Estruturação e validação diagnóstica de instrumento de autorrelato e perfil epidemiológico e indicadores de risco para variáveis endodônticas: uma série de estudos observacionais

Abstract

Aim: This thesis aimed to structure and validate a self-report instrument in oral health capable of detecting the prevalence of pulpal and periapical pathoses. As a secondary objective, this thesis evaluated the epidemiological profile and risk indicators associated with dental emergency visits (DEV) in different countries and was divided into eight articles: 1- a cross-sectional study evaluating the accuracy of the self-report of endodontic treatment (SRET) as a predictor of endodontic treatment (ET) and apical periodontitis (AP) in a sub-population in South Brazil; 2- a multicentre cross-sectional study evaluating the accuracy of SRET as a predictor of ET and AP in different population contexts, in sub-populations in Brazil and Malaysia; 3- a study on the construction and validation of a self-report questionnaire aimed at AP prediction (SRAP); 4- a cross-sectional study analyzing the accuracy of SRAP in a sub-population in South Brazil; 5- a cross-sectional study analyzing the prevalence of DEV involving pain relief and associated clinical and socio-economic indicators in a representative sample of dentists from the Australian private service; 6- a cross-sectional study analyzing the profile of DEV and the association with clinical and environmental parameters in a sample from the public service in Uruguay; 7- a cross-sectional study analyzing the association between socioeconomic aspects and endodontic variables in a sub-population in South Brazil; 8- a cross-sectional study analyzing the relationship between self-reported oral health (SROH) and lifestyle and sociodemographic, medical and dental indicators in a sub-population in South Brazil. Methods and Results: In article 1, 136 individuals who responded to a self-report questionnaire (exposure: SRET) and underwent clinical and periapical radiographic examination (outcome: radiographic observation of ET and AP) were included. The diagnostic accuracy was calculated separately for ET and AP. Accuracy, sensitivity (SS), specificity (SP), positive (PPV) and negative (NPV), efficiency and positive (PLR) and negative (NLR) likelihood ratios were observed. SRET showed high SS and SP for ET (0.960; 0.835), but not for AP (0.757; 0.631). The values referring to PPV and NPV were, respectively: ET (0.777; 0.972) and AP (0.396; 0.890). The values referring to PLR and NLR were, respectively: ET (5.853; 0.046) and AP (2.057; 0.383). In article 2, 338 patients from a multicenter sample (Brazil and Malaysia) were included. SRET (main exposure) was obtained through a questionnaire, and the diagnoses of ET and AP (outcomes) were obtained through the analysis of panoramic radiographs. Accuracy, SS, SS, PPV, NPV, PLR and NLR values were observed. The SRET values for the presence of ET and AP were, respectively: accuracy (0.879; 0.572); SS (0.938; 0.646); SP (0.824; 0.544); PPV (0.830; 0.453); NPV (0.936; 0.710); efficiency (0.881; 0.595); PLR (5.329; 1,418); NLR (0.075; 0.650). In article 3, initially, a questionnaire with 48 items was developed after a literature review about aspects related to AP. The instrument was submitted to content and face validation through the analysis of a committee of specialists and the application of the questionnaire in a pilot sample. Finally, the questionnaire was answered by 199 patients from two dental centres (PUCRS and UFSM), and their results were subjected to exploratory factor analysis (EFA) aiming at reducing items and resulting in the final self-report instrument for AP (SRAP). The face and content validation resulted in a 38-item questionnaire. After EFA, a questionnaire with 8 items remained, distributed over 3 factors. All items had a factor load> 0.5 and the model was responsible for 66.63% of the variance. The reduced scale had a Kaiser-Meyer-Olkin test (KMO) of 0.659 and Cronbach's alpha coefficient of 0.85 and an intraclass correlation coefficient of 0.82. In article 4, the accuracy of the SRAP instrument developed in the previous article was analyzed. The SRAP was answered by 182 patients who underwent panoramic radiographs at the screening clinics of PUCRS and UFSM, from which the diagnostic findings of AP and the presence of ET were collected. The analysis of the area under the receiver operator curve (AUROC) was calculated to establish the best discriminatory cut-off point for PA diagnosis for SRAP scores. Accuracy, SS, SP, PPV, NPV, PLR and NLR values were analyzed. The SRAP values for the diagnosis of AP were: AUROC (0.672), accuracy (0.615); SS (0.636); SP (0.590); PPV (0.0649); NPV (0.576); PLR (1.207) and NLR (0.615). In article 5, data were collected regarding the reason for the dental appointment (emergency / non-emergency) and clinical characteristics (number of teeth, number of decayed teeth, diagnosis) and sociodemographic (sex, age, health insurance) of 6504 patients from of a representative sample of 1148 dentists from the private sector in Australia, during the period between 2009-2010. The associations between DEV and clinical and sociodemographic indicators were calculated using Poisson regression models, with the values of prevalence ratios (PR) and confidence intervals (95% CI) being calculated. In the adjusted analysis for sex, age, health insurance, number of teeth and decayed teeth, the diagnosis of pulp / periapical disease was independently associated with DEV when compared to the other diagnoses (failure of restoration: PR = 0.45; periodontal: PR = 0.51; caries: PR = 0.34; others = PR 0.19), except for the diagnosis of dental trauma (PR = 1.37). In article 6, sociodemographic (age, sex, season and shift) and clinical (diagnosis) data were collected from 32168 patients treated at DEV in a public service in Uruguay, from January 2014 to January 2019. Poisson regression models were used to analyze the associations between the exposure variables (sociodemographic) and the outcome (diagnosis at the DEV) by calculating the PR and 95% CI. The adjusted regression model revealed an independent association between DEV due to endodontic diagnosis and male gender (PR = 1.05; 95% CI = 1.04 - 1.06), age groups <60 years (0 - 18 years PR = 1.36; 95% CI = 1.31 - 1.41 and 19 - 59 years old PR = 1.33; 95% CI = 1.28 - 1.38), afternoon (PR = 1.01; 95% CI = 1.00 - 1.02) and night (PR = 1.03; 95% CI = 1.01 - 1.04) shifts. Compared to summer, all other seasons had lower PR with respect to endodontic diagnosis in the DEV (spring PR = 0.97; 95% CI = 0.96 - 0.99, winter PR = 0.95; 95% CI = 0.93 - 0.96 and autumn PR = 0.97; 95% CI = 0.96 - 0.99). In article 7, 239 individuals who underwent panoramic radiographs at the PUCRS dental centre were included. Data regarding gender, age, smoking, frequency of alcohol consumption, educational level and monthly family income (exposure variables) were collected from questionnaires. Information regarding the presence of ET, quality of ET (QET) and AP diagnosis (outcome variables) were obtained from the analysis of panoramic radiographs. Poisson regression models were used to analyze the association between socioeconomic and endodontic variables (PR and 95% CI). The multivariate models demonstrated a higher frequency of ET in individuals with higher levels of education (complete high school: PR = 1.41 and university education: PR = 1.50) and with ≥60 years of age (PR = 1.31). AP was associated with the male gender (PR = 1.84). None of the exposure variables was associated with QET. In article 8, 172 patients answered a structured self-report questionnaire. The outcome variable consisted of the self-reported oral health (SROH) and the exposure variables consisted of sex, age, number of self-reported teeth, smoking habit, frequency of alcohol consumption, body mass index (BMI) - calculated from of patients' weight and height reports - frequency of tooth brushing and history of chronic non-communicable diseases (CNCD). The association between variables was analyzed using Poisson regression models with results of PR and 95% CI. The regression models adjusted for covariates demonstrated an association between poor SROH and smoking (PR = 1.10; 95% CI = 1.00 - 1.22) and obesity (PR = 1.15; 95% CI = 1.04 - 1.27). Conclusions: Article 01 demonstrated that SRET is a good predictor of the presence of ET, however, it does not sufficiently predict the presence of AP in this sub-population. Article 02 demonstrated that the accuracy of SRET in identifying patients with TE and/or AP is context-dependent, and may vary according to the profile of the population. In general, SRET proved to be a valid method for predicting ET but not AP in the two analyzed populations. Article 03 generated a succinct self-report instrument for endodontic variables, presenting acceptable levels of reliability and stability. Article 04 demonstrated that the SRAP instrument showed a predictive power for AP in up to 65% of cases, being easy to apply and low cost, especially for use in large-scale population screening. Article 05 revealed a strong association of endodontic diagnoses (pulp and periapical disease) and dental trauma with DEV and revealed that socioeconomic (gender, age and health insurance) and clinical factors (tooth loss, caries, endodontic diseases and dental trauma) are risk indicators for DEV in the Australian population. In article 06, seasonal trends were identified as risk indicators for DEV. Sociodemographic factors (male gender and younger age group) and seasonal patterns (summer and evening and evening shifts) were associated with a higher prevalence of DEV related to pulp and periapical disease. Article 07 demonstrated that a higher prevalence of AP was associated with males, whereas the presence of ET was associated with age and educational level, suggesting a presumed diversity in access to more conservative treatments such as ET, which can be modulated by socioeconomic factors such as access to education. Article 08 revealed that lifestyle-related factors are associated with SROH. Obesity and smoking were associated with poor SROH in the analyzed sub-population. This thesis proposes the use of a self-report instrument to detect AP in order to contribute to an increase in the volume of studies addressing epidemiological findings in endodontics, further demonstrating that the investigation of the prevalence of endodontic diagnosis, especially in the scenario of dental emergencies, may reveal patterns of distribution and association with different risk indicators.