Identifying factors that help improve existing decomposition-based PMI estimation methods.

Area of Science:

• Forensic Science
• Taphonomy
• Biostatistics

Background:

• Accurate postmortem interval (PMI) estimation is crucial in forensic investigations but remains challenging.
• Existing regression models for PMI and accumulated degree days (ADD) often lack precision due to small sample sizes and limited predictors.

Purpose of the Study:

• To develop more accurate outdoor PMI and ADD estimation models.
• To investigate the impact of larger sample sizes, advanced statistical models, and additional demographic/environmental predictors on estimation accuracy.

Main Methods:

• Utilized a dataset of 213 human subjects for outdoor decomposition analysis.
• Evaluated existing PMI/ADD formulae and developed new models incorporating total decomposition score (TDS), demographic factors (age, sex, BMI), and weather data (season, humidity).
• Compared prediction errors (RMSE) of new models against existing formulae and TDS-only models.

Main Results:

• Models combining TDS, demographic, and weather factors reduced PMI and ADD prediction errors by over 50%.
• The best PMI model achieved an adjusted R-squared of 0.42 and 15% lower RMSE than TDS-only models.
• The best ADD model achieved an adjusted R-squared of 0.54 and 10% lower RMSE than TDS-only models, outperforming previous formulas.

Conclusions:

• Incorporating readily available demographic and environmental data significantly enhances the accuracy of postmortem interval and accumulated degree day estimations.
• Developed models offer substantial improvements over existing methods for forensic applications.
• The study highlights the potential of sophisticated statistical approaches with comprehensive predictor variables for more reliable forensic estimations.