Driving under the influence of cannabis: a 10-year study of age and gender differences in the concentrations of tetrahydrocannabinol in blood

Addiction. 2008 Mar;103(3):452-61. doi: 10.1111/j.1360-0443.2007.02091.x. Epub 2008 Jan 8.


Background: Delta(9)-Tetrahydrocannabinol (THC) is the major psychoactive constituent of cannabis and its various preparations. Increasing use of cannabis for recreational purposes has created a problem for road-traffic safety. This paper compares age, gender and the concentrations of THC in blood of individuals apprehended for driving under the influence of drugs (DUID) in Sweden, where a zero-tolerance law operates.

Measurements: Specimens of blood or urine were subjected to a broad screening analysis by enzyme immunoassay methods. THC positives were verified by analysis of blood by gas chromatography-mass spectrometry (GC-MS) with a deuterium-labelled internal standard (d(3)-THC). All toxicology results were entered into a database (TOXBASE) along with the age and gender of apprehended drivers.

Findings: Over a 10-year period (1995-2004), between 18% and 30% of all DUID suspects had measurable amounts of THC in their blood (> 0.3 ng/ml) either alone or together with other drugs. The mean age [+/- standard deviation (SD)] of cannabis users was 33 +/- 9.4 years (range 15-66 years), with a strong predominance of men (94%, P < 0.001). The frequency distribution of THC concentrations (n = 8794) was skewed markedly to the right with mean, median and highest values of 2.1 ng/ml, 1.0 ng/ml and 67 ng/ml, respectively. The THC concentration was less than 1.0 ng/ml in 43% of cases and below 2.0 ng/ml in 61% of cases. The age of offenders was not correlated with the concentration of THC in blood (r = -0.027, P > 0.05). THC concentrations in blood were higher when this was the only psychoactive substance present (n = 1276); mean 3.6 ng/ml, median 2.0 ng/ml compared with multi-drug users; mean 1.8 ng/ml, median 1.0 ng/ml (P < 0.001). In cases with THC as the only drug present the concentration was less than 1.0 ng/ml in 26% and below 2.0 ng/ml in 41% of cases. The high prevalence of men, the average age and the concentrations of THC in blood were similar in users of illicit drugs (non-traffic cases).

Conclusions: The concentration of THC in blood at the time of driving is probably a great deal higher than at the time of sampling (30-90 minutes later). The notion of enacting science-based concentration limits of THC in blood (e.g. 3-5 ng/ml), as discussed in some quarters, would result in many individuals evading prosecution. Zero-tolerance or limit of quantitation laws are a much more pragmatic way to enforce DUID legislation.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accidents, Traffic
  • Adolescent
  • Adult
  • Age Factors
  • Automobile Driving*
  • Dronabinol / blood*
  • Female
  • Humans
  • Male
  • Marijuana Smoking / blood*
  • Marijuana Smoking / epidemiology
  • Middle Aged
  • Sex Factors
  • Substance Abuse Detection / methods
  • Sweden / epidemiology


  • Dronabinol