How Low Should a Benzodiazepine Taper Go Before You Jump Off?

Key Takeaways

• The last step off a benzodiazepine should feel like just another step, not a leap.
• A practical rule: stop when the dose that is left is no bigger than the steps already being handled comfortably.
• Receptor occupancy is the science behind well-built schedules, not something a patient can feel or measure.
• Symptoms after each step are the real guide to whether to continue, hold, or go back up.
• Cutting a fixed share of the current dose makes the steps shrink forever, so near the bottom the steps should stay the same small size.
• With diazepam oral solution, that usually means steps of about 0.1 mg and a final step from 0.1 mg to zero.
• A crumb of Klonopin, Xanax, or Ativan is still a big step to zero, so the end usually needs a liquid or a crossover to Valium.

How low should a benzodiazepine taper go before you jump off?

How low is low enough? When is the dose small enough to simply stop? After months of careful cuts, this is the question that keeps many people awake near the end of a benzodiazepine taper.

A common version comes from families. Someone has been helping a family member step down a Klonopin (clonazepam) dose, the dose is now very small, and nobody is sure whether the last crumb can just be dropped. Unfortunately, the answer from a doctor is often a shrug, or a flat instruction to stop.

The answer applies to every benzodiazepine, whether Klonopin, Xanax (alprazolam), Ativan (lorazepam), or Valium (diazepam). Valium has its own article on the last few milligrams of a Valium taper, including palpitations and BIND (benzodiazepine-induced neurological dysfunction) near the end. So, how does a person know when the final step is safe to take?

What does “jumping off” mean?

Jumping off is the final step of a taper, going from the last small dose to zero. Every other step leaves some of the drug behind. The jump leaves none.

The Maudsley Deprescribing Guidelines (chapter 3, p. 334) put it plainly: “Final doses before complete cessation will sometimes need to be very small (often less than 1mg of diazepam equivalent).” For many people, the end really does feel like standing at an edge. Yet, the jump does not have to be a leap.

Why does the last step cause so much trouble?

There are two reasons, and both are built into how tapers are usually done. The first is tablets. Most people finish on the smallest tablet, or a piece of it, and then stop. The Maudsley Deprescribing Guidelines (chapter 3, p. 338) warn that “Reducing from the smallest tablet to zero would cause a large reduction in effect, engendering severe withdrawal symptoms for some patients.”

The second is percentages. Many tapers cut a fixed share of the current dose, which is a sound way to begin, because each cut shrinks as the dose shrinks. The catch is that a fixed share never reaches zero. A share of a small dose is a smaller dose, and a share of that is smaller still, forever. At some point the person has to stop, and that last step, from whatever is left to nothing, is far bigger than the steps before it.

While this is not a perfect analogy, think of a staircase. Imagine walking down a long flight where every step is a comfortable height, and then the last step drops a full floor to the ground. Nobody would call that a staircase. They would call it a staircase with a hole at the bottom.

The fix is not more courage at the bottom. The fix is building the last few steps the same height as the ones before them.

Receptor occupancy is the science behind the schedule, not something to measure.

Many people hear the term receptor occupancy and find it confusing. That is understandable, and nobody needs to master it to follow the advice here.

Benzodiazepines work by boosting gamma-aminobutyric acid (GABA), the brain’s main calming messenger, at the GABA-A receptor. The Maudsley Deprescribing Guidelines build their schedules so that each step removes about the same slice of the drug’s effect on those receptors. The first milligrams removed from a high dose barely change that effect. The last ones change it a lot.

That is why the steps get smaller toward the end. It is also why, near the very bottom, where dose and effect run close to a straight line, steps of the same size in milligrams remove roughly the same slice of effect.

Occupancy cannot be felt or measured at home. It is the engineering inside a good schedule. What a patient actually uses is a schedule that already builds it in, their own symptoms, and a plain rule for the last step.

The last step should feel like just another step.

Dr. Leeds explains the principle this way: stop when the dose that is left is no bigger than the steps already being handled comfortably. That is his plain-language translation of the Maudsley guidance, which reads, “Reduce dose to zero when the reduction is no larger than previously tolerated reductions in terms of receptor occupancy” (chapter 3, p. 351).

In practice, the end of a taper has three parts.

1. Higher up, each cut is a proportion of the current dose, so the steps shrink as the dose shrinks.
2. Once the steps reach the smallest amount that can be measured reliably, they stop shrinking, and the person keeps taking steps of that same size.
3. When the dose that is left is one of those steps, the final step is simply the next one.

Done this way, there is no hole at the bottom of the stairs. The jump off is the same height as the step before it.

How do the steps change in a real schedule?

How the steps change in the slowest diazepam schedule in the Maudsley Deprescribing Guidelines (chapter 3, pp. 409 to 410). It is an illustration, not a plan.

• Around 29 mg: steps of 1 mg, about one thirtieth of the dose.
• Around 6 mg: steps of 0.4 mg, about one fifteenth of the dose.
• Around 4 mg: steps of 0.2 mg, about one twentieth of the dose.
• From 2 mg down to zero: 0.2 mg every step, which is one tenth of the dose at first, rising to one half, then all of it.

