Severe tooth loss influences the perception of racial and gender discrimination among Brazilian older adults

Abstract

Objective

To analyse whether severe tooth loss influences perceived racial and gender discrimination in Brazilian older adults.

Methods

The data were obtained from the Brazilian Longitudinal Study of Ageing, a nationally representative sample of community-dwelling individuals aged 50 and older. The outcome measured was the perception of discrimination in five different settings. The main explanatory variables were self-identified/self-perceived race (Whites and Non-whites) and gender, while covariates included education level, equivalized per capita income, depression, and the use of dental prostheses. Negative binomial and logistic regressions with interaction terms between race, gender and severe tooth loss were used to estimate the intersections between gender and race-related perceived discrimination inequities.

Results

The analytical sample comprised 8,556 respondents with complete variables. While the prevalence of perceived discrimination in white Brazilians was 14.3% (95%CI: 12.2;16.3), it was 18.8% (95%CI: 16.8;20.6) among non-Whites. The interaction of being female, non-white and with severe tooth loss showed higher chances of perceived discrimination in 50–59 years [RR = 1.53(95%CI: 1.01;2.28)]. Moreover, being non-white with severe loss in 60–69 years [RR = 1.74(95%CI: 1.13;2.69)] had higher chances of perceived discrimination adjusted by all covariates. A sensitivity analysis of all older adults confirmed the findings.

Conclusions

The findings highlight the intersection between severe tooth loss, race, and gender, with more pronounced associations among women and non-white individuals. These results reinforce the importance of equity-oriented oral health policies and interventions within the Family Health Strategy, particularly for populations experiencing severe tooth loss.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-26384-7.

Introduction

A growing body of evidence indicates that racial discrimination is becoming a significant risk factor for various diseases and plays a role in perpetuating racial health disparities [1]. Despite the ongoing debate regarding the inclusion of race in biomedical research [2], addressing racial disparities in oral health is crucial for promoting equity and social justice. Discrimination can manifest in scenarios where individuals perceive themselves as being treated with hostility, rejection, or disdain due to attributes like skin colour, age, gender, or physical appearance [3]. Previous accounts on racial inequalities in health have assumed that excessive levels of chronic morbidity and disability are widespread among black women in the United States, as well as that early health deterioration results from the cumulative impact of repeated experiences with social, economic, and political exclusion [4]. In this sense, race and gender variables are important predictors for suffering discrimination, at earlier stages of life [5]. Black men and women reported more frequent experiences of each type of discrimination than their White counterparts [5], and demonstrate that the health-harming effects of discrimination on psychological distress begin early in the life course and suggest that resilience-based coping mechanisms found in older samples of Black adults may also exist for younger adults [5].

According to the World Health Organization (WHO), having at least twenty permanent teeth is essential for individuals to engage in social activities and maintain proper chewing function [6]. This threshold is referred to as a functional dentition [7]. On the other hand, experiencing severe tooth loss, defined as having fewer than ten teeth remaining in the mouth [8], can have a more profound effect on a person’s quality of life. Brazilian adults and older adults with multimorbidity have a higher chance of severe tooth loss and a lower chance of having functional dentition [9, 10] reinforcing the importance of oral public policies to identify those individuals. This finding reinforces the importance of oral public policies aimed at identifying, monitoring, and effectively addressing the oral health needs of these individuals.

Recent evidence highlights that racism itself must be recognized as a structural driver of oral health inequities, reinforcing that experiences of discrimination are not merely individual events, but part of broader social mechanisms that shape tooth loss and access to dental care [11]. In one Brazilian study, with data from the national health survey, authors have shown that among adults, 11.5% reported being discriminated against in health services and regarding dental characteristics, the higher the tooth loss was, the higher the prevalence of discrimination [12]; They considered the use of dental prostheses as a protective factor; however, the study did not apply the classification of severe tooth loss [8]. By analysing the association between severe tooth loss and the perception of discrimination among Brazilian older adults, including race and gender covariates, this work aims to provide evidence for healthcare professionals, researchers, and policymakers to improve the environment considering the older adults, aiming for their well-being. Understanding these relationships can inform interventions and strategies that promote not only oral health but also the mental and emotional health of older adults, contributing to the construction of a more inclusive and equitable society. The hypothesis to be tested is that severe tooth loss could influence the perception of discrimination between race and gender, independently of dental prosthesis use.

Methods

Study population

The Brazilian Longitudinal Study of Ageing (ELSI-Brazil) is a large cohort study based on a population sample that aims to represent Brazilians aged 50 and above. The primary goal of the study is to explore the dynamics and determinants of ageing in the Brazilian context. Baseline data collection was carried out in 2015-16, with further details available in previous publications [13]. Individual variables were collected through face-to-face interviews, conducted with a structured questionnaire at the participants’ homes.

