These two are not equivalent options at different price points. One is a licensed treatment for alcohol use disorder with a long trial record; the other is an early and interesting off-label finding. Anyone comparing them should start from that asymmetry rather than from which drug is newer. This article is part of our GLP-1 and Alcohol hub.
What makes the comparison worth making at all is that they appear to do something similar by different routes, which is genuinely intriguing.
# The headline difference
Naltrexone is approved for alcohol use disorder. It has been licensed in the US since 1994 and is available across Europe, the UK, Australia and Canada. Its use is supported by a large body of randomized trials and by Cochrane review evidence. It is the drug behind the Sinclair Method.
No GLP-1 agonist is approved for alcohol anywhere. The evidence is a strong animal literature, supportive population data, and one small positive randomized trial, alongside one older trial in a different drug that was largely negative.
That is not a close contest on evidence, and it should not be presented as one. It is a mature treatment against a promising early signal.
# How each one works
They converge on the same target from different directions, which is the interesting part.
Naltrexone blocks opioid receptors. Alcohol triggers endorphin release, endorphins activate those receptors, and the result is the pleasurable reinforcing hit that makes the next drink appealing. Block the receptor and the reinforcement is blunted. Drinking still happens; it just stops paying off in the same way.
GLP-1 agonists activate GLP-1 receptors, including those in the ventral tegmental area and nucleus accumbens, the same reward circuitry. The apparent result is a dampened dopamine response to alcohol and, in patient accounts, reduced wanting rather than blocked enjoyment.
The distinction people report is subtle but consistent. Naltrexone users often describe drinking and finding it unsatisfying. GLP-1 users more often describe not thinking about it in the first place. That difference is anecdotal and should be held loosely, but it recurs often enough to be worth mentioning.
# Practical comparison
Access. Naltrexone is a cheap generic tablet, though prescribers are often unfamiliar with it for alcohol, which is covered in why naltrexone is underprescribed. GLP-1 agonists are expensive, frequently supply-constrained, and will generally only be prescribed if you have a diabetes or weight indication. Getting one prescribed for drinking alone is unlikely.
Administration. Naltrexone is a daily or as-needed oral tablet. Most GLP-1 agonists are weekly injections, with an oral semaglutide option.
Side effects. Naltrexone commonly causes nausea early on, which usually settles, and it is absolutely contraindicated with opioid medication because it will precipitate withdrawal. GLP-1 side effects are predominantly gastrointestinal and dose-related, with rarer pancreatitis and gallbladder concerns. See naltrexone side effects and can you drink alcohol on a GLP-1.
Weight. GLP-1 agonists produce substantial weight loss, which for some people is the point. Naltrexone’s weight effect is indirect, mostly through removing alcohol calories.
Track record. Thirty years against roughly three.
# Which suits whom
Nobody has run a head-to-head trial, so what follows is reasoning from what each drug is licensed and evidenced for, not a research finding.
If reducing drinking is the goal in itself, naltrexone is the option with the license and the evidence. It is inexpensive, well characterised, and has an established protocol behind it.
If you already have obesity or type 2 diabetes and also drink too much, a GLP-1 becomes a reasonable thing to raise with your prescriber, since you may qualify for it on its licensed indication and the alcohol evidence, thin as it is, clusters in exactly that population.
If you are on a GLP-1 already and your drinking has dropped, that is a useful observation to record and report, not a reason to change anything.
If you want a GLP-1 specifically to stop drinking and have no other indication, the current evidence does not support that, and a prescriber declining is following the evidence rather than being unhelpful.
# Naltrexone comes with a method, and that is not a small difference
One asymmetry gets missed in every comparison of these two drugs. Naltrexone has an established protocol for how to use it. GLP-1 agonists, for alcohol, have nothing of the kind.
Under the Sinclair Method, naltrexone is taken an hour before drinking and only on days you drink, with the deliberate aim of letting the reinforcement fade over months of drinking on the drug. That is a specific, testable procedure with a mechanism behind it and a body of evidence, and it produces a gradual reduction rather than an immediate one. The details are in the Sinclair Method explained.
There is no equivalent for GLP-1 agonists. You take the weekly injection for your diabetes or your weight, and any alcohol effect is a passenger. Nobody has established whether timing relative to drinking matters, whether the effect needs continuous dosing, or whether anything is being unlearned in the way the Sinclair Method proposes.
That gap matters for anyone deciding between them. With naltrexone you get a drug and a plan for using it. With a GLP-1 you currently get a drug and an observation.
# Cost and access, which often decides it
The pharmacology is rarely what determines which drug someone ends up on. Availability usually does.
Naltrexone is off-patent and cheap almost everywhere. In the UK it is not routinely commissioned for alcohol in every area, and many people obtain it privately at modest cost. In the US it is inexpensive as a generic tablet, with the extended-release injection considerably more expensive. Across most of Europe, Canada and Australia it is available and affordable. The obstacle is rarely money; it is finding a prescriber who is comfortable with it for alcohol.
GLP-1 agonists are the opposite problem. They are expensive, and coverage is generally tied to a diabetes or obesity indication. If you meet the criteria for one of those, cost may be manageable through insurance or a national system. If you do not, you are looking at a substantial monthly private cost for a drug that is not licensed for the reason you want it.
There is a third option people forget in this comparison. Acamprosate is also licensed for alcohol use disorder, works differently again, and suits people aiming at abstinence rather than reduction. It is covered alongside the others in naltrexone vs acamprosate vs disulfiram. Framing this as a two-horse race between the drug you have heard of and the drug in the news leaves out a licensed treatment that might fit better than either.
# Could they be combined?
The question comes up, and the answer is that nobody knows.
There is no trial data on taking a GLP-1 and naltrexone together for alcohol. The mechanisms are different enough that a combination is not obviously redundant, and both are used with other medications routinely, so there is no evident reason to expect a dangerous interaction. But “no obvious reason to expect a problem” is a long way from evidence of safety or benefit.
There is one wrinkle worth knowing: naltrexone already appears in a combination weight-loss product alongside bupropion, sold as Contrave, which is discussed in naltrexone and weight loss. That is a different combination to the one being asked about here, and it is easy to conflate them.
Anyone considering both should be doing it with a prescriber who knows they are on both.
# How AlcoLog helps
Whichever route you are on, the same thing makes it legible: a record of what you took and what you drank.
AlcoLog tracks medication doses, including naltrexone and GLP-1 agonists, alongside every logged drink. For naltrexone in particular, dose timing relative to drinking is the whole method under the Sinclair protocol, and the app is built around that. For a GLP-1, the useful pattern is what happens in the weeks after a dose step.
The AlcoScore deliberately excludes medication, so nothing you take is graded. Your data stays on your device with no account or email, and export gives your prescriber something concrete rather than a recollection.
