Should benzodiazepines still have a role in treating patients with anxiety disorders?

D. Baldwin,Bilal Talat

DOI: https://doi.org/10.1002/hup.1278

2012-05-01

Human Psychopharmacology

Abstract:There are at present no ideal drugs for treating patients with anxiety disorders, so there is much room for improvement in the development of novel treatments with enhanced effectiveness and greater acceptability (Baldwin, 2011). The hazards of benzodiazepines should be familiar and include cognitive and psychomotor impairment, reduced driving performance, abuse liability, and withdrawal effects in a substantial minority of patients. But it is uncertain whether drowsiness, clumsiness, and memory problems associated with benzodiazepines are more burdensome than the nausea, headache, and sexual dysfunction seen with selective serotonin reuptake inhibitors or more upsetting than the weight gain and risk of glucose intolerance seen with antipsychotic drugs. Furthermore, the upset caused by side effects of medication has not been compared with the distress experienced whilst waiting for psychological interventions of limited availability or with the misery arising from relapsing after stopping previously effective treatment. Some benzodiazepines have been found efficacious in some anxiety disorders, but in generalised anxiety disorder, high drop-out rates due to adverse effects limit their effectiveness (Martin et al., 2007; Baldwin et al., 2011a); in social phobia, little is known about their long-term value (Blanco et al., 2003); their relative efficacy and tolerability, when compared with other treatments, are uncertain over the long-term in panic disorder (Kumar & Malone, 2008); and they have uncertain efficacy in either acute or continuation treatment in post-traumatic stress disorder (Zhang & Davidson, 2010). Hence, benzodiazepines would seem to have only a limited role in managing patients with anxiety disorders. This being so, how frequently are they prescribed? Prescribing data from the UK suggest that approximately 6 million prescriptions for benzodiazepine anxiolytic drugs are issued each year (NHS Information Centre, 2010). In the USA, recent findings indicate that alprazolamwas the most commonly prescribed psychotropic drug (Grohol, 2010). A significant proportion of European psychiatrists recently endorsed the use of benzodiazepines as potential first-line treatments in patients with generalised anxiety disorder (Baldwin et al., 2011b), and in a large prescribing survey conducted in Asian countries, over half of inpatients with the diagnosis of schizophrenia were prescribed benzodiazepines, often in high doses (Tor et al., 2011). These observations may be perplexing, as most guidance recommends that benzodiazepines should be prescribed for only short periods and in only a minority of patients (British National Formulary, 2011; NICE, 2011), but evidence from pharmacoepidemiological studies and prescribing practice surveys shows that doctors are still prepared to prescribe for longer periods, to large numbers of patients, even when guidance recommends against this. Why might this be? An answer may be provided through understanding the emerging concept of ‘mindlines’. It has been argued that medical practitioners absorb knowledge from many disparate sources and rarely access evidence-based guidelines during actual practice. Instead, they typically follow internalised guidelines (‘mindlines’) that are built up over an entire career, which seem more suitable for the complexities of everyday clinical work, which blend formal, informal, tacit, and experiential evidence and which are sufficiently broad and malleable to resolve resultant tensions (Gabbay & Le May, 2011). When making treatment decisions, it seems probable that doctors prefer to follow their own mindlines rather than the recommendations of received guidelines. It seems timely to reassess the relative merits and drawbacks of benzodiazepines. There is a persisting need for placebo-controlled combination studies in acute treatment; for placebo-controlled augmentation studies after non-response, for example, after unsuccessful treatment with a selective serotonin reuptake inhibitor or cognitive behaviour therapy; for welldesigned relapse prevention studies in most anxiety