Statistical Method for Dental Clinics for Determining Presence and Stage of Periodontitis with aMMP-8 Mouth Rinse...

Background/Objectives: This study proposes a framework for building a statistical prediction model for dental clinics to facilitate the diagnosis of periodontitis and its stages. The method is based on active-matrix metalloproteinase-8 (aMMP-8) mouth rinse point-of-care testing (POCT). Methods: A complete model was created within a three-step modeling scenario: (i) the first function differentiates healthy patients from those with periodontitis; (ii) the second function differentiates stage I and II patients from stage III patients; and (iii) the third function separates stage I and II patients from each other. The model was developed using logistic regression analysis, and the aMMP-8 POCT results utilized in the predictive functions were obtained from an Oralyzer digital reader. Sample data comprised 149 adult patients who visited dental clinics in Thessaloniki, Greece. Results: Patients without periodontitis were identified in 74.2% of cases (95% CI: 55.1-87.5%). Patients with periodontitis were revealed with a success rate of 94.1% (95% CI: 87.7-97.4%), and of these, the correct stage was determined in 71.2% of cases (95% CI: 61.7-79.2%). The complete model was tested on the same patient data from which it was formed. Conclusions: The results of the study showed that logistic regression can be used in the development of a model for dental clinics to reveal and stage periodontitis with sufficient accuracy. In the complete model created, aMMP-8 mouth rinse POCT results in ng/mL, visible plaque index (VPI), and the information on the patient's missing teeth were statistically important factors in determining the presence and stage of periodontitis.