If you're researching naltrexone — for yourself or someone you care about — you're probably trying to answer a few basic questions: Does it actually work? What does it feel like to take it? Is it safe? And is it the right fit compared to other options? Those are exactly the right questions, and this page answers them directly.

Naltrexone is a medication that reduces the brain's reward response to alcohol. It doesn't make you sick if you drink, it doesn't require you to be sober before you start, and it comes in two forms: a daily pill and a once-monthly injection. It's FDA-approved, backed by decades of research, and dramatically underused — not because it doesn't work, but because the healthcare system hasn't made it easy to access. If your doctor hasn't brought it up, it's entirely reasonable to ask.

How does naltrexone actually work in the brain?

When you drink, alcohol triggers the release of natural opioid chemicals — like β-endorphin — in your brain's reward circuitry. Those chemicals bind to μ-opioid receptors, which sets off a dopamine surge. That surge is the "buzz" — the feeling that makes drinking reinforcing and worth repeating.

Naltrexone is a μ-opioid receptor antagonist. It occupies those receptors without activating them, so when alcohol triggers the release of endogenous opioids, they have nowhere to land. The reward signal is blunted. People who drink while taking naltrexone often describe alcohol as feeling less exciting, less pleasurable, less worth continuing.

A meta-analysis of 16 studies involving 686 participants quantified these effects: naltrexone significantly reduces alcohol craving (Hedges g = −0.252), reduces stimulation from alcohol, increases sedation, and increases negative mood during drinking [1]✓ Verified knowledgeRay et al. (2019) — Naltrexone effects subjective. These are modest but consistent effects — and they're mechanistically coherent. Block the opioid-mediated reward, and the pull toward drinking weakens.

Importantly, this mechanism operates across a range of severity — not just in people with a formal alcohol use disorder diagnosis, but also in heavy drinkers who haven't crossed that threshold [1]✓ Verified knowledgeRay et al. (2019) — Naltrexone effects subjective. Naltrexone doesn't change the taste of alcohol or cause illness when combined with it. It quietly reduces the neurological payoff.

What are the two forms of naltrexone, and how do they compare?

Naltrexone comes in two FDA-approved formulations for alcohol use disorder. Here's how they compare across the dimensions that matter most for your decision:

The ADOPT randomized controlled trial — which started naltrexone at hospital discharge in 248 people with AUD — found that both forms produced striking improvements. Heavy drinking days fell by 39.3 percentage points in the oral group and 46.9 percentage points in the injectable group over three months. The difference between the two was not statistically significant [2]✓ Verified knowledgeMagane et al. (2025) — Oral extended release. The trial's conclusion: patient preference and insurance coverage, not formulation superiority, should guide the choice [2]✓ Verified knowledgeMagane et al. (2025) — Oral extended release.

The adherence advantage of the injectable form

The clearest practical argument for [3]✓ Verified knowledgeVivitrol(/treatment/medication/vivitrol/) is adherence. A meta-analysis of 42,268 patients found treatment persistence significantly higher with the injectable form compared to oral naltrexone at both three months (OR = 2.24) and six months (OR = 1.59) [4]✓ Verified knowledgeNunez et al. (2026) — Comparative effectiveness naltrexone. In a VA cohort, long-acting injectable naltrexone was associated with a median time to relapse of 150.5 days versus 50.5 days for oral naltrexone [5]✓ Verified knowledgeFarabee et al. (2020) — Randomized comparison extended.

That persistence advantage is real — but it doesn't automatically translate into fewer hospitalizations or emergency visits. The same meta-analysis found no statistically significant differences between the two forms in all-cause ED visits or inpatient admissions [4]✓ Verified knowledgeNunez et al. (2026) — Comparative effectiveness naltrexone. Better adherence is meaningful, but it's not the whole picture.

The cost reality of the injectable form

Vivitrol's wholesale cost exceeds $1,000 per monthly injection, and prior authorization requirements from insurers create delays that oral naltrexone — far less expensive — doesn't face. For many people, the pill is the more accessible starting point. If cost or insurance is a barrier, that's a practical factor worth discussing openly with your prescriber, not something to work around in silence.

