01. Benzodiazepine Tapering: Key Takeaways From New Guidelines

Text version

When I recommend to a resident or a fellow that we might initiate a benzodiazepine for a patient on the consult service, I’m always amazed by their response. It’s usually either complete shock like I’ve just suggested we shoot the patient or it’s a sense of embarrassed relief as if to say, wait, that’s an option? It could really be that simple?

Our trainees are learning psychiatry at a time when there’s a lot of discussion about deprescribing controlled substances, particularly in populations that may be susceptible to side effects, especially older patients. In general, it’s a really good thing for us to be mindful of the potential negative effects of our medications. But I worry sometimes that the message trainees are hearing is not “be cautious and thoughtful” but instead “don’t ever, ever prescribe a benzodiazepine for anyone and if you see anyone already prescribed a benzodiazepine, get them off of it immediately.”

Into this landscape comes a joint clinical practice guideline on benzodiazepine tapering issued last year by a combination of organizations including the American Psychiatric Association and published in the Journal of General Internal Medicine.

Guidelines Target Physical Dependence

The guidelines were led by the American Society of Addiction Medicine and the authors are a mix of MDs, APPs, and pharmacists including some psychiatrists. Importantly, these guidelines are not specifically targeted towards older patients, though the document does note that older adults actually have the highest rates of benzodiazepine prescriptions despite the increased risk in that population.

The stated goals of the guidelines are to help clinicians decide when it is appropriate to taper benzodiazepines and how to appropriately taper them in the setting of physical dependence. The guidelines are extensive, nearly 45 pages long, so we won’t be able to unpack everything today.

Risk-Benefit Analysis Before Tapering

The first major focus of the discussion is the idea of conducting a careful risk-benefit analysis before deciding whether to taper or discontinue benzodiazepines for a given patient. Understandably, the authors aren’t able to provide concrete and specific recommendations about which patients should be deprescribed. But they recommend considering risks, benefits, and alternatives to ongoing benzo treatment as well as considering risks of the taper itself.

This latter point is important. Tapering is not without risks. A retrospective cohort study even indicated that the mortality rate among patients who discontinued benzodiazepines over a six-month period was 1.6 times higher than that for patients who continued use.

The authors do specifically recommend a taper rather than abrupt discontinuation for any patient likely to have developed physical dependence and they provide some practical guidelines for determining the likelihood of physical dependence. They note, for example, that even some patients taking benzodiazepines for less than a month will be physically dependent while others taking the medication for up to six weeks may not be.

Likely factors besides length of treatment that may influence the risk for dependence include:

• The specific benzodiazepine and the dose
• Frequency of use
• Age
• Co-occurring conditions
• The use of other medications

Practical Taper Recommendations

In situations where a decision has been made to initiate a taper, the guidelines provide some recommendations for conducting the taper. These include things like consideration of replacing a short-acting benzodiazepine with a longer-acting benzo and then tapering the longer-acting agent, closely monitoring for withdrawal symptoms, considering a pause at times during the taper, and offering adjunctive interventions or medications. As is the case with many guidelines, a lot of this stuff is pretty common sense.

Emphasis on Slow and Gradual Tapers

So what are some important take-home points for psychiatrists and what are some places where the guidelines could be better? I like the fact that the authors emphasized that tapers should be extremely slow and gradual, especially for patients who had been on benzos for years.

In the outpatient setting, they recommend that the initial pace of the taper should generally include dose reductions of 5% to 10% every two to four weeks, typically not ever exceeding 25% every two weeks.

Inpatient Tapers Risk Bad Outcomes

As a CL psychiatrist, one of the trends that concerns me is the desire to get every patient completely off of benzodiazepines during their brief medical admission. I get it. I was this way as a fellow too. It drove me crazy to see an older patient come in on Xanax 1 mg q.i.d.

