Oral health of quilombola and non-quilombola rural adolescents: a study of hygiene habits and associated factors

This study evaluated oral hygiene habits and their association with sociocultural, environmental factors and factors related to the use of dental services among Quilombola and non-Quilombola rural adolescents from inland Bahia. Thiswas a cross-sectional study conducted in 2015. Prevalence and prevalence ratios were estimated for the outcomes and multiple Poisson regression analysis with robust variance was performed. We interviewed 390 adolescents, of which 42.8% were Quilombola. Poor tooth brushing and non-flossing were found in 33.3% and 46.7% of adolescents, respectively. Poor tooth brushing was increased by male gender (PR = 1.45), not living with both parents (PR = 1.45), lower hand hygiene habit (PR = 1.72) and worse oral health self-assessment (PR = 1.38). Non-flossing was associated with economic level E (PR = 1.54), older age (PR = 0.91), lower hand hygiene habit (PR = 1.53) and worse oral health self-assessment (PR = 1.33). Different associated factors were observed between Quilombola and non-Quilombola. It is necessary to consider the specificities of the populations for the promotion of adolescent oral health and the importance of the intersectoriality between education and health and care to families.


Introduction
Adolescence, the stage of life between childhood and adulthood, is a period in which differentiated attention is required regarding general and oral health 1 .Physiological, psychological and social changes, typical of this stage make this population group more vulnerable to health risk situations 2 .
Oral health is a relevant aspect of adolescent life.An adverse perception of oral health influences the self-esteem and socialization of this group 3,4 .Among the oral problems that may interfere with adolescents' quality of life are untreated tooth decay, severe occlusion, tooth pain and loss, and gum bleeding 5,6 .
One of the ways of preventing diseases and oral problems is proper oral hygiene, where tooth brushing and flossing are essential tools 1 .Brazilian studies show that the prevalence of tooth brushing (≥ 3 times/day) ranges from 49.8% to 77.0% among adolescents.However, flossing is still not a consolidated habit when compared to brushing, with frequencies varying from 0.5% to 51.0% 7,8 .
Oral hygiene habits are influenced by different factors related to individuals and their families.A higher prevalence of poorpracticesis noted in male adolescents, with lower economic level/ maternal level of schooling and who report feeling alone 7,9,10 .International studies have revealed that low prevalence of tooth brushing and flossing are associated with dwelling in rural areas 11,12 .In Brazil, a study on rural children and adolescents aged 6-12 years showed tooth brushing (≥ 3 times/day) of 56.4% and flossing of 26.4% of respondents 13 .
The place of residence is a realm establishing the oral health situation of social groups.Individuals with similar social profiles may evidence different oral health levels, depending on the location and attributes of the dwelling place 14 .
Quilombola adolescents mostly reside in rural communities with specific territorial relationships related to resistance to the historical oppression suffered by this population 15,16 .They coexist with social vulnerabilities related to their places of residence, such as lower coverage of water supply through a general distribution network and regular collection of garbage, low schooling and household income, when compared with other rural communities 17 , as well as difficult access to the dental services 18 .
Thus, this study aimed to evaluate oral hygiene habits and its association with sociocultur-al and environmental factors, as well as factors related to the use of dental services among rural Quilombola and non-Quilombola adolescents from a rural area in inland Bahia.