Do you have to stop drinking before you start naltrexone?

No — and this is one of the most important things to understand. The evidence does not support requiring abstinence before starting naltrexone.

The ADOPT trial initiated naltrexone at hospital discharge, not after a period of proven sobriety [2]✓ Verified knowledgeMagane et al. (2025) — Oral extended release. A meta-analysis of extended-release naltrexone across seven RCTs found that trials not requiring lead-in abstinence actually showed larger reductions in heavy drinking days [6]✓ Verified knowledgeMurphy et al. (2022) — Effect extended release. Waiting for abstinence before prescribing isn't evidence-based — and it may cost people the window of motivation that a hospitalization or a difficult moment creates.

If you're in alcohol detox or have just been hospitalized, that's actually an ideal time to start the conversation about naltrexone. The medication can begin working while you're still figuring out the rest of your recovery plan.

Who is most likely to benefit from naltrexone?

Naltrexone works across a broad range of people with alcohol dependence and AUD, but a few factors suggest who may respond especially well.

Family history of alcohol problems. People with a family history of AUD tend to show heightened reward sensitivity to alcohol — a stronger opioid-mediated dopamine response to drinking. Because naltrexone works by blocking that opioid-mediated reward, people with this background have more reward to block, and therefore more to gain from the medication. This is one of the few concrete examples of treatment-matching in AUD pharmacotherapy. Asking about family history isn't just a routine intake question — it's a clinically meaningful predictor of naltrexone response.

People who experience strong cravings or a pronounced "buzz." The mechanism makes intuitive sense here: if alcohol's rewarding effects are particularly strong for you, blunting those effects pharmacologically has more room to make a difference.

People with adherence challenges. If remembering a daily pill feels unrealistic given your current situation, the monthly injectable form removes that barrier entirely.

Adolescents with alcohol problems. Research shows naltrexone disrupts the loop between negative emotions and drinking in adolescent drinkers — during treatment, higher estimated blood alcohol concentration was associated with greater negative affect, which in turn predicted reduced subsequent drinking [7]✓ Verified knowledgeCarpenter et al. (2022) — Naltrexone moderates association. Clinical trial data in this age group are still limited, but the biological rationale is established.

What about genetics — the OPRM1 gene?

You may have read about a gene variant called OPRM1 A118G. Carriers of the G allele appear to have a μ-opioid receptor that responds more strongly to endogenous opioids, which theoretically makes them more responsive to naltrexone's blocking effect. Early research generated real excitement about genotype-guided prescribing.

The honest clinical position: the mechanistic rationale is compelling, but the clinical evidence is inconsistent. OPRM1 genotype-guided prescribing is not yet ready for routine use. You should not be denied naltrexone because of your genetic status, and your prescriber shouldn't wait for genetic testing before starting treatment.

What are the side effects?

Naltrexone is generally well-tolerated at standard doses. The most common side effects — occurring in more than 10% of patients in clinical trials — are:

• Nausea. Most prominent in the first week and typically diminishes with continued use. Taking the medication with food helps significantly.
• Headache. Usually mild and transient.
• Fatigue. Can be more noticeable early in treatment.

Less common side effects include insomnia, anxiety, and abdominal discomfort. With the injectable form, injection site reactions — pain, tenderness, and occasionally nodule formation — are common but usually mild and resolve on their own.

The liver toxicity concern, in context. Naltrexone carries a boxed warning for hepatotoxicity. This is important to understand accurately: the liver injury observed in studies occurred at doses of 300 mg/day or higher — six times the standard therapeutic dose. At the standard 50 mg oral dose or the 380 mg monthly injection, clinically significant liver injury is rare. Baseline liver function tests are reasonable, and periodic monitoring is appropriate, but this warning should not be a reason to withhold naltrexone from people who need it.

For people with existing liver disease from heavy drinking, the clinical calculus is straightforward: the risk of continued heavy drinking to a damaged liver almost always exceeds the risk of standard-dose naltrexone. When liver impairment is severe, [8]✓ Verified knowledgeacamprosate(/treatment/medication/acamprosate/) — which is renally cleared and doesn't require hepatic metabolism — is often preferred.

What if you also use opioids?

