Representation of Benzodiazepines in Treatment Guidelines: The Paradox of Undesirable Objectivity

Various treatment guidelines (e.g., the American Psychiatric Association (APA) practice guidelines for the treatment of panic disorder [5], guidelines from the British Association for Psychopharmacology (BAP) [6], Canadian clinical practice guidelines [7], and the Royal Australian and New Zealand College of Psychiatrists (RANZCP) clinical practice guidelines [8]) seem to be unanimous when it comes to pharmacotherapy of anxiety disorders. They all recommend SSRIs and SNRIs as the preferred treatment or first-line and even second-line treatment for panic disorder, generalized anxiety disorder, and social anxiety disorder. The role of benzodiazepines varies somewhat between the guidelines, but it is always secondary to SSRIs and SNRIs. The reason for this status of benzodiazepines is not in their efficacy as most guidelines acknowledge that these drugs are effective. Rather, according to the guidelines, the key problem with benzodiazepines is in their adverse effects (i.e., sedation, interference with psychomotor performance, falls among the elderly, and negative effects on cognitive functioning) and association with dependence issues, such as difficulty discontinuing [5-8]. Some guidelines also mention that tolerance develops with benzodiazepine use [6, 8], that benzodiazepines can be abused [5, 7, 8], and that there is a significant risk of relapse after their discontinuation [8]. Therefore, guidelines suggest that SSRIs and SNRIs are preferred over benzodiazepines because these medications are likely to be less risky, i.e., they are better tolerated and without a dependence-producing potential.

In accordance with this general attitude towards benzodiazepines, guidelines make restrictive recommendations about their role in anxiety disorders. The usual suggestion is that benzodiazepines can be administered to patients who have not responded to the treatment of choice or are otherwise treatment-resistant. However, more often than not, treatment resistance is not clearly defined, resulting in different recommendations. For example, APA guidelines [5] suggest that benzodiazepines can be used in patients with panic disorder after a failure to respond to two different SSRIs or SNRIs. In the BAP guidelines [6], benzodiazepines are considered for generalized anxiety disorder “after a non-response to SSRI, SNRI, pregabalin and buspirone treatment” (p. 417). The RANZCP guidelines [8] go a step further and suggest two SSRIs, one SNRI, combination with cognitive-behaviour therapy, and yet another medication (a tricyclic antidepressant for panic disorder, irreversible inhibitor of monoaminoxidase for social anxiety disorder, and pregabalin or agomelatine for generalized anxiety disorder) before administering a benzodiazepine.

Another common recommendation about benzodiazepines is that they should generally be used short-term [6-8]. In other words, guidelines advise against long-term treatment of anxiety disorders with benzodiazepines. Finally, guidelines stipulate that benzodiazepine monotherapy for panic disorder can be used “only in the absence of a co-occurring mood disorder” ([5], p. 23), that benzodiazepines may be warranted to manage severe anxiety, agitation or panic attacks [5, 7, 8] and that they can be combined with SSRIs to alleviate exacerbation of anxiety caused by SSRIs [5, 7, 8] or “target residual symptoms” ([5], p. 12).

As already noted, the key argument about benzodiazepines in treatment guidelines is that despite their efficacy, they are troublesome because of their adverse effects and dependence-related issues, especially when compared with SSRIs and SNRIs. This argument could be supported by two lines of evidence. The first would be based on demonstration of poorer tolerability of benzodiazepines than SSRIs and SNRIs. The second would relate to a finding that benzodiazepines are more difficult to discontinue than SSRIs and SNRIs or that withdrawal symptoms after benzodiazepine discontinuation are more severe than those that occur upon cessation of SSRIs and SNRIs. Only a few studies have directly compared benzodiazepines, SSRIs, and SNRIs for anxiety disorders, but there is no evidence to support the central argument about benzodiazepines in the guidelines. In fact, the evidence suggests the opposite – that benzodiazepines are associated with fewer adverse effects and lower dropout rates and that they are better tolerated than SSRIs and SNRIs in the treatment of anxiety disorders [9-12], whereas withdrawal symptoms after cessation of SSRIs and SNRIs are similar to benzodiazepine withdrawal symptoms [13] and can be at least as severe [14-16].

Other concerns about benzodiazepines that do not make them a preferred treatment option according to the guidelines can similarly be questioned. Thus, in the absence of a history of substance abuse or concurrent substance abuse, there is substantial evidence that tolerance to anxiolytic or antipanic effects of benzodiazepines during a long-term treatment of pathological anxiety does not tend to occur and therapeutic benefit is maintained without increasing the dose [10, 17-22]. Likewise, benzodiazepine abuse in patients with anxiety disorders is rare in the absence of other substance abuse [23, 24]. Furthermore, it has not been confirmed that relapse rates after discontinuation of benzodiazepines are higher than relapse rates after discontinuation of SSRIs or SNRIs, and one study showed that clonazepam in the long-term treatment of panic disorder predicted lower relapse rates than paroxetine [25]. Finally, studies of long-term treatment of anxiety disorders with benzodiazepines did not indicate that such use of benzodiazepines was problematic [10, 18, 20, 25].

