Using two consecutive blood samples to calculate backward to THC levels at the time of driving is unreliable due to analytical error overwhelming the small actual concentration changes.
Forensic toxicologists, DUI prosecutors, defense attorneys, and drug-impaired driving policymakers.
87% not significantof THC concentration changes between two blood samples were within analytical measurement uncertainty
What the researchers found
In 87% of cases, concentration changes between two blood samples were not statistically significant due to analytical measurement uncertainty. The two-sample approach to retrograde extrapolation of THC cannot be recommended.
Why it matters
Forensic toxicologists and courts need reliable methods to estimate THC levels at the time of driving. This study definitively shows that the two-sample approach, used in some jurisdictions, is unreliable.
The numbers in context
548 cases from Norwegian apprehended drivers (2000-2020). Median time between samples: 31 min. THC decreased in 62% but increased in 38% of cases. Mean apparent t1/2=6.0 hours. 87% of changes not statistically significant.
How the study worked
Retrospective analysis of 548 cases with two consecutive blood samples from apprehended drivers in Norway. THC measured by GC-MS and UHPLC-MS/MS. Linear regression of rate constants against THC concentration.
Who was studied
548 apprehended drivers in Norway (2000-2020) with THC detected in two consecutive blood samples taken at least 20 minutes apart.
What this study cannot tell us
Retrospective data from a specific legal jurisdiction. Only apprehended drivers (not general population). Cannot know true blood THC at time of driving for validation.
How to read the evidence
Large forensic dataset with rigorous pharmacokinetic analysis; definitive finding against the two-sample approach.
When this study was published
Uses 20 years of Norwegian forensic toxicology data (2000-2020).
The bigger picture
This study has direct implications for drug-impaired driving prosecution and policy, demonstrating that a commonly used forensic approach is scientifically unreliable for THC.
Replication
Definitive analysis of a commonly used forensic approach; applicable across jurisdictions.
Funding
Not specified in abstract
Conflicts of interest
Not specified in abstract
Questions still open
- What alternative approaches could forensic toxicology use for THC retrograde extrapolation?
- Should per se THC limits replace retrograde extrapolation in DUI prosecution?
Read the original research
A two-sample approach to retrograde extrapolation of blood THC concentrations - Is it feasible?
Forensic science international, 352, 111833
Citation
Holman, Peder Olai Skjeflo; Høiseth, Gudrun; Bachs, Liliana; Thaulow, Cecilie H; Vevelstad, Merete S; Mørland, Jørg; Strand, Maren Cecilie. (2023). A two-sample approach to retrograde extrapolation of blood THC concentrations - Is it feasible?. Forensic science international, 352, 111833. https://doi.org/10.1016/j.forsciint.2023.111833