J D Bader1, B A White, O Olsen, D A Shugars. 1. Sheps Center for Health Services Research, University of North Carolina, Chapel Hill 27599-7590, USA. jim_bader@unc.edu
Abstract
OBJECTIVES: The reliability of practicing dentists' classifications of patients' caries risk and periodontal disease risk and reason for treatment for individual teeth were determined. The risk classification protocols had been in use in a group practice for more than a year, and the reason-for-treatment protocol had been introduced six months previously. METHODS: Eight dentists' classifications for caries (n = 66) and periodontal disease risk (n = 66), and six dentists' classifications for reason for treatment (n = 73) were compared to those of a nominal standard examiner. Reliability was expressed as percent agreement and kappa values. RESULTS: Percent agreement was 76 percent, 83 percent, and 74 percent for caries, periodontal disease, and reason for treatment, respectively, with kappa values of 0.56, 0.70, and 0.69. CONCLUSIONS: Dentists can attain reasonable levels of reliability using simple classification protocols with little formal training, although misclassification may be problematic for specific administrative or research-related purposes.
OBJECTIVES: The reliability of practicing dentists' classifications of patients' caries risk and periodontal disease risk and reason for treatment for individual teeth were determined. The risk classification protocols had been in use in a group practice for more than a year, and the reason-for-treatment protocol had been introduced six months previously. METHODS: Eight dentists' classifications for caries (n = 66) and periodontal disease risk (n = 66), and six dentists' classifications for reason for treatment (n = 73) were compared to those of a nominal standard examiner. Reliability was expressed as percent agreement and kappa values. RESULTS: Percent agreement was 76 percent, 83 percent, and 74 percent for caries, periodontal disease, and reason for treatment, respectively, with kappa values of 0.56, 0.70, and 0.69. CONCLUSIONS: Dentists can attain reasonable levels of reliability using simple classification protocols with little formal training, although misclassification may be problematic for specific administrative or research-related purposes.