Color/race inequalities in oral health among Brazilian adolescents*
Cor/raça e desigualdades em saúde bucal entre adolescentes brasileiros
João Luiz BastosI; José Leopoldo Ferreira AntunesII; Antonio Carlos FriasII; Maria da Luz Rosário de SouzaIII; Karen Glazer PeresIV; Marco Aurélio PeresIV
IPost-graduate Program in Epidemiology, Federal University of Pelotas
IISchool of Dentistry, University of São Paulo
IIISchool of Dentistry of Piracicaba, State University of Campinas
IVPost-graduate Program in Public Health, Federal University of Santa Catarina
This study assessed oral health outcomes (perceived dental treatment need, untreated dental caries, gingival bleeding, periodontal pockets, and pain in teeth and gums), in relation to color/race inequalities among adolescents in each Brazilian region. The database included dental examination and interview of 16,833 15-19-year-old adolescents, surveyed by the Brazilian health authority, from May 2002 to October 2003, in accordance with international diagnostic criteria standardized by the World Health Organization. Prevalence ratios estimated by Poisson regression, and controlled by socioeconomic status and access to fluoridated piped water, assessed oral health differentials among color/race groups and country's regions. Except for periodontal pockets, prevalence figures were higher in the North and Northeast: perceived dental treatment needs, untreated dental caries, gingival bleeding at probing and pain in teeth and gums varied between 80-83%, 75-76%, 38-43%, and 17-18%, respectively, in these regions. Adolescents living in the Southeast - the richest Brazilian region - presented a better general profile of oral health than their counterparts living in the remaining regions; they had a lower prevalence of untreated dental caries (54%) and unfavorable gingival status (29%). However, the Southeast presented color/race inequalities in all oral health outcomes, with a poorer profile systematically affecting browns or blacks, depending on the oral health condition under consideration. These results reinforce the need for expanding the amplitude of health initiatives aimed at adolescent oral health. Socially appropriate health programs should concurrently aim at the reduction of levels of oral disease and its inequalities.
Keywords: Oral health. Dental caries. Adolescent. Race relations. Brazil.
O presente estudo avaliou desfechos de saúde bucal (necessidade de tratamento dentário autopercebida, cárie dentária não tratada, sangramento gengival, bolsa periodontal e dor nos dentes e gengivas) com o intuito de identificar desigualdades por cor/raça entre adolescentes em cada uma das macro-regiões brasileiras. O banco de dados incluiu informações sociodemográficas e de exames bucais de 16.833 adolescentes entre 15-19 anos de idade, investigados pelo Ministério de Saúde entre maio de 2002 e outubro de 2003, conforme critérios diagnósticos preconizados pela Organização Mundial da Saúde. Foram utilizadas razões de prevalência, calculadas por regressão de Poisson e ajustadas para variáveis socioeconômicas e acesso a água fluoretada, para estimar desigualdades nos desfechos de saúde bucal entre os grupos de cor/raça e as macro-regiões brasileiras. Exceto para bolsa periodontal, as estimativas de prevalência foram mais altas no Norte e no Nordeste: necessidade de tratamento dentário autopercebida, cárie dentária não tratada, sangramento gengival e dor nos dentes e gengivas variaram entre 80-83%, 75-76%, 38-43% e 17-18%, respectivamente, nestas regiões. Os adolescentes do Sudeste - a macro-região mais rica do país - apresentaram, em geral, melhores condições de saúde bucal, quando comparados com seus pares das demais macro-regiões; os adolescentes do Sudeste apresentaram menores prevalências de cárie dentária não tratada (54%) e de sangramento gengival (29%). Entretanto, o Sudeste demonstrou desigualdades por cor/raça em todos os desfechos investigados, com piores condições afetando sistematicamente pardos ou pretos, a depender da condição de saúde bucal investigada. Estes resultados reforçam a necessidade de expandir as ações dirigidas à saúde bucal dos adolescentes brasileiros. Programas/intervenções em saúde socialmente sensíveis devem visar a redução de níveis de morbidades bucais, bem como de suas desigualdades.
