Alcoholism and anxiety go hand in hand. The extent of this comorbidity is clear from the numbers: as many as 45% of patients with alcohol disorders meet diagnostic criteria for a co-occurring anxiety disorder. And alcoholic patients with a comorbid anxiety disorder—particularly panic disorder or social phobia—are three to seven times more likely to relapse than those without concurrent anxiety (Kushner MG et al, Alcoholism: Clin Exp Res 2005;29:1432–1443).

It’s not clear why anxiety and alcoholism so commonly co-occur, but there are at least three potential, and not mutually exclusive, explanations for this phenomenon. First, anxiety may lead to alcohol abuse. According to this theory, patients use alcohol as a way to “self-medicate” their anxiety (Morris EP et al, Clin Psychol Rev 2005;25:734–760). Second, and conversely, excessive alcohol use may generate an anxiety disorder, via a “kindling” effect of repeated withdrawal cycles or disruptions to the stress response system. Finally, there may be no clear primary disorder, but rather a common underlying vulnerability to both anxiety and to alcohol abuse (Kushner MG et al, Current Psych 2007;6(8):55–64). This may be psychological, like high anxiety sensitivity, or biological, like GABA receptor dysfunction or a gene polymorphism.

In most cases, it’s not important to determine which disorder came first. In my experience, in fact, this chicken-and-egg problem is irrelevant, because by the time most patients present for treatment, they’re stuck in a mutually reinforcing cycle of anxiety, self-medication with alcohol, and increased distress when they attempt to stop or curtail their alcohol use. As a result, it’s best to address both disorders in treatment. Precisely how to do this, however, is open to debate.

Historically, substance use treatment and psychiatric treatment have remained distinct. One common approach has been—and remains—to treat the conditions sequentially, usually with an initial focus on alcohol use, to determine whether anxiety symptoms remit with consistent sobriety. While this makes sense intuitively—particularly when the anxiety may be at least partially substance-induced—these efforts often fail because of the vicious cycle of sobriety leading to more anxiety and subsequent alcohol relapse.

Another strategy is parallel treatment, in which the anxiety is managed psychiatrically (with medications or therapy directed specifically at the anxious symptoms) while a specialist in addiction simultaneously treats the alcohol disorder. Many chemical dependency (CD) treatment centers take this approach, and it’s also the basis of the oft-used strategy of referring patients to AA or CD treatment while you, the psychiatrist, treat the anxiety disorder as you would with any other patient. However, effective treatment requires close coordination and communication among providers, and patients can be frustrated by the division in treatment strategies, especially when they get contradictory messages. For instance, some exposure-based treatment methods for anxiety require the patient to confront situations which they are simultaneously taught to avoid because of their high relapse risk.

A third alternative is an integrated approach. While such a strategy is most frequently recommended by experts and has theoretical appeal, there are few standardized, integrated treatment strategies and the data on outcome is both sparse and unimpressive (see for example, Tiet QQ and Mausbach B, Alcohol Clin Exp Res 2007;31(4):513–536). This may speak to the lack of trained professionals who can recognize and treat both addiction and anxiety disorders, but also, in my experience, to wide variability in patients’ histories, motivations, and circumstances.

Psychotherapy for Anxiety and Alcoholism
Cognitive behavioral therapy (CBT) is well suited for the combined treatment of alcoholism and anxiety, as it can help patients to recognize environmental triggers and identify maladaptive coping strategies and practice new ones, just as CBT for anxiety teaches patients to challenge their own self-destructive thoughts and behaviors. I have actually found that 12-step and other group work teaches patients to identify distorted thoughts and beliefs, and to learn from others’ successes and failures.

The “Relapse Prevention” (RP) model (Marlatt GA and Gordon JR, Eds. Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors. New York: Guilford Press, 1985) is widely used chemical dependency technique based on CBT. The model identifies three general stimuli that precede relapse: negative emotional states, social pressure, and interpersonal conflict. Each of these can heighten anxiety (eg, social pressure can aggravate social phobia and negative emotions can worsen generalized anxiety disorder).

Often, patients with worsening anxiety use avoidant coping strategies, specifically relying on alcohol to relieve the anxiety as a form of “negative reinforcement”—that is, it reinforces the value of drinking by taking away the unpleasant anxiety. The RP model can be adapted for the management of anxiety in the context of substance abuse, but it has not yet been empirically tested for comorbid conditions.

