RESUMO
Clinical forensic assessments of injuries' life-threatening danger may have an impact on the legal aftermath following a violent assault. The pursuit of evidence-based guidelines should ensure a user-independent and reproducible forensic practice. However, does it? The aim of this study was to evaluate the forensic life-threatening danger assessments after a protocol implementation in 2016. The evaluation concerned usability and reproducibility of the protocol, and its influence on assessment severity. We analyzed the level of inter- and intra-rater agreement using 169 blinded, prior-protocol cases that were reassessed by two forensic specialists. We compared assessment made the year before and after protocol implementation (n = 262), and the forensic specialists' reassessments with the prior-protocol cases' original assessments (n = 169). Whether to make an assessment, the levels of agreement varied between weak agreement (inter-rater, Κ = 0.43; assessor 1, Κ = 0.57) and strong agreement (assessor 2, Κ = 0.90). Regarding severity, the levels of agreement varied between strong agreement (inter-rater, Κ = 0.87; assessor 1: Κ = 0.90) and almost perfect agreement (assessor 2: Κ = 0.94). The assessments were statistically significant redistributed after the implementation (chi-square test: p < 0.0001). The proportion of cases assessed as having not been in life-threatening danger increased from 9 to 43%, and moderate severity assessments decreased from 55 to 23%. Of the moderate severity assessments, 55% were reassessed as having not been in life-threatening danger. The protocol ensured independent and reproducible assessments when the forensic specialists agreed on making one. The protocol resulted in less severe assessments. Future studies should examine the reliability of the protocol and its consequences for legal aftermaths.