Outcome

The outcome was self-perceived discrimination. The Perceived Discrimination Scale was used [14] and adapted. This scale is widely used to measure the frequency of everyday experiences of discrimination [14]. It has been adapted for different contexts and populations, including the ELSI-Brazil, to assess discrimination across multiple life domains. Instruments like the Everyday Discrimination Scale are often adapted to cover different aspects of people’s lives, such as social encounters, access to healthcare services, work, etc., to better understand where and how discrimination is perceived, based on yes and no responses. The five questions used to assess discrimination were: “In the last 12 months, have you felt like a victim of some kind of discrimination, when ” 1) used medical or health services; 2) social gatherings; 3) work ambient; 4) in the family; 5) place that you live. We scored it from 0 to 5 points if the elders felt discrimination in all places. Zero is considered no self-perceived discrimination in any place and five was considered in all places.

Tooth loss assessment

Tooth loss was measured by asking: " How many natural teeth do you have at the…? 1) upper mouth; 2) lower mouth. Upper and lower self-reported natural teeth count was analyzed and classified into the severity of tooth loss, regardless of the tooth position. Severe tooth loss was considered the loss of 23–32, i.e., lower than 10 natural teeth in the mouth [8].

Main exposures - Racial and gender classification

The self-reported race was based on the classification of the Brazilian Institute for Geography and Statistics (IBGE) [15], which includes Whites, Browns, Blacks, Yellow and Indigenous. We stratified into Whites and Non-Whites for this analysis because Whites are the more privileged racial group in Brazil. Self-report gender was classified into male or female.

Covariates

Age (divided into groups of 50–59, 60–69, 70+), schooling (0, 1–4, 4–7, 7–11 and 11 + years of formal education) and income (equivalised per capita income, grouped into under and above one minimum wage) [16]. In 2016, the Brazilian minimum wage was considered BRL$ 880 or USD$ 275,00. Self-reported depression was assessed with the following question: “Has any doctor ever told you that you have depression?” (yes; no); and the Use of dental prosthesis was assessed with the following question: “Do you use any type of removable prosthesis to replace the teeth in the part of low mouth/upper mouth?” (yes/no).

We followed the STROBE guidelines for human observational studies [17].

Statistical analysis

First, we estimated an unadjusted negative binomial regression due to the variance of the score two times higher than the mean of perceived discrimination. The model aimed to estimate the magnitude of racial, gender and severe loss inequities in self-perceived discrimination. Then we adjusted for education, equivalised per capita income, self-reported depression and use of dental prosthesis. Associations between race, gender and severe loss were expressed as Risk Ratios (RR), following the assessment of goodness-of-fit criteria through AIC and BIC. Analyses were run using Stata 14.2, taking into consideration the complex sampling design and the sampling weights. Age groups were also analysed separately to identify potential differences among them.

We made one sensitivity analysis considering all older adults of all age groups (Appendix1).

Ethics and consent to participate

The Fiocruz Research Ethics Committee, Minas Gerais (CAAE 34649814.3.0000.5091), approved the ELSI-Brazil study protocol [12]. All participants signed separate informed consent forms for each of the research procedures.

Results

The analytical sample comprised 8,556 ELSI-Brazil participants aged 50 years and over who had information on all variables included in the study. A total of 856 participants were excluded due to missing data in one or more key variables required for the multivariable models. Table 1 shows the characteristics of these complete cases. The prevalence of discrimination was higher among Females, Non-Whites, those with depression and under one minimum wage. Moreover, severe tooth loss was higher in Non-whites, Females, 0 years of formal education, those earning less than one minimum wage and those who use removable prostheses.

Table 1.

Sociodemographic characterístics and covariates. Elsi Brazil (n=8556)

