Medical marijuana opponents are taking to the streets to oppose Amendment 2, citing statistics that drugged driving would be a major side effect to legalizing cannabis.
Don't Let Florida Go To Pot, a coalition of more than 40 organizations fighting against the proposed medical marijuana amendment, says on its website that the drug is implicated in a fourth of all fatal accidents.
"Twenty-five percent of all drug-related fatal vehicle accidents in the U.S. involve marijuana," the group says under the header "Statistics."
PolitiFact Florida hit the books to see what Don't Let Florida Go To Pot was driving at.
Eric Pounders, spokesman for the Florida Sheriffs Association, said the 25 percent stat came from an October 2011 White House Office of National Drug Control Policy report that measured the rate of positive results among drivers that had been tested for drugs in fatal crashes between 2005-2009. The study used National Highway Traffic Safety Administration's Fatality Analysis Reporting System data of accidents across 50 states in which the driver was killed and subsequently tested for drugs.
The report noted that of the roughly 127,000 fatal crashes in that time frame, almost 78,000 drivers were tested for drugs. Winnowing down those results, positive tests for cannabinoids rose steadily from 22.6 to 26 percent between 2005 and 2008, dropping slightly to 25.3 percent in 2009.
There are some problems with that data. First, there is no set procedure or uniform level of toxicity for drug testing among the states, as laws and protocols vary widely. The report says "a positive test result does not necessarily imply impairment or causation," and says testing often is inaccurate. Furthermore, data in which a driver may have been using drugs but survived a crash in which someone else died is not included.
There is often no universally accepted threshold of impairment for illicit drugs, the White House report says. There also is some question as to whether the specific presence of cannabinoids, including the main psychoactive chemical component, tetrahydrocannabinol, is an accurate indicator of impairment.
The NHTSA states, "It is difficult to establish a relationship between a person's THC blood or plasma concentration and performance impairing effects." That's in part because cannabinoids linger in a person's system long after they've ingested the drug.
"Marijuana, unlike alcohol and most other drugs, stays in a person's system for up to 30 days," says Ben Pollara, spokesman for United for Care, the group promoting Florida's Amendment 2. "So its presence in a person's blood is not an indicator that they were impaired at the time of the accident."
That 25 percent figure seems to track with other studies of drug-related fatal crashes, however. One 2011 study from the Pacific Institute for Research and Evaluation in Calverton, Md., examined fatal single-vehicle accidents between 1998 and 2009 that involved speeding, failure to obey or yield, inattention and failure to use a seat belt.
That study found about 23 percent of those crashes involved cannabinoids, although the study adds the caveats that the drivers' drug levels were not available, and those deaths most often involved speeding and the driver not using a seat belt.
When you change the parameters of the research, the numbers vary. A 2014 Columbia University study looking at six states reported marijuana was present in 12 percent of accidents in 2010, up from 4 percent in 1999.
A University of Colorado School of Medicine study released in May 2014 comparing Colorado to 34 other states concluded marijuana was present in 10 percent of fatal accidents by the end of 2011, as opposed to 4.5 percent at the beginning of 1994.
But yet another study from 2012 from universities in Oregon, Montana and Colorado measured crashes in medical marijuana states. The study found that although instances of drugged driving went up in those places, total fatal accidents dropped somewhere between 8 and 11 percent overall.
The possible reason, in a nutshell? Most people probably smoke their weed at home instead of driving home drunk from a bar, the study's authors hypothesized.
Alcohol still remains the most abused substance involved in fatal drug-related accidents — as high as 60 percent, depending on the year and study.
Montana State University economics professor D. Mark Anderson, one of the authors of the study showing a decrease in fatalities, said that the numbers didn't supply a direct link between marijuana use and accidents. Indeed, none of the research PolitiFact Florida cited above claimed to establish a definite cause and effect.
"Maybe 25 percent of all people involved in fatal vehicle accidents also drank milk for breakfast," Anderson said.
Don't Let Florida Go To Pot said, "25 percent of all fatal drug-related vehicle accidents in the U.S. involve marijuana."
On balance, we rate the statement Half True.
Edited for print. Read the full version at PolitiFact.com/florida.