ACCURATE MEASUREMENT OF ENAMEL FLUOROSIS IN THE FACCT STUDY, REPUBLIC OF IRELAND

Accurate measurement of the prevalence of enamel fluorosis is challenging but essential in countries with community water fluoridation as a caries preventive method. Aim: To train and calibrate dentists to measure enamel fluorosis using Dean’s Index in the FACCT (Fluoride and Caring for Children’s Teeth) oral epidemiological study. Methods: Ethical approval was obtained for a training and calibration programme. Ten dentists were trained to use Dean’s Index during a 4-day training programme (Sept-2013) and a 2-day training programme in Nov- 2013. Training was led by an experienced (benchmark) examiner who provided training in Dean’s Index for previous international, national and regional surveys and comprised small group teaching sessions, discussion of fluorosis images and schoolbased examination of 12-year-old-children. Additional training to maintain consistency was provided via an online fluorosis training tool and sets of fluorosis images both previously scored by the benchmark examiner. Intra- and inter-examiner reliability were finally determined using the Kappa statistic. Examiners had to demonstrate proficiency and achieve a Kappa score ≥0.6 (substantial agreement) before fieldwork commenced. Results: During training and calibration, there was substantial agreement for Dean’s Index (overall Kappa ≥0.6). In the FACCT fieldwork 2,378 12-year-old-children were examined for fluorosis. Intra- and inter- examiner reliability were each determined for 5% of the sample. For Dean’s Index almost perfect intra- examiner agreement was found overall for the weighted kappa (0.8) and substantial agreement was found overall for the unweighted kappa (0.7); at least fair agreement was found for each examiner (0.3-1.0). For inter- examiner reliability, almost perfect agreement was found overall for both weighted (0.9) and unweighted kappa (0.8), and a level of substantial agreement at least for each examiner (0.6-1.00). Conclusions: Indices used to measure oral health are largely subjective. Considerable resources must be expended to ensure that measurement of oral conditions such as enamel fluorosis is valid and reliable.