Health inequalities in Brazil: race matters


Desigualdades en salud en Brasil: la raza importa



Dóra Chor

Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz, Rio de Janeiro, Brasil




"...111 presos indefesos, mas presos são quase todos pretos/Ou quase pretos, ou quase brancos quase pretos de tão pobres/E pobres são como podres e todos sabem como se tratam os pretos" (from the song "Haiti", by Caetano Veloso).

"Race and health" is a controversial theme in Brazil. I am convinced of its importance, based on the scarcity of scientific studies on both the size of racial inequalities and their impact on disease. The issue is not to discuss the hierarchy of conditioning factors for health inequalities in Brazil. I believe it is a consensus that the origin of such inequality is social. However, in a world where conservative thinking continues to make inroads, it is necessary to combine the universal and the specific by including scientific evidence on discrimination, racism, race, and health in Brazil.

I have been asked by both colleagues and students why some Brazilian researchers have taken a stance against health studies with a racial focus. I have never been able to answer this question directly. There was a heated discussion in the 2000s, the theme was presented at a round table during the 6th Brazilian Congress of Epidemiology in 2004, and the resulting papers were published in Cadernos de Saúde Pública (see Forum on Race, Racism, and Health in Brazil 1,2,3,4) in 2005. More recently, the theme has disappeared not only from congresses on Epidemiology and Public Health (at least as a priority activity), but even from scientific meetings dedicated to Social Determinant of Health.

Despite the interest by foreign researchers on racial inequalities in health in Brazil, it was not possible to incorporate sufficient scientific studies on the theme in a broad panel on Brazil published in 2011 in The Lancet. The exceptions pointed consistently to the disadvantage of the same racial groups – an alarming mortality rate from violence affecting young black and brown Brazilians 5, higher under-five mortality among black and brown children 6, more difficulties and greater dissatisfaction towards prenatal and obstetric care among black women, and high rates of chronic illnesses among indigenous peoples 7.

Running against this trend towards the diminishing visibility of racial issues, sectors of the Brazilian press have spotlighted cases of racism, such as that of a black boy who was nearly thrown out of an automobile dealership in Rio de Janeiro while his white adoptive parents were shopping for a car. In his news story, veteran journalist Zuenir Ventura 8 asks whether "what many prefer to call 'cordial racism' is not really the opposite: veiled, camouflaged, and – when detected in the act – glossed over, written off as a mistake or a misunderstanding". By way of curiosity, Ventura concludes his news story by reporting that mortality from infectious diseases is 43% higher among black Brazilian children as compared to white. What are the health consequences of the treatment that the young boy and others like him receive? How can one tell whether healthcare is permeated by discrimination or not?

The Brazilian press recently ran a story on a dubiously impartial "first come, first served" rule during triage in waiting lines for health services. If all the patients on line in the Brazilian public health system are poor, are blacks treated last? How can one know without the proper research? And if the topic is so sensitive and difficult to measure, do we turn it into a taboo for scientific studies? Some authors have said, no, the Unified National Health System is not racist 9. On the contrary, its ideals are those of universal and equal treatment for all Brazilians. However, one cannot assume (based on these ideals) that racist treatment only happens in other institutions.

Some Brazilian studies provide evidence of racial iniquities in health (inequitable because they are unfair and avoidable). In Pelotas, Rio Grande do Sul State, the mortality rate for children of white mothers in 1982 (30 per 1,000 live births) was only attained by children of black and brown mothers in 2004! That year, the infant mortality rate for children of white mothers was already on the same level as in high-income countries (13.9 per 1,000 live births) 10. A study of data on mammograms from the National Household Sample Survey (PNAD 2008) showed that breast cancer screening prevalence was higher in white women and those with higher income, more schooling, or living in metropolitan areas with higher socioeconomic standards 11.

The voluminous international literature on social inequalities in health does not rule out the influence of race, while it relates such influence to socioeconomic conditions, highlighting the complexity and challenges of this cross-analysis. The Brazilian public health field needs a more in-depth effort in this direction.

The results of Brazilian studies and comparison with the international literature reinforce the hypothesis that in Brazil as well, racial inequality in health (and not only socioeconomic and gender inequality) deserves investigation. These three forms of inequality can act jointly and create groups that are particularly exposed to risks. In many cases, economic adversity will provide the most important explanation. In other cases the economic focus will not suffice, and understanding the role of race and gender will be indispensable for explaining outcomes and contributing to the elaboration of public policies.

Quoting Edward Telles 12 (p. 38), "I share the concern that use of the term 'race' may bolster social distinctions that have no biological value; however, race is still immensely important in sociological interactions and should thus be taken into account in sociological analyses." Studies on race have a tradition in the Brazilian social sciences. In education, racial iniquities have already been diagnosed as "heavy", meriting immediate action 13. What about the field of health? I think we lack sufficient evidence to rule out race as one of the distal determinants of diseases and injuries, since it influences the "causes of the causes": the opportunity to attend good schools, the "choice" of occupation (also conditioned by the possibility of entering the most heavily disputed professions), and income from one's work.

Considering race as a social construct that combines various dimensions of the life history of individuals and generations, studies on race and health conducted in other contexts are not sufficient to explain the Brazilian reality. The absence of legally approved segregation throughout Brazil's history, miscegenation, and the so-called "prejudice against being prejudiced" have shaped a unique picture for Brazil in the world, different from both the United States and South Africa. In the field of research on discrimination, adequate instruments for our context are necessary and can provide measurable results that are closer to the Brazilian reality 14.

