UCSF

Background

To prove that Caries Management by Risk Assessment (CAMBRA) can be successfully implemented in dental practices outside of the university setting, dentists in the San Francisco Bay Area (CA) were approached to participate in a Practice Based Research Network (PBRN) study. The overall goal of the CAMBRA-PBRN study was to recruit 30 dentists to perform a two-year study involving approximately 900 patients. Goal of the calibration study was to standardize and calibrate dentists potentially participating in the CAMBRA-PBRN study.

Methods

To minimize inter-examiner variability in data collection, including classification of carious lesions and recording of existing restorations, participating dentists were trained and calibrated in accurate DMFS (decayed, missing, filled surfaces) charting. Dentists were also trained and calibrated to diagnose and differentiate between sound surfaces and non-cavitated caries lesions (International Caries Detection and Assessment - ICDAS scores 1 and 2) for posterior occlusal surfaces. Thirty dentists were calibrated to a single gold standard examiner (BJ) during 6 calibration sessions, between 2011 and 2014. Kappa statistics were used to determine inter-examiner reliability on 13 or more patients, aged 12-63 (average age 38 ± 15 years), per examiner during each session, resulting in 94 patient encounters over the course of all 6 sessions. To participate in the main study, examiners needed to achieve a minimum required kappa of 0.75. During the calibration process, examiners scored between 1036 and 2220 tooth surfaces.

Results

The kappa values (unweighted kappa) of the participating dentists compared to the gold standard examiner ranged from 0.75 to 0.90, with an average kappa of 0.84 ± 0.03. 90% of the examiners achieved overall kappa values above 0.8. However, separate reliability for assessment of non-cavitated lesions, as in other studies, was lower (0.55 ± 0.15). Multiple subcategories were evaluated. All dentists reached sufficient reliability values to proceed into the study; nevertheless, one dentist discontinued with the study due to scheduling conflicts.

Conclusions

The high inter-examiner reliability results have shown that dentists who work in primarily non-research based practices can be effectively standardized and calibrated in data collection, based on specific guidelines created to anticipate potential research study scenarios.