This Month’s Expert: Mark Willenbring, M.D. Medications for Alcohol Dependence

This Month’s Expert: Mark Willenbring, M.D. Medications for Alcohol DependenceTCPR: Dr. Willenbring, there are now several medications with evidence for treating alcohol problems. It gets confusing to know which ones to use in which situations.

Dr. Willenbring: I agree, and the real problem is that psychiatrists are not using any of these medications enough. This is unfortunate, because psychiatric patients have a higher prevalence of alcohol use disorders than general medical patients, and about 90% of people with alcohol dependence receive no focused alcoholism treatment.

TCPR: It seems that many psychiatrists don’t consider themselves expert enough in treating alcoholism and end up referring these patients to treatment programs.

Dr. Willenbring: It would be useful if psychiatrists were more comfortable treating these patients in their practices. Experience has shown that patients are often willing to address their drinking with their existing providers, but that they are often not willing to go to a specialized treatment program. One of the problems is that most of the studies of alcoholism treatment enroll the most severely dependent patients, and clinicians may get the false impression that medications are primarily helpful for these most serious cases. But in the real world, most people don’t have that level of impairment. In fact almost ¾ of people who develop alcohol dependence have only a single episode lasting on average 3 to 4 years. Many of these people have milder forms of the disorder, and these are the patients who can derive a lot of benefit from office-based treatment.

TCPR: A couple of years ago, the NIAAA sponsored the COMBINE trial, which was largest clinical trial ever done on alcoholism (JAMA 2006;295:2003-2017). What were some of the take-home points from that trial?

Dr. Willenbring: The big news was that the most effective treatment was naltrexone combined with brief medication management sessions. Acamprosate, on the other hand, had essentially no effect vs. placebo, and the patients who were assigned to the cognitive behavior therapy only group, without any pill, did the worst.

TCPR: Does that imply that cognitive therapy is worse than no treatment at all?

This article originally appeared in The Carlat Psychiatry Report -- an unbiased monthly covering all things psychiatry.
Want more, plus easy CME credit?
Subscribe today!

Dr. Willenbring: No, you have to be cautious interpreting this, because there might have been a disappointment effect. Eight of the nine study arms involved some type of pill, either active or placebo, and most patients who enrolled in the trial were expecting to get a medication. Those who got therapy alone may have been disappointed and may have done worse because of this.

TCPR: Tell us more about the patients who did the best.

Dr. Willenbring: These patients received naltrexone 100 mg/day, and in addition were seen by physicians or nurses for nine 20- minute sessions over 16 weeks. These sessions involved reviewing how things were going in terms of drinking, medication side effects, encouraging adherence to the medication, encouraging abstinence and encouraging attendance at AA meetings. We have condensed this manual into a two-page format that is included in our Clinician’s Guide: Helping Patients Who Drink Too Much, and it covers exactly what to do and what to say during these visits. Also, over the next month we’ll be releasing an online training using videotaped scenarios showing how physicians can be most effective during brief medication management sessions. All of these tools, as well as patient handouts and a pocket version of the Guide, are available on our website,

TCPR: Based on your personal clinical experiences, how do you suggest we go about deciding which medications to choose with which patients?

Dr. Willenbring: When I decide on medications, I look first at the strength of the evidence, and then I look at other factors like side effects, patient preference, drug interactions and so on. At this point, the medication with the strongest evidence base is naltrexone, and I recommend it as the first-line choice for most patients. I tell my patients that it reduces craving and makes it easier to stop drinking if they slip, rather than going on to a full relapse.

TCPR: And how do you typically dose naltrexone?

Dr. Willenbring: I often start with 50 mg daily, but sometimes I’ll prescribe it on an as needed basis at 25 mg, with instructions to take it immediately before a potential drinking situation. Many patients find that this works for them. There’s no evidence that increasing dose beyond 50 mg works any better, and higher doses increase the risk of elevated liver enzymes. In fact, in my practice, I’ve never seen any patient with elevated liver enzymes at 50 mg, but the data indicate that 10% of patients on 100 mg will have elevated LFTs.

TCPR: Do you routinely monitor LFTs in these patients?