Methodological limitations have been cited as obstacles to epidemiological studies on race and health. For example, in disease-specific mortality rates according to race, the fact that the numerator and denominator come from different data sources purportedly limits the results' accuracy (the numerator – number of deaths – comes from the Ministry of Health's Mortality Information System (SIM), while the denominator comes from the National Population Census). However, even more important problems such as the coverage of the SIM did not prevent Brazilian researchers from studying the country's mortality profile in the 1980s. In fact, these and other problems only surfaced to the extent that the data were analyzed and their limitations identified. In other words, this process helped improve the system and provided highly relevant information on mortality in the country.

Another alleged problem is the variability with which some individuals classify themselves. As in other countries, racial identity in Brazil is not immutable, and its validity and reliability are limited. Although it is plausible to imagine that persons do not tend to alter their racial self-classification in an epidemiological study (as they might feel pressured to do in a job application), this is a frequent limitation for epidemiologists interested in the effect of psychosocial exposures. There is no gold standard in this case, and measurement cannot be done with equipment, molecular biology, or computed tomography. Even so, there is strong evidence of the effects of race on health.

I recently participated in a review panel for a doctoral student's qualifying exam, during which I asked the candidate why he intended to adjust the statistical model for race, since it did not make sense to me. He replied that such analyses had been presented in other articles... but then realized that all the studies had been conducted in the United States. The scarcity of conceptually and methodologically well-based Brazilian studies with consistent discussions of the results may induce young researchers to imitate lines of reasoning that originate in other social contexts.

The available evidence indicates that health inequalities reflect the broader inequalities in society. Race, socioeconomic status, and gender influence Brazilians' health through different relations and with diverse magnitudes, depending on the target question. The study of exposures situated closer to the outcome on the causal chain is not inconsistent with the investigation of distal factors, including race. It is not wise to raise barriers to this line of research without sufficient grounds. According to an implicit consensus (not stated explicitly), studying racial inequalities in health creates social divisions in Brazil rather than exposing them, potentially reinforces biological determinism, distorts the scope of other aspects of discrimination, and fails to help understand the origins of social inequalities in Brazil; yet this line of argument appears to be as polarized as restricting the discussion to the creation of specific health programs exclusively targeting blacks and browns. I close this paper by quoting a colleague who does not share my point of view, in his dedication to a book on the theme: "Above all, let us encourage the debate!"



To Álvaro Cesar Nascimento, for his assistance in editing the manuscript.


A versão em inglês deste texto está disponível online no Portal SciELO (

The English version of this text is available online in the SciELO (



1. Faerstein E. Fórum: raça, racismo e saúde no Brasil. Cad Saúde Pública 2005; 21:1584-5.         

2. Chor D. Lima CRA. Aspectos epidemiológicos das desigualdades raciais em saúde no Brasil. Cad Saúde Pública 2005; 21:1586-94.         

3. Lopes F. Para além da barreira dos números: desigualdades raciais e saúde. Cad Saúde Pública 2005; 21:1595-601.         

4. Cardoso AM. Santos RV, Coimbra Jr. CEA. Mortalidade infantil segundo raça/cor no Brasil: o que dizem os sistemas nacionais de informação? Cad Saúde Pública 2005; 21:1602-8.         

5. Victora CG, Aquino EML, Leal MC, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. Lancet 2011; 377:29-42.         

6. Reichenheim ME, Souza ER, Moraes CL, Jorge MHPM, Silva CMFP, Minayo MCS. Violence and injuries in Brazil: the effect, progress made, and challenges ahead. Lancet 2011; 377:69-82.         

7. 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; 377:1949-61.         

8. Ventura Z. Brasil: um país com racismo, mas sem racistas. O Globo 2013; 26 jan.         

9. Maio MC, Monteiro S, Rodrigues PHA. O SUS é racista? In: Fry P, Maggie Y, Maio MC, Monteiro S, Santos RV, organizadores. Divisões perigosas: políticas raciais no Brasil contemporâneo. Rio de Janeiro: Editora Record; 2007. p. 235-40.         

10. Matijasevich A, Victora CG, Barros AJD, Santos IS, Marco PL, Albernaz EP, et al. Widening ethnic disparities in infant mortality in southern Brazil: comparison of 3 birth cohorts. Am J Public Health 2007; 97:692-8.         

11. Chor D, Oliveira EXG, Melo ECP, Pinheiro RS, Carvalho MS. Desigualdade socioeconômica afeta a chance de realizar mamografia no Brasil. (acessado em 27/Mar/2013).         

12. Telles E. Racismo à brasileira: uma nova perspectiva sociológica. Rio de Janeiro: Editora Relume Dumará; 2003.         

13. Arbix G. Apresentação. In: Soares S, Beltrão KI, Barbosa MLO, Ferrão ME, organizadores. Os mecanismos de discriminação racial nas escolas brasileiras. Rio de Janeiro: Instituto de Pesquisa Econômica Aplicada; 2005.         

14. Bastos JL, Faerstein E. Desafios metodológicos para a mensuração da discriminação interpessoal no Brasil. In: Bastos JL, Faerstein E, organizadores. Discriminação e saúde: perspectivas e métodos. Rio de Janeiro: Editora Fiocruz; 2012. p. 75-102.         



D. Chor
Departamento de Epidemiologia e Métodos Quantitativos em Saúde, Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz.
Rua Leopoldo Bulhões 1480, Rio de Janeiro, RJ
21041-210, Brasil.

Submitted on 31/Mar/2013
Approved on 01/Apr/2013

Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz Rio de Janeiro - RJ - Brazil