Individual variables	n = 8,556	% (95%CI)	Severe tooth loss	Discrimination
			% (CI95%)	% (CI95%)
Ethnic Group
Whites	3430	43.1 (37.7–48.8)	54.3 (48.7; 59.8)	14.3 (12.2; 16.3)
Pardos/Blacks/Yellow/Indigenous	5126	56.9 (51.2–62.3)	59.2 (52.8; 65.3)	18.8 (16.8; 20.6)
Sex
Female	4839	54.0 (50.8; 57.1 )	60.9 (42.9; 50.0)	17.6 (15.8; 19.5)
Male	3717	46.0 ( 42.9; 49.2)	46.4 (58.0; 63.7)	16.0 (14.1; 18.0)
Schooling
no formal education	1335	12.6 (10.3; 15.3)	76.1 (72.5; 79.3)	15.1 (12.3; 18.4)
1 to 4 years	1730	18.9 (17.2; 20.7)	69.7 (66.9; 72.5)	14.6 (12.3; 17.3)
4 to 7 years	2606	31.3 (28.8; 34.1)	58.4 (55.2; 61.5)	17.6 (15.7; 19.6)
7 to 11 years	962	12.2 (10.9; 13.6)	42.7 (36.6; 49.0)	18.2 (15.2; 21.5)
Above 11 years	1923	24.9 (22.7; 27.3)	24.5 (22.0; 27.2)	17.8 (15.2; 20.7)
Dental prosthesis use
no	2645	31.1(28.7; 33.7)	25.8 (22.8; 29.1)	16.4 (14.4; 18.6)
yes	5911	68.9 (66.3; 71.3)	66.4 (61.0; 67.7)	17.0 (15.3; 18.8)
Depression
no	7010	81.7 (79.7; 83.5)	51.9 (48.6; 54.6)	14.4 (12.6; 16.0)
yes	1546	18.3 (16.7; 20.3 )	56.1 (52.2; 60.0)	27.8 (25.1; 30.8)
Income (equivalized)
under 1 minimum wage	2321	24.9 (21.8; 28.3 )	60.9 (57.8; 63.8)	18.5 (16.3; 20.9)
above 1 minimum wage	6002	75.1 (71.7; 78.2)	50.1 (46.7; 53.5)	16.2 (14.7; 17.9)

Table 2 showed that the unadjusted negative binomial regressions models showed only significative associations only between non-whites in all age groups. After adjusting for socioeconomic status, depression and use of dental prostheses, we still only observe racial inequities in self-perceived discrimination. However, the interaction terms showed that being female, non-white and with severe loss (3-interaction terms) had higher chances of perceived discrimination in 50–60 years [RR = 1.53(95%CI: 1.01;2.28)] and 70+ [RR = 1.60(95%CI: 1.02;2.81)] adjusted for all covariates. We only observe the 3-interaction terms surpassing the risks of one individual exposure or 2-interactions in the age group 50–60 years (younger elders). Moreover, the intersection of being non-white with severe loss in 60–70 years [RR = 1.74(95%CI: 1.13;2.69)] has higher chances of perceived discrimination adjusted by all covariates than all other interaction terms.

Table 2.

Regression models of discrimination, race, gender and interaction terms with severe loss, Elsi - Brazil

		Discriminated in the last 12 months?
		Unadjusted RR (CI95%)	Adjusted RR (CI95%)*	Two Interaction Terms
		50–59 years
Gender		prevalence discrimination 18.5%
	Male	1		gender/tooth loss
	Female	1.30 (1.08; 1.57)	1.12 (0.92; 1.36)	1.25(0.95 ;1.64)
Race
	Whites	1		race/tooth loss
	Non-whites	1.23 (1.00; 1.50)	1.33 (1.06; 1.60)	1.48 (1.10; 2.00)
Severe Loss				three interaction terms (race/gender/tooth loss)
	No	1
	yes	1.16 (0.95; 1.40)	1.11 (0.94; 1.33)	1.53( 1.01; 2.28)
		60–69 years
Gender		prevalence discrimination 17.5%
	Male	1	1	gender/tooth loss
	Female	0.98 (0.75; 1.29)	0.88 (0.75; 1.22)	0.93 (0.57; 1.49)
Race
	Whites	1		race/tooth loss
	Non-whites	1.54 (1.21; 1.96)	1.56 (1.21; 2.00)	1.74 (1.13; 2.69)
Severe Loss				three interaction terms (race/gender/tooth loss)
	No	1
	yes	1.10 (0.83; 1.46)	1.00 (0.72; 1.45)	1.46 (1.01; 2.66)
		≥ 70 years
Gender		prevalence discrimination 12.5%
	Male	1		gender/tooth loss
	Female	1.00 (0.78; 1.29)	0.98(0.79; 1.34)	0.80 (0.53; 1.19)
Race
	Whites	1		race/tooth loss
	Non-whites	1.65 (1.27; 2.12)	1.63 (1.26; 2.12)	1.39 (0.87; 2.25)
Severe Loss				three interaction terms (race/gender/tooth loss)
	No	1
	yes	0.82 (0.62; 1.08)	0.78 (0.62; 1.08)	1.60 (1.02; 2.81)

* adjusted for income, schooling, depression and use of dental prosthesis. Bold values significative associations

The sensitivity analysis showed similar results, considering all age groups. The 3-interaction term surpasses the risk of one or 2-interaction terms, in adjusted analysis, corroborating our findings.