Palavras-chave: Saúde bucal. Cárie dentária. Adolescentes. Relações raciais. Brasil.
Brazil is acknowledged as a continental country because of its large territory and heterogeneous geographic and environmental conditions. Socioeconomic status also varies widely across the Brazilian population, both at the inner-city and macro-regional levels. The Southeast is the most densely populated and industrialized region of the country. It comprises the two largest Brazilian cities (São Paulo and Rio de Janeiro), and together with the South comprises the country's richest regions, with the highest rankings in the Gross Domestic Product (GDP). A dynamic economy based on manufacturing, agribusiness and exports mostly benefits the Southeast, the South and the Center-West regions. The North region is mostly occupied by the Amazon rainforest, and presents a low population density. The Northeast is a densely populated region, chronically affected by drought, poverty and underdevelopment; towns in the North and the Northeast regions suffer the poorest overall profile of socioeconomic standings in Brazil1.
The color/race composition of the Brazilian population, according to a pre-coded five-category item included in the 2000 national census2, was 54%, 6%, 0.4%, 39% and 0.4% of self-classified whites, blacks, yellows (people of Asian ancestry), browns (or pardos, the official term for the admixed population in Brazil) and indigenous, respectively. Interestingly, in spite of the peculiarities and fluidity involved in the color/race characterization in Brazil as highlighted by different authors3-7, the richest Brazilian regions concentrate whites (84% in the South, 62% in the Southeast), whereas the North and Northeast present, respectively, 69% and 66% of browns and blacks1.
In terms of "racial relations", absent open racial conflict and cordiality have been reported to be outstanding sociologic features of Brazilian society8,9, and miscegenation a prominent demographic characteristic, since browns make up the second largest group in the country1. Notwithstanding, and restricting our focus to the three major color/race groupings in Brazil, there is documented evidence on the poorer status of browns and blacks, in comparison to whites, as regards income, education, labor market, law enforcement and health conditions10-12.
Importantly, disparities in color/race in Brazil have recently been examined through the aggregation of browns and blacks in a single category, a controversial procedure that has been targeted by criticism from different Brazilian scholars13. In public health, some authors argue that this procedure may be part of a more general process of construction of the field of "health of the black population"14, and, also, that collapsing browns and blacks into one stratum may blur important differences between these two groups.
Color/race inequalities in health have been discussed from different perspectives, ranging from their causes to how to reduce or control them15. Although health inequalities among color/race groups may be attributed to genetic characteristics, disparities in the distribution of individual health-behaviors, such as diet, exercise, and tobacco use, as well as the over-representation of some groups in lower socioeconomic strata, emerging paradigms to explain this type of health inequality are the psychosocial stress and the structural-constructivist models16. The first model emphasizes the stresses associated with minority group status, the experience of racism and discrimination, while the latter focuses, on the intersection of racially stratified social structures with goals and aspirations constructed within racial and ethnic groups as the causes of differences in morbidity and mortality among color/race groups16.
As regards oral health, color/race inequalities in the distribution of dental caries and periodontal disease have previously been reported in Brazil17-20. These studies documented the poorer dental status of browns and blacks, sometimes classified in a single group. However, color/race inequalities in oral health have not been assessed from a geographic perspective, accounting for differences in color/race composition and living standards among regions of the country.
The assessment of inequalities in oral health is particularly relevant to adolescents. The most prevalent oral health outcomes are cumulative and chronic, as well as a source of avoidable pain, functional and aesthetic limitations, which all impact negatively on day-to-day life and self-esteem21,22. The objective of this study was to compare selected oral health outcomes of brown, black and white Brazilian adolescents, and to assess the magnitude of differences among macro-regions of the country.
