CBD and Addiction Research

This article is for information only and is not medical advice. Talk to a doctor before using cannabis for a health condition, and do not stop or change prescribed treatment on your own.

Plain-language summary

Cannabidiol (CBD) is a non-intoxicating compound in cannabis that researchers have begun studying as a possible aid for addiction. The interest stems from preclinical work suggesting CBD may dampen anxiety and reduce responses to stress and drug-related cues—and stress is a major trigger for craving in recovery ^[1].

The strongest human result comes from a 2019 double-blind trial by Hurd and colleagues: a single oral dose of 400–800 mg CBD reduced cue-induced craving and anxiety in 42 people with heroin use disorder, with effects lasting up to one week Weak / limited ^[2]. This is a well-designed study, but it is one trial measuring craving rather than long-term abstinence or overdose prevention.

For most other drugs (alcohol, cocaine, methamphetamine, nicotine), the human evidence is much smaller, older, or entirely absent. CBD is not an FDA-approved addiction treatment anywhere. The only FDA-approved CBD product is Epidiolex, for rare forms of epilepsy ^[3].

⚠️ This article is not medical advice. Addiction is a serious, treatable medical condition. If you or someone you know is struggling, contact a qualified clinician or in the US, SAMHSA's National Helpline: 1-800-662-HELP (4357).

What probably works (relatively speaking)

Honestly: nothing about CBD for addiction has yet reached the standard of evidence supporting established treatments like buprenorphine or naltrexone for opioid use disorder.

The strongest single finding is the Hurd et al. (2019) trial: in abstinent individuals with heroin use disorder, a single 400 mg or 800 mg dose of CBD reduced both cue-induced craving and anxiety, with effects persisting at one-week follow-up Weak / limited ^[2]. The trial was double-blind, placebo-controlled, and well-conducted.

Why "weak" and not "strong"?

• Small sample (n = 42)
• Primary outcome was subjective craving, not sustained abstinence, reduced overdose, or return to use
• No independent replication at scale to date
• Participants were already abstinent and not receiving agonist therapy (methadone or buprenorphine), which is how most people with opioid use disorder are actually treated
• A single positive trial does not establish clinical efficacy

A promising signal justifies further investigation. It does not yet justify clinical adoption.

What might work

Cannabis use disorder (CUD). Freeman et al. (2020) conducted a Phase 2a randomized trial of 400 mg and 800 mg CBD daily in people attempting to quit cannabis. Both doses reduced urinary THC-COOH levels and increased days of complete abstinence compared to placebo, while 200 mg did not differ from placebo Weak / limited ^[4]. The finding is noteworthy but based on a single trial with moderate sample size, and the clinical significance of urinary THC reduction without long-term abstinence data remains unclear.

Tobacco/nicotine. Morgan et al. (2013) conducted a small pilot study in dependent smokers given ad-hoc inhaled CBD, which reduced cigarette consumption by approximately 40% over one week Weak / limited ^[5]. A follow-up trial measuring attentional bias to smoking cues showed mixed results ^[6]. The evidence is suggestive but underpowered; no adequately sized confirmatory trial has been published.

Alcohol use disorder. As of October 2026, no human randomized controlled trials of CBD in alcohol use disorder have been published. Preclinical rodent studies show CBD reduces self-administration of alcohol and may mitigate alcohol-induced liver and brain injury Weak / limited ^[7]. However, preclinical drug effects in rodents have a poor track record of translation to human alcohol use disorder treatment.

Anxiety and sleep as indirect targets. Some clinicians theorize that if CBD reduces anxiety or improves sleep, it might indirectly support recovery and reduce relapse risk. Acute anxiety reduction has been shown in some experimental human paradigms Weak / limited. However, the leap from "reduces anxiety in a lab" to "helps maintain long-term abstinence" lacks direct evidence and represents a substantial inference.

What doesn't work or lacks evidence

Over-the-counter CBD at consumer doses (10–50 mg/day). No clinical addiction evidence exists for these doses. All published trials showing any effect used 400–800 mg of pharmaceutical-grade CBD No data.

CBD for stimulant use disorders (cocaine, methamphetamine). No adequately powered human trials support benefit No data ^[8].

CBD as a stand-alone opioid replacement or alternative. There is no evidence that CBD substitutes for or approaches the efficacy of methadone, buprenorphine, or extended-release naltrexone No data ^[9]. These medications reduce all-cause mortality by 50% or more; CBD has not been shown to reduce overdose death or mortality in any population.

