Education and income-based inequality in tooth loss among Brazilian adults: does the place you live make a difference?

Background Socioeconomic inequalities in tooth loss might be minimized or potentialized by the characteristics of the context where people live. We examined whether there is contextual variation in socioeconomic inequalities in tooth loss across Brazilian municipalities. Methods Data from the 2010 National Oral Health Survey of 9633 adults living in 157 Brazilian municipalities were used. The individual socioeconomic indicators were education and household income. At the municipal level, we used the Municipal Human Development Index (HDI) as our contextual indicator of socioeconomic status (low:< 0.699 versus high: > 0.70). The Relative (RII) and Slope (SII) Indexes of Inequality, Relative (RCI), and Absolute (ACI) Concentration Indexes were calculated to compare the magnitude of education and income-based inequalities among municipalities with low versus high HDI. Multilevel Poisson regression models with random intercepts and slopes were developed. Results At the individual level, adults with lower education & income reported more tooth loss. The mean number of missing teeth was 9.52 (95% CI: 7.93–11.13) and 6.95 (95% CI: 6.43–7.49) in municipalities with low and high HDI, respectively. Municipalities with high HDI showed higher relative and absolute education-based inequality. For income-based inequalities, higher SII and RCI was observed in municipalities with lower HDI. A significant cross-level interaction indicated that high-education adults reported fewer missing teeth when they lived in municipalities with high HDI compared to adults with the same education level living in low HDI municipalities. For individuals with the lowest education level, there was no difference in the number of teeth between those from municipalities with high and low HDI. Conclusions There was a social gradient in tooth loss by education and income. Living in disadvantaged municipalities cannot overcome the risk associated with low schooling. The protective effect of higher education can be reduced when people live in disadvantaged areas.


Background
Tooth loss is an important oral health problem with consequences for physical and psychosocial health as well as the quality of life [1][2][3]. It is a good proxy for the cumulative oral health status of individuals, [4] summarizing the impacts of adverse circumstances throughout the life course [5].
Despite a signi cant decline in the prevalence and incidence of severe tooth loss in the past two decades [6], socioeconomic inequalities in this condition persist across the globe, including Brazil [4,[7][8][9].
The socioeconomic inequalities in tooth loss have been studied considering compositional (individual) and contextual effects based mainly on the pathways linking income inequality to health [10][11][12][13][14].
Metanalysis research with 11 studies showed that adults with lower levels of income presented greater chance of tooth loss [13]. Considering contextual-level, collective evidence has shown associations between high income-based inequality and worse oral health [15]. Compositional explanations suggest that the effect of income inequality is a result of how individual incomes affect oral health. The proposed mechanisms involve the material resources (eg. the ability to purchase higher quality diets, to afford preventive and regular dental care due to the treatment cost), access to symbolic resources (status and rank within one's community) and behavioral/cultural explanations, which stress the role of poor health behaviors due to low income (tobacco use, high sugar consumption, infrequent and symptomatic dental visits and poor oral hygiene practices) [14,16]. Contextual explanations sustain that income inequality has broader psychosocial effects on oral health via stress-induced oral-health-related behaviors [17]. Income inequality also may result from lower social spending on public services and infrastructure, including dental care services and water uoridation [18].
The income inequality had been a frequent contextual social determinant of oral health in the epidemiologic studies. However, it may not capture all dimensions in which social inequality can occur. Different levels of education can also explain such disparities. Education is a socioeconomic indicator that captures the long-term in uences of both early life circumstances on adult health, as well as the in uence of adult resources (for example, through employment status) on health [19]. Previous studies have shown a consistent independent effect of lower individual education on tooth loss among adults [20][21][22][23] and elders [24]. Education equips individuals with knowledge and skills that are useful for the prevention of disease and attitudes toward healthy behavior. Also, higher educational attainment confers greater prestige and status within the community as well as serving as a credential for employment [14].
Beyond the interest in distinguishing the individual (compositional) source of variation from the contextual on socioeconomic inequalities studies, it is crucial to evaluate whether contextual factors affect every income and education group in the same manner [25]. Hence, it is reasonable to assess whether the variations among areas are similar for different income or education groups or whether the variations differ for different education and income groups. This approach is crucial from an equality policy perspective to target for interventions from speci c groups living in speci c areas. Additionally, it takes into account that socioeconomic diversity may minimize or potentialize the adverse oral health consequences of those living in an area with concentrated disadvantage [26].
A fewer number of studies assessed the interaction between oral health and contextual level factors, and most of them were focused on income inequality [20,23,31,35,36]. Differential effects of public policies on oral health among individuals with higher education and income had been observed [28,35]. These results have conducted to discuss if the public policy may both decrease or increase the social gap in oral health, considering the aim, the target population, and the approach of implementation of them (e.g. prioritization of municipalities with better socioeconomic). Can the disadvantage resulting from low education and income be decreased in areas that present more opportunities and conditions for healthy living? Or Do public policies have more effect on groups that are already socially advantaged? The double disadvantage, characterized by concomitantly presenting a worse social situation in the individual and contextual level, has been reported on oral epidemiology [37,38].
For that, it is crucial to investigate the income and education-based inequality considering the social characteristics of context level. A composite indicator of social development can re ect the contextual differences as a result of public policy. The starting point was to estimate the magnitude of income and education-based inequality according to areas with different social development. The aims of this study were therefore: i) to evaluate income and education-based inequalities in tooth loss comparing Brazilian municipalities with high and low Human Development Index (HDI); ii) to investigate the association between tooth loss and socioeconomic indicators on individual (income and education) and contextual (HDI) levels; iii) to determine if the social environment modi es the relationship between income/education and tooth loss. We hypothesized that tooth loss would be inversely associated with an individual's household income and education, but that the magnitude of the association would be more pronounced in a uent municipalities (high HDI) than in poor municipalities (low HDI).

