Incorrect overestimation of age in criminal proceedings can have serious consequences and should be prevented as far as possible. A specificity well above 99% is to be aimed for, even if the higher specificity is associated with lower sensitivity.

However, in recent years, several studies conducted in multiple different countries have reported cases of individuals below the age of 21 years that reached ossification stage 4 [7,8,9,10,11,12]. Two studies conducted in Thailand and Japan discussed a possible difference in bone maturation between individuals of Caucasian and Asian descent [11, 12], even though studies by Schmeling et al. and Meijerman et al. had concluded no significant impact of ethnicity on bone maturation [16, 17]. Most of these recent studies present with insufficient exclusion criteria, which is why they are not used as reference studies for forensic age diagnostics [18].

According to the current method, when applying the minimum age concept, the assessed characteristic attribute corresponds to complete ossification of the medial clavicular epiphysis on at least one side [6]. This study indicates that in young men, only attainment of stage 4 or 5 on both clavicles can be taken to indicate an age over 21 years with a probability bordering on certainty. If stage 4 or 5 is only reached on one side, age may be overestimated.

Although adjusting the investigated attribute to bilateral attainment of stage 4 would be associated with a reduced sensitivity, with a higher probability of age underestimation, due to the effect of socio-economic status on the time of ossification of the clavicle, which was investigated by Schmeling et al., a high scatter range of age limits with a high probability of age underestimation is generally to be expected [16, 19]. In this study, a higher amount of subjects presented with stage 1 ossification than it would be suspected for this age group. The beforementioned impact of socioeconomic status might have had an impact, as subjects with and without a migration background were included in the study.

A possible limitation of this study is that neither the presence of hormonal diseases nor the use of certain medications were included in the exclusion criteria. It is not suspected that the prevalence of these criteria would be high amongst this age group (especially as the main indication for the CT scans was trauma), but a possible impact on the results cannot be excluded. Particularly with the one subject that stood out, it was described in the radiological findings that he was “very muscular”, which might indicate the use of anabolic substances.

In order to determine the exact percentage of overestimations, studies would have to be carried out with significantly larger study populations. At present, it can be assumed that the specificity of thin-section CT imaging as a test method for age estimation is about 99% or higher, but a specificity of 100% would be desirable.