Study design and participants

This was a cross-sectional study involving 12-year-old children from a region with low socioeconomic indicators in the city of Manaus (East Zone), Brazil. The city of Manaus is the capital of Amazonas state, located in the North region of Brazil. The Human Development Index of Manaus and the East Zone of the city was 0.737 and 0.659 in 2010. A randomized stratified sample of public schools with children in the 7th year from primary school classes was selected in the 11 neighbourhoods that make up the eastern region of the city of Manaus. In the two-stage probabilistic sampling, the schools were proportionally selected according to the corresponding number of schools in each neighbourhood.

All 12-year-olds in the 7th year from all classes of the selected schools were invited to participate in the study. Children in the 7th year of selected schools who were not 12 years-old and those aged 12 years but not attending the 7th year were not invited to participate. Exclusion criteria were use of orthodontic appliances, children with syndromes and those with special care needs.

Study power calculation

The present study included a final sample size of 400 schoolchildren, which resulted in a study a power of 90% with 5% of statistical significance to detect effects of 0.10 using structural equation modelling analysis with four observed variables and two latent variables [12]. In addition, a sample of 400 schoolchildren would lend a power of 95% in a multiple regression model including 3 predictors considering a statistical significance of 5% to detect effects of 0.05.

Theoretical model

The theoretical model used in this study was adapted from the conceptual model proposed by Wilson and Cleary model [10]. The model included biological and physiological factors (number of decayed teeth and number of teeth presenting clinical consequences of untreated dental caries), symptom status (dental pain), individual characteristics (sense of coherence), environmental characteristics (social support) and quality of life related to oral health (OHRQoL). Dental pain, SOC and social support were considered the possible mediators and effect modifiers on the relationship between dental caries and OHRQoL (Fig. 1).

Fig. 1
figure 1

Theoretical model adapted from Wilson and Cleary [10]

Data collection and measures

Data were collected through dental clinical examinations and pre-tested questionnaires carried out by five previously calibrated dentists from October 2016 to June 2017.

A questionnaire on socioeconomic characteristics, including parent’s/guardian’s level of education (years of study) and monthly family income (Up to ½ Brazilian Minimal Wage [BMW], ½ to 1 BMW, > 1 BMW) was completed by children’s parents/guardians. One BMW corresponded to U$271.09 in 2016. Children answered a self-completed questionnaire to assess sex, dental pain, SOC, social support and OHRQoL.

The schoolchildren were examined under natural light on the premises of the selected schools. First, supervised dental brushing using toothbrush, fluoride dentifrice and dental floss was performed. Then, clinical examination was conducted using a plain dental mirror No. 5 (Duflex®) and a ball point OMS (Stainless®) probe. Schoolchildren were examined sitting in school chairs.

Oral health-related quality of life

Child Perceptions Questionnaire (CPQ11-14) Impact Short Form (ISF: 16) was used to evaluate children’s OHRQoL [13]. The questionnaire consists of 16 items grouped into 4 dimensions: oral symptoms, functional limitations, emotional state and social well-being, that evaluate the frequency of life events during the last 3 months. A 5-point Likert scale was used for each response: 0 = Never; 1 = Once or twice; 2 = Sometimes; 3 = Often; 4 = Every day or almost every day. The CPQ11-14 total score is obtained by summing the items and can range from 0 to 64. The higher the score the greater the impact of oral health status on children’s quality of life.

Dental caries

Dental caries was measured according to the number of decayed teeth [14] and clinical consequences of untreated dental caries (PUFA/pufa index) [15]. The former is obtained by adding the total number of permanent teeth presenting clinical cavities due to caries, based on the component decayed of the decayed, missing and filled teeth (DMFT) index [14]. PUFA/pufa index was used to assess the occurrence of dental conditions resulting from untreated dental caries including visible pulp, ulceration of the oral mucosa due to root fragments, fistula, or abscess. The number of teeth with clinical consequences of untreated dental caries are summed to obtain the PUFA/pufa index [15].


The mediators included dental pain, SOC and social support. Dental pain was assessed according to the following question: “Did you experience toothache during the last 6 months?”, using the response options 0 = No or 1 = Yes. The 6-months period prevalence of dental pain was used because OHRQoL measure (CPQ11-14) referred to the last 3 months. Thus, dental pain should be reported for period longer than 3 months according to the theoretical model (Fig. 1.) A period prevalence of dental pain longer than 6 months would possibly result in recall bias due to the age of the participants.

Children’s SOC was measured using the SOC-13 scale [16] transculturally adapted to the Portuguese language [17]. SOC-13 scale consists of 13-item questionnaire using a 5-point Likert scale. The scores of the items related to negative SOC were reversed before adding the scores to obtain the final SOC score. The final score can range from 13 to 65 points. The higher the final score the greater the SOC.

