aMMP-8 self-testing and self-administered questionnaires for periodontitis screening: A diagnostic trial in a Chinese...

Limited awareness, clinical barriers, and persistent high prevalence make large-scale detection of periodontitis challenging, creating demand for simple, affordable, and population-level screening strategies outside traditional dental settings. This study evaluates the diagnostic performance of self-test activated matrix metalloproteinase-8 (aMMP-8) for identifying periodontitis and assesses the feasibility of a fully self-reported screening model compared with models based on quantitative aMMP-8 point-of-care test (POCT) results.

In this cross-sectional diagnostic study, we evaluated the diagnostic performance of three index tests: self-performed or POCT aMMP-8 alone on an oral rinse sample, or in combination with a questionnaire. Periodontitis status was determined using radiographic bone loss according to the 2017 classification, as stages III-IV or stages II-IV. Binary logistic regression models were developed using least absolute shrinkage and selection operator regularization and 5-fold cross-validation. Model performance was assessed by area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and accuracy.

A total of 566 participants were included, and 41.7% were diagnosed with stages II-IV periodontitis. Both self-test and quantitative aMMP-8 results were significantly associated with periodontitis. In univariable models, self-test aMMP-8 showed lower AUROC than quantitative results. However, adding questionnaires and demographic factors in multivariable models improved the diagnostic performance of self-test and minimized the gap between the two. The final multivariable models showed comparable performance for self-test aMMP-8-based (AUROC = 0.955) and quantitative aMMP-8-based (AUROC = 0.957) models in stages III-IV periodontitis prediction, with similar results for stages II-IV periodontitis.

This study showed that a self-test aMMP-8-based multivariable diagnostic model can perform similarly to a quantitative aMMP-8 POCT-based one. Once further developed and validated, a fully self-reported diagnostic model can provide a potential tool for periodontitis screening.

This study examined whether a simple, self-administered mouth rinse test could aid in identifying individuals with advanced gum disease. Over 550 adults participated, and approximately 42% were found to have severe gum disease. We discovered that the self-test alone was less accurate than a more detailed machine-based quantitative reading. However, when combined with a questionnaire about dental health and basic personal information, the accuracy of detection improved a lot. These combined methods performed well and were similar to using only the machine results. Our results suggest that a simple screening approach, which includes a self-test, a questionnaire, and personal details, could be an inexpensive and easy-to-use tool for community health screenings to identify severe gum disease early. Further research is needed to test this method in various populations and confirm its effectiveness in real-world settings for identifying gum disease and promoting early dental care.