Skip to Main Content


Tackling addiction with treatment and predictive outcomes modeling

Yale Medicine Magazine, 2019 - Spring


Researchers provide insights into the biological profiles of the underlying risk of substance use relapse, while treatment methods offer patients new options.

From the predictive to the prescriptive, advances in the science of addiction treatment are bringing us closer to delivering on the long-implicit promise that functional recovery from substance use disorders is possible. Mapping neural profiles to identify individuals at risk of addiction to certain drugs, investigating the underlying causes of relapse, and expanding the reach of successful behavior modification techniques all give patients with substance use disorders cause for hope.

Your own neurological map

Researchers in the lab of Kathleen Carroll, PhD, the Albert E. Kent Professor of Psychiatry, combined machine learning and neuroimaging to build a brain-behavior model that predicted cocaine abstinence among a group of 53 people who were entering treatment for cocaine use. The researchers applied this model to neuroimaging data from a group of 45 individuals who were also entering treatment, and found their prediction of who would abstain from cocaine during treatment to be 70 percent accurate. Led by Sarah Yip, PhD, FW ’15, assistant professor of psychiatry, and in the Yale Child Study Center; in collaboration with Dustin Scheinost, PhD ’13, FW ’15, assistant professor of radiology and biomedical imaging, and in the Yale Child Study Center, the team laid the groundwork for evaluating mechanisms that could guide the treatment of patients.

“We don’t have a good predictor of who will benefit from a course of treatment,” says Yip, “Using these data analysis approaches, we hope to one day be able to tailor treatment to individuals.”

Yip’s research points in the direction of a personalized neural map that may one day save patients valuable time, money, and resources—not to mention heartbreak. Published online in the American Journal of Psychiatry on January 4, 2019, the prediction of cocaine abstinence based on the connectome—a complete map of neural connections in the brain—also provides insight into the neurobiology of recovery, which may speed the development of novel interventions to further improve outcomes.

Rerouting the road to relapse

While researchers like Yip develop predictive tools to aid in addiction treatment, other clinicians and researchers are focusing on another problem area—the threat of relapse once a patient successfully completes a course of treatment. “Our targets are channels that influence dopamine activity in the brain,” says Nii Addy, PhD ’07, associate professor of psychiatry and of cellular and molecular physiology. “Certain cues can trigger drug relapse—think about the smoker who’s attempting to quit, and then returns to a bar to socialize with friends, only to find themselves smoking again. We’re using rodent models to understand what happens in the brain that allows cues to bring about that relapse. A great deal of research has shown that brain dopamine levels increase in the presence of drug cues. So channels that decrease brain dopamine concentrations could be potential therapeutic targets for drug relapse.”

Addy’s project, funded in part by a grant from the National Institutes of Health, focuses on L-type calcium channels in cells, which are channels that allow calcium to enter the cells. These channels affect how dopamine (the neurotransmitter most responsible for stimulating the reward system within the brain) acts neurologically. At present, scientists mostly understand the roles dopamine plays in motivating reward, regulating mood, and fostering learning. Figuring out when and how to regulate dopamine could allow scientists to develop treatments that short-circuit addictive behavior as well as lower the risk of relapse.

Addy’s interest in the neurological makeup of substance use disorder was sparked when he was an undergraduate at Duke University. He worked in a lab where researchers studied how nicotine could reverse the cognitive deficits associated with schizophrenia and antipsychotic medications.

One discovery that surprised Addy during his most recent research is that blocking the calcium channel decreased the frequency of relapse among subjects but increased dopamine levels. “That was the opposite of what we expected, given current understanding of how dopamine operates,” Addy says.

Reprogramming neural pathways

Carroll has spent nearly 30 years developing a program called Computer Based Training for Cognitive Behavioral Therapy, or CBT4CBT. The program takes a model proven to facilitate positive changes in a person’s actions—cognitive behavioral therapy—and offers it online to patients with a substance use disorder. Developed and tested with support from the National Institute on Drug Abuse and the National Institute on Alcohol Abuse and Alcoholism, CBT4CBT makes a standardized, effective form of treatment available to people wherever and whenever they need it; it is particularly promising in reaching underserved populations.

CBT4CBT started when Carroll saw the potential for technology-assisted treatment in the early 2000s. “My daughter loved learning to read while sitting on my lap using the Clifford the Big Red Dog program on the Public Broadcasting Service website,” says Carroll. “I realized that CBT could be taught and practiced remotely using those new multimedia tools. We think that by demonstrating cognitive and behavioral self-control skills and strategies in an engaging, fun way—for example, by including a series of videos that show people using the skills in real-world challenging situations—we’ve developed a very powerful tool. Here we are nearly 20 years later, and the technological opportunities for treatment have far exceeded early expectations.”

Asked how CBT4CBT compares to other forms of treatment for substance abuse disorders, Carroll is optimistic. “We spend billions in the U.S. on treatments that haven’t been proven or even shown not to work, like detoxifying people without follow-up,” she says. “We have tested CBT4CBT extensively in over seven randomized clinical trials, and have shown that it can have better outcomes than standard outpatient treatment while costing a lot less. Plus, patients really like it, which is the bottom line.”

There is still much to learn about how bodies respond to treatment for substance use disorders. Yip, Carroll, and Addy are all bringing medical science and the clinical application of that science closer to the day when substance use disorder is seen as a treatable chronic disorder that can be managed, enabling people to resume happy, productive lives.