Assessing unhealthy alcohol use in hospitalized patients

The National Institute on Alcohol Abuse and Alcoholism (NIAAA) estimates that 3 in 10 Americans drink enough to put them at risk for health problems.

The National Institute on Alcohol Abuse and Alcoholism (NIAAA) estimates that 3 in 10 Americans drink enough to put them at risk for health problems (see Table). Often, these problems are severe enough to lead to hospitalization. Researchers have speculated that inpatients could be an ideal audience for alcohol intervention because hospitalization, especially for an alcohol-related health problem, seems to be an ideal venue for a “teachable moment.” But are brief interventions enough to provoke long-lasting change in hospitalized patients who drink too much?

Richard Saitz, FACP, professor of medicine and epidemiology at Boston University and director of the Clinical Addiction Research and Education (CARE) Unit at Boston Medical Center, recently led a study that attempted to answer this question. In the government-funded randomized, controlled trial, which was published in the Feb. 6, 2007, issue of Annals of Internal Medicine, Dr. Saitz and his coauthors assigned 341 patients identified at hospitalization as “risky drinkers” to receive 30 minutes of motivational counseling about their drinking, or usual care. The researchers followed up with the patients after the intervention to see whether and how much their drinking behavior had changed. The good news was that over 40% of patients in each group had sought treatment for unhealthy alcohol use at three months. The bad news? The difference between groups wasn't statistically significant, and the authors couldn't prove that the counseling had any effect.

Dr. Saitz recently spoke to ACP Hospitalist about alcohol assessment in the hospital and what his study's results might mean for hospitalists' practice.

Q: Is assessment for alcohol dependence and risky drinking something that's routinely done for inpatients?

A: It's getting more and more common, and it's certainly recommended. The U.S. Preventive Services Task Force recommends that we screen for alcohol problems in patients in general. But most studies that either survey physicians or examine medical records find that alcohol problems or unhealthy alcohol use is underdiagnosed, and hospital settings are not an exception to that.

Q: Do you think that this screening should be done?

There's ample evidence for it. Because our study of inpatients was a negative study-we didn't find that we were able to improve referral to any kind of help, and we also didn't find that we were able to decrease drinking-one conclusion could be that maybe nothing should be done in the hospital setting at all. But there are a lot of reasons that doctors should still screen and provide feedback.

First, unhealthy drinking is common and is often related to the reason for admission. Second, patients should be told if they have an alcohol-related condition, even something that isn't necessarily caused by alcohol. Drinking might be related to medication adherence for any condition, which might occasion a hospitalization. At a minimum, screening lets doctors prevent or treat alcohol withdrawal and make sure that they're not prescribing something [in the hospital or during discharge planning] that's going to be affected by recent alcohol use.

Q: What have you learned from the results of your study?

Even though our study was negative, that doesn't mean that brief intervention is worthless. When people with addictions eventually quit for some prolonged period of time, they often attribute it to a comment someone made about their health. They might say “Well, three years ago a doctor told me I was drinking so much that I'd damaged my heart. At the time I didn't do anything about it, but now I think that was a catalyst.”

It's a process. Hospitalists and general internists are familiar with this whole issue because they all know backwards and forwards about the stages of change. It's probably worth talking to patients over time because you can sort of move them along. It doesn't mean that you're going to change health behavior tomorrow, but you're helping patients take baby steps.

Q: Is there anything hospitalists can do when screening identifies a patient with unhealthy alcohol use? What course of action would you recommend?

For the smaller proportion [of patients] that the hospitalists are going to identify, those who have nondependent unhealthy alcohol use, giving feedback, like “You're drinking above recommended limits” or “Your drinking puts you at risk for a heart problem,” is the first step. The next step is to provide some specific advice, like “In my medical opinion, people in your situation should abstain from alcohol.” Then, finish with some sort of negotiation or goal setting. Ask the patient what he thinks, and negotiate what he's going to do. The patient may either choose to follow your advice, or he might say something like, “Gee, I don't know if I can cut down that much. But I could cut down to X drinks a day and then follow up with my doctor. How's that?” Some randomized trials in the inpatient setting have shown that the fairly simple combination of feedback, advice and goal setting, which can be done in maybe 15 minutes, effectively decreases alcohol use and alcohol-related problems in nondependent adults with unhealthy alcohol use.

Q: What about people who are alcohol-dependent?

In our study of inpatients, we identified mostly people in this group. This is the reverse of the outpatient setting, where nondependent alcohol use would be more common. What we can say from our study is that a one-time intervention didn't increase linkage to Alcoholics Anonymous or to specialty substance abuse or alcohol treatment in alcohol-dependent patients, and it also didn't affect their drinking. One short answer is that in this group, hospitalists can't rely on a brief intervention alone. But, then, what can they do?

The probable implication of our study is that these patients really do need to be followed up in the long term. Hospitalists can help by connecting with the care that patients will receive after the hospitalization. For many patients, hospitalization is really a seminal event. I've often seen people come back from a hospitalization and really remember what a hospital physician told them. Maybe the first step needs to be making sure that these patients are engaged in longitudinal medical care from their primary care physician, who can then, over several visits, talk to them and hopefully get them the care that they need over time.

Q: What should be the focus of future research?

No one has studied disease management for alcohol dependence in a controlled fashion. We've designed and implemented a disease management program for alcohol dependence and are testing that in a randomized trial, just like others have tested disease management programs for congestive heart failure. Any interventions that would help prolong an intervention beyond the hospital stay should be tested. The solution to this problem is going to require a more long-term view.