J Forensic Sci. 2026 Jun 30. doi: 10.1111/1556-4029.70401. Online ahead of print.
ABSTRACT
Determining the cause of death using postmortem fentanyl concentrations in femoral blood remains a major challenge in forensic toxicology. This is due to substantial concentration overlap between overdose deaths and incidental non-drug-related deaths. Existing literature largely reports descriptive statistics such as ranges and medians. These statistics do not provide an objective framework for evaluating the evidentiary value of a specific fentanyl concentration. Using distribution analysis, nonparametric tests, and likelihood ratios, we developed a four-zone statistical model to quantify overdose probability based on fentanyl concentrations in femoral blood. We also determined how polysubstance use alters these probabilities, leading to adjustments in the interpretive windows. Logistic regression modeling showed that age and sex contribute minimally to overdose probability and identified which drug classes best predict overdose classification. Together, these results provide a rigorous and objective method to determine postmortem overdose certainty in fentanyl-only and multi-drug intoxication cases.
PMID:42376689 | DOI:10.1111/1556-4029.70401
