Abstract: Objective To investigate the prediction effect of the self-reported periodontal health on moderate and severe chronic periodontitis and its feasibility.Methods A total of 330 subjects from five districts of Shenyang city attended the survey.Each subject answered questions about self-reported periodontal health and risk factors of periodontitis and was given a periodontal examination.The predictiveness was assessed by logistic regression modeling measuring receiver operator characteristic (ROC) curve and area under curve(AUC),sensitivity and specificity.Results Self-reported periodontal health for the prediction of moderate and severe chronic periodontitis was moderately effective,with AUC of 0.837,sensitivity of 74.8%,specificity of 71.3%.Self-reported periodontal health combined with risk factors of periodontitis was highly effective in prediction of moderate and severe chronic periodontitis,with AUC of 0.912,sensitivity of 82.1%,specificity of 85.4%.Conclusions Self-reported periodontal health can be used for epidemiological investigation of moderate and severe chronic periodontitis.Self-reported periodontal health combined with risk factors of periodontitis can improve the accuracy of the results of the survey.