IADR Abstract Archives
Dental Plaque Score Reliability from Photographs in Preschoolers: BEECON Trial
Objectives: To assess the accuracy and reliability of measuring dental plaque in preschoolers via extraoral photographs.
Methods: The study’s Coordinating Center used extraoral photographs from the completed BEhavioral EConomics for Oral health iNnovation (BEECON) pilot trial in preschoolers (ages 9-39 months) to train and calibrate plaque raters for consortium controlled trials. Disclosing liquid was applied to primary maxillary incisor (MI) facial tooth surfaces and extraoral photographs were collected. The debris index (Greene & Vermillion, 1965) of the simplified oral hygiene index (OHI-S), for 4 primary MI facial tooth surfaces (OHI-MIS) was used for scoring. A protocol, webinar training materials, and customized REDCap electronic data capture system were developed. Two experienced dentist clinical researchers individually rated photographs, adjudicating discrepancies to establish gold standard consensus. After the first calibration based on rater feedback, criteria were refined and raters recalibrated with another photograph set. In the ongoing main BEECON trial, the study dentist-rater re-rated 30 randomly selected screening exam photographs at least 2 weeks later. Lin’s concordance correlation (LCC) for mean OHI-MIS score was calculated with 95% confidence intervals (CIs) using a 0.75 criterion for inter- and intra-rater reliability.
Results: LCC for mean OHI-MIS score vs gold standard consensus was 0.73 (95%CI: 0.52-0.86) after the first training (N=29, excluding an unscoreable photograph) and 0.89 (95% CI: 0.86-0.91) after the second training (N=30). Intra-rater reliability (N=30) was 0.97 (95% CI: 0.95-0.99).
Conclusions: Reliability versus gold standard consensus and within the same rater over time using one rater to evaluate photographic images to score MI facial tooth surface plaque in preschool children was clearly demonstrated. Thus, dental clinician raters can be calibrated remotely using digital photographic images to score plaque in preschool children. Benefits of this system include the ability to train raters until reliability is achieved and the ability to efficiently rate images in batches.
Methods: The study’s Coordinating Center used extraoral photographs from the completed BEhavioral EConomics for Oral health iNnovation (BEECON) pilot trial in preschoolers (ages 9-39 months) to train and calibrate plaque raters for consortium controlled trials. Disclosing liquid was applied to primary maxillary incisor (MI) facial tooth surfaces and extraoral photographs were collected. The debris index (Greene & Vermillion, 1965) of the simplified oral hygiene index (OHI-S), for 4 primary MI facial tooth surfaces (OHI-MIS) was used for scoring. A protocol, webinar training materials, and customized REDCap electronic data capture system were developed. Two experienced dentist clinical researchers individually rated photographs, adjudicating discrepancies to establish gold standard consensus. After the first calibration based on rater feedback, criteria were refined and raters recalibrated with another photograph set. In the ongoing main BEECON trial, the study dentist-rater re-rated 30 randomly selected screening exam photographs at least 2 weeks later. Lin’s concordance correlation (LCC) for mean OHI-MIS score was calculated with 95% confidence intervals (CIs) using a 0.75 criterion for inter- and intra-rater reliability.
Results: LCC for mean OHI-MIS score vs gold standard consensus was 0.73 (95%CI: 0.52-0.86) after the first training (N=29, excluding an unscoreable photograph) and 0.89 (95% CI: 0.86-0.91) after the second training (N=30). Intra-rater reliability (N=30) was 0.97 (95% CI: 0.95-0.99).
Conclusions: Reliability versus gold standard consensus and within the same rater over time using one rater to evaluate photographic images to score MI facial tooth surface plaque in preschool children was clearly demonstrated. Thus, dental clinician raters can be calibrated remotely using digital photographic images to score plaque in preschool children. Benefits of this system include the ability to train raters until reliability is achieved and the ability to efficiently rate images in batches.