The use of benzodiazepines in populations with SUDs and co-occurring psychiatric disorders is controversial. This issue has been explored almost exclusively in populations with anxiety and alcohol use disorders. The prevalence of benzo-diazepine use among patients with alcohol use disorders is greater than in the general population but comparable to psychiatric disorder populations (Ciraulo, Sands, & Shader, 1988). Clinicians are often understandably concerned that prescribing benzodiazepines to these patients may lead to either a worsening of the alcohol use disorder, the development of a benzodiazepine use disorder, or potentiation of the benzodiazepine effect when combined with alcohol. Preliminary evidence from case reports (Adinoff, 1992) and a prospective naturalistic study (Mueller, Goldenberg, Gordon, Keller, & Warshaw, 1996) suggests that there may be a carefully selected subpopulation of patients with co-occurring alcohol use and anxiety disorders for whom long-term prescription of benzo-diazepine may not affect sobriety or result in benzodiazepine misuse. However, it may not improve outcomes either. For example, a retrospective naturalistic study of veterans with PTSD and SUD found that physicians were less likely to prescribe benzodiazepines for those with SUD (Kosten, Fontana, Sernyak, & Rosenheck, 2000). While those with prescribed benzodiazepines did not have worse outcomes, chronic benzodiazepine treatment (independent of a co-occurring SUD) did not improve anxiety or social functioning in these patients either. Similarly, Brunette, Noordsey, Xie, and Drake (2003) followed SPMI patients with SUDs annually for 6 years and found that the rate of benzodiaze-pine prescribing was high (up to 63%) but not associated with differences in substance use remission, hospitalization, or, interestingly, reductions in anxiety or depression. Also, unsurprisingly, patients prescribed benzodiazepines were more likely to abuse them than those who were not. While controlled trials are needed to explore these issues more fully, the findings from these reports add further to concerns that the long-term use of benzodiazepines in these populations perhaps offers the risk of abuse or dependence without great potential for clinical benefit.
Another pharmacological alternative in this population is buspirone, which does not have abuse potential. Thus far, there have been three doubleblind, placebo-controlled studies of buspirone in patients with alcohol dependence and anxiety—generalized anxiety disorder (GAD) (Tollefson, Montague-Clouse, & Tollefson, 1992), GAD and "other nonpanic anxiety" (Malcolm et al., 1992), or "anxious alcoholics" (Kranzler et al., 1994). Two of the studies found buspirone to be associated with improvements in anxiety and alcohol use outcomes (Kranzler et al., 1994; Tollefson et al., 1992). Although there have been concerns that buspirone's antianxiety effect is more limited in patients with a prior history of benzodiazepine use (Schweizer, Rickels, & Lucki, 1986), a pooled analysis of eight placebo-controlled randomized trials of patients with GAD (DeMartinis, Rynn, Rickels, & Mandos, 2000) found that patients with either remote (defined as at least 1 month duration) or no prior benzodiazepine treatment experienced improved anxiolysis, fewer adverse events, and clinical improvement similar to that with benzodiazepines compared to patients with recent benzodiazepine treatment. Thus, patients who have not received benzo-diazepines for at least 1 month may benefit from buspirone.
Was this article helpful?