In a study published earlier this month in the journal PLOS-One, scientists at the University of Lyon looked at data from all fatal accidents that occurred in France during 2011. They estimated the heightened risk of driving under the influence of various substances and found that “drivers under the influence of alcohol are 17.8 times more likely to be responsible for a fatal accident,” when compared to completely sober drivers. Drivers under the influence of cannabis, by contrast, are 1.65 times more likely to be responsible for causing a fatal accident.

Those findings are in line with most recent studies of alcohol and cannabis and driving risk. Earlier this year David Bienenstock investigated the science behind drugged driving estimates, and found that THC-positive drivers have a 5% greater crash risk than drivers with no drugs or alcohol in their system. That figure came from the largest domestic case-control study to date, which was published by the National Highway Traffic Safety Administration, a federal agency. That same study found that drivers with a blood alcohol level of .08, the legal limit in most states, were 293% more likely to be involved in a crash. Texting drivers were 310% more likely to crash. 

Past investigations “have shown a decreased capacity of drivers under the influence of cannabis, in particular a decrease in attention, increased reaction time and reduced ability to control direction,” the French researchers noted. Drivers under the influence of alcohol “tend to drive faster, which goes hand in hand with an over-estimation of their own capacities,” the researchers added, “whereas drivers under the influence of cannabis tend to drive more cautiously.”