This is the most critical safety issue with naltrexone, and it requires honest communication with your care team.

Naltrexone precipitates acute opioid withdrawal in anyone who has opioids in their system. This isn't a typical drug interaction — it's a direct pharmacological consequence of blocking occupied opioid receptors. Withdrawal can begin within minutes and include agitation, sweating, vomiting, muscle pain, and cardiovascular stress.

Before starting naltrexone, every person must be assessed for opioid use:

• Short-acting opioids (heroin, oxycodone, hydrocodone): must be opioid-free for 7–10 days before starting.
• Long-acting opioids, especially methadone: must be opioid-free for 14 or more days.
• When there's any uncertainty: a naloxone challenge — a small test dose of naloxone administered and observed for withdrawal signs — can confirm it's safe to proceed.

Pain management also requires planning if you're on naltrexone. Because opioid receptors are blocked, standard opioid pain medications will be ineffective or require much higher doses. Non-opioid approaches — NSAIDs, acetaminophen, regional anesthesia, ketamine — should be the primary strategy. Carry documentation of your naltrexone use and tell every provider who treats you, including emergency physicians and surgeons.

What is the Sinclair Method?

The Sinclair Method is a different approach to using naltrexone. Instead of taking it daily, you take naltrexone approximately one hour before you plan to drink — and only then. The theory is pharmacological extinction: by blocking the opioid reward from each drinking episode, the medication gradually extinguishes the conditioned reinforcement that drives compulsive drinking. Over time, the desire to drink is expected to diminish through a learning-based process.

This approach has an evidence base primarily from European and Finnish research, and it's particularly relevant for people whose goal is controlled drinking rather than complete abstinence — a goal that remains more accepted in European clinical practice than in the U.S.

The honest assessment: the Sinclair Method is a legitimate, evidence-informed option for some people, particularly those with controlled-drinking goals who are unlikely to accept daily naltrexone or abstinence-focused treatment. The evidence quality is moderate — not as robust as the daily-dosing RCT literature — and more replication studies in diverse populations are needed. It should be discussed openly with a prescriber rather than dismissed.

How does naltrexone compare to other AUD medications?

Naltrexone is one of three FDA-approved medications for alcohol use disorder. Here's how they compare:

You can explore [8]✓ Verified knowledgeacamprosate(/treatment/medication/acamprosate/) and [9]✓ Verified knowledgedisulfiram(/treatment/medication/disulfiram/) in more detail on their respective pages. The broader medication treatment overview covers how these options fit into a full treatment plan.

Medication is most effective when combined with behavioral support. Alcohol rehab programs typically integrate pharmacotherapy with counseling, and peer support through programs like Alcoholics Anonymous can complement medication treatment for many people.

What about low-dose naltrexone?

Some people searching online encounter references to low-dose naltrexone (LDN) — typically 1.5 to 4.5 mg per day — sometimes described as an alternative approach to alcohol problems. It's worth being direct about what the evidence does and doesn't support here.

LDN is not recommended by the American Society of Addiction Medicine for AUD treatment. The doses studied in the randomized controlled trials that established naltrexone's effectiveness are 50 mg daily (oral) and 380 mg monthly (injectable). The receptor-blocking effects that reduce alcohol's rewarding properties have been established at these doses — not at sub-therapeutic LDN doses. Controlled clinical trial data supporting LDN specifically for AUD are absent from the published literature.

LDN has been explored experimentally for chronic pain, fibromyalgia, and certain autoimmune conditions — a different proposed mechanism in a different clinical context. Whether that mechanism has any relevance to alcohol craving reduction is speculative. The low-dose naltrexone page covers this topic in more detail if you want to understand the distinction fully.

How long should you take naltrexone?

Most clinical trials studied 12 to 26 weeks of treatment — a reflection of trial design constraints, not a clinical recommendation to stop at six months. Real-world practice increasingly supports longer treatment, often six to twelve months, and sometimes longer for people with high relapse risk or limited recovery support.

A meta-analysis found that trials lasting more than three months showed larger reductions in heavy drinking days than shorter trials [6]✓ Verified knowledgeMurphy et al. (2022) — Effect extended release, suggesting sustained treatment produces sustained benefit. There is no evidence that stopping naltrexone after a fixed period is better than continuing it in people who are doing well.

