Cannabis and Alcohol Use: What a Randomized Trial Tells Us — and What It Doesn’t
A Clinical Perspective for PRC+ Healthcare Professionals
Cannabis and alcohol are among the most commonly co-used psychoactive substances in the United States, and many patients today experiment with cannabinoids as a way to moderate or replace their alcohol use. The culturally pervasive “California sober” trend — wherein individuals reduce alcohol in favor of cannabis — has lacked rigorous clinical investigation. A new randomized, placebo-controlled crossover trial published in The American Journal of Psychiatry offers the first experimental evidence addressing whether acute cannabis use changes subsequent alcohol craving and consumption.
Study Design at a Glance
Researchers led by Jane Metrik, PhD (Brown University), enrolled 157 adults (ages 21–44) who reported heavy alcohol use and used cannabis at least twice weekly. Across three lab visits, participants smoked cannabis cigarettes with either:
7.2% THC
3.1% THC
0.03% THC (placebo)
After cannabis exposure, participants completed an Alcohol Choice Task in a simulated bar setting, where they could drink freely or earn small cash amounts for declining drinks — a validated laboratory measure of real-time drinking behavior.
Key Results
Reduced alcohol consumption:
Participants smoked cannabis with THC prior to having access to alcohol. Compared to placebo, they consumed significantly less alcohol — about 19% less at 3.1% THC and 27% less at 7.2% THC.Lower immediate urge and delayed initiation:
After inhaling active cannabis, participants reported lower immediate urges to drink, and those in the higher THC condition took longer to take their first sip of alcohol.
What the Lead Investigator Says
Dr. Jane Metrik framed the findings in the context of the substitution hypothesis that many consumers have already embraced:
“What we found was consistent with this idea of the substitution effect popularized by the California sober trend,” she said. “Instead of seeing cannabis increase craving and drinking, we saw the opposite. Cannabis reduced the urge for alcohol in the moment, lowered how much alcohol people consumed over a two-hour period and even delayed when they started drinking once the alcohol was available.”
At the same time, Metrik cautions strongly against over-interpretation:
“We saw that cannabis reduces the urge in the moment. What we don’t know from this study is what the long-term effect is.”
And in clear clinical terms:
Clinical Context and Interpretation
This study’s strength lies in its controlled laboratory design and within-subject crossover methodology, enhancing causal inference. It demonstrates that, under tightly regulated conditions, acute cannabis exposure can reduce immediate motivational drivers of alcohol drinking in frequent cannabis users.
However, several important clinical caveats must govern interpretation:
1. Short-term, not long-term outcomes
The effects were observed only during a two-hour lab session. There is no evidence yet on lasting changes in drinking behavior or alcohol use disorder (AUD) outcomes.
2. Frequent cannabis users
Participants were regular cannabis users, many of whom may have developed tolerance and behavioral patterns distinct from occasional users — limiting generalizability.
3. THC-dominant cannabis only
The trial tested smoked cannabis with fixed THC potencies. Effects of other cannabinoids (e.g., CBD), edible forms, or high-potency products remain unknown.
4. Labeling vs. recommendation
Although reductions in alcohol consumption were observed, these findings do not justify recommending cannabis as a therapeutic tool for AUD or harmful drinking patterns — a point explicitly underscored by Metrik and colleagues.
Clinical Takeaways for Practice
Recognize substitution behaviors: Patients may report reduced alcohol use on cannabis use days. These behaviors may reflect real subjective and motivational shifts, but clinicians should contextualize them within broader patterns of substance use and risk.
Assess use patterns carefully: Ask when, how much, and why patients are using cannabis relative to alcohol. Simultaneous use — especially in social settings — may yield different outcomes and associated harms.
Prioritize evidence-based interventions for AUD: Established treatments (e.g., pharmacotherapy with naltrexone, psychosocial support models) remain the standard of care.
Monitor for cannabis use disorder: Cannabis itself carries risk for problematic use and addiction, particularly among individuals with polysubstance use histories.
How Does This Compare to Naltrexone? A Useful Clinical Contrast
For clinicians, the most relevant comparator to this cannabis study is not another recreational substance—but naltrexone, one of the most well-studied, FDA-approved medications for alcohol use disorder (AUD).
Naltrexone is an opioid receptor antagonist that reduces the rewarding effects of alcohol and is associated with reductions in heavy drinking days and alcohol craving. Despite strong evidence of safety and efficacy, it remains dramatically underused, prescribed to fewer than 2% of individuals with AUD in the United States Naltrexone for AUD. Recent commentary has even argued that oral naltrexone could be safely made available over the counter as a low-threshold harm-reduction tool, analogous to nicotine replacement therapy Naltrexone for AUD.
Mechanistic Parallels—and Key Differences
The contrast between THC and naltrexone is instructive:
Naltrexone works by blocking reward—blunting alcohol’s reinforcing effects at the level of the endogenous opioid system.
THC, in the Metrik et al. study, appeared to reduce alcohol consumption without consistently reducing craving, possibly by satiating intoxication or altering motivational salience through endocannabinoid signaling.
Both interventions resulted in less alcohol consumed when alcohol was readily available, but they do so via very different neurobiological routes.
Importantly, the magnitude of effect observed in the cannabis study—a ~20–27% reduction in alcohol consumption over a short window—falls within the same order of magnitude as reductions seen with targeted or as-needed naltrexone use in laboratory and real-world studies Naltrexone for AUD. However, similarity in effect size does not imply equivalence in clinical readiness.
Evidence Thresholds Matter
Naltrexone’s clinical status rests on:
Decades of randomized trials
Long-term outcome data
Known safety parameters (including liver safety, even in compensated cirrhosis)
Clear prescribing guidance
By contrast, the cannabis findings:
Are acute, not longitudinal
Reflect behavioral laboratory outcomes, not clinical endpoints
Involve a population with high baseline cannabis exposure
Do not address long-term harms, dependence risk, or functional outcomes
As Jane Metrik herself cautioned in discussing this work, these results do not support recommending cannabis as a therapeutic substitute for alcohol, particularly when FDA-approved options already exist.
A Clinical Framing That May Help
For clinicians navigating patient conversations, it may be useful to frame cannabis not as a treatment analog to naltrexone, but as a behavioral modifier whose short-term effects resemble what we expect from evidence-based AUD medications—less drinking without forced abstinence.
The key difference is that naltrexone was designed, tested, and regulated for this purpose, whereas cannabis was not.
Bottom Line for Practice
If a patient asks whether Cannabis “works like naltrexone,” the most accurate answer is:
Cannabis may reduce drinking in the moment, but it is not a medication, not standardized, and not proven safe or effective long-term.When patients are seeking moderation rather than abstinence, targeted naltrexone remains a far more defensible, evidence-based option.
The cannabis findings should sharpen—not replace—clinical conversations about why FDA-approved treatments for AUD remain underutilized, even as patients experiment with unregulated alternatives.
Looking Forward
This trial is an important step toward understanding acute behavioral interactions between cannabis and alcohol, but longer-term, ecologically valid research is essential before clinical practice guidelines can even consider incorporating cannabinoid-related strategies into AUD care.