In relation to the road-safety implications of driving after using cannabis, the following are questions policy makers need answers to, if they are to develop evidence-based policies.
Q1. Does research show that risks of crashing when there is THC in the bloodstream are significantly higher than when there is no THC in the bloodstream?
A1. No – when study biases are taken into account. In other words, the presence of THC in the blood is not associated with a higher risk (odds ratio) of crashing.
Q2. Does research show that driving performance is degraded when there is THC in the bloodstream compared to when there is no THC in the bloodstream?
A2. Yes, so we have a problem. Why does the degradation of performance (impairment) not increase the cannabis-crash odds ratio? There are many things to be taken into account. Cannabis users are not inclined to drive soon after using. Cannabis impairment is generally weak and probably normally below a threshold where driving performance is dangerously degraded. Furthermore, while there could be a small increase in crash risk due to impairment, it is probably balanced by a small decrease in crash risk due to more cautious driving (as demonstrated by slower travelling speeds and greater headway). Given the role of speed in crash causation, a little extra caution could go a long way.
Q3. If medicinal cannabis has effects on driving performance, then is there a time window after ingestion of THC when impairment is most likely to degrade driving performance?
A3. Yes, there is a time window. Its duration has been explored in a couple of recent reviews. The most conclusive evidence we have so far: For smoked cannabis, probably about three hours from time ingested. For edibles, such as cookies, probably a bit longer, starting after a delay of at least half-an-hour. The reviews indicate low-level impairment can, in some cases, last beyond three or four hours.
Q4. Does a performance-based test exist which can measure any decrease in performance that might arise from having THC in the bloodstream?
A4. No. There is too much individual variation in any performance tests and too much individual variation to be able to accurately gauge levels of THC impairment. However, the answer could possibly be ‘yes’ if baseline levels of performance could be established for individuals. Still, the measured performance decrement due to THC would probably be irrelevant to crash risk.