From May 2002 to October 2003, the Brazilian health authority performed a nationwide survey of oral health (the Brazilian Oral Health Survey), following international diagnostic criteria set up by the World Health Organization in 1997. The original report of the survey** described detailed information on methods of data collection: its conformity to national and international standards of ethics in research involving human participants; forms; how to perform dental examinations and reproducibility of diagnostic methods.
A total of 16,833 adolescents (15 to 19 years of age) living in 250 towns and five Brazilian macro-regions participated in this study. Data collection comprised dental examinations and interviews, using a questionnaire on socio-demographic characteristics, utilization of dental services and self-rated oral health. Dental exams were performed at the household, using natural light, periodontal probes and plane mouth mirrors.
The assessment of color/race followed criteria set forth by the 2000 Brazilian national census2. Adolescents were required to classify their color/race according to the following pre-coded categories, presented in a circular chart: white, brown, black, yellow and indigenous. Due to their low frequency, indigenous (1%) and yellow (3%) participants were excluded from the current analysis, so that all descriptive and analytical issues focused on the comparison among browns, blacks and whites.
Oral health outcomes
- Dummy variables assessed the following oral health outcomes:
- Perceived dental treatment need: adolescents answered a direct question asking whether they considered being in need of any dental treatment.
- Prevalence of untreated dental caries: having at least one permanent tooth diagnosed with caries (World Health Organization, 1997), including filled teeth that also presented further tooth decay.
- Gingival bleeding at probing: assessed by the Community Periodontal Index (equal to 1 in at least one mouth sextant, World Health Organization, 1997).
- Periodontal pockets: assessed by the Community Periodontal Index (equal to 4-5mm - shallow pockets - or > 6 - deep pockets - in at least one mouth sextant, World Health Organization, 1997).
- Pain in teeth and gums: direct question referring to the previous six months and to pain of "moderate" to "severe" degree.
In addition to sex, age, and color/race, other socio-demographic conditions were analyzed in this study. Living area refers to the comparison between rural and urban areas. Fluoridated water distinguished towns that added fluoride to their piped water supply for at least 5 subsequent years from those that did not. Family income was the only non-dichotomous variable: no income at all; lower than one Brazilian minimum wage; from one to two minimum wages; and more than two minimum wages. The Brazilian minimum wage is a national standard for this measurement, which roughly corresponded to US$70 during the period of data collection.
Household crowding - a proxy for socioeconomic status in epidemiologic studies23 - was assessed by the ratio between the number of dwellers and rooms in the household. The comparative analysis was adjusted by age and household crowding as parametrically assessed; i.e. these variables were not categorized.
School delay was assessed in terms of a less than expected number of years of study for the corresponding age; this variable was incorporated in the overall assessment of the human development index in Brazil1. The current study considered delayed in school, adolescents with at least one year of delay in relation to the expected number of years of study for their corresponding age (11 years of study for adolescents aged 18 and 19 years; 10 years of study for 17-year-olds; 9 for 16 and 8 for those aged 15). Adolescents with school delay included those that, for any reason, interrupted formal school attendance before concluding high school.
The distribution of adolescents by socio-demographic characteristics and oral health outcomes was described for each Brazilian macro-region. The comparative analysis considered the Southeast region - the richest in the country - as reference. The assessment of oral health outcomes among regions used the prevalence ratio, 95% confidence intervals and p-values, as estimated by multivariable Poisson regression models24 and adjusted by sex, age, color/race, housing area, fluoridated piped water, school delay, household crowding and family income.
The comparison of oral health outcomes among browns, blacks and whites in each region used the same analytical strategy, with adjustment for: (1) age only; and (2) sex, age, living area, fluoridated piped water, school delay, household crowding and family income. This assessment considered whites as the reference category and browns and blacks as comparison groups.
