Abstract

Introduction

Considered as real manifestation of prejudice,(1)1. Oliveira CLP, Barreto PCS. Percepção do racismo no Rio de Janeiro. Estud Afro-Asiat. 2003 jan;25(2):183-213. the most common forms of discrimination correspond to unfair and unfavorable(2)2. Dovidio JF, Hewstone M, Glick P, Esses VM. Prejudice, stereotyping and discrimination: theoretical and empirical overview. In: Dovidio JF, Hewstone M, Glick P, Esses VM, editors. The SAGE handbook of prejudice, stereotyping and discrimination. London: SAGE; 2010. p. 3-28. treatments addressed to specific individuals or social groups, and which impacts are expressed in different dimensions of life, such as health, especially in the care provided to health services users. The means by which discrimination can be expressed in the context of health care range from delays, negligence or treatment refusal, to harassment,(3)3. European Union Agency for Fundamental Rights. Inequalities and multiple discrimination in access to and quality of healthcare. Luxembourg: European Union Agency for Fundamental Rights; 2013. as well as provision of different or less conservative treatment options, depending on the patient.(4)4. Cabral ED, Caldas Jr AF, Cabral HAM. Influence of the patient's race on the dentist's decision to extract or retain a decayed tooth. Community Dent Oral Epidemiol. 2005 Dec;33(6):461-6.

Discriminatory practices may be based on characteristics such as ethnical identification, sex, age, physical appearance, social class and others, socially attributed or acquired. These multiple types of discrimination can be frequently found in the individuals' reports and can be simultaneously associated, according to the context, the place and the historical period studied.(5)5. Bastos JL, Goncalves H, Faerstein E, Barros AJ. Experiências de discriminação entre universitários do Rio de Janeiro. Rev Saude Publica. 2010 fev;44(1):28-38. The recognition and the study of the multiple forms of discrimination implies the adoption of a broader perspective, which takes into account multiple background factors and the complexity of discrimination, considering, in a more specific way, its effects on the population.(6)6. European Commission. Tackling multiple discrimination: practices, policies and laws. Luxembourg: Office for Official Publications of the European Communities; 2007.

Strong and consistent associations between discriminatory experiences (mostly racially motivated) and adverse health conditions have been demonstrated in the literature, including worse mental health conditions,(7)7. Shavers VL, Fagan P, Jones D, Klein WM, Boyington J, Moten C, et al. The state of research on racial/ethnic discrimination in the receipt of health care. Am J Public Health. 2012 May;102(5):953-66.^-(10)10. Bastos JL, Barros AJ, Celeste RK, Paradies Y, Faerstein E. Age, class and race discrimination: their interactions and associations with mental health among Brazilian university students. Cad Saude Publica. 2014 Jan;30(1):175-86. systemic arterial hypertension,(8)8. Pascoe EA, Smart Richman L. Perceived discrimination and health: a meta-analytic review. Psychol Bull. 2009 Jul;135(4):531-54.^,(9)9. Williams DR, Mohammed SA. Discrimination and racial disparities in health: evidence and needed research. J Behav Med. 2009 Feb;32(1):20-47. use/abuse of alcohol and tobacco,⁽⁹⁾ on the other hand, there are still few studies that deal, more broadly, with the discriminatory phenomenon in the context of health services. Particularly in Brazil, a prior study, carried out on a national scale, when evaluating users satisfaction with the public health care system, pointed out that approximately 10.0% of them felt victims of some type of discrimination perpetrated by health professionals, especially motivated by sex, age and social class.(11)11. Gouveia GC, Souza WV, Luna CF, Souza-Junior PR, Szwarcwald CL. Health care users' satisfaction in Brazil, 2003. Cad Saude Publica. 2005; 21 Suppl 1:109-18.

Despite these results, other research contexts must also be analyzed so that their particularities can be highlighted and, thus, provide improved consistency to what is documented in the country as a whole. The present study objective was to describe the prevalence of health service-related discrimination, its motivations and associated factors.

Methods

This is a cross-sectional analysis, carried out based on data from two epidemiological distinct population-based surveys. Individuals living in areas covered by the public primary health care services of Porto Alegre, Rio Grande do Sul State (RS), who had used this health service in the 24 months prior to the survey (the questionnaire, however, was not limited to the prior 24 months) were part of the first sample. The second sample comprised university students, regularly enrolled at the Federal University of Santa Catarina (UFSC), Campus João David Ferreira Lima, in Florianópolis, Santa Catarina State (SC). Data from both populations were collected between 2010 and 2012. The comparison between these two distinct groups is faced in this study as a positive aspect, because different contexts in the studied phenomenon can be assessed.

