I think benzos have their role including occasionally in the long term. It is important to know the indication, and choose patients appropriately. I don't use PRN benzos on an outpatient basis as this reinforces avoidance and dysfunctional beliefs about emotions being harmful or dangerous which only worsens or exacerbates anxiety states in the long run. Also PRNs aren't all that helpful as the anxiety/panic has already dissipated by the time the drug kicks in leading patients falsely to assume the drug was what helped. I do think the use should be very limited, but the problem is once you start someone on a prescription it can be hard to stop it, and that is why the best thing to do is to avoid prescribing them as far as possible. If you do use them, you should anticipate the possibility of having to continue them long-term and be okay with this. I have my long-term patients sign a contract and provide informed consent regarding the risks of increased mortality, accidents, dependency, withdrawal which can be fatal, and lack of evidence for long-term effectiveness, as well as no early refills and instances which will lead to tapering (diversion, doctor shopping, abuse of other substances, failure to attend appointments, lost/stolen prescriptions, escalating use) However there are a number of cases where it would be indicated to use benzos:
- patients with mania or hypomania, particularly to reduce the burden of neuroleptics which are far more neurotoxic than benzos. In patients declining other drugs it may be acceptable to use clonazepam alone
- catatonia, including periodic catatonia, and catatonia in the absence of major mental illness (for example in patients with neurodevelopmental disorders and intellectual disability)
- management of REM-sleep behavior disorder
- patients with severe OCD who have debilitating anxiety and have not responded to or been able to tolerate ERP and anafranil. Personally would prefer to use benzos than risperidone or other antipsychotic as adjunctive treatment
- patients with traumatic brain injury with significant agitation or aggression. There is risk of paradoxical reaction but it can often be helpful
- patients with agitation and aggression in the setting of advanced dementia (may be used as an adjunct or instead of neuroleptics- neither is ideal)
- patients with complex PTSD (for example from genocides, prisoners of war etc) who are still experiencing current threat. These patients do not really respond to PE or CPT and may not respond or tolerate SSRIs, SNRIs, mirtazapine, trazodone, or prazosin. They may respond somewhat to narrative approaches and then there are interpreter challenges too. We often manage these patients with significant improvement on as little as 0.5mg clonazepam which they can stay on for years without dose escalation
- patients with chronic psychotic illness who do not respond to, cannot tolerate, or choose not to take neuroleptics
Richard Balon wrote a short defense of Benzos in
JAMA Psychiatry