Where recreational marijuana is legal, data show minimal impacts on teen use and traffic deaths
PHOENIX — Since 2012, 11 states have legalized marijuana use for adults — which voters nationwide are considering on their ballots this year. Researchers are just beginning to understand the effects of those laws.
Colorado and Washington were the first states to legalize the drug, and California, the most populous state in the nation, followed them.
Among the most pointed concerns with legalization are whether it has caused more young people to use the drug and whether more people are dying in auto crashes caused by impaired drivers.
Data show little change in either area.
Surveys of young people in Colorado, for example, show a slight decline in the percentage of middle and high school students using the drug. In Washington, the rates have remained the same.
Opponents of legalization say the risk is too great if young people are given the impression that marijuana is not harmful, or if drivers become lax about getting behind the wheel when they are high.
States with legalized marijuana are finding more drivers impaired by the drug, but that comes in part because they are looking harder for it. Colorado, for example, did not track the level of marijuana impairment of drivers suspected of using it until 2016.
Washington saw an increase in drugged driving before legalization that continued after the drug was permitted, and has seen more fatal accidents with people on multiple substances. California data shows an increase in people driving while on drugs involved in fatal accidents.
But officials in those states are hesitant to peg the increases on legalization, and researchers haven't shown an increase in total traffic fatalities tied to the changes. For example, California had 8.3% fewer traffic fatalities in 2018, the year retail marijuana sales launched there, than it did in 2017.
In Arizona, recreational legalization is on the ballot this fall. Supporters say the best way to mitigate impacts on society are to license and regulate the industry, funneling some of the new tax revenue back into programs to reduce usage by young people and combat impaired driving.
The state's legislation would allow people 21 and older to possess one ounce of marijuana, not more than 5 grams of which can be a concentrated form of the drug. Adults also could grow six plants each in their homes, and the measure would license about 160 retailers to sell the drug.
South Dakota, New Jersey and Montana voters have similar measures on their ballots.
In Arizona, the proposed legislation also includes the ability for people previously convicted of minor marijuana crimes to have their records expunged, and changes requirements for law enforcement to cite drivers for operating a vehicle under the influence.
No increase shown in teen use, but a change in perception
Among the major concerns with legalizing marijuana is that it would create a more permissive attitude toward the drug and increase the number of young people who use it.
Colorado and Washington had some of the highest rates of marijuana use by young people in the nation from 2011-2015, according to the Centers for Disease Control, but state data don't show those figures increasing after legalizing the drug for adults.
In Washington, surveys of young people conducted in cooperation with the Office of the Superintendent of Public Instruction showed no increase in the numbers of sixth, eighth, 10th and 12th graders using marijuana after legalized sales began there in 2014.
In Colorado, similar surveys showed the number of middle and high school students using marijuana actually declined after legalization.
California hasn't yet released its biennial survey of young people since recreational sales began in that state in 2018, but the prior survey showed a 3.4% decline in young people in that state using the drug, even with its first-in-the-nation medical marijuana program that dates to 1996 and decriminalizing of the drug in 2011.
"The benefits outweigh the threats,” said Will Humble, former Arizona Director of the Department of Health Services and now executive director of the Arizona Public Health Association.
"The worst outcome for me would have been an increase in adolescent use, which the data doesn’t show,” he said.
But Lisa James, the spokeswoman for the opposition campaign known as Arizonans for Health and Public Safety, said that legalized states send the wrong message to young people.
James' group gets much of its funding from Center for Arizona Policy, a well-known conservative group at the Capitol that espouses "foundational values of life, marriage and family, and religious freedom."
"The youth are becoming desensitized because all the advertising and the abundance of it. There no longer is a stigma for underage use as there used to be," James said. "So teen use is starting to increase again."
Indeed, research from the National Institute on Drug Abuse shows an uptick in marijuana use nationally by eighth and tenth graders surveyed in 2019. Another study sponsored by the institute showed the perceived risk of using marijuana regularly has dropped among young people nationally, too.
For adults, data from the Centers for Disease Control and Prevention show a slight increase in adults in the U.S. using marijuana in recent years, mostly those aged 18-25.