Given that most recommendations about benzodiazepines in treatment guidelines are not evidence-based and therefore seem arbitrary, it is not surprising that there is a discord between these guidelines’ assertions and what occurs in clinical practice. In this context, it is telling that benzodiazepines remain widely used around the world [26-30]. While such data are often interpreted in an alarmist way, as an indicator that guidelines are not followed and that “dangerous” medications continue to be prescribed indiscriminately, they suggest that in the realm of benzodiazepines, guidelines may be irrelevant for clinical practice. Moreover, it seems that clinical practice is ahead of treatment guidelines in this realm [31], so that guidelines may need to follow clinical practice, not the other way around.

Most SSRIs and SNRIs were introduced in the 1990s and early 2000s. At that time, pharmaceutical industry organized and sponsored numerous activities to promote these agents [32] and thereby influence prescribers, healthcare organizations, and professional bodies. Moreover, SSRIs and SNRIs were specifically marketed as having advantages over benzodiazepines for anxiety disorders in terms of their alleged greater efficacy and safety and especially their purported inability to cause dependence [14, 33]. Thus, pharmaceutical industry directly contributed to the notion that SSRIs and SNRIs are superior to benzodiazepines in the treatment of anxiety disorders. This notion was subsequently incorporated into treatment guidelines, arguably because many of their authors were unduly influenced by the pharmaceutical industry and had undeclared or hidden conflicts of interest [34, 35]. As already noted, it is remarkable that these claims about SSRIs and SNRIs were made without any studies demonstrating their advantages.

SSRIs and SNRIs are no longer “new” and have been manufactured as generic drugs for more than a decade now. Profit-driven motivation to promote these medications has decreased somewhat, but a tendency to favour them over benzodiazepines persists, despite research evidence that does not support their superiority. Therefore, it would be illuminating to better understand this phenomenon.

Repudiation of benzodiazepines may have become a habit for many physicians, something that they had learned and feel reluctant to change, especially because there is no inducement from professional bodies to change. For decades, medical students have been primed to view benzodiazepines with suspicion and after graduating, this is very likely to persist as the “right” attitude. With any new data on benzodiazepines, confirmation bias quickly sets in, and attention is paid selectively to information that confirms a negative opinion about benzodiazepines, while information that suggests the opposite is usually dismissed as based on methodological or other flaws. Similarly, some physicians tend to overemphasize disadvantages of benzodiazepines and downplay their potential benefits, while underestimating difficulties with SSRIs and SNRIs. For example, sedation with benzodiazepines is seen as potentially dangerous and their quick onset of action is deemed unimportant or related to their propensity to be misused, while agitation and increased anxiety at the beginning of treatment with SSRIs or SNRIs are brushed aside as trivial and transient. Likewise, benzodiazepine withdrawal symptoms are portrayed to the patients as potentially dangerous, but withdrawal symptoms after cessation of SSRIs or SNRIs are considered mild or insignificant. Physicians have been fed these narratives long enough that they have become ingrained [36]. Perhaps uniquely, benzodiazepines are also stigmatized [37, 38], most likely because of the notion that they are “addictive” and therefore should be avoided. Consequently, some physicians refuse to prescribe them, others feel embarrassed admitting that they do prescribe benzodiazepines, and yet others conceal this prescribing practice altogether. Instead of fostering a well-balanced approach towards benzodiazepines, “opinion makers” and guidelines have adopted a negativistic and sometimes moralizing stance. This hinders objective evaluation of the role that benzodiazepines can play.

The ongoing “crusade” against benzodiazepines is even more remarkable at the time when many psychedelic drugs or drugs with psychedelic-like effects (e.g., ketamine) are “rehabilitated” for the purpose of being tested in a variety of treatment-refractory conditions, including anxiety disorders [39]. Thus, we have a paradoxical situation, whereby psychiatric establishment seems to endorse psychedelic strategies that involve dissociation and “ego dissolution” but rejects anxiety-alleviating and calming effects of benzodiazepines. This is yet another indication that much of the aversion towards benzodiazepines is so impervious to reason that it can be considered irrational.

Treatment guidelines are supposed to be objective in offering recommendations and its creators should not have conflicts of interest or be influenced by personal bias. These basic principles have been violated in treatment guidelines for anxiety disorders because they present selective evidence about benzodiazepines, SSRIs, and SNRIs and are clearly biased towards benzodiazepines. Instead of improving clinical practice by making recommendations that reflect evidence, these guidelines unjustifiably deprive patients of a useful therapeutic modality and promote a potentially harmful overuse of antidepressants [31, 37]. It is small wonder, then, that guidelines are often ignored by prescribing physicians.

Benzodiazepines are at least as effective for anxiety disorders as SSRIs and SNRIs, with some studies reporting that they may be even more effective [39-42]. There is no evidence that benzodiazepines are less safe than SSRIs and SNRIs in both short-term and long-term treatment [9-12, 43] and that long-term treatment with benzodiazepines is fraught with unacceptable levels of risk [10, 18, 20, 25, 43]. Therefore, no reason justifies withholding of the positioning of benzodiazepines as the first-line, long-term pharmacotherapy option for anxiety disorders, with due acknowledgment that they are not suitable for everyone. Suggesting that this should be incorporated into treatment guidelines may still seem bold in 2022, although similar calls have already been made [31, 39, 44]. Ultimately, the key criterion for evaluating treatment guidelines is their contribution to patients’ well-being. If the authors of the guidelines for anxiety disorders refuse to fully consider both the evidence on which they should be based and the reality of clinical practice, it is likely that these guidelines will sink into a long-lasting irrelevance.

The author has no conflicts of interest to declare.

None.

V.S. is the only author of this editorial.