Anxiety disorder treatment guidelines: Little by little a consensus emerges

‘The trouble with most of us is that we would rather be ruined by praise than saved by criticism’.

–Norman Vincent Peale

One of the benefits of being an expert reviewer of a scientific manuscript is early access to the findings and the data, keeping one up to date with the gust front of scientific research and modern practice. A second less well-discussed advantage of engaging in scientific peer review is the potential of influencing the final product, by constructive criticism and iterative manuscript revisions. Having engaged with versions of this document since 2016, it has been a validation of the scientific process to see how much the manuscript has matured, in part related to external review and feedback. The 2018 Royal Australian and New Zealand College of Psychiatrists Clinical Practice Guidelines for the treatment of panic disorder, social anxiety disorder and generalised anxiety disorder (GAD) (Andrews et al., 2018) are now a significant contribution to Australasian clinical practice.

A third, potential role of the peer reviewer is to provide editorial commentary and counterpoint. We congratulate the authors of the accompanying guidelines on taking the reasonable view that both medications and cognitive behavioural therapy (CBT) can be considered first line across the anxiety disorders considered. They have avoided the trap of addressing psychotherapy with ‘Uncritical Positive Regard’, a phenomenon which Nutt and Sharpe have described whereby trial evidence underpinning benefit of psychotherapy may be subject to lesser scrutiny than that supporting pharmacological approaches, while potential adverse effects may not be considered. It is also reassuring to note that the authors have acknowledged the difficulty that high placebo response rates in medication trials present. When pill placebos are indistinguishable from the real drug, many participants receiving them will be encouraged by the hope or expectation that they are receiving the active treatment, a phenomenon which is much less likely to occur for some control interventions used in trials to evaluate psychotherapy, most obviously the allocation to a ‘waiting list’. Bandelow et al.’s (2015) meta-analysis of all treatment arms in trials of three anxiety disorders (GAD, panic disorder and social anxiety disorder) reported that pre–post effect sizes of serotonin–noradrenaline reuptake inhibitors (SNRIs; 2.25, 23 trials) and selective serotonin reuptake inhibitors (SSRI; 2.09, 62 trials) are markedly greater than those for mindfulness therapies (1.56, 4 trials) and individual CBT (1.30, 93 trials) – the latter ranking a little above pill placebo, which had an effect size of 1.29 across 111 trials. Psychological ‘placebo’ interventions had a pre–post effect size of 0.83 (16 trials), but for ‘waiting list’ the figure was only 0.20 despite being a common control condition used in as many as 50 trials. These figures are illuminating, although we acknowledge that pre–post analyses are unconventional and the authors recognised several limitations, including heterogeneity and the small number of head-to-head trials comparing medication and psychotherapy.

Bandelow also reports an impressive pre–post effect size for pregabalin of 2.30 across eight trials in GAD or social anxiety disorder. Although pregabalin is discussed towards the end of the anxiety guidelines, it is surprising that the authors focused on SSRIs and SNRIs. Pregabalin is rated as a first-line treatment for GAD in the current Canadian clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive–compulsive disorders. It is described as an ‘alternative first-line treatment’ for GAD in the latest British Association for Psychopharmacology guidelines. It is licensed for this indication in the United Kingdom and many other European countries, although not in Australasia or North America. Readers of this journal will be aware of the issue of disconnect between clinical evidence and government licencing. Although there are some concerns about pregabalin’s potential for abuse, in our experience this is rare in people who have no history of substance abuse and we have found pregabalin to be generally well tolerated. It has trial evidence underpinning two further aspects of its utility in GAD. First, pregabalin can be used to augment the action of antidepressants; it was significantly more effective than placebo when added to an existing SSRI or SNRI drug that had been associated with partial response (Rickels et al., 2012). Second, introduction of pregabalin allowed benzodiazepine users to control anxiety more effectively than placebo when tapering off benzodiazepines (Hadley et al., 2012), although the difference in the proportion of individuals succeeding in stopping benzodiazepines entirely did not reach significance.

Benzodiazepines remain widely prescribed in Western countries and Bandelow’s data attest to their efficacy in anxiety disorders being comparable to the other classes of medication we have discussed (pre–post effect size 2.15, 42 trials). In our experience of treating anxiety disorders in secondary and tertiary care, a substantial minority of patients are already established as regular users of benzodiazepines at the time of referral. Often, therefore, the issue for the psychiatrist is not whether benzodiazepines should be started (as discussed in the guidelines in this issue), but rather whether their continuation can be justified, and if not, how ardently the prescriber should seek alternative ways of controlling anxiety symptoms that might allow the benzodiazepines to be tapered off. Some authors take a strict view that, despite their effectiveness, longer term benzodiazepine use is always undesirable for reasons of tolerance, propensity to dose escalation and increased risk of falls and fractures. Others (Baldwin et al., 2013) acknowledge that in certain circumstances continuation of benzodiazepines which have controlled anxiety disorders may be acceptable provided that there are periodic attempts to achieve dose reduction. We note with interest the supportive position the guidelines in this issue adopt with advice that ‘maintenance with benzodiazepines could be considered’ in treatment-refractory anxiety disorders.

Anxiety disorders are common and frequently disabling. We therefore hope that the guidelines published in this issue of the journal will encourage both careful diagnosis and judicious selection of treatment from the many effective strategies outlined. We also look forward to examining the growing evidence base of the effectiveness of guided digital CBT proposed by the authors of these guidelines.