Oral Health Ratings From Exfoliated Primary Teeth: Assessing Measurement Reliability
Résumé
Objectives: The objective of this study was to develop a measurement tool to assess indicators of oral health from deciduous teeth and evaluate the inter-rater reliability among researchers of varying backgrounds. Methods: Nine raters representing disciplines focused on studying teeth (dentistry, pediatric dentistry, and biological anthropology) and different experience levels (post-graduate, post-doctoral, resident) scored 30 exfoliated primary maxillary central incisors. These teeth were donated to the STRONG (Stories Teeth Record of Newborn Growth) study and collected from children whose mothers were pregnant during, or one year prior to, the 2013 Boston Marathon bombings. All raters were trained to identify and evaluate visible oral health indicators. A Standard Operating Procedure (SOP) was developed to score 12 indicators of oral health/ characteristics: Tooth type, crown color (using VITA classical shade guide), localized discoloration, surface roughness, chipping, enamel translucency, linear enamel hypoplasia, pitting enamel hypoplasia, calculus, decay, sealants and fillings. Some of the indicators were characterized as present versus absent, while others were characterized by the amount or the degree in which they were present. Percent agreements, as well as Kappa’s and Gwet’s AC1 values were calculated to evaluate inter-rater reliability. Results: 8 out of 12 indicators assessed (chipping, enamel translucency, pitting enamel hypoplasia, calculus, linear enamel hypoplasia, decay, fillings, sealants) had good agreement across raters (>%75; AC1> 0.75). Surface roughness, crown color, and localized discoloration had the lowest agreement among raters (values of 0.5 or lower). Conclusions: Raters of varying backgrounds were consistent in their assessment of many visible tooth characteristics. However, color, roughness, and discoloration appeared to be more subjective indicators. Individuals with various research/training backgrounds can be well-trained to consistently identify several observable features of exfoliated primary teeth. Measures described in this study can be used for future tooth collection studies to differentiate between oral health indicators.