In the regression analyses, the maximum number of participants corresponded to those with non-missing values in all variables included in the models. The variable with the highest percentage of missing observations was school delay (2%). Statistical analysis was performed with Stata, version 9 for Windows. The selection of participants in the Brazilian Oral Health Survey followed a multistage, probabilistic sampling design, but the sampling process was never concluded, so that sampling weights and the sample's structural variables (selection strata and primary sampling units) could not be correctly defined25. However, since the main objective of the present study is to evaluate the relationship between exposures (e.g.: color/race) and outcomes (e.g.: perceived dental treatment need) and not to produce precise prevalence estimates for the country as a whole, absence of sampling weights and structural variables in the analyses are not of major concern.
The study protocol was approved by the Brazilian National Human Research Ethics Council, document number 581/2000, on 21 July 2000. All participants or their guardians, in case of individuals under the age of 18, provided written informed consents.
The proportion of self-classified brown and black adolescents was higher in the North and Northeast regions; these regions also presented a worse profile of socioeconomic status; i.e. higher proportion of adolescents with school delay, household crowding and family income lower than two Brazilian minimum wages (Table 1). The proportion of adolescents living in towns without fluoridated piped water was expressively higher in the North and Northeast regions.
Perceived dental treatment need, untreated dental caries, and gingival bleeding were the most prevalent oral health outcomes nationwide; periodontal pocket was the least prevalent. Having had at least one episode of moderate to severe pain in teeth or gums during the previous six months affected from 14% to 18% of adolescents in each region. All oral health outcomes were unequally distributed among Brazilian regions; in general, the North and Northeast regions presented a poorer oral health profile than the remaining ones (Table 1).
The Southeast region presented a better overall profile of oral health than the North and, especially, the Northeast. The inter-region comparisons indicated a significant difference between the Southeast and the Northeast for perceived dental treatment need, untreated dental caries and gingival bleeding (Table 2); the difference of higher magnitude was observed for gingival bleeding at probing. Untreated dental caries was the only oral health outcome ranking significantly higher (p<0.05) for the South and Center-West regions than for the Southeast. For the remaining oral health outcomes, no statistically significant differences were observed.
The majority of oral health outcomes were unequally distributed according to color/race in each Brazilian region (Table 3). Overall, a color/race gradient in the prevalence of these outcomes was observed: the darker the participants' color, the higher the prevalence of unfavorable oral health conditions. Importantly, there were exceptions to this oral health gradient with, for instance, blacks in the Northeast scoring lower than browns and whites with regard to periodontal pockets. Another exception was that blacks showed prevalence estimates half the way between that of browns and whites, such as in the cases of perceived dental treatment need in the South, untreated dental caries in the Southeast, and pain in teeth and gums in the Northeast. For some oral health outcomes, though, the prevalence in blacks was equal to that of whites (e.g.: perceived dental treatment need in the Center-West and untreated dental caries in the Northeast, see Table 3).
Despite the fact that the Southeast region had a better overall profile of oral health, it was the single region presenting color/race inequalities in all oral health outcomes (Tables 3 and 4). In the Southeast region, black adolescents had a prevalence of periodontal pockets five times higher, as well as 48% more pain in teeth and gums, in comparison with whites (Table 3). When controlled by socio-demographic variables and access to fluoridated water, these differences were slightly reduced, but remained statistically significant (Table 4). Blacks from the Northeast were almost 60% less affected by periodontal pockets than whites (Table 3); further adjustment by socio-demographic factors and access to fluoridated water did not expressively affect the magnitude of this difference (Table 4). With regard to brown adolescents in the Southeast, these presented a 30%, 17% and 8% higher prevalence of gingival bleeding, untreated dental caries and perceived dental treatment need, respectively, when compared to whites, net of socio-demographics and access to fluoridated water (Table 4). Untreated dental caries was also more prevalent in browns in the North and Center-West regions even after full adjustment for covariates. In the last-mentioned region, the prevalence of untreated dental caries was also higher in blacks, when compared to whites (Tables 3 and 4). Compared to whites, browns from the South showed a 10% higher prevalence of perceived dental treatment need.
