Evaluation of Self-Reported Measures for Prediction of Periodontitis in a Sample of Brazilians

Background: Questionnaires including self-reported measures have become effective as a means of accessing many diseases. The aim of the present study is to evaluate the performance of a set of self-reported periodontal measures on estimating the prevalence of periodontitis. Methods: The sample comprised 284 individuals, aged 18 to 60 years, from Belo Horizonte, Brazil. Full-mouth periodontal examinations were performed and periodontal parameters were recorded. Periodontitis was categorized as no or mild, moderate, and severe. Each participant answered 18 questions covering sociodemographic variables, known risk factors, and self-reported periodontal measures. Questions were globally tested through logistic regression analysis. Results: The complete final model for moderate periodontitis included age, dental flossing, and gum disease (sensitivity = 23.1%; specificity = 98%; area under the receiver operating characteristic curve = 75.4%). The complete final model for severe periodontitis included all previously cited variables in addition to the number of teeth (sensitivity = 36.4%; specificity = 96.9%; area under the receiver operating characteristic curve = 85.3%). Conclusions: Self-reported periodontal measures showed a moderate predictive value for periodontitis prevalence. The use of these measures could be a good strategy in investigating prevalence of periodontal disease. J Periodontol 2011;82: 1693-1704.