In Porto Alegre-RS (Sample 1), the sample was originated from a study whose primary objective was to evaluate the differences in the attributes of the primary health care provided by different types of services. Fifteen (15) Primary Health Care units were selected, from which six units belong to the Conceição Hospital Group Community Health Services (GHC), five are common basic health units (UBS) and four, Family Health Units (USF). The inclusion criteria for the units in the study were as follows: to have an Oral Health team with a dental surgeon and a dental health care assistant or technician; and to have a dental surgeon working in the unit for at least two years.

In the coverage area of each unit, census tracts were randomly selected. In each one of these census tracts, the first 40 households (or enough number to complete the minimum sample of 30 interviews) were randomly selected. In each census tract, the blocks of houses were randomly numbered and the field coordinator went to the corner of the first block, walking on the left until he or she identified the first house. At that moment, he or she would throw a coin and decided whether that house (heads) or the second house of the sequence (tails) would be chosen. From there on, they would follow skipping a house in order to address the odd (heads) or the even (tails) houses until completing the minimum sample. In each household a public health service user who met the following inclusion criteria was interviewed: being between 18 and 59 years old; and having used the health unity for examination or dental treatment in the prior 24 months. In the households with more than one eligible adult, only one was included, by random selection. The selected users were interviewed by previously trained field workers, using structured questionnaires. The eligibility was determined by applying an initial questionnaire with questions about the household residents, including name and age of all the residents, relationship between them, full address and telephone number.

For not having, initially, data that allowed performing the sample calculation, 30 interviews were conducted in each unit. The information gathered from these interviews was the base for the sample size calculus used in the study, estimating the average values of the Primary Care Assessment Tool scale for the three health unit models under study: GHC, UBS and USF. The data collected in the pilot study were also included in the final sample. The database of the Porto Alegre-RS population was built by scanning the questionnaires, using the software Teleform. Thereafter, the data were submitted to an editing process to correct inconsistencies and possible reading problems from both the program and the scanner device.

In Florianópolis-SC (Sample 2), participants were selected within a population of 19,963 students, regularly enrolled in the first semester of 2012 at the Federal University of Santa Catarina, Campus João David Ferreira Lima, considering information on the association between discriminatory experiences and self-assessment of general health. The sampling procedure was complex, two-staged, and the sample size calculation used data (prevalence of self-reported health 'very bad / bad / regular' between discriminated and not discriminated) of a similar population. In the first stage, which determined the primary sampling units, undergraduate courses were selected, proportionally to their size. Then, the students were selected according to previously defined categories, which were: first phase students; enrolled in the semester; and last phase students. Undergraduate students enrolled in 12 of the 70 undergraduate courses selected in the sampling process were considered eligible for the study.

A pre-test with 17 students outside the study target population was conducted, followed by a pilot study with 43 undergraduate students, that subsidized the development of an instruction booklet in order to standardize the fieldwork. The self-administered questionnaires were applied in classrooms; the data were typed twice, with subsequent review to verify possible failures in filling, including automatic checks for consistency and amplitude by the program EpiData Entry version 3.1.

The sociodemographic characteristics were raised from questions on:

To evaluate the health service-related discrimination experiences, three questions were applied. They were taken from an instrument to investigate the effects of discriminatory experiences on health conditions and behaviors.(12)12. Bastos JL, Faerstein E, Celeste RK, Barros AJ. Explicit discrimination and health: development and psychometric properties of an assessment instrument. Rev Saude Publica. 2012 Apr;46(2):269-78.

The first question was ' When attending health units, hospitals, emergency rooms and other health services, have you been treated differently compared to the others present there?', with the following answer options:

The first two answer options - i and ii - were considered as 'no', and the last three options - iii, iv and v -, as 'yes'.

The second question was 'When such thing happened, which may have been the reasons for you to have been treated that way?', with the following answer options:

The third question was 'On those occasions, have you felt discriminated?', with the answer options being 'no' and 'yes'.

The individuals who reported different treatment in healthcare services and interpreted the situation as discriminatory, according to the corresponding items of the instrument were considered discriminated.

This study considered as 'non respondents' those who refused to participate in the study in both capitals or who were not regularly enrolled in the educational institution during the fieldwork period in Florianopolis-SC.

The first step of statistical analysis included the investigated samples analysis, according to their socioeconomic and demographic characteristics. After that, we evaluated the association between the health service related-discrimination report and socioeconomic and demographic variables, using the chi-square test for heterogeneity in case of categorical variables or for linear trend in cases of ordinal variables (considering samples the complex outline, as well as the sample weights, in Florianopolis-SC case); or using the Fisher's exact test, when chi-square assumptions were violated (small sample). The global frequencies of discrimination were also estimated by calculating their 95% confidence intervals based on binomial distribution. At last, the most frequent motivations reported for the health services-related discrimination were described, according to the same sociodemographic characteristics of participants in both studies. All analyses were carried out in Stata 11.2, considering values of two-tailed probability below 5% as statistically significant.

Both projects were approved by the respective ethics committees in research, under the numbers 10-120 (Ethics Committee in Research of Hospital Group Conceição) and 459,965 (Ethics Committee in Research of the Federal University of Santa Catarina).

Results

A total of 1,943 individuals were contacted for interviews in Porto Alegre-RS; from which, 81 declined and 1,445 did not meet the inclusion criteria (mainly for not having consulted with the dentist of the analyzed services), ending up with 428 individuals who lived in the areas covered by the public primary health care services of Porto Alegre-RS. In Florianópolis-SC, 1,023 undergraduate students were interviewed (Table 1). The response rate in Porto Alegre-RS was 84.0%, and 81.0% in Florianópolis-SC.

Among the participants of the Porto Alegre-RS sample, most of them were women (79.0%), older than 51 years old (49.5%), self-declared white (58.9%), from the categories B2 to C2 according to the CCEB (85.4%), and who have never smoked (52.8%). The Florianópolis-SC sample consists, in its majority, of men (54.1%), aged between 21 and 30 years old (62.3%), self-declared white (81.1%), from the categories A2 to B2 according to the CCEB (77.3%), and who have never smoked (55.1%).

From the 428 individuals interviewed in Porto Alegre-RS, 13.6% (95%CI: 10.5;17.2) reported having been discriminated in health services, while among the university students, this proportion was lower, 7.4% (95%CI: 5.8;9.1) (tables 2 and 3). In Porto Alegre-RS, the categories with more reported discrimination were: individuals aged 31-40 years old (22.7%); with skin color/ethnicity black/dark (13.8%); men (16.5%); from the economic categories A1 to B2 according to the CCBE (16.0%); and smokers (17.6%). In Florianópolis-SC, the highest frequencies were: individuals aged 31-40 years old (17.5%); with skin color/ethnicity black/ dark (20.0%); women (8.6%); from category C1 of the CCEB (13.9%); and smokers (16.1%).

In Porto Alegre-RS, a statistically significant difference regarding age (p = 0.04) was noted, showing that individuals between 31-40 years old reported more discrimination. In Florianópolis-SC, on the other hand, the significant results in terms of difference in treatment noticed by respondents were also regarding age (p = 0.01), in addition to skin color/ethnicity (p = 0.01), economic classification (p = 0.01) and tobacco use (p = 0.02).

Among the reported motivations, the discrimination for social reason was the most cited, both in Porto Alegre-RS (34.5%, or 20 of 58 cases) and in Florianópolis-SC (26.7%, or 20 of 75 cases). In both samples, only the skin color/ethnicity variable was statistically associated with the type of discrimination reported (tables 4 and 5). In Porto Alegre-RS, no individuals with white or indigenous skin color reported discrimination for racial reason, while 36.0% (4 of 11 cases) of the individuals with black skin reported discrimination for this reason (p=0,03). In Florianópolis-SC, 30.0% (3 of 10 cases) of the individuals with black skin reported discrimination for racial reason, whilst individuals with white, brown or indigenous skin color did not report this (p=0,02). Socioeconomic position was not associated with the discrimination for social motivation in any of the two capitals (p>0,05). The young age was associated with the discrimination for age (p=0,03) in the university students sample: 19.3% (12 of 62 cases) reported discrimination for this reason, while older ages did not report having suffered such discrimination.