This study documented and assessed the magnitude of inequalities in adolescents' oral health outcomes among Brazilian geographic regions, and among color/race groups in each geographic region. The most relevant result is the identification of the Southeast - the richest Brazilian region - as concurrently presenting an improved overall profile of oral health in the country and color/race inequalities in all oral health outcomes under consideration.
It is important to mention, though, that the magnitudes of the majority of color/race inequalities, specifically for perceived dental treatment need, untreated dental caries, gingival bleeding and pain in teeth and gums, was low and involved brown-white comparisons more frequently. Browns perceived the need for dental treatment more, were more affected by untreated dental caries, and showed a higher prevalence of gingival bleeding. In these cases, in spite of having effects of similar magnitude also reported for blacks, statistical significance was mostly observed for browns due, probably, to the higher sample size of browns in all Brazilian regions.
The poorer oral health status of brown and black adolescents had already been reported in Brazil, as regards a higher prevalence of decayed teeth26. The prevalence of untreated dental caries and gingival bleeding was previously documented as higher for brown and black schoolchildren, also categorized in only one group countrywide, than for their white counterparts18,19. With regard to pain in teeth and gums, a similar finding was observed in adults from a Southern Brazilian city: blacks living in Pelotas showed a 30% higher toothache prevalence than whites, while the prevalence in browns was not statistically different from that of whites27.
The color/race inequality of highest magnitude was observed for periodontal pockets and was statistically higher for blacks living in the Southeast and lower for blacks in the Northeast, when compared to whites. The latter represents an unexpected finding. Although not statistically significant, browns and blacks in the North and South were equally more affected by periodontal pockets than whites. Peres et al.20 reported a higher prevalence of periodontal disease in brown and black than in white adults, and stated that the odds of blacks was slightly higher, after adjustment for schooling level, income, sex, age and geographic region. Therefore, the current study confirms and reinforces previous observations on color/race inequalities for several oral health outcomes in the Brazilian context.
This study also reported that the North and Northeast regions had the poorest indices of oral health, which is not surprising, since these regions present the poorest profile of socioeconomic status in the country, and the association between deprivation and oral health has been extensively evidenced for several Brazilian contexts. Therefore, the identification of poorer oral health in deprived regions is consistent with studies showing socioeconomic inequalities in the prevalence of dental caries in adolescents living in different Brazilian contexts for the country as a whole18; for the states of São Paulo21; Bahia28; Santa Catarina29,30; Rio Grande do Sul31; and Paraná32, in particular.
Following a major strand of research in racial inequalities in health, the present study assessed oral health differentials among color/race categories, adjusting for possible confounders in the context of multivariable modeling. Such a methodological approach has strengths and weaknesses, which all deserve to be mentioned and appreciated more carefully.
First, it is clear that this strategy can provide important evidence on societal patterns of distribution of health and disease. As long as color/race is recognized as an important dimension to study variability in oral health-related outcomes, detection of racial inequalities in oral health may be used to assess ongoing public policies, actions and/or to inform new oral health interventions. Indeed, this is a polarized topic of discussion in Brazil, since the need to promote race-based policies and interventions to tackle social inequalities in health is fulfilled with controversies among scholars, politicians and the lay public in general14,33.
The strength of this methodological approach also lies in identifying the whole spectrum of possible confounders to adjust for in multivariable analyses assessing the relationship between the race variable and a selected oral health outcome. For example, the fluoridation of water supplies is acknowledged as an effective measure for the prevention of dental caries34. However, its unequal implementation in the country was reported as a source of inequality in the experience of dental disease35. Thus, besides including socio-demographic variables, regression models in this study also considered the adjustment for the availability of fluoridated piped water in each town. Therefore, the analytical scheme adopted in this study attempted to minimize the possibility of residual confounding36 due to the omission of an important variable in the model37 as a possible explanation for the persistence of color/race inequalities in oral health.
