When Belgium switched from urine screening to oral fluid screening for roadside drug testing, the rate of false positives dropped from 17% to 8%.
Readers interested in how drugged driving is detected and how different testing methods compare.
False positive rate dropped from 17% to 8% after switching to oral fluid testing
What the researchers found
Belgium changed its roadside drug testing protocol in 2010, moving from urine-based screening to oral fluid-based screening. Researchers compared the results from both periods across roughly 8,000 drivers tested.
Under the old urine system, 88% of cannabis-positive screens were confirmed by blood testing. Under the new oral fluid system, 66% of cannabis-positive screens were confirmed. However, the overall false positive rate (drivers who screened positive but had no confirmed drugs above legal limits in blood) dropped from 17% to 8%.
The newer oral fluid approach also shifted how police identified potential impairment, moving from requiring visible signs of impairment to looking for signs of recent drug use.
Why it matters
Accurate roadside drug testing is essential for road safety policy. This real-world comparison shows that switching to oral fluid testing reduced false positives, meaning fewer sober drivers were subjected to unnecessary blood draws while still identifying impaired drivers.
The numbers in context
About 4,100 urine and 3,900 oral fluid datasets analyzed. Cannabis confirmation rates: 88% (urine) vs. 66% (oral fluid). Cocaine: 21% vs. 30%. Amphetamines: 20% vs. 28%. Overall false positive rate dropped from 17% to 8%.
How the study worked
The study analyzed approximately 4,100 urine screening datasets (April 2008 to September 2010) and 3,900 oral fluid screening datasets (October 2010 to March 2013). All positive screens were confirmed against plasma drug concentrations using established legal cutoff values.
What this study cannot tell us
The two periods used different screening criteria (signs of impairment versus signs of recent use), which makes direct comparison complex. Different testing matrices detect different windows of use. The legal cutoff values also changed between periods, which could affect confirmation rates independently of the screening method.
How to read the evidence
This is a large cross-sectional comparison of two enforcement periods with real-world data from approximately 8,000 drivers, providing moderate-quality evidence on testing method performance.
When this study was published
Published in 2015, covering data from 2008-2013. Roadside testing technology has continued to evolve since then.
The bigger picture
Many countries are grappling with how to test drivers for cannabis impairment, especially as legalization expands. Belgium's experience with transitioning testing methods provides useful data for other jurisdictions designing or refining their drugged-driving enforcement programs.
Questions still open
- How do oral fluid THC levels correlate with actual driving impairment? Would combining behavioral assessment with oral fluid testing further improve accuracy?
Common questions
Which testing method was better at catching cannabis use?
Does a positive roadside test prove impairment?
Read the original research
Roadside drug testing: comparison of two legal approaches in Belgium.
Forensic science international, 249, 148-55
Citation
Van der Linden, T; Wille, S M R; Ramírez-Fernandez, M; Verstraete, A G; Samyn, N. (2015). Roadside drug testing: comparison of two legal approaches in Belgium.. Forensic science international, 249, 148-55. https://doi.org/10.1016/j.forsciint.2015.01.034