In terms of the drug’s effect on the brain, though, each of those last steps, including the final one, is about the same size.

What does this look like with diazepam liquid?

Diazepam oral solution is a manufactured product, not a compounded one, with 1 mg in each mL. On an oral syringe, 0.1 mL is 0.1 mg, which is about the smallest amount most people can measure reliably.

So, in Dr. Leeds’ practice the last stretch of a diazepam taper is usually steps of about 0.1 mg, and the final step is 0.1 mg to zero. That is slightly gentler than the Maudsley schedule above, which stops from 0.2 mg. The stopping point is agreed in advance, so the tail has an end.

Professor C. Heather Ashton, whose Ashton Manual taught a generation of patients how to taper, advised: “Take the plunge when you reach 0.5mg daily” (Chapter II). By Dr. Leeds’ rule, her final jump was no bigger than her last steps, which were also half a milligram. In Dr. Leeds’ view, her advice reflected the 2 mg diazepam tablets she taught from.

The problem is that half-milligram steps at that level are large for many people. At 1 mg, a half-milligram cut is half the dose. Professor Ashton’s jump remains an option for someone who is doing well and chooses it, since diazepam’s long half-life softens the last step, and the companion article on the last milligrams of Valium weighs that choice. It is not the default.

What about Klonopin, Xanax, and Ativan?

Unfortunately, the more potent benzodiazepines make the last step harder. A crumb of Klonopin, Xanax, or Ativan is still a big step to zero. The Maudsley Deprescribing Guidelines (chapter 3, p. 338, Table 3.8) show that a quarter of the smallest clonazepam or lorazepam tablet still acts on more than twice as many GABA-A receptors as a quarter of the smallest Valium tablet.

Fortunately, there are two ways to make the end gentle. The first is a crossover taper to diazepam for the last stretch, then the diazepam liquid, which is Dr. Leeds’ usual route and follows the reasoning behind why the Ashton Manual recommends switching to diazepam. The second is a liquid of the person’s own medication.

In the United States, alprazolam and lorazepam are both made as oral solutions. Clonazepam is not, so the choices are a compounded liquid or a crossover to diazepam. The slowest clonazepam schedule in the Maudsley Deprescribing Guidelines (chapter 3, p. 394) finishes on liquid at a tiny fraction of a milligram, which shows how low the tail can go, not a plan to copy.

Symptoms decide the pace, not the calendar.

Schedules are drawn in advance. Bodies are not. After each step, the Maudsley Deprescribing Guidelines (chapter 3, pp. 351 to 353) describe three choices: continue if the step was tolerable, hold if symptoms need time to settle, or go back up if they do not.

Holding is not failure. Neither is going back up a little after a step that proved too large. Physical dependence is the nervous system’s adaptation to a medicine taken as prescribed, and dependence is not addiction. Needing a slow finish says nothing about character.

None of this means the end can be rushed. The 2020 boxed warning update from the United States Food and Drug Administration (FDA) warns that stopping abruptly or cutting the dose too quickly can cause serious withdrawal reactions, and it calls for a gradual taper. That is also why even a few weeks of regular use usually calls for a short taper rather than stopping cold turkey.

Work with your prescriber on every step, including the last one. Seizures, sudden confusion, or hallucinations are emergencies at any point, so call 911 or go to the nearest emergency room. Anyone in emotional crisis can call or text the 988 Suicide & Crisis Lifeline.

What else do people ask about the last step?

How does someone know the dose is low enough to stop?

The dose is low enough when what is left is no bigger than the steps they have already handled comfortably. If the last few steps were 0.1 mL of diazepam liquid and each one settled well, a final 0.1 mL is just another step. If the remaining crumb is much larger than any recent step, it is not the last step yet.

Can Klonopin be tapered to zero without switching to Valium?

Yes, with a compounded clonazepam liquid, which allows steps far smaller than any piece of a tablet. Many patients find the crossover to diazepam easier, but that is a choice made with the prescriber, not a requirement.

Is it normal to feel symptoms after the final step?

Often, yes. Symptoms after the last dose tend to ease with time, usually in windows and waves, where better stretches alternate with harder ones. The article on life after the benzodiazepine taper describes what the months after the last dose can look like.

The end of a taper can be quiet.

Most people picture the jump off as a dramatic moment. When the steps are built well, it is closer to an ordinary Tuesday. The dose that was there yesterday is simply not there today, and the body barely notices.

Dr. Leeds is an osteopathic physician and deprescribing specialist in Fort Lauderdale who helps patients throughout Florida taper off of benzodiazepines by telemedicine, working with each patient directly at weekly video visits. In his practice, the stopping point is planned together, the last steps are sized to what each patient has already handled, and the pace follows the patient’s symptoms rather than a calendar. Contact Dr. Leeds to talk about planning the end of a taper.

This article is educational. It is not medical advice, and reading it does not create a doctor-patient relationship. Decisions about starting, continuing, or tapering any medication should be made with your own physician.