Accuracy of pediatric primary care providers' screening and referral for early childhood caries.

PURPOSE: Tooth decay is one of the more common diseases of childhood. Slightly >40% of US children are already affected by the time they reach kindergarten. Primary care physicians can play an important role in prevention and control of this disease because of their ready access to this population. Unlike dentists, they see a large percentage of children during their infant and toddler years. However, few studies have been conducted on oral screenings and referrals by primary care physicians or the effectiveness of their oral health preventive activities. The purpose of this study was to determine the accuracy of pediatric primary care providers' screening and referral for Early Childhood Caries.
METHODS: We sought to compare independent, blinded oral screening results and referral recommendations made by primary care providers with those of a pediatric dentist, considered for purposes of the study to be the reference gold standard. The study was conducted at a private pediatric group practice in North Carolina. The practice was selected because it serves a large volume of Medicaid patients and includes a large number of pediatric primary care providers (11 pediatricians and 1 nurse practitioner). Study participants included Medicaid-eligible children younger than 36 months of age with erupted teeth. The pediatric primary care providers in this practice received 2 hours of training in infant oral health. The training consisted of a review of the study methods and clinical slides illustrating dental caries in various stages of progression. Specific instructions were given to the providers on how to recognize a cavitated carious lesion and how to determine when a dental referral is needed. Providers were instructed to refer any child with 1 or more cavitated carious lesions, soft tissue pathology, or evidence of trauma to the teeth or mouth. Before commencing the study, calibration and a comparative analysis were performed to establish reliability and validity of the examinations performed by the pediatric dentist. Both a pediatric dentist and a pediatric primary care provider conducted a dental screening on each child and recorded carious teeth and whether a dental referral was needed. Sensitivity and specificity were calculated to compare the pediatric primary care providers' screenings to the gold standard (pediatric dentist) in 3 categories: caries at the tooth level, caries at the patient level (1 or more affected teeth), and need for referral.
RESULTS: The final study sample consisted of 258 preschool-aged children (122 males and 136 females) with a mean age of 21.2 months (standard deviation [SD]: 9.13). One hundred eighty-four (71.3%) of the participants were white, 58 (22.5%) were black, and 16 (6.2%) were Hispanic. Tooth-Level Analysis: The pediatric dentist reported an average of 0.30 (SD: 0.005) cavitated teeth per child, whereas the pediatric primary care providers reported a mean of 0.25 (SD: 0.004). This difference was not statistically significant (t test). The pediatric dentist identified 80 (2.4%) teeth with cavitated carious lesions, whereas the pediatric primary care providers identified 64 (1.9%), 25 of which were false-positives. Their screening results include 41 false-negative teeth. Thus, the primary care providers tended to under-count the number of teeth with carious lesions. They achieved a sensitivity of 0.49 (95% confidence interval [CI]: 0.47-0.51) and a specificity of 0.99 (95% CI: 0.99-1.0) when their screening results for individual teeth were compared with the gold standard. Patient-Level Analysis: At the patient level, the pediatric dentist identified 25 (9.7%) children with 1 or more teeth affected by cavitated lesions. The pediatric primary care providers collectively identified 30 (11.6%) children who had cavitated lesions. They achieved a sensitivity of 0.76 (95% CI: 0.71-0.81) and a specificity of 0.95 (95% CI: 0.93-0.98) in identifying those children with cavitated carious lesions. There were 6 false-negatives and 11 false-positives when the pediatric primary care providers' findings were compared with the gold standard. At the patient-level, the positive predictive value of the dental screening was 0.63 and the negative predictive value was 0.97. Dental Referral: The pediatric dentist referred a total of 27 (10.5%) children to a dentist. Two of these children were referred for trauma and the other 25 were referred for cavities. The pediatric primary care providers referred a total of 23 (8.9%) children to a dentist. Two referrals were made because the provider was concerned about stains on the teeth, whereas the remaining 21 were referred for cavities. The pediatric primary care providers achieved a sensitivity of 0.63 (95% CI: 0.57-0.69) and a specificity of 0.98 (95% CI: 0.96-0.99) when their recommendations for referral were compared with the gold standard. The number of children receiving a referral from a pediatric primary care provider for cavities (N = 21) was less than the number of children they identified as having cavities (N = 30). The providers as a whole tended to under-refer, and only 70% of children with evidence of dental disease received a referral.
CONCLUSIONS: After 2 hours of training in infant oral health, the pediatric primary care providers in this study achieved an adequate level of accuracy in identifying children with cavitated carious lesions. Additional training and research would be needed to optimize pediatric primary care providers' identification of carious teeth if that were the goal of screening. However, the purpose of screening by nondental personnel generally is to accurately identify those in need of referral, which does not require a tooth-by-tooth identification of cavities. Additional research is also needed to determine how to improve dental referrals by pediatric primary care providers. Results of our study suggest that dental screenings can easily be incorporated into a busy pediatrics practice and that pediatric primary care providers can significantly contribute to the overall oral health of young children by the identification of those children who need to be seen by a dentist.