Implementation of recreational cannabis laws correlated with an increase in traffic fatalities in Colorado, no significant change in Washington State

1. Compared to a simulated synthetic model of Colorado, in which recreational cannabis laws (RCLs) had not been implemented, the state of Colorado experienced a mean increase of 75 additional traffic fatalities per year from 2014-2017.

2. There was no statistically significant difference in traffic fatalities in Washington state compared to a simulated Washington state without RCLs.

Evidence Rating Level: 2 (Good)

Study Rundown: Laws pertaining to cannabis use, both on a recreational and medicinal basis, have been widely implemented across much of the United States over the past decade. In an effort to clarify some of the public health implications of these new policies, the authors of this study modeled the traffic fatalities in Colorado and Washington, two of the first states to legalize the recreational use of cannabis, and compared the trend in fatalities to a synthetic model, in which RCLs had not been passed. Analyses revealed that the traffic fatality rate per 1 billion vehicular miles traveled (VMTs) increased significantly in Colorado, but not in Washington. Further, when the authors removed states with medicinal cannabis laws and neighboring states (to control for traffic spillover), the gap between observed and expected traffic fatalities widened further in Colorado, but not in Washington state. The results of this study indicate that the legalization of cannabis may have further reaching public health implications, including potential correlation with traffic accidents and fatalities. However, the fact that Washington states’ rate of traffic fatalities did not diverge significantly from the synthetic model indicates that there may be other mitigating factors, both related to discrepancies in cannabis use and otherwise, affecting traffic fatalities. Additional study is warranted.

Traffic fatality data was acquired from the National Highway Traffic Safety Administration’s FARS database from 2004- 2017. Data from states in which RCLs were implemented prior to 2017, and Hawaii were excluded. Sensitivity analyses were also conducted in which data from neighboring states was excluded (to control for traffic spill over), excluding states with medicinal cannabis laws (MCLs), and by starting the observed model from the time in which the RCL was passed as opposed to the date it was implemented.

Click to read this study in JAMA

Relevant Reading: Cannabis Effects on Driving Skills

In-Depth: The authors of this study utilized a synthetic group approach to create an algorithm traffic from data from similar states that modeled the predicted trend in traffic fatalities in both Washington state and Colorado, had RCLs not been implemented. In Colorado, there was an increase of 1.46 deaths per 1 billion VMTs as compared to the synthetic non-RCL model (MSPE 2.44). This translates into an average of 75 excess fatalities per year. Washington state, on the other hand, did not see a statistically significant increase in traffic fatalities as compared to the expected non-RCL model. When neighboring states were excluded from the models, the gap between actual fatalities and expected fatalities in the non-RCL modeled Colorado widened. The authors report a difference 1.84 fatalities per 1 billion VMT, which translates to an average of 94 excess deaths per year in Colorado. When states with MCLs but not RCLs were excluded from the data pool, there was a difference of 2.16 fatalities per 1 billion VMTs, an average of 111 excess deaths per year, when comparing Colorado and synthetic- Colorado. There was no significant difference between observed and expected deaths using the Washington state and synthetic Washington state models in the sensitivity analyses.

Image: PD

©2020 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.