Discussion

Prevalence of health service-related discrimination among the surveyed individuals was 13.6% in Porto Alegre-RS and 7.4% in Florianópolis-SC. These results are similar to outcomes found in a national study on discriminatory experiences in health services, with a 9% prevalence (11% among the National Health System - SUS - users),(11)11. Gouveia GC, Souza WV, Luna CF, Souza-Junior PR, Szwarcwald CL. Health care users' satisfaction in Brazil, 2003. Cad Saude Publica. 2005; 21 Suppl 1:109-18. and in studies conducted in the cities of Belo Horizonte, Minas Gerais State (MG)(13)13. Macinko J, Mullachery P, Proietti FA, Lima-Costa MF. Who experiences discrimination in Brazil? Evidence from a large metropolitan region. Int J Equity Health. 2012 Dec;11(80):1-11. (2.3%) and São Paulo, São Paulo State (SP) (12.0%).(14)14. Barata RB, Ribeiro MC, Cassanti AC. Social vulnerability and health status: a household survey in the central area of a Brazilian metropolis. Cad Saude Publica. 2011;27 Suppl 2:S164-75. A study carried out in an adult population of Rio de Janeiro State showed a global prevalence of discrimination of 34.7%;(15)15. Celeste RK, Goncalves LG, Faerstein E, Bastos JL. The role of potential mediators in racial inequalities in tooth loss: the Pro-Saude study. Community Dent Oral Epidemiol. 2013 Dec;41(6):509-16. however, a population of university students reported a rate of approximately 77.6%.(10)10. Bastos JL, Barros AJ, Celeste RK, Paradies Y, Faerstein E. Age, class and race discrimination: their interactions and associations with mental health among Brazilian university students. Cad Saude Publica. 2014 Jan;30(1):175-86. Some of these differences can be explained by the fact that each study used instruments with a distinct number of items to evaluate discrimination; moreover, there are regional differences between the states, in terms of sampling, as well as differences in the time frame measurement of discrimination (discrimination in the past year versus discrimination throughout life, for example) which may reduce the comparability between these frequencies.

The prevalence of reported discrimination on health services was not statistically associated with any variable in Porto Alegre-RS sample. However, in Florianópolis-SC, individuals aged over 30 years, economically poorer, with black skin color and smokers reported higher prevalence of discrimination in health services. Similar results to Porto Alegre-RS were observed in Belo Horizonte-MG.(13)13. Macinko J, Mullachery P, Proietti FA, Lima-Costa MF. Who experiences discrimination in Brazil? Evidence from a large metropolitan region. Int J Equity Health. 2012 Dec;11(80):1-11. It can be inferred that the absence of such association is explained by the fact that the Porto Alegre-RS sample is composed by users of the public health system,⁽¹³⁾ - possibly they are users exactly for not suffering or noticing discrimination experiences in this area. Internationally, there are few reports on the prevalence of health services-related discrimination. In New Zealand, for example, it was found that the experience of racial discrimination committed by a health professional and noticed by the users varied from 1.2% to 7.1% among the different studied groups. The results of the New Zealander study pointed out that this discrimination can lead to underutilization of health services, besides altering the way medical advice is interpreted and practiced by users.(16)16. Harris R, Cormack D, Tobias M, Yeh LC, Talamaivao N, Minster J, et al. Self-reported experience of racial discrimination and health care use in New Zealand: results from the 2006/07 New Zealand Health Survey. Am J Public Health. 2012 May;102(5):1012-9.

With regard to discrimination reasons reported in the context of the investigated samples, although individuals claim to have lived discriminatory experience, a significant number of respondents (44.9% in Porto Alegre-RS and 53.3% in Florianopolis-SC) was unable to inform the reason for such discrimination and did not point out another reason for the fact. The identification of this reason would be important, as it would characterize if the experience was unfair and discriminatory or not.(17)17. Jamieson LM, Steffens M, Paradies YC. Associations between discrimination and dental visiting behaviours in an Aboriginal Australian birth cohort. Aust N Z J Public Health. 2013 Feb;37(1):92-3.

Thus, among the main results presented, the fact that in both capitals, most of the individuals were not able to identify the discrimination reason, corroborating with the previous study conducted in Rio de Janeiro,(10)10. Bastos JL, Barros AJ, Celeste RK, Paradies Y, Faerstein E. Age, class and race discrimination: their interactions and associations with mental health among Brazilian university students. Cad Saude Publica. 2014 Jan;30(1):175-86. deserves attention. In the present study, the most reported reason for health services-related discrimination was the socioeconomic position (20 individuals in Porto Alegre-RS and 20 in Florianopolis-SC).(11)11. Gouveia GC, Souza WV, Luna CF, Souza-Junior PR, Szwarcwald CL. Health care users' satisfaction in Brazil, 2003. Cad Saude Publica. 2005; 21 Suppl 1:109-18. The highest prevalence of discrimination reports was observed among individuals with black/dark skin color, smokers and aged 31 to 40 years old.

