The association and potential pathways between common mental disorders and oral health among Finnish adults

There are behavioural and physiological alterations associated with depression and anxiety which may affect oral health. Nonetheless, there is limited research on this topic and results are inconclusive. This study explored the association of depression and anxiety with clinical and perceived oral health and possible pathways underlying these associations. The first hypothesis was that depression and anxiety negatively affected clinical oral health via oral health-related behaviours, medication use and physiological response. The second was that depression and anxiety negatively affected perceived oral health via clinical oral health and use of medications. This is a secondary analysis of data on adults who participated in the nationally representative Finnish Health 2000 survey. Depression and anxiety were assessed with the Beck Depression Inventory and the Composite International Diagnostic Interview, respectively. The association of each mental disorder with clinical (dental caries, periodontal disease and tooth loss) and perceived oral health were tested in regression models adjusted for confounders and potential mediators. Findings indicated that depression and anxiety were associated with the number of decayed teeth. These associations were not entirely explained by oral health-related behaviours and medication use. The associations of depression with numbers of teeth and filled teeth were fully explained by those mediators. Perceived oral health was strongly related with depression and anxiety. The association between anxiety and perceived oral health was completely explained by clinical oral health status and use of anxiolytic medication, whereas the association between depression and perceived oral health was attenuated, but remained significant. In conclusion, there were significant associations between depression and anxiety with some clinical measures of oral health and with perceived oral health. However, these associations were largely explained by socio-demographic factors, and to a lesser extent by the mediators assessed