Decisions about when to stop should be individualized — based on the stability of your recovery, your support network, your history of relapse, and your own preferences. And if a relapse happens after stopping, restarting naltrexone is entirely appropriate. A relapse is a signal that continued support is needed, not a reason to abandon pharmacotherapy.

Why aren't more people getting naltrexone?

This is one of the most troubling facts in addiction medicine. In U.S. substance use treatment facilities, naltrexone was prescribed to only 0.49% of AUD admissions in 2000 — and by 2018, that figure had risen to just 1.64% [10]✓ Verified knowledgeCowan et al. (2025) — Emergency department initiated. In emergency departments, only 0.5% of treatment-eligible encounters resulted in a naltrexone prescription [11]✓ Verified knowledgeResearch et al. (2000) — Research refines alcoholism.

The data are specific about who gets it and who doesn't. Prescribing is more likely among people with private insurance, those seen in academic medical centers, and those referred through specialty addiction care [research-2000-research-refines-alcoholism, snyder-2024-pharmacist-driven-alcohol]. The people least likely to receive naltrexone are those with the fewest resources.

Critically, the data challenge the assumption that patients are the main barrier. Among people who received a naltrexone prescription in an ED setting, 45% actually filled it [11]✓ Verified knowledgeResearch et al. (2000) — Research refines alcoholism. A pilot study of ED-initiated oral naltrexone found the approach feasible and acceptable, with high patient satisfaction [10]✓ Verified knowledgeCowan et al. (2025) — Emergency department initiated. When providers offer naltrexone, a substantial proportion of people engage. The bottleneck is upstream — at the level of provider behavior, system workflows, and insurance coverage.

A pharmacist-driven screening protocol on an inpatient psychiatric unit illustrates what structured identification can accomplish: of 641 screened patients, 66 were identified as candidates for injectable naltrexone, and administration increased from 2 to 8 patients compared to the prior year [12]✓ Verified knowledgeSnyder et al. (2024) — Pharmacist driven alcohol. Small numbers, but a proof of concept that systematic protocols move the needle.

If your doctor hasn't mentioned naltrexone, it is entirely appropriate to ask about it directly.

Special situations worth knowing about

• Pregnancy. Naltrexone is Category C in pregnancy — animal studies have shown adverse effects, but adequate human studies are lacking. The risks of untreated AUD during pregnancy must be weighed against uncertain risks of naltrexone exposure. This requires individualized counseling and shared decision-making with your provider.

• Older adults. Standard dosing applies, but age-related changes in metabolism and fall risk warrant attention. Fatigue and dizziness — potential naltrexone side effects — can increase fall risk, and this should be part of the prescribing conversation.

• People with HIV. Research suggests naltrexone is feasible in HIV-positive individuals [13]✓ Verified knowledgeLeighty et al. (2019) — Treatment outcomes long, though the evidence base is limited and drug interactions with antiretroviral therapy should be reviewed with your prescriber.

• People experiencing homelessness. Naltrexone has been studied in harm-reduction contexts for people experiencing homelessness [14]✓ Verified knowledgeCollins et al. (2021) — Combining behavioral harm, where the combination of behavioral support and pharmacotherapy may be especially important given structural barriers to consistent medication access.

• Criminal justice settings. People involved in the justice system have disproportionately high rates of AUD and stand to benefit substantially from pharmacotherapy. When a program offers only one formulation — often the injectable form because it's observable — people have the right to know what other evidence-based options exist and why alternatives aren't available in that setting. Informed consent means transparency about choices, not compliance under pressure.

One important note about urine drug screens

Oral naltrexone produces a metabolite called noroxymorphone that can cause a false-positive result for oxycodone on urine drug screens [15]✓ Verified knowledgeBeauregard et al. (2024) — Opioid positive urine. This has real clinical and legal implications. If you're taking oral naltrexone and a drug screen comes back positive for oxycodone, that result should be confirmed with more specific testing before any action is taken. Make sure anyone reviewing your drug screens knows you're taking naltrexone.