Haavikko’s method and reference centile curves for age estimation in Bosnian-Herzegovian children aged 4-15

The aims of this study were to evaluate adopted Haavikko method (1974) based on Finnish sample for dental age estimation in Bosnian-Herzegovian children and to create reference centile curves with LMS method for clinical, legal and forensic usage. Materials and methods In cross-sectional, retrospective study, the panoramic radiographs (OPGs) taken from 805 girls and 636 boys in Bosnia and Herzegovina, aged between 4 and 15 years, were studied. Method is based on scoring of four permanent teeth, 41, 44, 46, and 47 up to 10 years of age and and 13, 43, 44, and 47 in children over 10 years of age, with one of 12 mineralization stages. Dental age is than computed from the Haavikko’s tables as the mean of all four teeth. Kappa statistic and intraclass correlation coefficient (ICC) were used for testing intra- and inter-observer repeatability of mineralization stages and observed dental age by assessments of 10% (N=144) of OPGs. Results showed that Haavikko’s method underestimate the dental age in the BH children. Mean underestimation was -0.33 (SD 0.72) years in girls and -0.12(SD 0.82) years in boys. Cohen kappa scores were 0.79 for intra- and 0.80 for inter-observer agreement for mineralization stages and average measures for ICC for dental age were 0.98 for intra- and 0.90 for inter-observer agreement. In addition, 1st, 3rd, 5th, 50th, 95th, 97th and 99th centile curves of chronological age against the dental age were constructed for girls and boys separately using the LMS method by Cole and Green (LMS ChartMaker Software, Medical Research Council, UK), which could be used for age estimation in the BH children.