Rationale and objectives : With the current large computerized payment systems and increase in the number of claims, unusual dental practice patterns to cover up fraud are becoming widespread and sophisticated. Clustering the characteristic of dental practice patterns is an essential task for improving the quality of care and cost containment. This study aims at providing an easy, efficient and practical alternative approach to developing patterns of dental practice profiles. This will help the third-party payer to recognize and describe novel or unusual patterns of dental practice and thus adopt various strategies in order to prevent fraudulent claims and overcharges. Methodology : Knowledge discovery (or data mining) was used to cluster the dentists' profiles by carrying out clustering techniques based on the features of service rates. It is a hybrid of the knowledge discovery, statistical and artificial neural network methodologies that extracts knowledge from the dental claim database. Results : The results of clustering highlight characteristics related to dentists' practice patterns, and the detailed managerial guidance is illustrated to support the third-party payer in the management of various patterns of dentist practice. Conclusion : This study integrates the development of dentists' practice patterns with the knowledge discovery process. These findings will help the third-party payer to discriminate the patterns of practice, and also shed more light on the suspicious claims and practice patterns among dentists.
All Science Journal Classification (ASJC) codes
- Health Policy
- Public Health, Environmental and Occupational Health