Skeletal age estimation in a contemporary South African population using two radiological methods (Bo/Ca and TW2)

Hand-wrist X-rays are the most objective tools for forensic age estimation. The aim of this study is to compare the accuracy of skeletal age (SAg) versus chronological age (CAg) assessments using Bo/Ca and TW2 methods in a South African sample of 224 individuals, aged between 6 and 16 years, focusing on ancestry and biological sex. The differences between estimated SAg and CAg were considered the method performance indicators. Accuracy, sensitivity and specificity have been calculated based on the age limit of 13 years (yrs). Bo/Ca and TW2 correctly classified Europeans (Eu) and Africans (Af) (-0.08 and 0.18 yrs vs -0.07 and -0.20 yrs, respectively) as well as boys and girls (-0.19 and 0.19 yrs vs -0.03 and -0.21 yrs, respectively). The standard error (SE) was calculated for each method according to biological sex and ancestry in children younger than 13 yrs (SE range for Bo/Ca: +/- 0.123-0.179 and for TW2 +/- 0.095-0.140) and in subadults 13 yrs or older (SE range for Bo/Ca: +/- 0.143-0.157 and for TW2: +/- 0.092-0.183). Bo/Ca seems to be less influenced by ancestry and biological sex in respect to TW2. According to previous studies, both methods showed greater inaccuracy in subadults aged 13 yrs or older compared with younger.