
Triazolam, sold as Halcion, is a benzodiazepine sleeping pill, and it is one of the shortest-acting members of the class. Its United States Food and Drug Administration (FDA) label approves it for the short-term treatment of insomnia, generally 7 to 10 days, and says prescriptions should not exceed a one-month supply.
That is the whole indication. It is not an anxiety medication, and it was never meant to be taken every night for years.
Yet, many people have taken it every night for years. The prescription was renewed, the sleep held for a while, and then it did not, and by the time anyone asked why, the body had adapted to the drug.
Stopping at that point is not a matter of willpower. It is a medical problem with a medical answer, and for a drug this short-acting the answer is usually not to reduce the triazolam itself.
A person who takes triazolam as prescribed and can no longer sleep without it does not have an addiction. In fact, the FDA said, when it added the boxed warning to the whole benzodiazepine class in 2020, that physical dependence can occur when these drugs are taken steadily for several days to weeks, even as prescribed.
Dependence is the nervous system’s adaptation to a drug that is always there. Addiction is a pattern of behavior, and most people on a nightly sleeping pill have none of it.
The distinction matters because the treatment has to match the condition. Physical dependence is not addiction, and a person with physical dependence needs a gradual medical taper, not a program, a meeting, or a bed in a facility.
People who search for “Halcion detox” usually mean one thing, which is how to come off the drug safely. The safe way is a supervised taper over months, not a rapid detox on a fixed timeline.
The Halcion label gives triazolam a mean elimination half-life of roughly 1.5 to 5.5 hours, and Professor C. Heather Ashton’s equivalence table in the first chapter of the Ashton Manual lists it at about 2 hours. A dose taken at bedtime is largely gone by morning.
Every night is therefore a complete cycle: the drug arrives, it boosts the calming effect of gamma-aminobutyric acid (GABA) at its receptors, the drug leaves, and the receptors are left waiting. With continued use the nervous system adjusts to that rhythm, and the adjustment shows up as withdrawal between doses.
The label itself notes an increase in daytime anxiety after as few as 10 days of continuous use and says that in some patients this may be interdose withdrawal. Rebound insomnia is the same effect at night: skip or cut a dose and the sleep is worse than before the drug was started. So, the drug ends up treating a problem it is now helping to cause.
A long-acting benzodiazepine keeps a fairly steady level in the blood, and a small reduction lowers that level a little for the whole day. Triazolam has no steady level to lower. Cutting the bedtime dose does not make the night gently lighter, it makes the second half of the night uncovered, and the person wakes at three in the morning in withdrawal.
The tablet does not help either. Halcion is supplied as a 0.25 mg scored tablet, so even a careful split is a large fraction of the dose. Splitting finer, or having the drug made into a liquid, changes the size of the step but not the shape of the night, because the drug is still gone before breakfast.
Think of triazolam as a light switch, on at bedtime and off by morning, and diazepam as a dimmer that can be turned down one notch at a time. A dimmer is what a taper needs.
The label is direct about this. Abrupt discontinuation or rapid dosage reduction of Halcion after continued use may precipitate acute withdrawal reactions, which can be life-threatening, and it names seizures among them. The 2020 FDA boxed-warning update says the same of the whole class and calls for a gradual taper.
This is why a fixed timeline of days or weeks is the exact harm to avoid, and why a person who has run out of tablets should call the prescriber rather than tough it out. A seizure is a medical emergency: call 911, and anyone in crisis or thinking about suicide can call or text 988 at any hour.
Chapter II of the Ashton Manual says there is absolutely no doubt that anyone withdrawing from long-term benzodiazepines must reduce the dosage slowly, and that abrupt or over-rapid withdrawal can bring on convulsions, psychotic reactions, and acute anxiety states. For most short-acting benzodiazepines, Professor Ashton’s answer to the half-life problem was to change the drug before changing the dose.
For triazolam itself she wrote, in the same chapter, that a 2-hour half-life leaves a person practically withdrawn every morning, so triazolam can be stopped without substitution of a long-acting benzodiazepine. If withdrawal symptoms follow, her remedy is a short course of diazepam from about 10 mg, reduced as in her Schedule 2.
Diazepam has a half-life of 20 to 100 hours, and its active metabolite lasts longer still, so a dose taken today is still working tomorrow. That is what makes it the Manual’s fallback even for triazolam.
