What GLP-1 Research Means for Addiction Care

Research on GLP-1 medications is raising a new question for addiction care. Could drugs developed for diabetes and obesity also help reduce alcohol cravings and consumption?

Randomized trials now provide encouraging evidence. They also raise practical questions about who might benefit, how to monitor treatment, and how medication fits alongside behavioral care.

For providers, the opportunity extends beyond reducing a craving score. It includes helping patients recognize triggers, practice effective responses, and build progress that carries over into daily life.

What the alcohol research actually shows

GLP-1 receptor agonists influence appetite and affect reward-related pathways. Researchers are investigating whether these effects can also alter alcohol consumption and craving. The clinical findings are promising, but their interpretation depends on the population, study design, and outcome measured.

A 2025 randomized trial enrolled 48 adults with alcohol use disorder who were not seeking treatment. Over nine weeks, low-dose semaglutide reduced alcohol consumed during a laboratory drinking task, weekly craving, and drinks per drinking day compared with placebo. It did not significantly reduce the number of drinking days or average drinks per calendar day. That distinction matters: improvement in some drinking measures does not imply improvement across all measures.

A 2026 randomized trial followed 108 treatment-seeking adults with alcohol use disorder and obesity for 26 weeks. Participants received semaglutide or placebo, and both groups received standard cognitive behavioral therapy. Both groups reduced heavy drinking, but the semaglutide group improved more. The estimated treatment difference in reducing heavy-drinking days was 13.7 percentage points.

The second study strengthens the case for further investigation. It also provides a more useful frame for the discussion: medication was evaluated within a treatment program that included behavioral care.

It does not establish that semaglutide works equally well for people without obesity, that benefits persist after treatment ends, or that a specific habit-change program enhances its effects. Alcohol findings also should not be generalized to opioids, stimulants, or all medications in the GLP-1 class.

Patient selection and tolerability remain central

Semaglutide use for alcohol use disorder remains off-label in the United States. Its prescribing information includes gastrointestinal adverse effects, potentially serious risks, and contraindications that warrant individualized assessment.

In the 2026 alcohol trial, gastrointestinal adverse events were more frequent with semaglutide. Providers evaluating this emerging evidence should consider both potential benefits and whether a patient can tolerate treatment.

Established addiction medications also belong in the conversation. For opioid use disorder, methadone and buprenorphine have substantial evidence of effectiveness. Interest in GLP-1s should expand research without displacing treatments already shown to help.

Habit change addresses a different part of care

Alongside medication research, behavioral approaches examine how cues, actions, and expected rewards reinforce each other.

A habit-loop framework makes that sequence easier to analyze:

• Trigger: What happened right before the urge?
• Behavior: What did the person do next?
• Reward: What relief or benefit did they expect, and what actually happened?

Research on an anxiety-focused mindfulness program describes teaching people to recognize these patterns, notice their physical and emotional experiences, and explore alternative responses. That work helps explain the framework. It should not be treated as proof that an anxiety-focused program treats substance use disorders.

For example, a clinician might help a patient examine a familiar sequence. Conflict at home or work, escalating tension, an urge to drink, and an expectation of relief. The work can then become more specific. What did the patient notice? What response was available? What happened when they tried it?

This is an illustrative care-planning example, not a reported patient outcome. Its value lies in making the next step concrete enough to practice and revisit.

Addiction is more complex than a habit loop. The framework can help organize part of the work without explaining away withdrawal, co-occurring conditions, or the circumstances surrounding substance use.

Behavioral promise deserves the same scrutiny

Mindfulness-based approaches have their own evidence base and limitations.

A 2021 Cochrane review identified 40 randomized trials on substance use disorders. Compared with other treatments, mindfulness-based interventions may slightly reduce the number of days of substance use. Evidence remained uncertain for several other outcomes, including abstinence and craving intensity.

That supports careful evaluation rather than broad claims that mindfulness prevents relapse or that any app teaching awareness yields clinical benefit.

The same standard applies to the proposed connection with GLP-1s. Reduced craving could make behavioral practice more manageable for some patients. Demonstrating that a specific combination improves outcomes requires studies designed to test that combination.

Make progress specific enough to follow

Provider conversations bring this discussion back to daily care.

One addiction-treatment team described measurable coping goals as an important part of treatment planning. This includes identifying skills a patient can use in difficult moments, setting a timeframe, and reviewing what changed. The team also emphasized that progress notes should link the intervention, the patient’s response, and the treatment plan goal.

Another provider raised concerns about outdated treatment-plan language, missing intervention explanations, and follow-ups that didn't carry over from one session to the next.

These observations suggest a practical priority for programs evaluating new treatments: define what improvement would look like for the individual, then ensure follow-up is useful.

For a patient receiving medication and behavioral support, this could include three questions:

1. What is changing? Track cravings and substance use separately, alongside patient-defined goals and treatment tolerability.
2. What is the patient practicing? Identify the trigger or situation, the response being tried, and whether it was usable.
3. What needs to happen next? Record remaining difficulties, plan adjustments, and who will follow up.

A lower craving score is useful information. By itself, it does not establish improved functioning or sustained recovery. Likewise, attending a session or completing a digital module does not show whether a patient can apply a skill when needed.

Design support around the moment it will be used

Between-session support should give patients something manageable to practice and clinicians something meaningful to revisit.

A brief exercise tied to a known trigger offers a clearer task than a broad instruction to “use coping skills.” A follow-up conversation can explore whether the patient remembered it, could use it in that setting, and found it helpful. If it was impractical, adjust the plan.

Digital tools can support that process, but engagement measures require interpretation. Opening an app, completing content, practicing a skill, and improving a clinical outcome are distinct measures. Programs should determine which question each measure answers.

The emerging GLP-1 evidence makes this work even more relevant. As medication options evolve, care teams need ways to understand both changes in substance use and the patient’s experience between appointments.

The next phase of research should clarify who benefits from GLP-1s, for how long, and in which treatment approaches. Providers can strengthen care now by setting specific goals, reviewing actual responses, and following through on patients' remaining needs.