6.4 33.0 27.4 – 38.5 2.1 1.5 – 3.0

*Adjusted for age and sex.

Comparing the afrodescendants with history of racial discrimination experiences to the whites, the adjusted association with HT was stronger for those with elementary education (PR = 3.0; 95%CI 1.3 - 6.7) than for those with a college degree (PR = 1.7; 95%CI 1.0 - 3.1), or with complete high school in an intermediate position (Figure 1).

Figure 1 Age and sex-adjusted prevalence ratios (PR) of hypertension by race/perceived racism, according to education. Pró-Saúde Study, Rio de Janeiro, Brazil, 1999 - 2001 (reference category: white people). 


The graded associations observed in our study suggest that among afrodescendants in Brazil, a higher lifetime experience of racial discrimination may substantially increase the risk of hypertension. The data are also consistent with the possibility that socioeconomic adversity - which is influenced by societal racism - may further potentiate this increased risk in a similarly graded pattern. In other words, a combination of these two axes of social disadvantage (race and socioeconomic position) may be operating. To our knowledge, this is the first study conducted with a Brazilian population to report such evidence.

Hypertension may be one of several adverse health consequences of race-based discrimination. In this study population, whereas 34% of the participants reported a history of discrimination in at least one life domain, afrodescendants reported histories more frequently than whites across five studied life domains (data not shown)12. Hypothesized mechanisms may be indirect, through structuring of socioeconomic disadvantage, which may influence dietary and physical activity patterning; through differential access to health care; or directly, through psychoneuroendocrine mechanisms6 , 7.

The Pró-Saúde study pioneered the adaptation, to a Brazilian population, of one among several existing questionnaires aimed at the measurement of interpersonal discrimination12. An additional strength of the study was the use of responses to an open-ended question that allowed participants to classify themselves by race using their own words; because we also included a close-ended question - with response options being the Brazilian Census Bureau (IBGE) list of racial categories - it was possible to observe that a sizeable number of afrodescendants chose "whitening" categories when "forced" to choose from the IBGE list14.

There are several potential limitations to consider when interpreting our results. First, we were restricted to prevalent cases of HT. In theory, psychological effects of hypertension "labeling" might include an enhanced perception of racial discrimination. However, we investigated this possibility by stratifying the associations we report according to prior awareness of HT, and the results did not differ. Second, even though we studied a population of civil servants with stable jobs and relatively limited socioeconomic variability, we found HT prevalence similar to estimates from statewide and nationwide Brazilian surveys with similar occupational groups. The existing social variability also allowed us to observe distinct associations according to schooling level. Third, we had no available data on which to assess potentially important aspects of our respondents' discriminatory experiences (e.g. intensity, coping behavior), and we were unable to analyze some of the available data (e.g. life domain, timing) due to small numbers.

Given the severity and pervasiveness of race- and ethnicity-based discrimination and violence in the world today, we believe it is critical to explore and document the health effects of these injustices in one of the world's most diverse societies. Further efforts should focus on better measurement, which may result from ongoing refinement of scales dealing with specific aspects of Brazilian race relations15. Studies should be replicated across larger and more socially and geographically diverse populations. With specific regard to hypertension, further analyses should explore the association of discriminatory experiences with levels of systolic, diastolic, and pulse pressure, as well as with ambulatory blood pressure and preclinical indicators, e.g. vascular reactivity and endothelial dysfunction. Importantly, the range of health-related outcomes of interest should be expanded, and whenever possible investigated with longitudinal data specifying life domains, types of discrimination, and coping mechanisms. Advances in cognitive research, e.g. Implicit Association Tests (IAT), may prove especially helpful16.

More broadly, it will be important to shed light on the web of proximate factors mediating the associations between race discrimination and health, e.g. lifestyle, health care, and the direct effect of chronic stress. A combination of research methods will allow better investigation at levels of society (e.g. institutions) where complex mechanisms of discrimination beyond interpersonal experiences are known to operate16 , 17. In particular, a detailed life course perspective, including concepts such as sensitive and latency periods and age-period-cohort analyses18, would benefit further studies of the complex relationships between discriminatory experiences and socioeconomic disadvantage.


. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. Lancet 2005; 365(9455): 217-23. [ Links ]

. Hajjar I, Kotchen JM, Kotchen TA. Hypertension: trends in prevalence, incidence, and control. Annu Rev Public Health 2006; 27: 465-90. [ Links ]

. Schmidt MI, Duncan BB, Azevedo e Silva G, Menezes AM, Monteiro CA, Barreto SM, et al. Chronic non-communicable diseases in Brazil: burden and current challenges. Lancet 2011; 377(9781): 1949-61. [ Links ]

. Cooper R, Rotimi C. Hypertension in blacks. Am J Hypertens 1997; 10(7 Pt 1): 804-12. [ Links ]

. Sichieri R, Oliveira MC, Pereira RA. High prevalence of hypertension among Black and Mulatto women in a Brazilian survey. Ethn Dis 2001; 11(3): 412-8. [ Links ]

. Spruill TM. Chronic psychosocial stress and hypertension. Curr Hypertens Rep 2010; 12(1): 10-6. [ Links ]

. Brondolo E, Love EE, Pencille M, Schoenthaler A, Ogedegbe G. Racism and hypertension: a review of the empirical evidence and implications for clinical practice. Am J Hypertens 2011; 24(5): 518-29. [ Links ]

. Moura C. História do negro brasileiro. São Paulo: Ática, 1994. [ Links ]

. Faerstein E, Chor D, Lopes CS, Werneck GL. Estudo Pró-Saúde: características gerais e aspectos metodológicos. Rev Bras Epidemiol 2005; 8(4): 454-66. [ Links ]

. Faerstein E, Chor D, Griep RH, Alves MGM, Werneck GL, Lopes CS. Aferição da pressão arterial: experiência de treinamento de pessoal e controle de qualidade no Estudo Pró-Saúde. Cad Saúde Pública 2006; 22(9): 1997-2002. [ Links ]

. Williams DR, Yu Y, Jackson JS, Anderson NB. Racial differences in physical and mental health: socio-economic status, stress and discrimination. J Health Psychol 1997; 2(3): 335-51. [ Links ]

. Faerstein E, Bastos JL, Fonseca MJ, Lopes CS. Padrões de discriminação no Estudo Pró-Saúde: Quem? Onde? Por quê? Comunicação oral ao VIII Congresso Brasileiro de Epidemiologia - Abrasco. São Paulo, novembro de 2011. [ Links ]

. Bastos JL, Celeste RK, Faerstein E, Barros AJ. Racial discrimination and health: a systematic review of scales with a focus on their psychometric properties. Soc Sci Med 2010; 70(7): 1091-9 [ Links ]

. Maio MC, Monteiro S, Chor D, Faerstein E, Lopes CS. Cor/raça no Estudo Pró-Saúde: resultados comparativos de dois métodos de autoclassificação no Rio de Janeiro, Brasil. Cad Saúde Pública 2005; 21(1): 171-80 [ Links ]

. Bastos JL, Faerstein E, Celeste RK, Barros AJD. Explicit discrimination and health: development and psychometric properties of an assessment instrument. Rev Saúde Pública 2012; 46(2): 269-78. [ Links ]

. Krieger N. Methods for the scientific study of discrimination and health: an ecosocial approach. Am J Public Health 2012; 102(5): 936-45. [ Links ]

. Bastos JL, Faerstein E. Conceptual and methodological aspects of relations between discrimination and health in epidemiological studies. Cad Saúde Pública 2012; 28(1): 177-83. [ Links ]

. Gee GC, Walsemann KM, Brondolo E. A life course perspective on how racism may be related to health inequities Am J Public Health 2012; 102(5): 967-74. [ Links ]

Financing source: none

Received: May 04, 2012; Revised: February 05, 2013; Accepted: June 12, 2013

Corresponding author: Eduardo Faerstein. Universidade do Estado do Rio de Janeiro. Instituto de Medicina Social, Departamento de Epidemiologia. Rua São Francisco Xavier, 524, bloco D, 7º andar, sala 7009, Maracanã, CEP: 20559-900, Rio de Janeiro, RJ, Brasil. E-mail:

Conflict of interests: nothing to declare

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.