Her table lists 0.5 mg of triazolam as approximately equivalent to 10 mg of diazepam, which puts the usual 0.25 mg bedtime dose near 5 mg. The Manual adds that these figures are based on clinical experience and vary between individuals.
The Manual’s exception was written for a two-hour drug taken as directed for a short time. The Halcion label, which now carries the 2020 boxed warning, says abrupt discontinuation after continued use may precipitate life-threatening withdrawal reactions and calls for a gradual taper.
When a person has taken triazolam nightly for years, Dr. Mark Leeds sides with the label, and with the Manual’s own fallback of diazepam, and does not stop the drug abruptly. The crossover is the taper’s first move, not a substitute for it.
The changeover is stepwise, one dose or part of a dose at a time. For a once-nightly drug, that means part of the triazolam dose is replaced with diazepam first, the new arrangement is held until it feels stable, and then the rest is replaced.
This is a crossover taper. The Manual’s reason for switching one dose, or part of a dose, at a time is that it also reveals the equivalent dose for that particular person, and the same logic applies to a single bedtime dose.
The Manual’s warning is about doing it all at once: too much diazepam too soon brings heavy daytime sleepiness, too little brings withdrawal, which is why the change is made in steps. A person who feels either during the switch should tell the prescriber, because the next step is adjusted from there.
Chapter II of the Manual describes the shape of the reductions rather than a fixed rate. Each decrement may be up to one tenth of the current dose, the steps come every week or two in the Manual’s example, and the person taking the drug is the best judge of when the next one is due.
The Maudsley Deprescribing Guidelines add a refinement. The relationship between dose and effect at the GABA receptor is hyperbolic, meaning the last few milligrams carry more effect than the first few, so reductions should get smaller as the dose falls.
Diazepam comes in small tablet strengths and as a liquid, so small steps are practical. None of this is a schedule to copy from a web page. It is a shape to work through with the prescribing physician.
Months, and sometimes longer. The Manual says withdrawal is usually carried out over a period of some months, and that whether it takes 6, 12, or 18 months is of little significance to a person who has taken benzodiazepines for years. Trying to beat that clock is one of the surest ways to raise the risk of protracted withdrawal.
Holds are part of the method, not a failure of it. The label’s own instruction is that if withdrawal reactions appear, the taper may be paused or the dose returned to the previous level, and then decreased more slowly.
If the pace is too fast, the doctor must help by advising a slower taper. If it is going well, patient and doctor may agree to go a little faster, as tolerated.
While a taper is in progress, sleep will be uneven, and the ordinary rules help. A fixed wake time, a dark and cool bedroom, no screens in the last hour, and no caffeine after midday all make the nervous system’s job easier, but they are support and should be treated as support.
Unfortunately, no bedtime routine will cure rebound insomnia on its own, and a person told to try harder at sleep hygiene instead of being offered a proper taper has been given the wrong thing. Fortunately, natural sleep does tend to return as the nervous system readjusts, usually lagging the taper, so bad nights in the middle do not mean the plan is failing.
Dr. Leeds is an osteopathic physician and deprescribing specialist in Fort Lauderdale, Florida, and his practice does one thing: medically supervised tapering of benzodiazepines and z-drugs. He works directly with each patient, without a team or clinic staff, on a concierge basis rather than through insurance, and only for patients in Florida.
For triazolam, Dr. Leeds generally prefers a diazepam crossover, then hyperbolic reductions, paced by the patient, rather than an abrupt stop.
Informed consent comes first: the risks of the taper, the alternative of holding at the current dose, an exit plan from the outset, and the right to decline or change course at any point. Patients direct the pace.
Details of the method are on the practice’s pages on medically supervised benzodiazepine tapering and the Ashton Manual approach.
Anyone in Florida who has taken triazolam far longer than the label intended can work through that plan with Dr. Leeds’ practice. Contact Dr. Leeds through the contact form to ask about a triazolam 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.

Mark Leeds, D.O. is an osteopathic physician and deprescribing specialist in Fort Lauderdale, Florida. He helps patients come off benzodiazepines and sleeping pills with a slow, patient-directed taper guided by the Ashton Manual and the Maudsley Deprescribing Guidelines. Physical dependence is not addiction, and a careful taper is measured in months, not days.
Contact Us for expert guidance on safely tapering off benzodiazepines and reclaiming your life.