Methods
This is a cross-sectional, population-based and home-based study conducted with adolescents aged 10-19 years of age, living in 21 Quilombola rural communities recognized by the Palmares Cultural Foundation (FCP) 15 and non-Quilombola communities from Vitória da Conquista, Bahia.This study is part of the study Adolescer: Adolescent Health of the Rural Area and its Conditions, approved by the Human Research Ethics Committee of the Multidisciplinary Health Institute, the Federal University of Bahia.
The sampling strategy considered the territorial extension and the population of adolescents living in rural communities, to ensure the representativeness and feasibility of the research.The following sampling principles were used: to select households proportionally to the number of adolescents per community and interview only one adolescent per household.Also, the sample was calculated separately for each stratum to facilitate valid estimates for Quilombola and non-Quilombola populations.
The research population was estimated at 811 adolescents, divided into two strata: Quilombola (n = 350) living in the Quilombola communities recognized by the FCP, and non-Quilombola (n = 461).This sample universe was obtained from a survey carried out at the Family Health Units (USF) and confirmed in a subsequent mapping of the communities.
The following criteria were considered for the sample design: prevalence of 50%, given the heterogeneity of the events measured, an accuracy of 5%, a confidence level of 95% and design effect equal to 1.0, totaling 184 Quilombola and 210 non-Quilombola adolescents.An additional 15% was added to cater for possible losses.However, considering that only one adolescent per household would be interviewed and, among the Quilombola, the number of households would be exceeded, 7.1% were added for losses in this group.
Sampling for the non-Quilombola stratum was performed by randomly selecting households that contained adolescents, followed by their proportional distribution by community.Subsequently, adolescents were randomly selected by household.Concerning the Quilombola stratum, we only proceeded with the random selection of adolescents by household.The presence of severe mental disorders with cognitive impairment among adolescents was an exclusion criterion.
The tool used to conduct interviews was a structured questionnaire based on surveys conducted in Brazil.For this study, questions were drawn from the National School Health Survey (PeNSE) 19 , the National Health Survey (PNS) 20 and the National Oral Health Survey (SBBrasil) 21 .
The questionnaire was divided into two blocks, the first one containing questions about household and economic characteristics was answered by the adolescent (when aged ≥ 18 years) or by the person responsible.The second block was responded to only by adolescents, containing information about them.
A pilot study was conducted in December 2014 with rural adolescents not belonging to the primaryresearch.Data were collected between January and May 2015.Re-interviews were carried out in 5% of the sample households within seven days after the first interview, toensure data quality.
Oral hygiene habits were evaluated based on the frequency of tooth brushing and flossing.The rate of tooth brushing was obtained from the question In the last 30 days, how many times a day did you usually brush your teeth?With the following answer options: a) I have not brushed my teeth in the last 30 days; b) Once a day; c) Twice a day; d) Three times a day; e) Four or more times a day; and f) I did not brush my teeth daily.The flossing was obtained from the following question What do you use to clean your mouth?, where one of the secondary questions was Dental floss?(No;Yes).For purposes of bivariate and multivariate analyses, the following were considered as outcomes: (1) unsatisfactory tooth brushing frequency when less than three times a day; (2) non-flossing.
The independent variables were selected considering a conceptual model based on Petersen 22 , organized in two blocks.The first block had sociocultural and environmental factors and the second block had variables related to the use of dental services.Sociocultural factors were: economic level 23 (B and C -higher levels; D and E -lower level); gender; skin color/ethnicity, classified as black (brown and black) and non-black (white, yellow and indigenous); age; schooling; one dose of alcoholic beverage; tobacco use; trying out illicit drugs; physical activity; number of close friends; family composition; feeling lonely in the last 12 months; parental understanding of personal problems in the previous 30 days.
Having as the primary source of water supply the general distribution network was considered an environmental factor, as well as the habit of hand hygiene before meals.The second block was composed of the following variables: oral health self-assessment; perception of the need for dental treatment; dental pain in the last six months; and dental consultation in the last year.
A descriptive analysis of oral health-related characteristics was made through the distribution of frequencies for the total sample and by stratum, Quilombola and non-Quilombola, as well as the estimation of the prevalence of tooth brushing and dental flossing.Differences between proportions were assessed by Pearson's chi-square test and Fisher's exact test.
A bivariate analysis was performed to identify the variables associated with poor tooth brushing and non-flossing.Estimates of prevalence ratios and confidence intervals were made using Poisson regression with a robust variance for the total sample and by stratum.All variables with a level of significance lower than 20% were included in the multiple analysis.
The hierarchical entry of variables in blocks was adopted in the multiple analysis.First block variables remained as adjustment factors for the second hierarchically lower block, and those with a p-value <0.05remained in the model.The comparison between models was made by the Akaike criterion and Bayesian information.The adequacy of values predicted by the models to the observed values was evaluated by chi-square.Stata, version 15.0 (Stata Corporation, College Station, USA) was used for data analysis.

