Premium
Validation of Periodontitis Screening Model Using Sociodemographic, Systemic, and Molecular Information in a Korean Population
Author(s) -
Kim HyunDuck,
Sukhbaatar Munkhzaya,
Shin Myungseop,
Ahn YooBeen,
Yoo WookSung
Publication year - 2014
Publication title -
journal of periodontology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.036
H-Index - 156
eISSN - 1943-3670
pISSN - 0022-3492
DOI - 10.1902/jop.2014.140061
Subject(s) - periodontitis , medicine , logistic regression , cohort , prospective cohort study , population , cohort study , chronic periodontitis , environmental health
Background: This study aims to evaluate and validate a periodontitis screening model that includes sociodemographic, metabolic syndrome (MetS), and molecular information, including gingival crevicular fluid (GCF), matrix metalloproteinase (MMP), and blood cytokines. Methods: The authors selected 506 participants from the Shiwha‐Banwol cohort: 322 participants from the 2005 cohort for deriving the screening model and 184 participants from the 2007 cohort for its validation. Periodontitis was assessed by dentists using the community periodontal index. Interleukin (IL)‐6, IL‐8, and tumor necrosis factor‐α in blood and MMP‐8, ‐9, and ‐13 in GCF were assayed using enzyme‐linked immunosorbent assay. MetS was assessed by physicians using physical examination and blood laboratory data. Information about age, sex, income, smoking, and drinking was obtained by interview. Logistic regression analysis was applied to finalize the best‐fitting model and validate the model using sensitivity, specificity, and c ‐statistics. Results: The derived model for periodontitis screening had a sensitivity of 0.73, specificity of 0.85, and c ‐statistic of 0.86 ( P <0.001); those of the validated model were 0.64, 0.91, and 0.83 ( P <0.001), respectively. Conclusions: The model that included age, sex, income, smoking, drinking, and blood and GCF biomarkers could be useful in screening for periodontitis. A future prospective study is indicated for evaluating this model's ability to predict the occurrence of periodontitis.