Discussion

This study revealed two key findings. First, our results indicate the existence of an intersecting system of racial oppression shaped by gender and severe tooth loss, which plays a crucial role in self-perceived discrimination among Brazilian older adults. Second, the intersectional effects were more pronounced among the younger elderly group (50–60 years), where the combined influence of race, gender, and tooth loss surpassed other risk combinations, regardless of dental prosthesis use. These findings suggest an association between earlier tooth loss and greater exposure to social stigma and discrimination, potentially reflecting cumulative oral health disadvantage and its aesthetic and identity-related implications.

Severe tooth loss has profound implications for older adults’ daily lives, from nutrition to self-image, and these consequences are often intensified by social stigma. As highlighted by Silva et al. (2010) [18], edentulism negatively affects social, psychological, and physical well-being, and our finding that non-white older adults, particularly women, report higher levels of perceived discrimination reinforces that these impacts are not equally distributed across population groups. Rather than a natural and inevitable process of ageing, severe tooth loss reflects long-standing social and health inequalities [19], in which marginalized racial groups face greater barriers to preventive and restorative care across the life course. Our results highlight the intersection between race, gender, and oral health as a key dimension for understanding vulnerability in older populations and emphasize the need for public policies that address these inequalities within a broader framework of social justice and health equity.

The association observed suggests that discrimination may operate both as a consequence and as a reinforcing mechanism of oral health inequities. Tooth loss can trigger stigmatization, which may further compromise mental health and limit social participation, particularly among women and non-white individuals [18, 19]. Another study [20] questioned whether severe tooth loss in older adults can be attributed solely to natural ageing or whether it reflects persistent social inequalities. Our findings are consistent with this argument, as the intersection between severe tooth loss and perceived discrimination reveals that edentulism is not merely an inevitable outcome of biological ageing but a manifestation of social and health inequities accumulated across the life course. In this sense, racism, and severe tooth loss intersect as structural and embodied dimensions of inequality that jointly shape social experiences and health vulnerabilities. The higher prevalence of discrimination among non-white older women with severe tooth loss demonstrates how these overlapping forms of oppression, acting through intersectionality, increase the likelihood of experiencing discrimination in later life.

Our findings are aligned with emerging literature that calls for acknowledging racism as a structural determinant of oral health inequities, influencing not only disease occurrence but also social reactions to dental conditions [11]. Recognizing racism as a driver of disparities highlights that the consequences of tooth loss extend beyond biological or behavioral origins, revealing social mechanisms that reinforce inequities among older adults in Brazil. Therefore, addressing severe tooth loss and its psychosocial impacts requires comprehensive and equity-oriented public policies that ensure access to care and actively combat discrimination in health and social environments.

This study has several strengths and limitations that should be noted. One major strength is the use of a large dataset from the ELSI-Brazil study, which is a nationally representative sample of individuals aged 50 and older. ELSI-Brazil is also part of a global network of harmonized ageing studies, including several countries such as the United States, England, Mexico, China, India, and South Africa. One limitation is the potential for reverse causality bias, as individuals with severe tooth loss may feel more vulnerable and perceive discrimination more frequently, regardless of whether their oral condition is the actual cause. Additionally, the subjective nature of perceived discrimination makes it challenging to establish a direct causal relationship and Internal consistency of discrimination questionnaire applied was not formally tested. Future research on discrimination and tooth loss should take these factors into account.

Conclusions

In summary, severe tooth loss was associated with exacerbated social inequalities and higher levels of self-perceived discrimination among Brazilian older adults. The findings highlight the intersection between severe tooth loss, race, and gender, with more pronounced associations among women and non-white individuals. These results reinforce the importance of equity-oriented oral health policies and interventions within the Family Health Strategy, particularly for populations experiencing severe tooth loss.

Abbreviations

CI

Confidence Interval

RR

Risk Ratio

ELSI-Brazil

Brazilian Longitudinal Study of Aging

WHO

World Health Organization

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

IBGE

Brazilian Institute for Geography and Statistics

OR

Odds Ratio

USD

United States Dollar

BRL

Brazilian Real

Authors’ contributions

Author 1 RND: Contributed to conception, design, drafted and critically revised the manuscript. Author 2 LCG: Contributed to conception, design, data analysis and interpretation, drafted and critically revised the manuscript. Author 5 RAB: Contributed to data analysis and interpretation, performed all statistical analysis, and drafted and critically revised the manuscript. All authors gave their final approval and agreed to be accountable for all aspects of the work.

Funding

No.

Data availability

The data that support the findings of this study are openly < (https://elsi.cpqrr.fiocruz.br/) >.

Declarations

Ethics approval and consent to participate

The Fiocruz Research Ethics Committee, Minas Gerais (CAAE 34649814.3.0000.5091), approved the ELSI-Brazil study protocol [12]. All participants signed separate informed consent forms for each of the research procedures.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.