Results
A total of 390 adolescents were interviewed, of which 42.8% (167) were Quilombola.Tooth brushing less than three times a day was observed in 33.3% of the total sample, 32.7% of non-Quilombola adolescents and 34.1% of Quilombola (Figure 1A).The prevalence of non-flossing was 46.7% for the total number of adolescents and 46.2% and 47.3% among non-Quilombola and Quilombola, respectively (Figure 1B).No significant differences were found between the two strata evaluated.
Most adolescents reported using toothbrush and toothpaste (99.7%) to perform oral hygiene, and 49.5% stated that they would replace the toothbrush with less than three months of use.Regarding oral health, 63.0% of adolescents considered their oral health to be good or very good, 48.1% reported a need for dental treatment and 2.6% for total dental prosthesis, 19.6% reported dental pain in the last six months and 73.1% said they were satisfied or very satisfied with their teeth and mouth (Table 1).
Of the total number of adolescents, 15.6% reported no dental appointments in their lifetime and 33.2% had no dental appointments in the last year.Of the adolescents who had already had a dental appointment, 67.8% used the public service and 66.8% were attended to in anFHU, 50.8% did so for cleaning, maintenance, revision or prevention purposes, 41.9% due to tooth pain, extraction, gum problems and dental treatment and 7.3% due to orthodontic appliance maintenance or budgets, and 85.4% considered the last appointment good or very good.Differences between Quilombola and non-Quilombola adolescents were observed only for the variable dentalappointment in the lifetime, where 22.7% and 10.3% reported no consultation, respectively (Table 1).
Unsatisfactory tooth brushing was significantly higher in male adolescents, who did not Non-quilombola Total sample Quilombola live with both parents, did not practice hand hygiene before meals often and considered their oral health to be either fair, poor or very poor (Table 2).Among non-Quilombola adolescents, a positive and significant association was observed between unsatisfactory tooth brushing and males, not living with both parents, with a lower frequency of hand hygiene before meals and no dental consultation in the last year.A negative association with parental understanding of their problems was sometimes observed in the last 30 days.In the Quilombola stratum, tooth brushing less than three times a day was more prevalent among those who did not wash their hands before meals frequently and those who reported worse oral health (Table 2).
Regarding the non-flossing, a positive and significant association with economic level E was observed in the total sample, lack of the general distribution network as the primary form of water supply, rare or no hand hygiene before meals, and fair, poor or very poor oral health.A negative association was observed with the highest age and schooling (Table 3).
The non-flossing was more prevalent among non-Quilombola adolescents of economic level E, who reported not washing their hands before meals frequently and did not consult a dentist in the last year, and less prevalent among older adolescents and higher schooling.Among the Quilombola, a positive association was observed with parents never o rarely understanding personal problems in the last 30 days and with the lack of the general network of distribution as the primary form of water supply and negative with the higher age and not having consulted the dentist in the last year (Table 3).