However important color/race may be for the abovementioned purposes, the complexity involved in racial classification in Brazil must also be highlighted. Many scholars have commented in detail on the difficulties of using the race variable in epidemiologic studies38-40, with some recent quantitative investigations exploring Brazilian specificities of racial classification6,7,41,42. In essence, these authors raise concerns about the validity and reliability of the race variable, and recommend that it should be used very cautiously in health-related publications.
Color/race may have different meanings in Brazil, on account of its large territory and historical processes of immigration and occupation. Browns in the North, for example, are of predominantly indigenous ancestry when compared to those equally categorized in the remaining regions. Regional specificity may undermine the process of generalization of results, as well as contribute to unexpected findings, such as the one involving a lower prevalence of periodontal pockets in blacks from the Northeast. This finding deserves to be investigated more deeply, and peculiarities of the racial classification system and racial relations that may be at play in the Northeast should be taken into account.
A final and important limitation of the present analysis is that persistence of color/race inequalities after adjustment for confounding does not provide a single or most probable interpretation with respect to their underlying causes. As already mentioned, these inequalities may be attributed to residual confounding, but also to consequences of unmeasured aspects of racial discrimination (e.g.: chronic psychological stress) or unmeasured factors that are associated with both color/race and the specified outcome but not related to either discrimination or socioeconomic position, such as culturally shaped patterns of health related behaviors36.
Some authors have proposed that discriminatory processes based on racial categorization may contribute to color/race inequalities in periodontal disease prevalence. Since this oral health outcome may be etiologically influenced by stress, and acknowledging that racial discrimination is commonly recognized as a stressful experience36, Peres et al.20 used color/race as a proxy for racial discrimination to assess the association between this variable and periodontal disease in Brazil. However, these authors also recognized that, due to the nature of the data at hand, they could not provide further evidence supporting the hypothesis of an association between race-based discrimination and periodontal disease. This is exactly the same case for the present study.
In conclusion, the inequality of oral health among browns, blacks and whites was predominantly low and systematically more frequent in the Southeast, the richest Brazilian region. Furthermore, the Brazilian Dental Council (www.cfo.org.br) acknowledges that 59% of dental practitioners in the country and 50% of dental schools are located in the Southeast region. In spite of being the region with the highest rankings by GDP and a higher provision of dental services, the Southeast region did not extend these benefits to brown and black adolescents to the same extent than for whites. In spite of being of low magnitude, this inequality may be considered avoidable and unjust.
The prevalence of unfavorable conditions of all oral health outcomes assessed in this study is amenable to reduction by educational programs, prevention and dental treatment. Supervised tooth brushing, dental plaque identification, expansion of access to fluoridated piped water and restorative dental treatments, among other initiatives of oral health promotion may reduce the burden of dental disease, prevent avoidable pain, and improve the quality of life of adolescents. These initiatives must be programmed with an effective universal scope in order to prevent the exclusion of extended population segments or target specific population groups that demand additional resources, when persistent and expressive inequalities have been identified.
Socially appropriate health programs should aim at addressing the overall improvement of health indices and the concurrent reduction of inequalities in the distribution of negative outcomes. Suppressing color/race inequalities in health is a socially relevant goal for health services. It should also be considered an important strategy for the promotion of social justice.
1. PNUD. Programa das Nações Unidas para o Desenvolvimento. Instituto de Pesquisa Econômica Aplicada. Relatório do Desenvolvimento Humano no Brasil. Brasília: Fundação João Pinheiro; 2003.
2. IBGE. Censo Brasileiro 2000. Rio de Janeiro: Instituto Brasileiro de Geografia e Estatística; 2001. Report No. 1.
3. Fry PH. A persistência da raça: ensaios antropológicos sobre o Brasil e a África Austral. Rio de Janeiro: Civilização Brasileira; 2005.
4. Sansone L. Negritude sem etnicidade: o local e o global nas relações raciais e na produção cultural negra do Brasil. Salvador: Edufba - Pallas; 2007.