Other psychotherapeutic approaches have been developed for specific anxiety disorders, such as “Seeking Safety” for concurrent PTSD and substance abuse (Najavits LM et al, J Subst Abuse Treatment 1996;13(1):13–22). However, in a three-month trial of 107 women with PTSD and substance abuse (not limited to alcohol), Seeking Safety was found tobe no more effective than outpatient cognitive therapy alone (Hien DA et al, Am J Psychiatry 2004;161(8):1426–1432). On the other hand, a two-week pilot trial of CBT-based integrated treatment for panic disorder and alcohol abuse in 48 patients showed benefit over alcoholism treatment alone (Kushner MG et al, J Mental Health 2006;15(6):697–707).

Psychopharmacology for Anxiety and Alcoholism
Despite the approval of medications for anxiety and for alcohol dependence, drug trials typically exclude subjects with concurrent disorders. (See the article by Dr. Frenz in this issue for a review of medications for alcoholism.) There are only a handful of published studies focusing specifically on the treatment of comorbid alcoholism and anxiety (reviewed in Smith JP and Book SW, Psychiatr Times 2008;25(10):19–23), and results are rather lackluster.

In one study of paroxetine (Paxil), 42 subjects with alcoholism and social anxiety were treated with paroxetine or placebo for 16 weeks. Patients taking paroxetine reported less anxiety but alcohol intake was about the same. However, they relied less on alcohol to engage in social situations (Thomas SE et al, Alcohol Clin Exp Res 2008;32(1):77–84).

Another study on the use of sertraline (Zoloft) for comorbid alcoholism and PTSD randomized 94 patients to 150 mg sertraline or placebo for 12 weeks. Both groups drank less at the end of the trial, although a subgroup analysis showed that sertraline was more effective in reducing alcohol intake in patients with less severe alcohol dependence at baseline (Brady KT et al, Alcohol Clin Exp Res 2005;29(3):395–401). But a similar 12-week trial comparing sertraline with placebo for PTSD and alcoholism found no significant differences at all (Labbate LA et al, Compr Psychiatr 2004;45:304–310).

At least three studies have examined buspirone (BuSpar) for management of alcohol dependence and anxiety symptoms. Two RCTs comparing 15 mg/day buspirone with placebo showed trends toward reduced anxiety and alcohol intake (Kranzler H et al, Arch Gen Psych 1994;51:720–731; Tollefson G et al, J Clin Psychopharm 1992;12:19–26) but these were not statistically significant. A third, using a higher buspirone dose (60 mg/day), showed no effect relative to placebo (Malcolm R et al, Alcohol Clin Exp Res 1992;16(6):1007–1013).

Using benzodiazepines to treat anxiety in alcohol-abusing patients is controversial. Most treatment guidelines recommend avoiding benzodiazepines in patients with alcohol dependence. Psychiatrists and internists frequently use benzodiazepines to treat alcohol withdrawal symptoms, but their use is time-limited (usually less than seven days) and usually in a highly supervised setting. Patients with known anxiety disorders sometimes require a longer taper (Edwards G et al, The Treatment of Drinking Problems, 4th ed. Cambridge, UK: Cambridge University Press, 2003). Continuous use of benzodiazepines should be undertaken with caution, as they may have a high abuse potential in alcoholics; in fact, alprazolam (Xanax), diazepam (Valium), and lorazepam (Ativan) have a mood-enhancing effect in alcoholics that is not observed in non-alcoholics (Ciraulo AD and Nace EP, Am J Addict 2000;9:276–284). On the other hand, there is evidence to suggest that a history of alcohol dependence may not necessarily result in greater abuse of benzodiazepines, particularly in those with less severe dependence (Lingford-Hughes A et al, Adv Psychiatr Treat 2002;8(2):107–116).

In my practice, I find benzodiazepines to be effective if used judiciously and with very close follow-up. Two red flags that make me suspect my patient is becoming addicted are: (1) requests for escalating doses and (2) overreliance on medications at the expense of other strategies that have proven effective for that patient—for example, a reluctance to engage in therapy or ongoing 12-step work.

The opiate blocker naltrexone (ReVia) is effective for preventing alcohol relapse, presumably because the euphoric effects of both alcohol and benzodiazepines may be mediated by the endogenous opiate system. While naltrexone is not approved for benzodiazepine addiction, one study found that it prevented the euphoric (but not theanxiolytic) effects of benzodiazepines when the two drugs were used together (Richardson DK et al, Pharmacol Biochem Behav 2005;81(3);657–663).

Controlled studies of all sorts of comorbid anxiety/alcohol treatments—whether separate or integrated, psychotherapy or medication—have been disappointing, most likely due to the wide variability among patients who present for treatment. Like most experts in the field, I have found that the most effective approach is an individualized one. Engaging the patient in treatment is more than half the battle, so when possible, I start with the condition that the patient identifies as most problematic (which may or may not be the substance abuse), while using therapies such as motivational interviewing to encourage him or her to address other problems.