Purpose: Predicting mortality of older patients is a difficult, yet crucial task, which has lead to development of several prognostic tools. We hypothesized that the evaluation of oral health, which is often neglected, might be helpful to improve 1-year mortality prediction in a cohort of patients > 65 years admitted to an acute geriatric unit. Methods: We assessed 200 patients using the Decayed, Missing or Filled Teeth tool, the Periodontal Screening and Recording tool and the Plaque Index. A comprehensive geriatric assessment was performed to measure functional, cognitive, nutritional and comorbidity status and to calculate the Multidimensional Prognostic Index (MPI). An age- and sex-adjusted regression model was used to identify independent predictors of 1-year mortality. Area under ROC curve (AUROC) was calculated to assess predictive accuracy of MPI risk groups before and after correction for dental examination results. Results: Plaque Index and Periodontal Screening and Recording tool were associated with 1-year mortality. Participants with periodontal disease had a fivefold higher risk of death (p < 0.001). Conversely, absence of periodontal disease identified with high specificity (96.6%) participants who were still alive 1 year after discharge. Periodontal disease was an independent predictor of 1-year mortality in a regression model adjusted for age, sex and other variables of the comprehensive geriatric assessment. The correction of MPI according to results of Periodontal Screening and Recording tool significantly improved AUROC from 0.62 to 0.68 (p < 0.003). Conclusion: Oral health evaluation, if integrated in the comprehensive geriatric assessment, improves estimation of 1-year mortality in hospitalized patients.
