Association between obesity and periodontitis in Australian adults: A single mediation analysis

Abstract

Background: Obesity and periodontitis are conditions with high burden and cost. This study aims to unfold the proposed pathways through which the effect of obesity in the presence of health behaviors (dental visiting behavior and diabetes) increases the risk of periodontitis?. Methods: The effect decomposition analysis using potential outcome approach was used to determine obesity-related periodontitis risk using the Australian National Survey of Adult Oral Health 2004 to 2006. A single mediation analysis for exposure, “physical-inactivity induced obesity,” mediator “dental visiting behavior (a de facto measure of healthy behaviors),” outcome “periodontitis,” and confounders “age, sex, household income, level of education, self-reported diabetes, alcohol-intake and smoking,” was constructed for subset of 3,715 participants, aged ≥30 years. Proposed pathways were set independently for each risk factor and in synergy. The STATA 15 Paramed library was used for analysis. Sensitivity analysis was conducted to detect unmeasured confounding using non-parametric approach. Results: The average treatment effect of physical inactivity induced obesity to periodontitis is 14%. Pathway effect analysis using potential outcomes illustrated that the effect of obesity on periodontitis that was not mediated through poor dental visiting behavior was 10%. Indirect effect of obesity-mediated through poor dental visiting behavior on periodontitis was 3%. Conclusions: The direct effect of physical inactivity induced obesity on periodontitis was higher than the indirect effect of obesity on periodontitis through dental visiting behavior. Establishing a pathway of causal relationship for obesity and periodontitis could help in developing management strategies that focuses on mediators.