Methods
Data from the present analysis came from the 2010 National Oral Health Survey (SB Brazil 2010) conducted by the Brazilian Ministry of Health in Brazilian urban areas [39] between February and November 2010. The sample was obtained through the random selection of municipalities and census sectors, via multi-stage cluster sampling with probability of selection proportional to population size. Detailed information on the methods is found in other publications [40,41]. Data for adults aged between 35 and 44 years were used in this study.
Individual interviews using a structured questionnaire was used to obtain demographic and socioeconomic characteristics. Oral health examinations were conducted in people's homes by calibrated dentists (kappa>0.65) under natural light following the guidelines of the WHO manual for epidemiological studies [42]. The DMFT (Decayed, Missing, and Filled Teeth) index was used to determine tooth status.

Outcome variable
The outcome variable was the number of teeth (discrete quantitative variable) that was missing for any reason, determined by the sum of codes 4 and 5 of the DMFT index.

Exposures
Education and income were used as measures of socioeconomic position at the individual level. Income was measured as total income received by all family members in the month preceding the survey (in seven categories from "R$250.00 or less" to "R$9500.00 or more"). For our analysis, the monthly household income was converted into multiples of the minimum wage, based on the current value at the time of survey (1 minimum wage = R$ 510.00, USD$303.57) and collapsed into four categories: up to 1, 1 to 2.9, 3 to 4.9, 5 or more times the minimum wage. This grouping was de ned to distingue mechanisms in which income exerts its effects. Education was asked as the number of years of formal schooling and classi ed as less than four years (insu cient education), 4-7 (incomplete elementary education), [8][9][10] (completed elementary, but incomplete secondary education), and 11 or more (completed secondary, incomplete university education, or college graduate), according to the formal education system in Brazil.
We used the Municipal Human Development Index (HDI) as an indicator of municipal area SES. The Brazilian HDI considers three dimensions: Longevity, Education (access to knowledge, based on average years studied of the population) and Income (living standards and purchasing power of the population according to the Municipal Gross Income per capita) [43]. The HDI was obtained from the 2013 Brazil Atlas of Human Development, which allows a selection based on data extracted from 2010 demographic. The values of HDI of sampled municipalities ranged from 0.481 to 0.847, and the groups are de ned as very low (0-0.499), low (0.500 -0.599), medium (0.600 -0.699), high (0.700 -0.799) and very high (0.800 -1.00). According to this classi cation, the frequency of sampled municipalities was: low and very low (3.93%), medium (7.88), and high and very high (88. 19). In this study municipalities were aggregated into low (<0.699) versus high (>0.70)

Covariates
The covariables at individual level were age (adult: 35-39 and 40-45 years old), sex (female, male), skin color/ethnicity (white, black, yellow and brown/Ameridians), time since the last dental visit (< 12 months, between 1 and 2 years, > 3 years, did not visit). Skin color refers to the classi cation adopted in the demographic census performed in Brazil: whites, blacks, browns, yellows, and Amerindians. In this study, this variable was dichotomized: white versus blacks, browns, yellows, and Amerindians.
At the contextual level, we included the presence of uoridated water supply (present or absent). The data regarding uoridation was obtained on the National Basic Sanitation Survey performed by the Brazilian Institute of Geography and Statistics (IBGE) in 2008 [44]. We also included the estimated coverage of the population by primary care oral health services, which corresponds to the mean monthly number of primary care oral health teams for every 3000 individuals to the total population of the municipality in the analyzed year. Higher oral health services coverage indicated higher potential access to basic dental services. The cut off for this variable was 40% that was the goal to be achieved in the biennium 2010/2011 [45]. Data about coverage were obtained from the website of the Department of Information Technology of the Uni ed Health System (DATASUS).

