Studies of young marijuana users show that they have lower cognitive performance and higher rates of psychotic disorders than their peers who aren’t users. Yet scientists cannot conclude from these studies that marijuana use is the cause. “People may have been different from the start, and that might have made them more likely to perform differently or use marijuana in the first place,” says psychiatrist Eden Evins, MD, MPH, Director of the Center for Addiction Medicine at Massachusetts General Hospital.

To learn more about the causes and effects of marijuana use in young people, Evins is devising new ways to study its use in adolescents. She must proceed carefully, though, because so far, studies have linked marijuana with harm, including worsening the course of schizophrenia and potentially contributing to its occurrence. To avoid doing a study that asks young people to use marijuana, she instead decided to pay young people who already use the drug to quit. In a pilot study of the idea, over 90 percent of the 50 young people Evins and her colleagues enrolled abstained. The next step will be to recruit a larger group and see if abstaining from use changes cognitive measures.

Such studies are important now that marijuana is legal for recreational or medical use in many states across the country. Sellers promote its safety, but the drug may be harmful and addictive, especially for young people. “We’ve got a growing for-profit industry that’s looking to encourage use of cannabis products, particularly among young people,” says Evins. “There’s a reduced perception of harm and increased prevalence of use.”

To design new studies of cannabis use, Evins draws on her previous work studying tobacco smoking in relation to schizophrenia. One longstanding question is whether nicotine helps patients manage the disorder. Years ago, Evins tried to answer this question by asking people with schizophrenia to quit smoking and then evaluating whether or not their symptoms got worse.

It turned out to be extremely difficult to get smokers with schizophrenia to quit smoking. Only about four percent succeeded without medications to help them quit, says Evins. Many were able to quit with a smoking cessation medication, which was great news for treatment of nicotine dependence in schizophrenia but complicated the research. “Essentially the study participants are still using a medication that acts in the brain like nicotine, so it is difficult to tell if that substitute also affects symptoms of schizophrenia,” she says.

Studies of abstention could help answer questions about the effects of marijuana, however. Marijuana users did, for the most part, stop using while receiving payments. The team verified that the participants quit by checking urine samples for markers of cannabis use.

The study also suggests that marijuana may be more addictive than perceived by the public. Almost all of the participants in the study said that they felt better while abstaining. They also said they weren’t going to go back to using the drug, or if they did, they’d use less. Yet within two weeks of the end of the study, they were back to using at the same levels as before, according to urine test results. “They resume use almost immediately once you stop paying them,” says Evins.

The team is now recruiting about 200 young people to participate in a larger study of the effects of abstinence. In the new study, Evins and Mass General neuropsychologist Randi Schuster, PhD, who recently was awarded a mentored training grant from the National Institute on Drug Abuse, will perform cognitive tests similar to those they’ve used in the past.

For instance, in an earlier study that compared cannabis users to non-users, Evins had participants learn a list of 16 words and then recall the words later. She found that those who had started using marijuana before age 16 not only had poorer recall, they also had a harder time learning the words in the first place. Older adopters of cannabis and non-users had less difficulty with the task.

The new study will determine whether cognitive performance improves after abstinence for a month. If performance does improve, it will suggest that cannabis impaired performance. If not, however, the question remains open. “There could be no effect, or it could be that the effect of use on cognition isn’t reversible in a month,” says Evins. “It’s not a perfect study, but it’s an innovative design.”

Evins is also looking for innovative ways to study medical marijuana. According to a recent report from the National Academies of Science, evidence supports the use of medical marijuana for nausea related to chemotherapy use and spasticity caused by multiple sclerosis, but no others. Yet, says Evins, “the momentum is in favor of cannabis use for all sorts of indications for which there is no evidence for efficacy and in some cases evidence for harm.”

One such condition is depression. To determine if marijuana use helps or harms people with depression, Evins, with colleague neuroscientist Jodi Gilman, PhD, is recruiting a group of people with this diagnosis who intend to use medical marijuana. They will compare two groups: one that begins use immediately, and another that delays use for three months. They plan to study the effects of the drug on psychiatric symptoms, addictive behavior and also on brain activity using brain scans. “What we’re trying to do is understand the risk/benefit ratio of medical marijuana for particular illnesses and, where there is evidence for harm, to prevent it,” she says. “People are turning to cannabis because they think it will help them, but it could unwittingly make some disorders worse.”