Dental identification by comparison of antemortem and postmortem dental radiographs: Influence of operator qualifications and cognitive bias

Dental Evidence in Forensic Identification – An Overview, Methodology and Present Status

“…The dental evidence is compared with the ante-mortem records available to the dentists for identification of the deceased [ 43 ] in DVI. These comparisons can be achieved by using dental radiographs [ 11 , 44 ] and dental records available with the dentists. Sometimes, the dental evidence can be compared with the ante-mortem photographs available with the family members usually showing the anterior teeth [ 7 ].…”

Accuracy of dental identification of individuals with unrestored permanent teeth by visual comparison with radiographs of mixed dentition

“…On the other hand, as the age of the AM radiograph increased, the accuracy also escalated ( Figure 3); it can be assumed that the advancement of age and development stage of the AM OPG resulted in a bigger number of similar traits with the corresponding PM radiograph, facilitating the identification. According to the literature, training and experience in forensic identification influence the performance [8,22,23]; for this reason, only practitioners with a forensic background were recruited (Table 1); in our study, the relation between accuracy and years of experience was very weak and negative ( Figure 1); this means that, although not statistically significant, there was a tendency of more experienced professionals to be less accurate. The webbased questionnaire allowed to contact volunteers worldwide, facilitating the communication with the Authors; this approach demonstrated to be a valid, accurate and reliable method to test error rate of forensic identification by comparison of dental radiographs [22].…”

“…When the number of comparable distinctive characteristics decreases and the development stages differ significantly, identification is expected to be more challenging [13,19]. The percentages of sensitivity, specificity and accuracy in our research were inconsistent with previous studies testing dental identification by visual comparison of radiographs; MacLean et al [20] concluded that identification using dental bitewing was 93% accurate even with few or no restorations and a time-lapse between 1 and 15 years; however, both AM and PM radiographs were of permanent dentition; according to Kogon et al [10], the methodology was 88% accurate, but the study was conducted exclusively on children with restored dentition; Sholl et al [21] estimated that the average sensitivity was 93,3%, however the samples were 22 dry skulls from the collection of the Royal College of Surgeons of Edinburgh, which could not show any anatomical difference or development change between the simulated AM and PM radiographs; more recently, Fridell et al [12] concluded that the sensitivity within samples of children with unrestored dentitions was 88,3%; Pretty et al [22] found that the mean accuracy was 85,5%, but the study included restored dentitions; Pinchi et al [8] estimated that accuracy ranged between 76% and 97%, but identifications were performed on samples with both restored and unrestored teeth. In our study, the increase of the time interval between AM and PM radiographs was weakly correlated to the decrease of accuracy of the examiners (Figure 2).…”

“…Overall, only a few examiners fully appreciated the various anatomical traits visible in the radiographs and dental characteristics were mentioned more frequently than non-dental features. Different assumptions can be elucidatory: firstly, 15 out of 19 examiners were dentists; supposedly, that they felt more comfortable in analysing mainly dental characteristics; it is also possible that not all examiners were confident with the type of radiograph used in this project: on an every-day clinical practice, dentists handle intra-oral radiographs (periapical and bitewings) more frequently; this would suggest that dentists should master their knowledge of all anatomical structures and their familiarity with any type of radiographic image; secondly, there is not a standard procedure for comparing dental radiographs with identification purposes [8] and the literature has not provided enough statistical evidence on the reliability of maxillo-facial traits other than teeth for identification when comparing panoramic and intra-oral radiographs [6,19,24]. As opposed to what expected by the Authors, FA did not contribute significantly in providing information on which bone characteristics had identification validity.…”

“…Visual comparison is the most inexpensive and commonly adopted method for the analysis of traditional films or digital radiographs for identification purposes. However, there are no standardized protocols and the final conclusion is susceptible to the personal interpretation of the operator, who might confirm or exclude identity based on a single trait [7,8]. Any scientific method that aims to produce evidence with medico-legal outcomes should follow the Daubert standard: be accepted by the scientific community, be repeatable, standardized and be subjected to peer-review and publication reporting an acceptable error rate [9].…”