Maine and marijuana: Sales of pot for recreational use start in October
What traffic data shows about fatalities
The other key point of debate for marijuana legalization is public safety, and mostly whether it adds danger to the roadways.
Early data shows that while more people overall may use the drug and get behind the wheel in legalized states, impaired driving does not appear to kill more of them.
And though more drivers test positive for marijuana, that comes with several caveats. One is that legalized states are testing more for marijuana than they did in the past. Another is that simply finding marijuana metabolites in a drug screen does not indicate impairment, and may show use that took place weeks in the past.
For many drivers found to be under the influence of alcohol, it's likely that arresting officers have not historically conducted additional "time consuming and costly" testing for drugs if a driver is clearly in violation of blood-alcohol limits, according to the Colorado Department of Transportation.
This could possibly undercount marijuana use in historical data, while the legalization of the drug in some states has likely prompted law enforcement to look harder for marijuana impairment, something the Colorado Department of Transportation acknowledges in its statistical reports.
"Legalization has heightened awareness of the need to gather data on marijuana and, in some cases, has led to improvements in data collection that then make analyzing historical trends difficult," the department says in one report.
Either way, several studies show significantly more drivers test positive for marijuana in places where the drug is legal. But simply testing positive doesn't mean impairment, because signs of the drug can show up on a drug screen weeks later and don't indicate whether the driver was actually impaired, which lasts only hours after using the drug.
Further complicating the issue: Marijuana does not have an agreed upon, testable limit for impairment. People are affected differently by the drug depending on how often they use it and other factors, such as body weight.
Some states, including Washington and Colorado, have set a limit to clarify when a driver is legally impaired, and simply testing positive for tetrahydrocannabinol, or THC, does not legally show impairment in those states.
They set impairment at having 5 or more nanograms of active Delta-9-THC per milliliter of blood, even though drivers who fall below this limit may be cited for being impaired based on other factors.
The data leaves the question of whether legalization makes roads more dangerous. Even looking just at fatality rates doesn't give a clear answer.
In 2018, for example, fatal accidents where drivers were impaired by marijuana were about 8% of Colorado's total of 632 fatal crashes, compared with 33% of those crashes where the driver was impaired by alcohol, said Sam Cole, traffic safety communications manager for the Colorado Department of Transportation.
"So the numbers are much smaller when compared to alcohol, but that doesn’t mean marijuana isn’t important," Cole said.
That's why the state now spends $500,000 to $1 million annually on educating drivers on the dangers of driving high, which is funded by taxes on marijuana, he said.
Colorado's data tracks with broader studies.
A study published in the American Journal of Public Health in 2017 "found no significant association between recreational marijuana legalization in Washington and Colorado and subsequent changes in motor vehicle crash fatality rates in the first 3 years after recreational marijuana legalization."
"Although our findings seem at odds with the known effects of marijuana impairment, and with previous studies finding associations between motor vehicle crashes and marijuana use, they are consistent with the most recent analysis of medical marijuana legalization and motor vehicle crash fatalities," the authors wrote.
Washington: More drivers on multiple substances
Washington officials have recorded a troubling trend of more drivers in fatal crashes testing positive for more than one substance, often alcohol and marijuana.
These "poly-drug" cases are now more common than fatal accidents with drivers impaired by a single substance, which has raised alarms for officials in that state.
The increase in these incidents began before recreational sales began in that state and have grown since then.
Meanwhile, the total number of fatal crashes has gone up and down but overall has remained relatively flat. There were 499 fatal crashes in 2015 and the same number in 2019. It hit 534 in 2017.
"Marijuana by itself is still a small number of the overall fatalities," said Shelly Baldwin legislative and media director for the Washington Traffic Safety Commission. "But poly-drug use, which marijuana and alcohol is the most common, that really, really increased."
Is that increase because of the state's legalization of marijuana?
"I can’t say that yet," Baldwin said, saying more studies need to be done. "It is not like marijuana just showed up on our roadways when we had the initiative. It has been out there forever."
Like in Colorado, alcohol remains the larger concern even after marijuana was legalized and made widely available.