Over time though, I have seen a lot of bad outcomes when benzodiazepines are tapered too quickly in the acute setting. Sometimes, those bad outcomes are even worse than the concerns that prompted the desire to get the patient off of benzos in the first place. Sometimes, those bad outcomes become obvious during the admission but other times they don’t occur until days to weeks after discharge and inpatient psychiatrists may never hear about them. For that reason, I’m a big advocate for slow tapers and I push back against the idea that we absolutely have to get a patient completely off of benzos before they leave the hospital, even in cases where the benzos are likely contributing to adverse outcomes.

Coordinate With Outpatient Prescribers

Instead, I think the first thing we should be doing on the inpatient side is speaking with the outpatient prescriber. If the outpatient prescriber isn’t on board with the idea of deprescribing, we are probably doing harm by subjecting the patient to the risks of a taper in-house knowing that they are just going to go back on the same dose when they leave.

Now, this doesn’t mean we should just acquiesce to the outpatient provider’s wishes. We should absolutely make the case for deprescribing and consider a slightly reduced dose in-house if the outpatient prescriber declines, but we probably shouldn’t pursue a more aggressive taper in that case.

If we do start to taper in-house, I tend to go pretty slowly, no more than 25% every few days to a week. That’s quicker than the outpatient guidelines but again we’re monitoring patients a little more closely. It may be that the only thing I can accomplish during an admission is going from 4 mg of Xanax daily to 3.5 mg, but I would still consider that a victory. As the guidelines note, it may take months to years to fully taper benzos, especially if the patient has been taking a high dose for an extended period of time.

Not Everyone Benefits From Full Discontinuation

I also appreciate that the authors emphasized that not everyone needs to be completely tapered off of benzodiazepines. Some people really benefit from maintenance benzos and a dose reduction may be a very appropriate goal in those patients.

Interestingly, when they mentioned cases in which long-term benzo use may be warranted, the authors highlight severe generalized anxiety disorder and bipolar disorder but not, for example, panic disorder.

I would argue that panic, agoraphobia, and specific phobias are some of the most important situations where patients may derive very unique benefit from benzos and may benefit from intermittent or continuous maintenance therapy, though recognizing that the general recommendation for the vast majority of patients with those conditions is for short-term use. I do of course appreciate that the authors include catatonia on the list of conditions which may warrant long-term benzos.

Not All Benzodiazepines Are Equal

One thing I wish the guidelines had done a better job of is acknowledging that not all benzodiazepines are created equal. I wish the authors had done more to help clinicians understand some of the unique properties of certain benzos.

As psychiatrists, I think it’s important for us to familiarize ourselves with all of the benzodiazepines available, even the ones that we rarely encounter. Knowing that triazolam may have unique properties as a treatment for initial insomnia because of its incredibly short half-life and lack of hangover is important. Someone who has taken triazolam for sleep for years may not get the same benefit from a longer-acting benzodiazepine like clonazepam and may immediately complain of insomnia if we attempt to switch them before tapering. In that case, a very gradual taper of the triazolam itself may actually be preferable.

It’s also important to understand that while most benzodiazepines are cross-reactive in terms of covering withdrawal, there are cases of benzos with an extremely high affinity for the GABA receptor, like alprazolam, not cross-reacting with other benzos, like lorazepam, and still precipitating withdrawal in the setting of a substitution taper.

I also often highlight for trainees that clonazepam withdrawal has burned me a number of times, both because symptoms can be very delayed, up to two to three weeks in some cases, and because the objective vital sign changes we expect to see are not always apparent as the presenting sign. I’ve had several cases of patients in clonazepam withdrawal who present with a bizarre delirium without other obvious signs of withdrawal and EEGs are consistent with benzodiazepine withdrawal and resolution only occurs when clonazepam is reinstituted.

Guidelines Minimize Benzo Benefits

Perhaps my biggest quibble with the guidelines though is this degree of hysteria surrounding the evils of benzodiazepines and minimization of their benefits. This comes back to what I was saying in the opening.

