Research in Context

Research in Context March 24, 2026

Treating addiction

Addiction is a serious public health problem in the United States. Nearly 80,000 people nationwide died of drug overdoses in 2024 alone. But drug overdoses are just one of the many harms caused by addiction and substance use disorders.

About 178,000 people in the United States die from excessive drinking each year. Alcohol use disorder increases the risk of unintentional injuries, car accidents, and suicide. It also contributes to cancer, heart disease, liver disease, birth defects, and developmental disabilities.

Cigarette smoking kills more than 480,000 Americans each year. Tobacco and nicotine addiction contribute to cancer, heart disease, respiratory diseases, and diabetes.

Intravenous drug use increases the spread of blood-borne diseases like HIV and hepatitis C.

Drug addiction is also associated with serious social harms. These include parental loss due to fatal drug overdoses and other forms of trauma. Dr. Nora Volkow, director of NIH's National Institute on Drug Abuse (NIDA), warns that substance use and substance use disorders can have a profound impact on individuals, families, and society as a whole.

To address addiction, NIH-funded researchers are working to develop better treatments, including medications, behavioral therapies, and integrated health services.

Drugs and the brain

"Addiction is a brain disorder," explains Dr. Lorenzo Leggio, an NIH clinical investigator and clinical director of NIDA. "It's not a moral failure. But that does not mean that non-medical, non-biological factors don't play a role."

Addiction is when you go from taking a drug for pleasure to taking it because you can't feel good without it. Alcohol, opioids, cannabinoids, nicotine, and stimulants are a few examples of drugs that when used can lead to addiction. Although these drugs act on the brain in different ways, they all cause bursts of the chemical dopamine in a brain region called the nucleus accumbens.

Dopamine bursts tell the brain that an activity is rewarding and worth repeating. This triggers brain rewiring that makes it easier to repeat that activity automatically. The brain links the reward to cues, such as sights, sounds, or situations. Encountering these cues will make someone want to do the activity to get the pleasure it causes. This process is called conditioning, and it helps create habits.

Some habits can be helpful or even vital for survival, such as eating when you're hungry. But drugs can produce much stronger dopamine surges than everyday rewards. The brain compensates by reducing activity in the reward circuit, like turning down the volume on a loud radio. This makes it harder to get pleasure from other activities. Over time, only the drug can provide a normal level of reward, and a person needs more of it to get the same effect. Meanwhile, drugs weaken the brain circuits responsible for self-control and stress management. This can make cravings stronger and relapse more likely.

Not everyone who uses addictive drugs will become addicted. Genetics and age play a role. So do many social and environmental factors, such as drug availability, family supervision, social network, and poverty. But no single factor can determine addiction risk.

Cognitive behavioral therapy

One way to treat substance use disorders is with behavioral treatments. These treatments aim to teach people to change harmful habits through psychotherapy or counseling. They may be used on their own but can also be combined with medication.

One effective behavioral treatment is called cognitive behavioral therapy (CBT). CBT is designed to change the thought processes that lead to unhealthy behaviors. In the case of alcohol use disorder (AUD), it might involve identifying feelings and situations that contribute to heavy drinking and developing skills to deal with those feelings and situations.

"Many times, as people start to develop problems, alcohol becomes their main way of coping with stressors of daily life, as well as other internal or external situations that come up," says Dr. Brian Kiluk of Yale School of Medicine. "And so the idea is to help people to build newer strategies and develop healthier ways of dealing with those stressors without having to turn to alcohol use."

Despite evidence showing that CBT and other behavioral treatments work, there may be challenges to getting the best care. Therapists may do CBT differently from the way it's done in clinical trials. Training therapists to follow standardized, research-backed protocols can be time-consuming and expensive.

To overcome these challenges, a research team led by Kiluk's colleague, the late Dr. Kathleen Carroll, developed a web-based CBT program called Computer-Based Training for Cognitive Behavioral Therapy (CBT4CBT). The program uses video, graphics, audio instruction, and interactive exercises to teach strategies for avoiding drug or alcohol use. It's not meant to replace in-person therapy but to complement it. By teaching CBT skills in a consistent and high-quality way, it could reduce some of the variability between therapists.

Kiluk led a clinical trial to test how effective CBT4CBT was for treating AUD. The team compared it with in-person CBT or standard (non-CBT) outpatient treatment. Ninety-nine participants were randomly assigned to one of the three treatments. Those assigned to receive CBT4CBT also had brief check-ins with a clinician each week. Treatment lasted for 8 weeks, and the team followed participants' alcohol use for 6 months after treatment.

During the 8 weeks of treatment, the percentage of alcohol-free days increased by a similar amount in all three treatment groups. But by the end of the follow-up period, the alcohol-free days for those in the CBT4CBT group increased by almost 75% compared to before treatment. The other treatment groups had a much smaller increase. This suggests that incorporating CBT4CBT into AUD treatment might help people more than standard treatment alone.