"Alcohol is the deadliest substance involved in fatal crashes," Washington's Traffic Safety Commission reports. "Drivers under the influence of alcohol, alone or in combination with other drugs, emerge as the most high-risk drivers ultimately being involved in fatal crashes."
The total percentage of drivers in fatal crashes who test positive for any substance is little changed since legalization, according to the same report.
Washington has tried to communicate that mixing alcohol and marijuana magnifies impairment, Baldwin said. This took on a new urgency after a 2018 survey conducted by Montana State University for the commission showed some drivers intentionally use marijuana after drinking alcohol. They believed that it "sobers them up," she said.
“They felt like it helped get them ready to drive," she said. "That is alarming. It magnifies impairment. You are impairing two different systems now."
California sees more drugged drivers, but fatalities decline
Data on California drivers from 2017-2018 shows a large increase in the number of drivers killed in crashes who test positive for drugs, but the data includes all drugs and does not differentiate between active and inactive metabolites for marijuana.
Overall, though, fatal crashes in California decreased 8.3% during that period.
And the 2018 mileage death rate, or fatalities per 100 million miles traveled, was lower in California, Colorado and Washington than in Arizona, where recreational use is still not legal, according to federal data.
Less precise studies have shown an increase in accidents in states with legalized marijuana, but have not shown the drug as the cause.
In 2018, the Insurance Institute for Highway Safety and Highway Loss Data Institute estimated that collision claims rose a combined 6% following the start of retail sales of recreational marijuana in Colorado, Nevada, Oregon and Washington, compared with the control, non-legalized states of Idaho, Montana, Utah and Wyoming.
Analysts for that study looked at collision data from January 2012 through October 2017.
Other problems in California
The California legalization has widely been condemned for allowing the black market to thrive, but it may be too new to draw conclusions about impaired driving and the rates at which youth use the drug, and the state still is tinkering with tax rates.
The generally lax attitude toward the drug there makes it even more difficult to assess the impact on society from allowing retail, recreational sales of the drug in 2018, because the state already had a more permissive legal stance.
California decriminalized marijuana possession in 2011, and with that change researchers saw impacts on things like the drop-out rate and drugged driving.
Beyond the roadways, a report last year from the California Cannabis Advisory Committee said that as much as 80% of marijuana sales in that state are still on the black market.
It cited a variety of factors, including state regulations that allow local municipalities to ban marijuana retailers, pushing more sales to the illicit market, as well as excessive taxes.
The advisory committee set up by the state said a new ballot initiative may be needed to fix those problems. And an educational campaign was launched to try to convince consumers to purchase marijuana from legally licensed retailers, and for retail locations to go through the proper channels to sell marijuana legally.
Even with about 40% of the state's municipalities banning retail sales, the state had more than 5,700 dispensaries a year ago, with thousands more seeking licenses.
Arizona measure tries to avoid pitfalls
The proponents of recreational legalization in Arizona said they wrote the initiative to avoid problems seen in other states like California and Colorado, where some people complain about the prevalence of retail stores selling the drug.
The act would essentially cap the number of retail dispensaries to the existing 130 medical marijuana retailers licensed in the state, plus a few additional licenses in rural counties and 26 new licenses to be issued to people who have been disenfranchised by marijuana laws.
“When we did our polling it was very clear that people did not want dispensaries on every corner, but they did by a significant majority want dispensaries to be available to people,” said Steve White, CEO of Harvest Health and Recreation, multistate marijuana retailer that is financially backing legalization legislation.
People wait in line outside Nature's Medicines in Phoenix while waiting in line to get into the store on March 20, 2020. Arizona Republic
He said the opponents all incorrectly assume that if marijuana remains illegal, none of the problems associated with it will affect society.
"I always am shocked by the arguments that, well, if you make it legal, people all of the sudden are going to do X. Or they tell you stories of personal traumas they’ve seen, family members who became addicted to other substances," White said.
"The thing that underlies all of that is that all of that (happened) under prohibition because it is not legal in Arizona. To me the question is always whether or not you are going to tax and regulate the product, or whether you are going to continue to see the illicit market meet the demands."
One final question Arizona voters might want ask residents in other states is whether they would do it again.
