Skip to main content
NaltrexoneReviews home

Studies

What the research on naltrexone shows

151 studies, each explained in plain English: 59 randomized trials, 29 reviews that pool many trials, and 63 other studies. Every page links to the paper.

The short version

Naltrexone helps with drinking, by a modest amount on average. The Cochrane review pooled 50 trials with 7,793 patients and found it lowered the risk of heavy drinking to 83% of the risk on placebo (Rösner 2010). A larger review of 122 trials found that 12 people need to take it for one of them to avoid a return to heavy drinking (Jonas 2014).

It works mostly by stopping a drink from turning into a heavy-drinking relapse. In the first trial, 19 of 20 people on placebo who drank went on to relapse, compared with 8 of 16 on naltrexone (Volpicelli 1992). The other 1992 trial found the same pattern alongside therapy (O'Malley 1992).

It doesn't work for everyone. In 627 veterans with long-term, severe alcohol dependence, it did no better than placebo (Krystal 2001). A review of trials in people who were still drinking when they started found no effect on total alcohol intake (Palpacuer 2018).

Taking it only when needed has some support. In one trial, people took naltrexone daily for 12 weeks, then only when they craved a drink for 20 more. With coping skills therapy, 27% on naltrexone had no relapse to heavy drinking over the whole 32 weeks, compared with 3% on placebo (Heinälä 2001). J.D. Sinclair, of the Sinclair Method, was one of its authors. In a 12-week trial of heavy drinkers trying to cut down, the benefit showed up in men but not women (Kranzler 2009).

The monthly shot works too. A 380 mg Vivitrol shot cut heavy drinking days by 25% compared with placebo over six months (Garbutt 2005).

Compared with acamprosate, naltrexone did better on heavy drinking and craving, and acamprosate did better at keeping people fully abstinent (Maisel 2013). In the large US COMBINE trial, naltrexone helped and acamprosate didn't (Anton 2006).

Serious side effects are no more common than on placebo. Across placebo-controlled trials of the pill, the risk of serious adverse events wasn't higher on naltrexone (Bolton 2019). Nausea and headache are the most common side effects (Croop 1997).

Every study

Filter by topic or type, or search. Each one opens a page with what it tested, what it found and its limits.

Showing 151 of 151 studies