Dr. Willenbring: I get baseline LFTs, and again in three months or so. But I practice fairly conservatively, and many physicians don’t get LFTs at all, especially if the dose is not higher than 50 mg. About 10-15% of patients will have some nausea, and I tell them to take it with food. A clinical pearl is that taking naltrexone in the morning doesn’t make sense, because most people drink in the evening. So I instruct my patients to take it in the late afternoon, or an hour or so before they might be tempted to drink.

TCPR: In my experience, I’ve had several patients who stop taking naltrexone because they’re not sure that it’s still necessary, and after stopping it they relapse.

Dr. Willenbring: This is a significant problem, and it’s always a challenge to encourage patients to continue using any medication for alcoholism. Especially with naltrexone, it can be hard to detect an effect, because its main activity is to prevent a relapse. So patients will say to me, “I’m taking this medication, and I don’t feel any different, why should I continue it?” I make sure they are aware of the data, namely, that there are 20-40% fewer relapses in the first 3 to 4 months on naltrexone compared to placebo, and I say, “If you’re not drinking, it’s working.” At the outset, I tell patients that they should plan to stay on it for a year, although I reevaluate the treatment every three months at a minimum.

TCPR: After naltrexone, what it your second choice?

Dr. Willenbring: I find topiramate (Topamax) to be quite effective. Two good placebo controlled trials have been published, and it has been shown to be effective in helping actively drinking patients achieve recovery, and not simply prolonging the time to relapse. About 20% of drinking patients on Topamax will recover vs. about 5% of patients on placebo.

TCPR: What is topiramate’s mechanism of action?

Dr. Willenbring: It works on the GABA and glutamate system. With alcohol dependence, there tends to be a hyperglutamergic state and a hypo-GABA state, and this may be part of the so-called “protracted abstinence syndrome,” characterized by a lack of pleasure, low grade anxiety, insomnia, and hyper-responsiveness to stress. Topiramate may help to restore the normal balance by increasing GABA and decreasing glutamate.

TCPR: And what about dosing and side effect issues with topiramate?

Dr. Willenbring: I start at 25 mg/bedtime for a week, and I increase by about 25 mg a week, up to a target of about 200 mg/day. I warn patients about paresthesias, a bad taste, and word-finding problems. It can also cause anorexia and weight loss, but this is an advantage for many patients. In my experience, the side effect patients complain about the most is the word-finding problem; they feel like they can’t think straight. So, while topiramate is effective, it has the worst side effect burden of the medications for alcoholism.

TCPR: What about disulfiram (Antabuse)?

Dr. Willenbring: I use disulfiram a fair amount. The clinical pearl here is that I always make sure somebody monitors the patient to make sure they are taking it, because it will have no effect at all if the patient is not adherent. Usually, the monitor is the spouse, and I simply ask that the spouse observe the patient taking the pill. But I try to emphasize that the spouse should be supportive, and should not play the role of a cop. So I have two rules: first, there are no unpleasant conversations during monitoring, and second, they should thank each other afterward. I ask the monitor to call to inform me if the patient misses two days, allowing me to bring up the issue at the patient’s next appointment.

TCPR: You haven’t mentioned acamprosate (Campral) yet.

Dr. Willenbring: There have been two large negative trials of acamprosate in the U.S. In my own practice, I use it rarely, but many researchers believe the jury is still out and that it may have some usefulness.

TCPR: Finally, I’d like to get your take on Vivitrol, the long-acting injectable version of naltrexone.

Dr. Willenbring: I think it’s a great option. It causes fewer side effects than oral naltrexone, it gives better coverage throughout the day, and it overcomes problems with medication adherence, because it is a monthly injection. The downsides are that a lot of people don’t like shots, and that it is expensive.

This Month’s Expert: Mark Willenbring, M.D. Medications for Alcohol Dependence

This article originally appeared in:

The Carlat Psychiatry Report
Click on the image to learn more or subscribe today!

This article was published in print 3/2008 in Volume:Issue 6:3.


APA Reference
Willenbring,, M. (2013). This Month’s Expert: Mark Willenbring, M.D. Medications for Alcohol Dependence. Psych Central. Retrieved on September 18, 2020, from


Scientifically Reviewed
Last updated: 3 Sep 2013
Last reviewed: By John M. Grohol, Psy.D. on 3 Sep 2013
Published on All rights reserved.