What limited research there is on the subject seems to show little buyers' remorse.
A study in Washington state found that two years after voters approved legalization, the same measure likely still would pass at the ballot — and by a wider margin. Just 5% of people who voted for legalization said they would change their votes in the 2014 study, while 14% of people who had voted against it said they would change their votes to support it.
What the research shows: Colorado
Recreational sales available: January 2014.
Taxes: 15% whether wholesale or retail.
Tax collections 2019: $302 million.
Impaired driving: Prior to 2016 the state does not have data on specific levels of Delta-9 THC, but since it began tracking that, the percent of fatalities with drivers who tested positive for Delta‐9 THC at the 5 ng/mL level has fallen. It was 8% in 2017, down from 13% in 2016.
DUIs: The total number of DUIs in Colorado issued by state troopers fell 15%, from 5,705 in 2014, to 4,849 in 2017, mostly because of a decline in those involving alcohol only. The number of cases where marijuana alone or in combination rose 5% in that same time.
Middle and high school students who reported current marijuana use in 2014: 20%
Middle and high school students who reported current marijuana use in 2017: 19%
Colorado Sources: Colorado Department of Revenue Taxation Division — Marijuana Regulation and 2019 tax collected; State of Colorado Marijuana Taxes, Licenses, and Fees Transfers and Distribution — 15% tax; Colorado Division of Criminal Justice, Department of Public Safety, Office of Research and Statistics — Impaired Driving and DUIs; Healthy Kids Colorado Survey Data Brief on Colorado Youth Marijuana Use in 2017.
Year recreational sales available: July 2014.
Taxes: Excise tax of 37% in addition to sales tax of 6.5%.
Tax collections in 2019: $395.5 million
Impaired driving: Washington has tracked not only an increase in drivers in fatal accidents testing positive for THC, but also for what researchers in one report describe as "poly drug drivers" — those who test positive for alcohol and at least one other drug, usually marijuana.
There were 97 "poly-drug" fatalities in 2013 before legalization and 137 in 2016 after legalization. The number of fatal crashes with drivers testing positive for THC increased from 7 to 27 in the same time frame, according to the Washington Traffic Safety Commission.
10th graders who reported currently using marijuana in 2014: 18%.
10th graders who reported currently using marijuana in 2018: 18% (Similarly flat trends were found for 6th, 8th and 12th graders, as well as all four grades for "lifetime use" and "heavy use.")
Washington sources: Washington State Liquor and Cannabis Board — 37% excise tax and 2019 tax collections; Washington Traffic Safety Commission — Impaired driving; Looking Glass Analytics, Healthy Youth Survey 2014 — 10th graders who reported marijuana use in 2014; Washington State Department of Health — 10th graders who reported marijuana use in 2018.
Recreational sales available: January 2018
Tax on marijuana: Additional excise tax of 15% on retail sales, in addition to state/local sales taxes and a local cannabis tax of 0-15% municipalities may impose. Growers pay a tax of $9.65/ounce of flower, $2.87/ounce on leaves.
2019 tax collections from marijuana: $629 million
Drugged driving prior to legalization: In 2017, 11% of all drivers killed in motor vehicle crashes who were tested were positive for legal and/or illegal drugs.
Drugged driving after legalization: In 2018, 42% of all drivers killed in motor vehicle crashes, who were tested were positive for legal and/or illegal drugs. Traffic fatalities decreased 8.3%, from 3,884 in 2017 to 3,563 in 2018.
11th graders who used marijuana in the past 30 days 2013-15: 20.1%
11th graders who used marijuana in the past 30 days 2015-17: 16.7% Note: Recreational sales did not begin until Jan. 1, 2018, which was not covered in this survey.
California sources: California Department of Tax and Fee Administration — 15% tax on marijuana; Institute on Taxation and Economic Policy — 2019 tax collections from marijuana; California Office of Traffic Safety — Drugged driving prior to legalization and after legalization; WestEd's California Healthy Kids Survey, School Climate, Substance Use, and Student Well-being in California, 2015-17 — 11th graders who used marijuana in the past 30 days from 2013 to 2015 and from 2015 to 2017.