NHANES III contains data from clinical dental examinations noting sealed, restored, decayed, extracted and unerupted teeth. Dental insurance claims record delivery of dental services including sealants, restorations and tooth extractions, but do not include oral health status information. From NHANES III data, we estimated sealant utilization (SU) and sealant-to-restoration ratio (SRR), for the United States and four regions among high- and low-income children, using full information from the clinical examination (SUExam and SRRExam) and using only information that would be available in claims data (SUClaims and SRRClaims). Validity of the estimates from claims data, with exam data as the gold standard, was evaluated based on bias (difference between claims-based and exam-based estimates), mean square error (MSE), and agreement on rankings. Limiting the analysis to first molar occlusal surfaces in children aged 5 to 8 years, we calculated 1) SUClaims as percentage of children with at least one sealant; 2) SUExam as percentage of children with 4 first molars present with at least one sealant; 3) SRRClaims as number of sealants over number of decayed surfaces; and 4) SRRExam as number of sealants over number of decayed or filled surfaces. SUClaims had smaller MSE than did SRRClaims. After adjusting SUClaims for bias, the MSE decreased for both low- and high-income groups to 1.16 and 4.63, respectively. However, after adjusting SRRClaims for bias the MSE for both groups remained large (low-income=840, high-income=1983). SUClaims performed better than SRRClaims as measured by bias, MSE and agreement of rankings.
Learning Objectives: N/A
Keywords: Sealant Programs, Health Care Delivery
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
Disclosure not received
Relationship: Not Received.