Social support was evaluated through of the Social Support Appraisals (SSA) scale validated for Brazilian participants [18]. The SSA scale is a 30-item questionnaire answered using a 6-point Likert scale using the following response options: 1 = I fully agree, 2 = I strongly agree, 3 = I agree a little, 4 = I disagree somewhat, 5 = I strongly disagree, 6 = I fully disagree. The items are grouped into four dimensions of social support: family, friends, teachers and others. The total SSA score results from the sum of the 30 items, ranging from 30 to 180. The items against social support are reversed before obtaining the SSA score. Thus, the higher the SSA score indicates greater social support.

Calibration study and reliability analysis during the main study

Initially, five dentists were calibrated for dental examinations previous to data collection of the main study. The calibration study included 20 children from the same schools who did not participate in the main study. They were examined twice on 7 days interval according to the same dental exam protocol and instruments used in the main study to obtain repeated measures of DMFT index and PUFA/pufa index. The intra-examiner Kappa coefficient for DMFT index ranged from 0.80 to 0.81 and for PUFA/pufa index ranged from 0.60 to 0.90. The Kappa coefficient for inter-examiner agreement ranged from 0.90 to 1.00 for DMFT index, and from 0.60 to 0.90 for PUFA/pufa index.

Cronbach’s alpha of CPQ11-14, SOC-13 and SSA scales was 0.674, 0.876 and 0.812, respectively. Dental examinations and questionnaires were replicated in 10% of the sample in the main study over a 14-day period. The intra-examiner Kappa coefficients for DMFT index and PUFA/pufa index were 0.93 and 0.87, respectively. Intra-Class Correlation Coefficient for CPQ11-14, SOC-13 and SSA scales were 0.83, 0.89 and 0.89, respectively.

Data analysis

Data analysis was carried out in three steps. First, the variables were described using frequencies and means (standard deviations) for categorical and continuous variables.

Second, confirmatory factorial analysis (CFA) and structural equation modelling (SEM) were used to evaluate mediation. CFA evaluated the measurement model concerning multidimensionality of the two latent variables and the respective indicators [19]. OHRQoL was a latent variable composed by 4 indicators represented by the CPQ11-14 dimensions: oral symptoms, functional limitations, emotional state and social well-being. Social support was a latent variable created from four indicators represented by the SSA dimensions: family, friends, teachers and others. SEM assessed mediation through the direct and indirect relationships between observed and latent variables according to the theoretical model (Fig. 1). The total effects representing the sum of the direct and indirect effects, were obtained using the maximum likelihood estimation method. Mediation was detected when the indirect effect between two variables (e.g. dental caries and OHRQoL) was significant. In this case, the variables representing the different pathways (e.g. SOC, social support, dental pain) between the exposure (e.g. dental caries) and the outcome (e.g. OHRQoL) were considered the mediators even when the direct relationship between the exposure and the mediator, or between the mediator and the outcome was not significant. Sex and monthly family income were included in the SEM for adjustment. Nine hundred samples via bootstrap procedure were re-sampled from the original dataset to estimate the 95% confidence intervals (CI) and more accurate standard errors [20]. The Chi-square test (χ2/df < 3.0) was used to assess the adequacy of the overall fit of the model. In addition, the following fit indexes and thresholds were used to assess the model fit. GFI (Goodness of Fit) ≥ 0.90, CFI (Comparative Fit Index) ≥ 0.90, SRMR (Standardized Root Mean Square Residual) ≤ 0.08 and RMSEA (Root Mean Square Error of Approximation) ≤ 0.06 [21]. The CFA and SEM analyzes were performed using AMOS 25.0.

Third, moderating effect analysis of the interaction of dental pain, SOC and social support with dental caries on OHRQoL were tested using negative binominal regression according to each moderation variable. Initially, the likelihood ratio test was used to compare the Akaike’s information criterion (AIC) of the null negative binomial regression model with the school-level variable (AIC = 2801.946) and without the school-level variable (AIC = 2802.842). Multilevel analysis accounting for school-level was not used since both models were not statistically different (P = 0.639). Three statistical models were tested for each moderation variable. The model 1 tested the crude association between of dental caries (number of decayed teeth and number of teeth with clinical consequences of untreated dental caries) and OHRQoL. In model 2, the moderation variable (dental pain, SOC or social support), sex and monthly family income were inserted in the regression model. Model 3 included the variables in model 2 and the interaction term “dental caries × moderation variable”. The assessment of the interaction effect was by comparing the Likelihood Ratio between model 2 (without the interaction term) and model 3 (with the interaction term) using the Chi-square test. Statistical differences between model 2 and Model 3 suggested moderating effect. The analyzes of moderation effect were performed using STATA 14.0.

Ethic aspects

This research was conducted in accordance with the Helsinki Declaration and approved by the Ethics Committee of the Federal University of Amazonas (Protocol No. 57273316.1.0000.5020). Informed consent was obtained from all parents/guardians before data collection.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (