A retrospective comparative study to evaluate the reliability of post-mortem interval sources in UK and US medico-legal death investigations

Post-mortem interval (PMI) information sources may be subject to varying degrees of reliability that could impact the level of confidence associated with PMI estimations in forensic taphonomy research and in the practice of medico-legal death investigation. This study aimed to assess the reliability of PMI information sources in a retrospective comparative analysis of 1813 cases of decomposition from the Allegheny County Office of the Medical Examiner in Pittsburgh, US (n = 1714), and the Crime Scene Investigation department at Southwest Forensics in the UK (n = 99). PMI information sources were subjected to a two-stage evaluation using an adapted version of the 3x5 aspects of the UK police National Intelligence Model (NIM) to determine the confidence level associated with each source. Normal distribution plots were created to show the distribution frequency of the dependent variables (decomposition stage and source evaluation) by the independent variable of PMI. The manner, location, and season of death were recorded to ascertain if these variables influenced the reliability of the PMI. A confidence matrix was then created to assess the overall reliability and provenance of each PMI information source. Reliable PMI sources (including forensic specialists, missing persons reports, and digital evidence) were used across extensive PMI ranges (1 to 2920 days in the US, and 1 to 240 days in the UK) but conferred a low incidence of use with forensic specialists providing a PMI estimation in only 35% of all homicide cases. Medium confidence PMI sources (e.g., last known social contact) accounted for the majority of UK (54%, n = 54) and US (82%, n = 1413) cases and were associated with shorter PMIs and natural causes of death. Low confidence PMI sources represented the lowest frequencies of UK and US cases and exclusively comprised PMI information from scene evidence. In 96% of all cases, only one PMI source was reported, meaning PMI source corroboration was overall very low (4%). This research has important application for studies using police reports of PMI information to validate PMI estimation models, and in the practice of medico-legal death investigation where it is recommended that i) the identified reliable PMI sources are sought ii) untested or unreliable PMI sources are substantiated with corroborating PMI information, iii) all PMI sources are reported with an associated degree of confidence that encapsulates the uncertainty of the originating source.