The Washington Traffic Safety Commission released new data this week that

shows that marijuana is increasing as a factor in deadly crashes. The number of drivers involved in deadly crashes who tested positive for marijuana increased 48 percent from 2013 to 2014.

Here’s the data. For comparison, from 2013 to 2014, the number of drivers involved involved in fatal crashes who had blood alcohol contents of 0.080 or higher declined by 12.8 percent. The total number of fatal crashes increased by 7.0 percent.

I-502 was approved by voters in Nov. 2012, but sales of recreational marijuana did not became legal until July 8, 2014. Crosscut notes,

The study’s unspoken implication is that the recreational legalization of marijuana is connected to its increased influence in fatal car crashes. . . . [But] the majority of the crashes took place in the first six months of the year, before recreational marijuana stores opened.

The chart below shows fatal crashes involving a driver who tested positive for marijuana (the data file notes that this test does not distinguish between active and inactive THC). The blue columns are the first half of the calendar year, and the green are the second half.

While the increase in marijuana-related fatal crashes in 2014 is notable, it’s not necessarily indicative of a new normal. As Crosscut mentions, these crashes declined in the second half of the year (but there were still more of them than in the same time frame in the previous several years). Further, the data file includes preliminary numbers for the first quarter of 2015. The number of total fatal crashes is the same in the first quarter of 2015 as it was in the first quarter of 2014, but the fatal crashes involving marijuana (and those involving alcohol) declined significantly in 2015 (each by about 50 percent).

Of course, the 2015 data is preliminary and could very well be revised upward. But it does caution against using the 2014 numbers as proof of an upward trend when they may instead turn out to be an anomaly.