"Full-spectrum" or hemp-derived CBD tinctures for addiction. The concentration, purity, and absorption of these products vary widely. None of the addiction trials used consumer-market products. Independent testing of retail CBD products frequently finds mislabeling, unlabeled THC, and contaminants—potentially harmful for anyone in recovery or subject to drug testing No data ^[10].

Marketing claims that CBD is a solution for addiction. The claim that CBD is "non-addictive" is largely accurate. The claim that it treats addiction is running far ahead of the evidence.

What we don't know

• Long-term clinical outcomes. Whether CBD's short-term reduction in craving translates to sustained abstinence, reduced overdose, hospitalization, or improved functioning.

• Optimal dose, duration, and formulation. Oral CBD bioavailability is low and variable (6–19%), and food intake dramatically affects absorption ^[11].

• Efficacy in typical treatment settings. The Hurd trial tested CBD in already-abstinent individuals not on agonist therapy. Whether effects persist or matter in people on methadone or buprenorphine—which is how most opioid use disorder is actually managed—is unknown.

• Long-term safety at 400–800 mg/day. Elevations in liver enzymes have been documented in Epidiolex (epilepsy) trials, particularly when combined with valproate ^[3]. The safety profile at addiction trial doses outside epilepsy populations is not well-characterized.

• Drug–drug interactions. CBD inhibits several cytochrome P450 enzymes (CYP3A4, CYP2C19, and others) and can raise levels of benzodiazepines, some SSRIs, warfarin, and other medications ^[12]—many of which are prescribed in addiction medicine and recovery.

• Subpopulation differences. Whether response varies by genetics, sex, concurrent anxiety or PTSD, or other factors.

Comparison with standard treatments

Opioid use disorder. The standard of care is medication for opioid use disorder (MOUD)—buprenorphine, methadone, or extended-release naltrexone—combined with psychosocial support. These medications have decades of evidence and are associated with 50%+ reductions in all-cause mortality ^[9]. CBD has no comparable evidence and is not a substitute.

Tobacco use disorder. First-line treatments are varenicline, combination nicotine replacement therapy, and bupropion, all supported by robust randomized controlled trial evidence ^[13]. CBD has one small pilot study.

Alcohol use disorder. Naltrexone, acamprosate, and disulfiram are FDA-approved and supported by meta-analyses of randomized trials ^[14]. CBD has no published human randomized trial data.

Cannabis use disorder. No FDA-approved medication exists; behavioral therapies (cognitive-behavioral therapy, contingency management, motivational enhancement) are first-line. In this area alone, CBD's Phase 2a data is noteworthy simply because nothing else is approved—but "competitive with nothing" means the bar is unusually low.

If CBD has a plausible near-term role, it would likely be as an adjunct to standard evidence-based treatment (e.g., for anxiety or craving alongside MOUD), not as a standalone or replacement therapy.

Risks and interactions

CBD is generally well-tolerated but not without risks, particularly at the 400–800 mg/day doses used in addiction trials.

Liver enzyme elevations. Documented in Epidiolex trials at higher doses, especially when combined with valproate or other hepatotoxic drugs ^[3].

Drug–drug interactions. CBD inhibits CYP3A4, CYP2C19, and other enzymes, potentially raising levels of benzodiazepines, some selective serotonin reuptake inhibitors (SSRIs), warfarin, and other medications ^[12]. This is particularly relevant in addiction medicine, where polypharmacy is common. See Cannabis and Polypharmacy Risks for more detail.

Sedation and fatigue. Reported at 400+ mg doses.

Product quality and unlabeled THC. Independent testing repeatedly finds mislabeling, contamination, and undisclosed THC in retail CBD products ^[10]. For individuals in early recovery, on probation, or subject to workplace drug testing, unlabeled THC exposure is a serious concern.

Cost. Pharmaceutical-grade CBD at trial-equivalent doses (400–800 mg/day) is expensive and uninsured. Consumer products at these doses may cost hundreds of dollars per month, with no guarantee of consistency or purity.

Interaction with substance-specific treatments. CBD's effect on CYP450 metabolism could alter blood levels of buprenorphine, methadone, naltrexone, acamprosate, or psychiatric medications commonly prescribed in recovery. This has not been systematically studied.

Sources

How this page was made

Related

• Cannabis and Polypharmacy Risks — How cannabis affects multiple-medication regimens, and which interactions are real.