Discussion
Satisfactory oral hygiene, tooth brushing and flossing habits were found in most of the study population.Quilombola and non-Quilombola adolescents did not differ about the prevalence of these habits.However, differences were found in sociocultural and environmental factors and factors related to the use of the dental services associated with these practices.
Freire et al. 10 , in a study with Brazilian ninth-graders of elementary school, found results similar to this study for tooth brushing ≥ twotimes/day in 2009 (95.2%) and 2012 (91.7%).However, brushing ≥ 3x / day, indicated by several studies such as the recommended daily rate for adolescence, had a lower prevalence than that found among adolescents from different Brazilian regions 4,7,8,24 .Flossing was higher than reported by studies with children and adolescents (9-12 years) in rural Pernambuco (33.3%) 13 , with rural and urban adolescents in Campina Grande (PB) (0.5%) 8 and urban adolescents from other Brazilian regions 4,7 .International studies showed a lower prevalence of tooth brushing and flossing 11,[25][26][27][28][29][30][31][32][33][34][35][36][37] .The literature evidences that Brazilians have a reasonable frequency in oral hygiene habits.However, the quality of this hygiene can leave something to be desired 1 .This situation contrib-utes to the fact that the eradication of oral diseases in the country is still a distant reality.It is emphasized that not only individual practices are responsible for oral health care but, when other types of social policies are lacking, these practices become even more crucial 1 .
Quilombola and non-Quilombola adolescents did not differ about the prevalence of tooth brushing and flossing, but differences were found in the use of dental services during lifetime and factors associated with poor oral hygiene habits, evidencing that there are specificities among these groups.
The similar prevalence of health-related behavioral aspects between Quilombola and non-Quilombola was also reported by Silva et al. 28 , in a study about tobacco experimentation in this same population.It is suggested that the social interaction of these adolescents in school could favor the mutual influence between them, transposing effects of the family, the environment and their neighborhood 11,29 .This interaction is even more evident in rural areas, where the supply of educational services is reduced when compared to urban ones.
The difference about performing dental consultations during lifetime can be explained, in part, by the location of USFs, which, are mostly based in non-Quilombola communities.The actions of health teams in Quilombola communities take place about once a month in satellite units that do not have the necessary infrastructure for dental consultation 29 .Thus, dental care is restricted to USF headquarters and educational actions are carried out in other communities.These facts add up to the obligatory presence of a responsible person over 18 years of age for the first dental care of the adolescent 30 , which increases the difficulty of the first access of Quilombola adolescents to dental services.When studying the use of health services by the adult Quilombola population of Vitória da Conquista (BA), Gomes et al. 31 suggested that there is a higher difficulty in accessing services by Quilombola due to the inequities faced by this population, especially worse social and economic conditions.*n: number of individuals; †P: prevalence of tooth brushing less than three times a day; ‡PR: prevalence ratio; § CI95%: 95% confidence interval.
Table 2. Prevalence, prevalence ratio and 95% confidence interval of tooth brushing less than three times a day for the total sample, Non-Quilombola and Quilombola.Bahia, 2015.Male adolescents had a higher prevalence of unsatisfactory tooth brushing, a result corroborated by other studies [7][8][9][10] .This behavior can be explained by a greater concern of adolescent girls regarding aesthetics and care with oral hygiene due to current social and cultural standards 32 .However, this difference between genders was not observedin the Quilombola stratum and more studies are required to understand the specificities of this group better.
The socioeconomic condition was only associated with the use of dental floss in the total sample and among non-Quilombola, a result similar to other studies with adolescents 7,33 .Dental floss, compared to toothbrushes, is a more expensive product and its use is a less consolidated habit, which may have influenced the greater use of better-off adolescents 36 .The National Oral Health Policy -Programa Brasil Sorridente (Smiling Brazil Program), provides for the availability of basic dental care appropriate to the population 34 .However, not only toothbrush and fluoride dentifrice would be necessary, but also the distribution of dental floss, which would possibly reduce the effect of economic conditions on habit-forming.
The greater homogeneity of the Quilombola population about the economic level may have contributed to the absence of this variable in the final model for this stratum.Also, it is local practice to carry out educational activities with the rural population on the making of dental floss in a handmade fashion (raffia) using lineage bags, material easily accessible in rural and Quilombola communities, by professionals of the oral health teams.
The one-year age increase reduced the prevalence of non-use of dental floss by around 9%, a relationship already evidenced in a study with adolescents from Paraíba 8 .The greatest attention to oral health is related to affectivity and social interaction 35 , aspects that can contribute to greater care among older adolescents.
Family aspects influenced oral hygiene habits of rural adolescents, a result corroborated by other studies.Davoglio et al. 7   parents do not understand adolescents, they may assume a disinterested posture vis-à-vis their appearance and self-care as a form of protest.Also, individuals who live alone or who have low sat-isfaction with their social relationships tend to adopt less preventive health behaviors 36 .
These findings show that health care with a focus on families can bring good results to ado- Oral hygiene is a physical hygiene component but requires learning 34 to be carried out correctly.Studies show that oral hygieneis significantly associated with regular hygiene habits 9,37 .In this study, dental brushing ≤ threetimes/day and non-use of dental floss were more prevalent among adolescents who reported not washing their hands before meals frequently.
Among the Quilombola, non-use of dental floss was more prevalent in adolescents who did not have the general distribution network as their primary source of water supply at home, which reinforces the vulnerability and need for self-care practices in this group, since it has less continuous access to fluoridated water.Public water fluoridation has been mandatory in Brazil since 1974.However, non-universal access to fluoridated water keeps an extensive number of people on the fringe of this admittedly effective and cost-effective relationship 38 .
Self-assessment of oral health as fair, poor or very poor was one of the variables associated with poor oral hygiene habits.The agreement between the clinical condition and self-perceived oral health usually occurs in more severe painful and aesthetic cases, while other milder oral problems are underestimated 39 .Poor oral condition has negative impacts on the daily activities of adolescents, among them, the greater difficulty in tooth brushing.Less frequent tooth brushing may, therefore, be a consequence of poor oral condition, which may influence the outcome found in this study.It should be emphasized that the perception of oral health is also related to socio-cultural and personal characteristics of individuals 39 .
The lack of dental consultation in the last year was a factor that increased poor oral hygiene habits only for the non-Quilombola stratum.The contact of the dental professional with the adolescent influences the adoption of satisfactory oral hygiene habits.However, although dental work is relevant, it solves individual problems recognized by those who use it.In population terms, oral health care results from a range of factors -biological, psychological and social 39 .
This work has some limitations because it is a cross-sectional study, which prevents us from inferring the temporality of some of the associations observed.The sample size was not planned to test differences between Quilombola and non-Quilombola, so there may not have been enough sampling power for some variables.However, this fact does not compromise the differences observed and described.Also, no information was collected on the clinical conditions of the adolescents, the tooth brushing and use of dental floss technique, as well as hygiene hours, hindering the analysis of these aspects.

