Halitose autorrelatada: prevalência, fatores associados e acurácia diagnóstica

Abstract

Self-reported halitosis has been used as an important diagnostic tool in epidemiological studies that can be easily applied, demands less time and resources to be performed, and can reflect the individual's perception of their condition. However, its prevalence and associated factors vary in the literature. The aim of this study was to evaluate the prevalence of self-reported halitosis and its associated factors, as well as to determine accuracy estimates with clinical evaluation of halitosis. A cross sectional study was conducted in a sample of 5420 individuals who answered, via e mail, a set of closed questions that addressed socioeconomic and demographic data, medical and dental history, oral hygiene habits, parameters of oral health and self reported measures of halitosis, being: measure 1 - "Put the palm of the hand in front of the mouth and blow. After that test, would you say you have bad breath? "; measure 2 - "Have you ever been diagnosed with bad breath by your dentist "; measure 3 - "Has any member of your family or friend ever told you that you have bad breath?". Subsequently, a clinical examination was performed in a subsample of 159 individuals for the clinical diagnostic of halitosis through the organoleptic method. Predictors associated with self-reported halitosis were determined by means of univariate and multivariate analyzes. Accuracy estimates of self-report were evaluated in the subsample. Prevalence of self-reported halitosis for measure 1 was of 14.6% (n = 792), for measure 2 of 4.1% (n = 221) and for measure 3 of 33.2% (n = 1799). Self-reported halitosis was mainly associated with socioeconomic variables (age, gender, educational level), parameters of oral health, (gingival bleeding, gingival infections, tongue coating general evaluation of oral health) and impacts on daily activities (family or social environment and intimate relations). The specificity values for self-reported halitosis measures were determined to be high for clinical (organoleptic score ≥2) and strong (organoleptic score ≥4) halitosis. Overall, combinations of self-reported measures improved the AROC of multivariate predictive models. It can be concluded that prevalence rates of self-reported halitosis may vary according to the self-reported measure, but can be considerate moderate. Estimates of diagnostic accuracy were determined to be useful and with good prediction for non-diseased individuals.