Whether one should prescribe BZs to patients with a history of alcoholism is a controversial question, and one that tends to polarize psychiatrists into the “purists” versus the “realists.” Most alcoholics take BZs at some point in the course of their disorder, sometimes illegitimately, but often as part of a bona fide treatment program. As examples, most detox settings prescribe BZs to achieve alcohol withdrawal safely; many alcoholics present to psychiatrists with genuine anxiety disorders and are given BZs; and some alcoholics may suffer from a poorly characterized “prolonged withdrawal” syndrome that can include anxiety along with dysphoria, for which BZs can be helpful.

The point is that most psychiatrists periodically find themselves in the difficult position of having to judge whether to initiate or to continue a BZ prescription in a person with a history of alcohol abuse. Given that this is such a common clinical issue, one might expect that there would be a large literature on it, but there isn’t, and most of what exists is pretty old.

The last time anyone really tried to tackle the issue was back in 1988, when a critical review of liability for BZ abuse among alcoholics was published (Am J Psychiatry 1988; 145:1501-1506). The authors located 11 experimental studies attempting to measure BZ use or abuse in alcoholics. However, because most of these studies did not do a good job of distinguishing between BZ “use” (that is, legitimate, prescribed medication) and “abuse” (getting it from the street), the conclusions were of limited usefulness. All the reviewers could really conclude is that about a third of alcoholics report using or abusing BZs at some point, and that this is about the same prevalence as is reported by patients with diagnosed psychiatric disorders other than alcoholism. If anything, these results run counter to the common impression that alcoholics are more likely to seek out BZs than are other psychiatric patients.

One small study was interesting and suggested that alcoholics get more of a high from BZs than non-alcoholics. In this study, 17 abstinent alcoholics (most had been sober for about two and a half months) and 12 controls (no psychiatric illness) were administered a single 1 mg dose of Xanax, and then were periodically queried about their sensations of feeling high for several hours after the dose. The alcoholic subjects reported significantly more post-dose euphoria than the controls, who reported only post-dose sedation and no euphoria. The authors concluded that this means that alcoholics “may be at high risk” to abuse BZs, which may be true, although experiencing euphoria from something doesn’t necessarily mean you’re going to abuse it (J Clin Psychiatry 1988; 49:333-337).

The most clinically relevant study yet published was a naturalistic study conducted as part of the Harvard/Brown Anxiety Disorders Research Program (HARP). Researchers recruited 711 outpatients with DSM-III-R anxiety disorders (panic disorder with or without agoraphobia, social anxiety disorder, and generalized anxiety disorder). At study entry, patients were evaluated with a number of structured interviews in order to ensure accurate diagnoses, and then they were simply sent back to their regular clinicians for ongoing treatment. The researchers did not assign patients to specific treatments, as this was a purely observational study (J Clin Psychiatry 1996; 57:83-89).

Of the original 711, 343 were being prescribed BZs at study entry. Ninetynine (29%) of these had a past or current history of alcoholism, while 244 (71%) did not. The researchers were interested in the following question: Do anxiety disorder patients with alcoholism tend to escalate their use of BZs in comparison with anxiety disorder patients with an alcohol-free history?

The results? Over the 12 months of the study, neither group escalated their average doses of BZs. In general, the anxiety/alcoholic group took a slightly higher dose throughout the study (4 mg Ativan equivalents vs. 3 mg Ativan in the pure anxiety group). What about prns? Did the anxiety/alcoholic group use more prn BZs than the pure anxiety group? During the first six months, there were no differences in prn use at all, and during the last six months, alcoholics were significantly less likely to take prns.

The authors concluded that the data is inconsistent with the theory that alcoholics show a lack of control over BZ use. They acknowledged, however, that the study has a major flaw: BZ use was ascertained purely by patient self-report, and it’s possible that alcoholics under-reported their BZ use. The researchers countered this by pointing out that self-report of drug use is a common methodology in research studies and is generally reliable, because drug abusers see little benefit in lying to researchers who promise confidentiality.

Even with this flaw, the study reassures TCR, and is certainly in line with common clinical experience, that anxiety patients with a history of alcoholism are usually able to use BZs appropriately.

TCR VERDICT: A history of EtOH abuse doesn’t preclude BZs.