About 4 percent of U.S. adults are dependent on alcohol. One drug that can help alcoholics cut back is naltrexone (or ReVia). Unlike the drug Antabuse, naltrexone doesn't make people feel sick when they drink. Instead, it reduces the craving for alcohol. Treatment with naltrexone isn't always successful, in part because taking a pill every day can be a nuisance, so people don't stick with it. In a new study, researchers tried giving naltrexone in a new formulation: a monthly injection.

What the researchers wanted to know: Does injectable naltrexone help people who are dependent on alcohol stop drinking?

What they did: 627 people joined the study at 24 hospitals and clinics. All of them were dependent on alcohol and had also drunk heavily at least twice in the month before they joined the study. They were randomly assigned to get six monthly injections of low- or high-dose naltrexone injections or a placebo. (Neither the patients nor the people giving the shots were told who was getting which—the syringes were even tinted so no one could tell the placebo and naltrexone were slightly different colors.) The subjects were also given supportive therapy by doctors, nurses, psychologists, therapists, and counselors at different study sites. Unlike most of the clinical trials of naltrexone pills, patients in this trial didn't have to stop drinking before the trial started, and most were drinking heavily.

What they found: People getting all three treatments—both naltrexone doses and the placebo—had many fewer heavy drinking days during treatment than before the treatment started. But people who got high-dose naltrexone lowered their heavy drinking by about 25 percent more than people who got the placebo. People on low-dose naltrexone lowered their heavy drinking by 17 percent more than those on the placebo. The researchers paid the most attention to heavy drinking because, of the many ways of drinking, heavy drinking is thought to be the most harmful, resulting in drunk driving, injuries, and conflicts with other people. Only 39 patients managed not to drink at all during the trial, and they were spread pretty evenly between the three groups.

What the study means to you: Injectable naltrexone may help some people cut back on drinking.

Caveats: Women didn't respond as well to the treatment as men, but they made up only 32 percent of the people in the trial, and the trial wasn't designed to look for differences in gender, so it's too early to decide whether this treatment works as well in women. A general problem with clinical trials of alcoholism is that patients who enroll in them might be more motivated to cut back on their drinking than the average alcoholic.

Find out more: Psychiatrists divide what the rest of us call "alcoholism" into two disorders—alcohol abuse and alcohol dependence.The Mental Health Channel includes the official definitions of both.

Check out USNews.com for a previous health brief on alcohol abuse versus dependence.

Read information about naltrexone from the American Academy of Family Physicians.

Read the article: Garbutt, J.C., et al. "Efficacy and Tolerability of Long-Acting Injectable Naltrexone for Alcohol Dependence: A Randomized Controlled Trial." Journal of the American Medical Association. April 6, 2005, Vol. 293, No. 13, pp. 1617-1625.

Abstract online: http://jama.ama-assn.org