You have tried to stop before, more than once. A month off that lasted eleven days. A promise made on a Sunday morning that held until Thursday. Each attempt ran on willpower, and each time it ran out, the lesson seemed to be that you needed more of it. Nobody mentioned that there are medicines for this.
Three medications are approved by the FDA for alcohol use disorder: naltrexone, acamprosate, and disulfiram, still widely known by its old brand name, Antabuse. NIAAA describes all three as nonaddictive. Yet in 2024, only 2.4% of US adults with past-year alcohol use disorder received medication for it, according to NIAAA's figures on alcohol treatment in the United States.
Below is what each one does, how it did in trials, and who should avoid it. Whether a medicine suits you, and at what dose, is a prescriber's call.
What these medicines do, and what they do not
Naltrexone blunts the rewarding effect of alcohol and can reduce craving. Acamprosate helps people who have already stopped drinking to stay stopped. Disulfiram makes drinking physically unpleasant: a deterrent, not a treatment for the urge. Our article on the science of alcohol cravings explains what that urge involves.
None of them is a cure, and their effects in trials are modest. A 2014 JAMA review of 122 randomized trials expressed the benefit as a number needed to treat: how many people must take a medicine instead of a placebo for one more person to avoid a return to drinking. For naltrexone and acamprosate, the figures ranged from 12 to 20. That is small for any one person, but 27.1 million US adults had alcohol use disorder in 2024, and at that scale it adds up.
They also do not treat withdrawal. If you drink heavily every day, stopping suddenly can cause dangerous withdrawal, including seizures and delirium tremens. It is managed medically, usually with benzodiazepines. If that describes your drinking, stopping safely is the first conversation. Our alcohol withdrawal timeline explains the warning signs.
Naltrexone: blunting the reward
Naltrexone blocks opioid receptors in the brain that are involved in the reward from drinking and in craving. Unlike disulfiram, it does not cause a reaction if you drink.
It comes as a daily 50 mg pill or as Vivitrol, a 380 mg injection given every four weeks. The Vivitrol label warns that injection-site reactions can be severe, and some have needed surgery.
In the 2014 review, oral naltrexone had a number needed to treat of 20 to prevent any return to drinking and 12 to prevent a return to heavy drinking, so its clearer effect was on heavy drinking. The evidence for the injection was weaker: a small reduction in heavy drinking days, but no difference in how many people returned to drinking.
Opioids are the main safety issue. Naltrexone must not be used by anyone taking opioids, including prescription painkillers and tramadol, or by anyone in opioid withdrawal, because it can trigger sudden, severe withdrawal. The naltrexone label calls for at least 7 to 10 days without short-acting opioids before starting; SAMHSA advises 10 to 14 days after long-acting ones. Naltrexone also blocks opioid pain relief, which matters after an injury or before surgery. Stopping it carries its own risk: opioid tolerance drops, the label warns that a previously tolerated dose can then be life-threatening, and fatal overdoses have been reported after treatment ended.
The most common side effects are nausea, headache, dizziness, and fatigue. A boxed warning about liver damage was removed from the label in 2013, but a liver warning remains, and the American Psychiatric Association's 2018 guideline advises against naltrexone for people with acute hepatitis or liver failure.
Timing is a prescriber's decision. NIAAA's guidance for clinicians says naltrexone can be started while someone is still drinking; the Vivitrol label is written for people who can stop before the first injection.
Acamprosate: help with staying stopped
Acamprosate is for people who have already stopped drinking. NIAAA describes it as easing the unpleasant symptoms some people feel during abstinence. The label says to start it as soon as possible after withdrawal and to continue it if the person drinks again, so a slip is something to raise with the prescriber, not a reason to abandon the plan.
It is taken three times a day; the label dose is two 333 mg tablets each time. The liver does not metabolize it and the kidneys clear it unchanged, so the dose is reduced in moderate kidney impairment, and it is not used when creatinine clearance is 30 mL/min or less. Diarrhea is the most common side effect.
The evidence is mixed rather than negative. In the 2014 review, acamprosate had a number needed to treat of 12 to prevent any return to drinking, and head-to-head trials found no significant difference from naltrexone. But in the US COMBINE trial of 1,383 people who had recently stopped drinking, acamprosate did no better than placebo, alone or in any combination.
Disulfiram (Antabuse): a deterrent you agree to
Disulfiram blocks the breakdown of alcohol at the acetaldehyde stage. Drink while taking it, and acetaldehyde builds up, causing flushing, a throbbing headache, nausea and vomiting, palpitations, and low blood pressure. Severe reactions can be life-threatening. Antabuse, the original brand, is no longer sold in the US, according to MedlinePlus, but generic disulfiram may be available.