Statistical analysis
Comparison of income and education-based inequalities between municipalities with high and low HDI Descriptive analysis was performed to obtain mean tooth loss for each municipality, and the results were shown separately according to the HDI level (high or low). The magnitude of relative and absolute educational and income-based inequalities in the tooth loss was calculated using the Relative Index of Inequality (RII), Slope Index of Inequality (SII) and Relative (RCI) and Absolute Concentration Index (ACI) for municipalities with high and low HDI. RII and SII are summary measures recommended when making comparisons across populations [46]. These indices are regression-based and take the whole socioeconomic distribution into account, rather than only comparing the two most extreme groups. For municipalities with high and low HDI, the population in each education or income category was assigned a modi ed ridit-score based on the midpoint of the range in the cumulative distribution of the participants in the given category. We used generalized linear models (log-binomial regression), with a logarithmic link function to calculate RIIs (rate ratios) and with an identity link function to calculate SIIs (rate differences) [47]. Both indices were estimated with 95% con dence intervals. The RII can be interpreted as the rate ratio and the SII can be interpreted as the rate difference at the bottom and the top of the educational or income hierarchy. If there is no inequality, RII assumes the value of 1.0. The further the value of RII from 1.0, the higher the level of inequality. RII assumes only positive values, with values larger than one indicating a concentration of the indicator among the advantaged and values smaller than one indicating a concentration of the indicator among the disadvantaged. If there is no inequality, SII takes the value of zero. Greater absolute values indicate higher levels of inequality. Positive values indicate higher coverage in the advantaged subgroups, and negative values indicate higher coverage in the disadvantaged subgroups. RCI and ACI were estimated based on the methodology described by Clarke et al. [48]. The RCI provides a measure of the relative differences among education/ income groups. Considering that the RCI was based on morbidity measure (tooth loss), a negative RCI indicates inequality favoring higher income groups. The ACI quanti ed the absolute differences in health between education and income groups, and this index is not affected by whether it is measured concerning health or morbidity. As RII/SII are mathematically related to RCI/ACI, they will produce the same rank ordering of health inequality between groups but will differ in scale.

Association between tooth loss and socioeconomic indicators
Multilevel Poisson regression procedures with unstructured covariances matrix were used to model the two-level structure of individuals (level 1) nested within municipalities (level 2). Multilevel techniques of analyses provide the overall relationship between the individual, compositional factors, and oral health ( xed part), and the variation between areas that cannot be accountable for such factors (randomintercept parameter). Besides, it is possible to assess the variation in certain individual relationships between municipalities (random slope parameters) and the interaction between individual and contextual characteristics (cross-level interactions). A ve-step sequential modeling strategy was adopted: i) model 1(empty model): a model without the inclusion of any covariates, in which the variance in tooth loss is inspected between municipalities. A signi cant random intercept variance indicates the presence of unexplained differences in tooth loss between municipalities. The Wald test evaluated the signi cance of random intercept, and the Median Rate Ratio measured the heterogeneity among municipalities, according to Austin et al. (2017) [49]. There is no variation between municipalities if the MRR is 1.0, but the higher the MRR, the greater the area-level variation. ii) model 2 (random intercept, xed effect): considers all the individual-level variables in the xed part. This model assessed the association between tooth loss and income and education adjusted for covariables at the individual level. The variation between municipalities is allowed for, conditional on the individual, compositional factors. iii) model 3: as model 2, but including the municipalities-level variables. The proportional change in variance (PCV) was calculated according to Merlo et al. (2005) [50], using the following formula: PCV = (variance model 1variance model 2)/variance model 1. iv) model 4 (random slope and random intercept model): as model 2, but the model allows both the intercept and slope to be random parameters. Therefore, each municipality has its intercept and slope in which the variability from the overall intercept and slope can be investigated with the addition of individual and municipality variables and their interactions (cross-level interactions). Random slopes for education and income were considered. The comparison of goodness t between model 2 and model 4 was performed using the LR-test. v) model 5: as model 4, including the municipalities-level variables, in which the cross-level interactions between education/income and HDI were considered. The interaction term represents the change in the slope of education/income on tooth loss across municipalities when HDI changed from low to high. The estimated mean of missing teeth according to individual socioeconomic variables for municipalities with high and low HDI was demonstrated using a graph of the predicted model.
Procedures for complex sample design were used to calculate the proportions of adults according to investigated variables, the age-and sex-adjusted estimates of the outcomes, SII, RII and ACI. Statistical analyses were performed using STATA version 15.0 (StataCorp LP, College Station, Texas, USA).
The Brazilian National Council of Ethics in Research approved the SBBrasil 2010 study, protocol no. 15498, January 7, 2010.