Early on, the authors highlight the many risks of benzodiazepines. These include falls, motor vehicle accidents, cognitive impairment, delirium, overdose and death, the latter occurring particularly when benzos are used in combination with opioids. The authors don’t really spend much time discussing the potential benefits.

In one of their 10 takeaway messages, the guidelines state clinicians should taper benzos in most adults over age 65 unless there are compelling reasons for continuation. I want to take a step back here. We have to be humble about what we think we know.

Evidence Challenges Conventional Wisdom

The repeated assertion that benzodiazepines increase fall risk, which seems to be pretty universally accepted at this point, was challenged pretty strongly by a study we looked at last year showing that patients actually had reduced falls in the time period immediately following benzo prescription and suggested that it may be insomnia itself rather than benzos that increases fall risk, highlighting a good example of indication bias.

And the idea that benzos cause cognitive decline was challenged by a 2020 study in the American Journal of Psychiatry that suggested that cumulative use may even be neuroprotective. That study was accompanied by an editorial by my former chief, Jerry Rosenbaum, who pointed out a tendency to trivialize the distress that accompanies anxiety and emphasized something that the guidelines today seem to miss.

While it’s all well and good to recommend CBT as an alternative to benzos for anxiety, the reality is that most patients can’t access CBT, even those who are treated at academic medical centers with large psychiatry departments. The guidelines make no mention of these studies challenging conventional wisdom about the downside of benzos and instead accept the risks as gospel. Now, I’m not saying those risks don’t exist. I’m just saying we have to be careful about our assumptions.

Bottom Line for Clinicians: Hold the Dialectic on Benzos

My biggest concern is that we now have at least two generations of psychiatrists who are completely afraid of an entire class of medications. I know because I was once one of them.

My perspective didn’t change until I took over in our HIV clinic from someone who was very comfortable prescribing benzodiazepines and I recognized that the clinical reality was much more nuanced than I had been taught, with huge upside for some patients and the vast majority of people using them responsibly, even in a population vulnerable to co-occurring personality disorders and substance use.

We have to be cautious about allowing the pendulum to swing too far in the opposite direction. Yes, benzodiazepines are addictive substances that may contribute to a variety of negative effects especially in some populations, but they are also unique and uniquely effective medications representing one of the most powerful classes of agents that we have in our armamentarium. As psychiatrists, we need to hold this dialectic and make sure that we are sending the right messages to our trainees. We can’t have a generation of psychiatrists afraid of ever prescribing a benzodiazepine.

Abstract

Joint Clinical Practice Guideline on Benzodiazepine Tapering: Considerations When Risks Outweigh Benefits

Brunner, E., Chen, CY.A., Klein, T. et al.

Description

The American Society of Addiction Medicine (ASAM) has partnered with nine other medical societies and professional associations representing a wide range of clinical settings and patient populations to provide guidance on evidence-based strategies for tapering benzodiazepine (BZD) medication across a variety of settings.

Methods

The guideline was developed following modified GRADE methodology and clinical consensus process. The process included a systematic literature review as well as several targeted supplemental searches. The clinical practice guideline was revised based on external stakeholder review.

Recommendations

Key takeaways included the following: Clinicians should engage in ongoing risk–benefit assessment of BZD use/tapering, clinicians should utilize shared decision-making strategies in collaboration with patients, clinicians should not discontinue BZDs abruptly in patients who are likely to be physically dependent and at risk of withdrawal, clinicians should tailor tapering strategies to each patient and adjust tapering based on patient response, and clinicians should offer patients adjunctive psychosocial interventions to support successful tapering.

Reference

Brunner, E., Chen, CY.A., Klein, T. et al. (2025). Joint Clinical Practice Guideline on Benzodiazepine Tapering: Considerations When Risks Outweigh Benefits. Journal of General Internal Medicine 40, 2814–2859.