Kiluk was also part of an NIH-funded research team that reviewed 30 randomized controlled trials that tested the effectiveness of CBT for substance use disorders. CBT consistently yielded better results than no or minimal treatment. But it didn't necessarily yield better results than other evidence-based, behavioral treatments.

"CBT may work for some; it may be less effective for others," Kiluk explains. "I think there are a range of other types of therapies that people should consider if one treatment isn't working out for them. Just like if they were taking a medication for depression or another condition that has a lot of treatment options. If this one isn't working, we might try something else, rather than giving up on the endeavor altogether."

Medications for addiction

For certain drugs, particularly opioids, medications can help people stop their use. One way these work is by reducing the withdrawal symptoms that are triggered when drug use stops. Medications can also reduce cravings. Those approved for treating opioid use disorder include buprenorphine, methadone, and naltrexone.

There are also a few approved medications for tobacco use disorder and alcohol use disorder. But like many medical disorders, not everyone responds to the same medications. And overall, the number of medications approved for opioid, alcohol, or tobacco use disorder is small. Furthermore, for stimulant or cannabis use disorders, there are no approved medications. That's why researchers at NIH and elsewhere are searching for new medications to treat addictions.

One promising class of drugs is the glucagon-like peptide-1 receptor agonists (GLP-1RAs). These drugs are currently used to treat obesity and type 2 diabetes. Examples include semaglutide, liraglutide, and tirzepatide (sold under the brand names Ozempic, Victoza, and Mounjaro for diabetes and Wegovy, Saxenda, and Zepbound for weight loss).

One of the ways GLP-1RAs work is to reduce food cravings by adjusting the dopamine signals associated with food cues. Since similar dopamine signals also motivate the desire for drugs, it makes sense that GLP-1RAs might reduce drug cravings as well. Indeed, people who have been treated with GLP-1RAs have reported less desire to drink alcohol. But it is likely that the ways these medications may help with addictions are more complex.

"People will tell you, 'I used to have to have two or three drinks of wine every now and then,'" Volkow says, "and now they say, 'I drink wine, and I cannot even finish it. I don't need any more. I have enough.'"

An NIH research team led by Leggio and Dr. Leandro Vendruscolo studied the effects of semaglutide on alcohol consumption in rodents. They found that the drug reduced binge-like alcohol drinking behavior in mice. The team also tested semaglutide in alcohol-dependent rats and found that it reduced alcohol consumption in these rats as well.

To better understand the mechanisms behind semaglutide's effect on alcohol drinking, the team measured electrical signaling in rat brains. They found that semaglutide altered signaling in some of the brain regions that are affected by chronic alcohol exposure. But it wasn't clear exactly how the changes caused by semaglutide led to reduced drinking.

These and other results in animal models supported clinical testing of GLP-1RAs for treating AUD. Further supporting the use of GLP-1RAs for AUD was a study of electronic health records conducted by a team led by Volkow and Dr. Rong Xu at Case Western Reserve University. The researchers looked at records from more than 80,000 people who were prescribed semaglutide or a non-GLP-1RA medication for obesity.

During 12 months of treatment, semaglutide was associated with a 50% lower risk of AUD diagnosis in people with no prior history of AUD. In people who had a history of AUD, semaglutide was associated with a similarly large reduction in the risk of relapse. These results held regardless of gender, race, or age, and whether people had type 2 diabetes.

To verify the association between semaglutide and AUD, the researchers also looked at health records for almost 600,000 people with type 2 diabetes. In this group, semaglutide was also associated with a much lower risk of AUD in people with and without a history of AUD.

Another study led by Leggio and Vendruscolo looked at alcohol use and GLP-1RAs in a different set of health records spanning a longer period. Instead of AUD diagnoses, the team looked at self-reported measures of alcohol use frequency and quantity. Consistent with Xu and Volkow's results, people taking GLP-1RAs had greater reductions in alcohol use than those who did not. In contrast, no such decline in alcohol use was seen in those taking a different type of anti-diabetes drug known as DPP-4Is. These findings suggest that GLP-1RAs might help treat AUD in people under real-world conditions. Leggio and his team have shown similar associations between GLP-1RAs and reductions in drinking in other health record systems.

Volkow and Xu's team also looked at semaglutide and other drug use. Semaglutide was associated with about a 50% lower risk of developing cannabis use disorder. The risk of relapses in people with a history of cannabis use disorder was 34-38% lower among people taking semaglutide.

The drug seemed to help with smoking, too. People taking semaglutide had fewer medical encounters related to tobacco use disorder, fewer prescriptions for smoking cessation medications, and less smoking cessation counseling. And among people with opioid use disorder, those taking semaglutide had a lower risk of opioid overdose. Thus, semaglutide might be effective at treating many substance use disorders.

"That's the way that science works, right?" Volkow says. "One thing leads to another and triggers something else, and then it's almost like it takes on a current of its own and it moves the field. And I think that's where we are right now with our understanding of GLP-1RAs."

Testing medications

Based on these and other results, researchers have launched several randomized clinical trials to test the effect of semaglutide on AUD. In one NIH-funded trial led by Dr. Christian Hendershot at the University of Southern California, 48 people with AUD were randomly assigned to receive increasing doses of either semaglutide or an inactive placebo each week.