Results
The sample was 9633 adults, residing in 157 municipalities; 146 individuals in 20 municipalities with a sample of fewer than 10 adults were excluded. The characterization of the sample and mean of missing teeth according to investigated variables are shown in table 1.  (Figure 1).
The inequalities indexes show a higher number of missing teeth in disadvantaged groups (lowest education and income levels). Higher relative and absolute education-based inequality were observed among municipalities with high HDI. For income-based inequalities, higher SII was observed in municipalities with lower HDI (Table 2). The multilevel crude estimates showed that all individuals variables were signi cantly associated with tooth loss (Table 3).  decrease in variability of tooth loss among municipalities (Table 4).   Table 5). The other random slopes were shown as an additional le (Additional le 1). The graph of tted lines shows how the municipal variation is more pronounced among individuals with higher education (Additional le 2).
Model 5 shows that when the association between education and tooth loss was allowed to vary across municipalities, the main effect of HDI was not signi cant. The social gradient in tooth loss by individual income and education continue to be observed. Municipalities with uoridated water showed a lower number of missing teeth (Table 5). The cross-level interaction between education and HDI was signi cant, indicating that the association between these two variables is not constant across municipalities, with steeper slopes observed in municipalities with high HDI. There is some evidence that the number of missing teeth is lower for the high-education group when they live in municipalities with high HDI compared with those with the same education level living in municipalities with low HDI. For individuals with the lowest education level, there was no difference in the number of teeth comparing those from municipalities with high versus low HDI. The xed effect of individual and contextual variables did not change with the inclusion of cross-level interactions ( Table 6).
When the variation was modeled as a function of individual income, the model was very similar to obtained with xed effect (Additional le 3). The graph of tted lines shows that the municipal variation is similar through income levels (Additional le 4). The cross-level interaction between individual income and HDI was not signi cant ( Figure 2). Exponentiated coefficients; *Model was adjusted for individual level variables: sex, age group, skin color and time since the last dental visit.

