Validation of self-reported oral health measures for predicting periodontitis among adult Filipinos with type 2 diabetes mellitus

Abstract

Background: Diabetes mellitus is currently being recognized as a global health problem. The likelihood of having periodontal disease among people with diabetes is about 3 times greater and progresses rapidly when uncontrolled. Adults with diabetes were less likely to have been seen or consulted a dentist than to seek consult with a health care provider for diabetes care. This provides an opportunity for health care providers to screen and educate patients regarding the possible oral complications that might develop. A cheap and easy way of clinical assessment via selfreported oral health questionnaire would be of great use especially in a developing country like the Philippines where there's limited resources for health care access. This study aims to validate self-reported oral health measures, sociodemographic and medical variables in predicting the presence and severity of periodontitis in Filipino adults with type 2 diabetes mellitus. Method: The validated self-reported oral health questionnaires created by the CDC Periodontal Disease Surveillance Project was translated into Filipino and used in this study. A cross-sectional study of 180 adult diabetic participants was conducted in a single institution. Socio-demographic and medical variables were obtained. Participants were given the self-reported oral health questionnaire and finally underwent a formal periodontal evaluation. Data analysis was done using the software Stata SE version 12. Multivariable logistic regression analyses was used to determine significant predictors that predicted the prevalence of total periodontitis and serious periodontitis. The predictive power of each variable was calculated and expressed using odds ratio. Result: In totality, 93.9% of the study participants had clinically defined periodontitis: 29.4% had mild periodontitis, 37.8% had moderate periodontitis and 26.7% had severe periodontitis. In general, understanding and responses to all oral health questions were very high and consistent. Socio-demographic and medical variables considered to be significantly predictive of serious periodontits were male sex [OR = 2.17], low educational status [OR = 2.98], poor glycemic control [OR = 2.58], less frequent dental visits [OR = 2.77] and teeth loss >6 [OR= 5.02]. Self reported oral health variables shown to be significantly predictive of serious periodontitis included gum disease -Q1 [OR = 8.33], state of gum health -Q2 [OR = 0.39], loose teeth -Q3 [OR = 63.0], brushing of teeth -Q4 [OR = 0.65], use of mouthwash -Q4 [OR = 0.69] and poor tooth appearance -Q5 [OR = 48.42]. A recommended set of questions and proposed scoring system based on the logistic regression analysis of each predictors' strength was then formulated. Conclusion: The use self-reported oral health questions appears to be a potentially useful screening tool for predicting the presence of serious periodontitis among type 2 diabetic patients in a local setting where resources are limited and routine clinical oral examinations are not feasible. This will provide a cost-effective and rapid method of identifying patients who are in need of immediate dental evaluation and would benefit most to a dental referral.