Final considerations
Oral health policies are still recent in Brazil and aim to promote, prevent diseases and restore the oral health of individuals.However, practices aimed at recovery are again emphasized, which makes it necessary to redirect the work process to create support for comprehensive care and the needs of different population groups 39 .
The National Oral Health Policy Directives provide for supervised oral hygiene activities so that autonomy is developedto self-care 34 .However, these activities are commonly geared to children, not covering adolescents, especially older ones.Collective measures facilitate savings regarding financial and human resources and are fundamental within a single health system to ensure the integrality of care, especially in more vulnerable places such as rural areas and traditional communities.
Oral health care demands intersectoral actions that consider aspects inherent to adolescents, in their social/household context, with the purpose of promoting comprehensive and resolutive adolescent oral health care.Health education actions should consider cultural traits and take into account that oral health is a component of health and transcends dentistry.The perception of oral health as a human right and not as a privilege should be increasingly strengthened, especially in populations of recognized vulnerability.

Table 1 .
Characteristics related to oral health in the total sample, Quilombola and non-Quilombola adolescents.Bahia, 2015.

Table 2 .
Prevalence, prevalence ratio and 95% confidence interval of tooth brushing less than three times a day for the total sample, Non-Quilombola and Quilombola.Bahia, 2015.

Table 3 .
Prevalence, prevalence ratio and 95% confidence interval of non-use of dental floss for the total sample, Non-Quilombola and Quilombola.Bahia, 2015.
argue that when

Table 3 .
Prevalence, prevalence ratio and 95% confidence interval of non-use of dental floss for the total sample, Non-Quilombola and Quilombola.Bahia, 2015.

Table 4 .
Factors associated with tooth brushing less than three times a day and non-use of dental floss, according to the regression model, for the total sample, non-Quilombola and Quilombola.Bahia, 2015.