5. Silva NV. Uma nota sobre "raça social" no Brasil. In: Hasenbalg C, Silva NV, Lima M. Cor e estratificação social. Rio de Janeiro: Contracapa; 1999. p. 107-125
6. Telles EE. Racial ambiguity among the Brazilian population. Ethn Racial Stud 2002; 25: 415-41.
7. Telles EE, Lim N. Does it matter who answers the race question? Racial classification and income inequality in Brazil. Demography 1998; 35: 465-74.
8. Guimarães ASA. Preconceito racial: modos, temas e tempos. São Paulo: Cortez; 2008.
9. Holanda SB. Raízes do Brasil. São Paulo: Companhia das Letras; 2007.
10. Chor D, Lima CR. Aspectos epidemiológicos das desigualdades raciais em saúde no Brasil. Cad Saúde Pública 2005; 21: 1586-94.
11. Hasenbalg C, Silva NV. Notes on racial and political inequality in Brazil. In: Hanchard M. Racial politics in contemporary Brazil. Durham: Duke University Press; 2005. p. 154-178
12. Telles EE. Racismo à brasileira: uma nova perspectiva sociológica. Rio de Janeiro: Relume Dumará; 2003.
13. Carvalho JM. Genocídio racial estatístico. In: Fry PH, Maggie Y, Maio MC, Monteiro S, Santos RV. Divisões perigosas: políticas raciais no Brasil contemporâneo. Rio de Janeiro: Civilização Brasileira; 2007. p. 111-5
14. Fry PH, Monteiro S, Maio MC, Bastos FI, Santos RV. AIDS tem cor ou raça? Interpretação de dados e formulação de políticas de saúde no Brasil. Cad Saúde Pública 2007; 23: 497-507.
15. Krieger N. Stormy weather: race, gene expression, and the science of health disparities. Am J Public Health 2005; 95: 2155-2160.
16. Dressler WW, Oths KS, Gravlee CC. Race and ethnicity in public health research: models to explain health disparities. Annu Rev Anthropol 2005; 34: 231-52.
17. Antunes JL, Pegoretti T, de Andrade FP, Junqueira SR, Frazao P, Narvai PC. Ethnic disparities in the prevalence of dental caries and restorative dental treatment in Brazilian children. Int Dent J 2003; 53: 7-12.
18. Antunes JL, Peres MA, de Campos Mello TR, Waldman EA. Multilevel assessment of determinants of dental caries experience in Brazil. Community Dent Oral Epidemiol 2006; 34: 146-52.
19. Antunes JL, Peres MA, Jahn GM, Levy BB. The use of dental care facilities and oral health: a multilevel approach of schoolchildren in the Brazilian context. Oral Health Prev Dent 2006; 4: 287-94.
20. Peres MA, Antunes JL, Boing AF, Peres KG, Bastos JL. Skin colour is associated with periodontal disease in Brazilian adults: a population-based oral health survey. J Clin Periodontol 2007; 34: 196-201.
21. Frias AC, Antunes JL, Junqueira SR, Narvai PC. Determinantes individuais e contextuais da prevalência de cárie dentária não tratada no Brasil. Rev Panam Salud Publica 2007; 22: 279-85.
22. Kallestal C, Dahlgren L, Stenlund H. Oral health behavior and self-esteem in Swedish adolescents over four years. J Adolesc Health 2006; 38: 583-90.
23. Antunes JL, Frazão P, Narvai PC, Bispo CM, Pegoretti T. Spatial analysis to identify differentials in dental needs by area-based measures. Community Dent Oral Epidemiol 2002; 30: 133-42.
24. Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003; 3: 21.
25. Queiroz RC, Portela MC, de Vasconcellos MT. Pesquisa sobre as Condições de Saúde Bucal da População Brasileira (SB Brasil 2003): seus dados não produzem estimativas populacionais, mas há possibilidade de correção. Cad Saúde Pública 2009; 25: 47-58.
