Cheyllyn Ranae Collinsworth, an 18-year-old Washington state resident, died in May following a car crash. The person responsible was driving under the influence of marijuana and has been charged with vehicular manslaughter. In states where marijuana is legal, car collisions are up 3%, according to the Highway Loss Data Institute. Although marijuana impairs driving ability, police knew the driver in Washington had been using the drug only because he confessed.
No reliable on-the-spot test for marijuana intoxication exists. Urine tests, used widely by employers, are not useful for testing impairment. They detect breakdown products of tetrahydrocannabinol, or THC, marijuana’s psychoactive component. Such metabolites can be found months after a marijuana high has worn off. Making the problem worse, there is very little data on the short- or long-term effects of marijuana on the brain and body.
The only test that currently shows any promise for detecting intoxication is blood-plasma analysis. But empowering police officers to draw blood on the roadside would set up a dangerous precedent for individual liberties. And the connection between blood THC levels and intoxication is not well-defined, as pattern of use and dose directly affect the impairment level, according to the National Highway Traffic Safety Administration.
How to move forward? Researchers must first identify a reliable biomarker of marijuana intoxication. For alcohol, 0.08 grams of ethanol per 100 milliliters of blood has been determined scientifically to cause impairment in a substantial number of people. That’s why 0.08 blood-alcohol content is the legal definition of alcohol intoxication in all states. A similar standard must be developed for THC.
Scientists also must determine a straightforward test to detect THC intoxication. One possibility is a “breathalyzer” test for marijuana. Scientific American reports that one company, Hound Labs, claims to have created a portable device that measures the amount of marijuana consumed. Other companies are racing to develop blood-, urine- or saliva-based tests. Alternatively, a company could consider developing a test that analyzes THC with small blood samples, the way diabetics test for glucose levels.
Another possibility is to develop a test that determines when THC was consumed. NHTSA says that impairment occurs for roughly three hours after cannabis is consumed. If the ratios of THC to its metabolites can be measured appropriately and over time, it may be possible to pin down the critical time when someone was first intoxicated.
As more states legalize marijuana and its availability increases, ensuring responsible use will become more important. We propose that the federal government loosen its restrictions on marijuana research so that Americans better understand how the drug affects their minds and bodies. This will help quantify a meaningful marijuana intoxication standard. That rule—along with the consequences for crossing that threshold—must be derived from scientific evidence, not ideology.
Mr. Siegel is the author of “Beyond the Galaxy” (World Scientific, 2015). Mr. Berezow is a senior fellow at the American Council on Science and Health.