Discussion
The tooth loss represents an "end state" in oral health, and the present study rea rmed the persistent concentration of tooth loss among the most disadvantaged individuals. Over and above individual disparities, our ndings also demonstrated that the magnitude of educational and income-based inequalities varied between municipalities with high versus low HDI. Our results showed that educational attainment was "less protective" against tooth loss when people live in disadvantaged areas. Conversely, adults with the lowest levels of education and income had greater tooth loss regardless of whether they lived in a municipality with high or low HDI.
Our study replicated the inverse association between socioeconomic status and tooth loss after adjusting for all covariates at the individual and municipality levels [10][11][12]. Using different outcomes related to tooth loss (< 9 remaining natural teeth; lack of functional dentition/ inadequate dentition: < 21 natural teeth; count of tooth loss or remaining teeth and edentulousness), previous studies have consistently shown more tooth loss among the more disadvantaged groups. Quasi-experimental evidence that worsened economic circumstances (measured by subjective economic deterioration or housing damage due to disaster damage) are associated with tooth loss was provided by a natural experiment following exposure to the 2011 Great East Japan Earthquake and Tsunami [51]. A meta-analysis of 10 cohort studies and two cross-sectional studies found a signi cant association between low income and tooth loss [13]. Those living in poverty suffer a greater burden of oral diseases, such as dental caries and periodontitis [52,53], the main causes of tooth loss in adults. They also have a higher prevalence of systemic conditions, such as diabetes, cardiovascular disease, and obesity [54,55], that are related to tooth loss. Besides these biological explanations, the conceptual mediators of the effect of SES on oral health include material deprivation, psychosocial distress, and behavioral factors. [14,17,19].
In addition to those factors, it has been demonstrated that economic constraints are associated with the type of dental treatment delivered. While subjects in the lower income brackets are more prone to dental extraction, individuals with higher income are more likely to seek periodic routine appointments and conservative dental treatment, resulting in a higher number of retained teeth [56,57]. In Brazil, the higher tooth loss associated with socioeconomic disadvantage can result from systemic factors related to the organization and delivery of oral health services. The national oral health policy, also known as Smiling Brazil ("Brasil Sorridente"), was implemented only in 2004, ensuring universal coverage [58]. This policy increased the access of adults to oral health services, who historically had signi cant ongoing unmet needs with only urgent care available in public services [58]. The lack of access to restorative and preventive services in public health services may have contributed to tooth loss, especially among the most disadvantaged individuals. Moreover, previous studies have shown that a change in access to oral health services does not necessarily result in reduced oral health disparities because it takes time for people to change their behaviors, e.g. seeking periodic preventive dental services [59].
Our results also showed differences in tooth loss inequalities according to municipal HDI, with higher absolute and relative education-based inequality found in municipalities with high HDI and higher absolute income-inequality found in municipalities with low HDI. The high education-based inequality among municipalities with high HDI shows the double disadvantage of adults who share the lowest education and living in places with the worst social indicators [38]. These results suggest that public policies may not reach the persons in the same way increasing the gap among social groups.
The association between education and tooth loss was not constant across Brazilian municipalities. The signi cant cross-interaction term reveals that the magnitude of the association between education and tooth loss was lower in municipalities with high HDI. This result suggests that contextual factors may weaken the potential protective effect of schooling at the individual level. Low HDI municipalities in Brazil are smaller cities with lower wealth, higher rates of urban violence, lower sanitation conditions, and urban infrastructure. These contextual aspects may represent fewer opportunities for healthy living for its residents (less access to health services, less availability of diversi ed healthy food, fewer leisure options, less culture, and social capital). Our result shows that where one lives has an effect on oral health and that this effect must be evaluated considering the characteristics of the people.
On the other hand, the signi cant cross-level interactions reinforce the hypothesis that public policies can increase the gap among social groups to the extent that it may not reach people in the same way. The cross-interaction was signi cant after adjusting by oral health policies (coverage of oral health services) and uoridation of water supply. DHI is an indicator that incorporating a life expectancy, education, and a modi ed measure of income in contextual level. However, these indicators may demonstrate the effect of other policies and environmental characteristics that favor access to primary health care in general, sanitation, access to cultural and social opportunities, more social capital, more work and economic security, and more healthy food available. All this contextual factors are potential mediators of the association between HDI and tooth loss and have been pointed out as determinants of oral health [34,60] in previous studies that had demonstrated the main effect of social-level indicators on tooth loss using multilevel approaches in Brazil, United States, European countries, Australia and Japan [18,[20][21][22][23][24][27][28][29][30][31][32].
The analysis performed in this study is advancing, as it may contribute to more targeted policies for the groups with the greatest need, in speci c contexts.
The only signi cant contextual factor we found was the uoridation of the water supply. Individuals living in municipalities with water uoridation presented lower tooth loss. The importance of this intervention was previously shown in maintaining functional dentition [21] and tooth retention [23]. Natural experiments applied in oral health context from Brazil showed that adults who accessed uoridate water <50% of their lifetime presented a higher mean rate ratio of DMFT index compared with those living >75% of their life with residential access to uoridated water. Longer residential lifetime access to uoridated water was associated with less dental caries, even in a context of multiple exposures to uoride [61].
Our models were intentionally parsimonious, adjusting only for key covariates. As with most multilevel studies, our choice of area-unit (municipalities) was made for reasons of sampling and analytic convenience rather than being underpinned by an explicit theory linking area disadvantage and oral health; hence associations among these variables are likely to be underestimated. However, the Brazilian municipalities have to deliver basic education, sanitation, and primary healthcare. Therefore, our chosen contextual factors (HDI, oral health services coverage, and uoridation of water supply) must t the area border. We used cross-sectional data where the distinction between current and past exposures cannot be made. Similarly, income can change throughout life resulting in upward and downward social mobility.

Conclusions
There was a social gradient in tooth loss by education and income. Living in disadvantaged municipalities cannot overcome the risk associated with low schooling. The protective effect of higher education can be reduced when people live in disadvantaged areas. This differential effect of education on tooth loss among municipalities with high and low DHI can increase the gap among social groups. The persistent inequality demands for innovative strategies and policies that address the wider social determinants of health, considering the differential effect of public policies among disadvantage and advantage part of society. These policies must protect oral health as a human right through, strengthen intersectoral strategies for poverty reduction, supporting scienti c research on the health social determinants, interact communities with public health managers and researchers, support community actions to promote oral health and eliminate barriers of access to oral health care. Authors'contributions: RCF analyzed and interpreted the data, wrote the manuscript. MIBS, LGR, FLC and AEBLM contribute to data analysis and interpretation. IK was the study supervisor, made substantial contributions to: data analysis and interpretation and made critical review of the intellectual content. All authors read and approved the nal manuscript. Figure 1 Mean of tooth loss (95% CI) in each Brazilian municipality according to the HDI Legend: HDI -Human Development Index