26. Gushi LL, Soares Mda C, Forni TI, Vieira V, Wada RS, de Sousa Mda L. Cárie dentária em adolescentes de 15 a 19 anos de idade no Estado de São Paulo, Brasil, 2002. Cad Saúde Pública 2005; 21: 1383-91.
27. Bastos JL, Gigante DP, Peres KG. Toothache prevalence and associated factors: a population based study in southern Brazil. Oral Dis 2008; 14: 320-6.
28. Cangussu MC, Castellanos RA, Pinheiro MF, Albuquerque SR, Pinho C. Cárie dentária em escolares de 12 e 15 anos de escolas públicas e privadas de Salvador, Bahia, Brasil, em 2001. Pesqui Odontol Bras 2002; 16: 379-84.
29. Bastos JL, Nomura LH, Peres MA. Dental caries and associated factors among young male adults between 1999 and 2003 in Southern Brazil. Community Dent Health 2007; 24: 122-7.
30. Peres MA, Peres KG, Traebert J, Zabot NE, Lacerda JT. Prevalence and severity of dental caries are associated with the worst socioeconomic conditions: a Brazilian cross-sectional study among 18-year-old males. J Adolesc Health 2005; 37: 103-9.
31. Peres MA, Peres KG, de Barros AJ, Victora CG. The relation between family socioeconomic trajectories from childhood to adolescence and dental caries and associated oral behaviours. J Epidemiol Community Health 2007; 61: 141-5.
32. Baldani MH, Vasconcelos AG, Antunes JL. Associação do índice CPO-D com indicadores sócio-econômicos e de provisão de serviços odontológicos no Estado do Paraná, Brasil. Cad Saúde Pública 2004; 20: 143-52.
33. Maio MC, Monteiro S. Tempos de racialização: o caso da 'saúde da população negra' no Brasil. Hist Cienc Saúde-Manguinhos 2005; 12: 419-46.
34. Centers for Diseases Control and Prevention. Achievements in public health, 1900-1999. Fluoridation of drinking water to prevent dental caries. Morb Mortal Wkly Rep 1999; 44: 1-40.
35. Peres MA, Antunes JL, Peres KG. Is water fluoridation effective in reducing inequalities in dental caries distribution in developing countries? Recent findings from Brazil. Soz Praventiv Med 2006; 51: 302-10.
36. Krieger N. Discrimination and health. In: Berkman LF, Kawachi I. Social epidemiology. New York: Oxford University Press; 2000. p. 36-75
37. Blank RM, Dabady M, Citro CF. Measuring racial discrimination: panel on methods for assessing discrimination. Washington: The National Academies Press; 2004.
38. Kaplan JB, Bennett T. Use of race and ethnicity in biomedical publication. JAMA 2003; 289: 2709-16.
39. Laguardia J. O uso da variável "raça" na pesquisa em saúde. Physis 2004; 14: 197-234.
40. Senior PA, Bhopal R. Ethnicity as a variable in epidemiological research. BMJ 1994; 309: 327-30.
41. Bastos JL, Peres MA, Peres KG, Dumith SC, Gigante DP. Diferenças socioeconômicas entre autoclassificação e heteroclassificação de cor/raça. Rev Saúde Pública 2008; 42: 324-34.
42. 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: 171-80.
Recebido em: 04/03/09
Versão final reapresentada em: 22/06/09
Aprovado em: 30/06/09
▲ This study was funded by the Brazilian Council for Scientific and Technological Development (CNPq), edict MCT/CNPq/MS-SCTIE-DECIT No. 026/2006 on Social Determinants of Health, Grant No. 40.3362/2004-0.
** Brasil. Projeto SB Brasil 2003. Condições de saúde bucal da população brasileira 2002-2003: resultados principais. Brasília: Ministério da Saúde, Secretaria de Atenção à Saúde, Departamento de Atenção Básica, Coordenação Nacional de Saúde Bucal, 2004.