After 8 weeks of treatment, the researchers tested the participants' alcohol consumption. Participants were allowed to drink their preferred alcoholic beverage at their own pace for two hours. Those treated with semaglutide drank significantly less during these sessions than those treated with a placebo.

The people treated with semaglutide also reported having fewer drinks per drinking day, lower weekly alcohol cravings, and fewer heavy drinking days. But the overall number of drinks per calendar day and the number of drinking days did not differ significantly between the two groups. Notably, among participants who smoked, those treated with semaglutide also reported a greater decrease in cigarettes smoked per day. Although this trial was small, the results justify larger clinical trials of semaglutide and other GLP-1RAs for treating AUD. Several clinical trials are ongoing, including one at NIDA led by Leggio and his team.

"The data we have are very promising," Leggio says of these new medications. "And it's actually not that common to see a pattern of scientific data across the board-different species, different labs, different databases, different models, different countries-where you consistently keep on seeing the same signal. So that gives me a lot of hope."

-by Brian Doctrow, Ph.D.

Related Links

• Scientists find common brain network for substance use disorders
• Telehealth improves treatment for opioid use disorder
• Treatment for opioid use disorder in jail reduces risk of return
• Opioid overdose deaths rise among Black Americans
• High-dose buprenorphine for opioid withdrawal
• Few are prescribed medications to treat alcohol problems
• Dopamine modification on chromatin linked to addiction
• Alcohol-related deaths increasing nationwide
• How opioid drugs activate receptors
• Dealing with drug problems: preventing and treating drug abuse
• Understanding alcohol use disorder
• Alcohol treatment navigator
• Treatment for alcohol problems: finding and getting help
• Psychotherapies
• Drugs, brains, and behavior: the science of addiction
• Facts about U.S. deaths from excessive alcohol use
• Semaglutide therapy for alcohol reduction (STAR)

References

A meta-analysis of cognitive-behavioral therapy for alcohol or other drug use disorders: Treatment efficacy by contrast condition. Magill M, Ray L, Kiluk B, Hoadley A, Bernstein M, Tonigan JS, Carroll K.J Consult Clin Psychol. 2019 Dec;87(12):1093-1105. doi: 10.1037/ccp0000447. Epub 2019 Oct 10. PMID: 31599606.

A digital cognitive behavioral therapy program for adults with alcohol use disorder: A randomized clinical trial. Kiluk BD, Benitez B, DeVito EE, Frankforter TL, LaPaglia DM, O'Malley SS, Nich C. JAMA Netw Open. 2024 Sep 3;7(9):e2435205. doi: 10.1001/jamanetworkopen.2024.35205. PMID: 39325452.

The glucagon-like peptide-1 (GLP-1) analogue semaglutide reduces alcohol drinking and modulates central GABA neurotransmission. Chuong V, Farokhnia M, Khom S, Pince CL, Elvig SK, Vlkolinsky R, Marchette RC, Koob GF, Roberto M, Vendruscolo LF, Leggio L. JCI Insight. 2023 Jun 22;8(12):e170671. doi: 10.1172/jci.insight.170671. PMID: 37192005.

Associations of semaglutide with incidence and recurrence of alcohol use disorder in real-world population. Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Nat Commun. 2024 May 28;15(1):4548. doi: 10.1038/s41467-024-48780-6. PMID: 38806481.

Association of semaglutide with reduced incidence and relapse of cannabis use disorder in real-world populations: a retrospective cohort study. Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Mol Psychiatry. 2024 Aug;29(8):2587-2598. doi: 10.1038/s41380-024-02498-5. Epub 2024 Mar 14. PMID: 38486046.

Association of semaglutide with tobacco use disorder in patients with type 2 diabetes : Target trial emulation using real-world data. Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Ann Intern Med. 2024 Aug;177(8):1016-1027. doi: 10.7326/M23-2718. Epub 2024 Jul 30. PMID: 39074369.

Glucagon-like peptide-1 receptor agonists, but not dipeptidyl peptidase-4 inhibitors, reduce alcohol intake. Farokhnia M, Tazare J, Pince CL, Bruns N 6th, Gray JC, Lo Re V 3rd, Fiellin DA, Kranzler HR, Koob GF, Justice AC, Vendruscolo LF, Rentsch CT, Leggio L. J Clin Invest. 2025 Mar 6;135(9):e188314. doi: 10.1172/JCI188314. eCollection 2025 May 1. PMID: 40048376.

Once-weekly semaglutide in adults with alcohol use disorder: A randomized clinical trial. Hendershot CS, Bremmer MP, Paladino MB, Kostantinis G, Gilmore TA, Sullivan NR, Tow AC, Dermody SS, Prince MA, Jordan R, McKee SA, Fletcher PJ, Claus ED, Klein KR. JAMA Psychiatry. 2025 Apr 1;82(4):395-405. doi: 10.1001/jamapsychiatry.2024.4789. PMID: 39937469.