As cannabis legalization spreads across the United States, an urgent public health issue continues to vex communities and policymakers alike—detecting cannabis-impaired driving and how to best respond. For more than seven years, my colleagues and I have been researching how to detect cannabis-impaired driving using tools such as a driving simulator, cannabis breathalyzers, and tests of reaction time and short-term memory. Several questions have surfaced during this research, including how to define cannabis impairment and how it varies by dose, method, or frequency of use. One thing is clear: as a necessary first step in responding to this challenge, public health leaders need more funding and stronger coordination across all levels of government.

Even before the surge in legal adult cannabis use, which now includes more than half of U.S. states, cannabis, after alcohol, is the most detected drug in drivers involved in crashes. While alcohol impairment can be quickly and reliably determined with a breathalyzer, no equivalent standard exists for cannabis. Adding even more complexity to the issue, the relationship between tetrahydrocannabinol (THC), the psychoactive compound in cannabis, and driving impairment is very complex. THC behaves differently in the body than does alcohol. THC can linger in the bloodstream long after the short-term effects have worn off, sometimes for days or even weeks. That means a driver could test positive for THC and they may be (or may not be) impaired at the time of driving. This alarming complexity should be drawing a swift public health and coordinated policy response. Instead, response efforts are stalled because our communities are waiting on better ways to measure real-time impairment.

Even more problematic, someone may be highly impaired shortly after cannabis use but a standard blood test may not be able to detect impairment. This makes enforcement of cannabis impaired driving much more challenging for law enforcement and the courts compared to alcohol-induced impaired driving. Unlike the coordinated approaches to curb drunk driving, the challenge of cannabis induced impairment has been left up to states to solve, including Colorado. However, a patchwork of state regulations is both impractical and haphazard. Federal leadership and investments in public health could help deliver a solution.

Now, enter the private sector. Several companies are racing to develop THC breathalyzers and other detection technologies, but product development lacks consistent scientific validation or regulatory oversight. As a researcher in this field, we have tested the accuracy of detection tools such as breathalyzers, ocular tests, and tablet-based tests; these findings are available for public review. Devices like breathalyzers can help narrow the window of time in which someone has used cannabis, to hours instead of days (as with blood), for example. Tablet-based tests can help discern if someone is impaired, no matter the cause—like fatigue or prescription drugs. However, these technologies are not yet reliable for high-stakes decisions like losing a driver’s license or one’s job. With more investments in public health and safety research, we can answer key questions about these devices and, at a minimum, support public safety professionals to detect driving under the influence and improve safety on our roads. This will require funding and stronger collaboration between scientists, industry, and government agencies to set standards and evaluate performance.

We also need to invest in public education and prevention campaigns. Many people who use cannabis mistakenly believe that driving under the influence of cannabis is not dangerous. But studies show that cannabis impairs attention, reaction time, and decision-making–all critical skills for safe driving.

Messaging around the risks of cannabis-impaired driving must catch up to the reality of legalization. Colorado has taken some steps to inform the public. A national, organized effort, like the decades-long campaigns that successfully reduced traffic deaths from drunk driving, would be a step in the right direction. “Friends Don’t Let Friends Drive Drunk” and “Buzzed Driving is Drunk Driving” are forever engrained in the American lexicon.

Public health experts, advocacy groups, and communications teams could join forces with federal leadership to initiate similar public awareness campaigns.

Every day that we delay, we leave law enforcement without the tools they need, courts without the evidence they require, and the public at risk when others get behind the wheel. This problem has been left to states for too long and federal leadership and resources are overdue. Public health promotes public safety, and by investing in public health we can make our roads and communities safer.

Ashley Brooks-Russell, Ph.D., is an associate professor at the Colorado School of Public Health and is the director of the Injury & Violence Prevention Center. The opinions expressed here are those of Dr. Brooks-Russell and do not reflect an official position on behalf of the University of Colorado or the Colorado School of Public Health.