Deprescribing BZRAs (benzodiazepines and nonbenzodiazepines) for insomnia is an important topic that is not well covered in medical education. The issue becomes increasingly relevant in older age groups, as shown by Maust et al, who found that the rate of new benzodiazepine visits was relatively constant across age groups. There was, however, an increase in the overall benzodiazepine visit rate among older adults, which was largely due to the increasing rate of continuation visits. Patients aged 80 years and older have much higher rates of BZRA treatment, particularly over the long-term, and long‐term BZRA treatment is associated with increased risk of falls, fractures, motor vehicle accidents, and cognitive impairment. Still, many prescribers hesitate to attempt deprescribing BZRAs, often because the process itself is time consuming and is associated with such withdrawal symptoms as short-term insomnia, anxiety, and restlessness. 

In their review of practice guidelines, Lee et al included the recommendation that, in community or long‑term care settings where BZRAs are used for insomnia, a slow taper of BZRAs should be attempted for all adults aged 65 years and older. While deprescribing is important, it is also recognized as challenging. Patients want to sleep, and they may become distressed by not being able to sleep. It is important to set realistic expectations and to help patients recognize that there may be unwelcomed effects during tapering. Referring patients for short-term cognitive behavioral therapy can also be helpful; however, data showing success with this approach are limited.

On the other hand, it is important to keep in mind that some patients may require long-term BZRA treatment for insomnia and that it is possible to become overzealous with deprescribing. While some practitioners may consider many, if not most, cases of insomnia to be symptomatic of some other underlying disorder, this is not always the case. The Diagnostic and Statistical Manual of Mental Disorders (DSM–5) recognizes that clinical insomnia may be a primary disorder, not simply a symptom of psychiatric, neurologic, or medical illness. The US Food and Drug Administration has set no specific limit on the length of time that BZRAs can be used to treat patients with insomnia. Still, given the risks associated with the long-term use of BZRAs, discontinuation should at least be attempted in most cases, with accompanying tools and appropriately set patient expectations to help maximize the odds of success.