The disulfiram label warns that a reaction can occur up to 14 days after the last dose and that hidden alcohol counts, naming sauces, vinegars, cough mixtures, and even aftershave lotions. Treatment should not start until at least 12 hours after the last drink, and the boxed warning says disulfiram must never be given to someone who is intoxicated or without their full knowledge. Serious liver injury is uncommon but has been reported, including liver failure that led to a transplant or death, and the label suggests liver tests before and during treatment.
A deterrent only works if you take it. A 2014 meta-analysis of 22 trials, published in PLoS One, found a benefit in open-label trials, where people knew they were taking disulfiram, and none in blinded trials, where the threat of a reaction applied to every group. That makes it a fit for people aiming for complete abstinence who want an external commitment, such as having each dose supervised by a clinic or by someone they trust, and a poor fit for cutting down.
Other medicines a doctor may mention
The American Psychiatric Association's guideline recommends offering naltrexone or acamprosate first to people with moderate to severe alcohol use disorder. For people who prefer another option, or who did not tolerate or respond to those two, it suggests disulfiram (for people aiming for abstinence), topiramate, or gabapentin, all with the same, lower-strength rating. Neither topiramate nor gabapentin is FDA-approved for alcohol use disorder.
Approvals also differ by country. In the European Union, nalmefene (Selincro) has been authorized since 2013 to help reduce drinking in adults with alcohol dependence who drink at a high-risk level, according to the European Medicines Agency's record for Selincro. Brand names and availability of the other medicines vary too, so ask what is approved where you live.
Semaglutide and other GLP-1 drugs, better known for diabetes and weight loss, are also being studied. In a 9-week trial of 48 people, semaglutide reduced drinks per drinking day and craving, but not the number of drinking days. In a 26-week trial of 108 people with alcohol use disorder and obesity, published in The Lancet in 2026, it cut heavy drinking days more than placebo, although the placebo group, which also received cognitive behavioral therapy, improved substantially as well. A third trial missed its primary outcome. None of these drugs is approved for alcohol use disorder.
The Sinclair Method (targeted naltrexone)
The Sinclair Method uses naltrexone on a different schedule. Instead of a daily pill, a person takes it only on days when they expect to drink, before drinking, aiming to drink less over time. In a 2001 paper, the researcher J. D. Sinclair argued that drinking while naltrexone blocks the reward gradually weakens the learned response, a process called extinction, and proposed that treatment continue indefinitely.
The trials behind it are small and short, with modest and mixed results. In a 2001 trial of 121 people who had not stopped drinking, naltrexone beat placebo only in the groups that also received coping-skills therapy. In a 2003 trial of 153 early problem drinkers, taking pills only in high-risk situations reduced drinking whether the pills were naltrexone or placebo. In a 2009 trial of 163 people, targeted naltrexone improved the main outcome, drinks per day, only in men.
The method is also off-label, since the label dose for alcohol dependence is 50 mg once daily. And it is a reduction approach: the aim is fewer and lighter drinking days, not the unbroken run of alcohol-free days a sobriety counter measures. Both goals are legitimate, but they are not the same.
Why so few people get these medicines, and how to ask
The 2.4% figure is not explained by a prescribing restriction. SAMHSA's naltrexone page notes that naltrexone can be prescribed by any practitioner licensed to prescribe medications. You can start with the doctor you already see. Bring the details that decide which medicine is safe for you:
- How much and how often you drink, honestly, and any shakes, sweats, or seizures after past attempts to stop.
- Any opioid use, including painkillers, tramadol, codeine cough medicine, methadone, or buprenorphine. This alone can rule naltrexone out.
- Liver and kidney history, including hepatitis, cirrhosis, kidney disease, and recent blood tests.
- Pregnancy, if you are pregnant, planning to be, or breastfeeding.
- Your goal: stopping completely or cutting down. Disulfiram only makes sense for the first, acamprosate is for people who have already stopped, and naltrexone has been studied for both.
How long naltrexone takes to work has no single answer. The trials in the 2014 review lasted at least 12 weeks, so ask how long your prescriber wants to try a medicine before judging it, and what would count as success.
These medicines tend to be used as part of a plan with some support: counseling, a mutual-help group, or check-ins with the prescriber. In COMBINE, naltrexone with medical management (meetings with a health professional) did well, but no combination beat naltrexone alone or the behavioral therapy alone, and every group, including the one on placebo pills, cut its drinking substantially. Our comparison of SMART Recovery and AA looks at two common sources of peer support.
Where Sober Tracker fits
Sober Tracker keeps your alcohol-free days visible while you follow the plan you and your clinician agree on. It does not track medication or doses, and it is not a substitute for medical care: a day count cannot tell you whether a medicine is working or whether it is safe for you. If you are still putting that plan together, our complete guide to quitting drinking covers the other pieces.
The bottom line
Three medicines are approved for alcohol use disorder, and very few people who have it receive one. The effects are modest, which argues for realistic expectations, not for skipping the conversation. Take your drinking pattern, opioid use, liver and kidney history, and goal to a prescriber, and ask which option fits.
This article is for general information and does not replace individualized medical advice.
