Examiner reliability in fluorosis scoring: a comparison of photographic and clinical methods.

OBJECTIVE

To assess examiner reliability when scoring dental fluorosis in Malaysian children using clinical (Dean's Index) and photographic methods.

METHOD

The upper central incisors of 111 children were examined both clinically and photographically for fluorosis status using Dean's index. Twenty children were re-examined after a two-week interval for intra-examiner reliability by a single examiner. In addition, two independent examiners and the clinical examiner scored 111 photographic images of the same children in a standardized manner. Fluorosis scores were compared individually between examiners for both clinical and photographic scoring. Examiner reliability was assessed using both simple and weighted kappa statistics at tooth level. Sensitivity, specificity, positive-negative predictive values and a Receiver Operating Characteristic (ROC) curve were also calculated to determine the accuracy of the test.

RESULTS

Across the three examiners, the prevalence of fluorosis (Dean's score ≥ 2) using photographs was lower (ranged from 23% to 26%) than the prevalence recorded by clinical examination (30%). The kappa score for intra-examiner reliability for the duplicate clinical examination was excellent (0.89). Inter-examiner reliability between the photographic method and the clinical examination (gold standard) for each examiner was substantial with weighted kappa values ranging from 0.74 to 0.77. The photographic method indicated higher specificity (99%) than sensitivity (79%) and the area under the ROC curve was also high (0.89) which suggests good accuracy of the diagnostic test.

CONCLUSION

These results suggest that photographic examination of fluorosis on central incisors can be recorded with good examiner reliability. The recorded fluorosis prevalence was lower using the photographic scores.