When verifying the frequency of each discrimination reason, according to the specific sample groups, the fact that the individuals who self-declared as black/dark or brown identify discrimination in health services as being racially motivated is noteworthy. Another national study also found this association.(18)18. Chor D, Lima CRA. Aspectos epidemiológicos das desigualdades raciais em saúde no Brasil. Cad Saude Publica. 2005 set-out;21(5):1586-94. Despite this, Brazilian studies (the study of Minas Gerais State (13)13. Macinko J, Mullachery P, Proietti FA, Lima-Costa MF. Who experiences discrimination in Brazil? Evidence from a large metropolitan region. Int J Equity Health. 2012 Dec;11(80):1-11. and another study, representative of the whole country population (19)19. Travassos C, Laguardia J, Marques PM, Mota JC, Szwarcwald CL. Comparison between two race/skin color classifications in relation to health-related outcomes in Brazil. Int J Equity Health. 2011 Aug;10(35):1-8.) did not find discrimination for racism in health services. The skin color, in Brazil as in other countries, is also related to the social class, educational level, income and geographic location,(20)20. Leal MC, Gama SG, Cunha CB. Racial, sociodemographic, and prenatal and childbirth care inequalities in Brazil, 1999-2001. Rev Saude Publica. 2005 Jan;39(1):100-7.^,(21)21. Lopes F, Buchalla CM, Ayres JR. Black and non-black women and vulnerability to HIV/AIDS in São Paulo, Brazil. Rev Saude Publica. 2007;41 Suppl 2:39-46. making it necessary to examine how these inequality axes act simultaneously, in the manifestation of the discriminatory phenomenon.

With regards to smoking, it must be considered that current smokers face discrimination of those who do not smoke, possibly resulting in a feeling of exclusion motivated by the fact of being a smoker.(22)22. Spink MJP. Ser fumante em um mundo antitabaco: reflexões sobre riscos e exclusão social. Saude Soc. 2010 jul-set;19(3):481-96. The unfair treatment motivated by age is a discrimination form that is usually related to the younger or older age groups.(23)23. Palmore EB. Ageism in Canada and the United States. J Cross Cult Gerontol. 2004 Mar;19(1):41-6. Other motivations reported in the discriminatory experiences also call the attention. The highest frequency among people in a higher socioeconomic position in Porto Alegre-RS suggests that this position can also constitute an important aspect for the established forms of treatment in health units. All these results reinforce the need to investigate other types of discrimination within the health services, and not just the types classically explored from the socio-anthropological point of view, such as the social class, ethnicity and sex discrimination.(24)24. Cecchetto F, Monteiro S. Discriminação, cor e intervenção social entre jovens na cidade do Rio de Janeiro (RJ, Brasil): a perspectiva masculina. Rev Estud Fem. 2006 jan-abr;14(1):199-218.

The studied populations present different characteristics: the profile of the residents in the areas covered by the public health services of Porto Alegre-RS differs from the profile of the undergraduate students of the Federal University of Santa Catarina, in Florianópolis-SC. This difference made a broader evaluation of the phenomenon possible, comparing the results observed between the two groups.

Nevertheless, some limitations to the development of this study must be highlighted. The fact that both samples refer to specific social groups makes it difficult to extrapolate the results to the general population of both cities. Both groups are socially and internally homogeneous, which can reduce the magnitude of the associations, influencing the statistical significance of the results for each group.

Another aspect to consider refers to the discrimination reported, that is representative of the experience lived by the individuals, possible to be affected by minimization or surveillance biases.(13)13. Macinko J, Mullachery P, Proietti FA, Lima-Costa MF. Who experiences discrimination in Brazil? Evidence from a large metropolitan region. Int J Equity Health. 2012 Dec;11(80):1-11. Thus, some discrimination cases may have been amplified or diminished, further or below of its real existence.(25)25. Pager D. Medir a discriminação. Tempo Soc. 2006;18(2):65-88.

Among the users of Porto Alegre-RS public health services, those who used these services in the prior two years were selected, probably, presenting a lower estimate of discrimination by health services. The non-participants individuals, if they had been discriminated with significant intensity and frequency, would possibly not use the services in the considered period and, therefore, they would not be included in the research.

We expect that the results presented here can subsidize researchers, managers and professionals in the elaboration and implementation of public policies that aim at improving the care provided by health services, seeking to mitigate social inequalities, reflected in them, and resulting injustices. We suggest that discussions on health service-related discrimination, even if they occur inside or outside the National Health System context, to be conducted by the approach of the multiple forms and different reasons - possibly combined or simultaneously experienced - of the discrimination that SUS users are victims.

Publication Dates

History