Benzodiazepines and Long-Term Anxiety Treatment: What the Evidence Actually Shows
Last reviewed: 09/10/2026
Reviewed by: Dr. Kiesa Kelly
You have a prescription for lorazepam, alprazolam, or clonazepam. You are about to start therapy that will ask you to approach the things you have been avoiding. Somewhere along the way — a forum, a friend, maybe a clinician — you heard that the two do not mix, and that the medication will quietly cancel out the work.
That claim is repeated confidently in a lot of places. When you follow it back to the research on benzodiazepines and long-term anxiety treatment, it turns out to be less settled than the confidence suggests. It is also not simply wrong. This article is an attempt to say what is actually known, what is still contested, and what you can reasonably do with that. One thing it is not: prescribing advice. Dr. Kelly is a clinical psychologist, not a prescriber, and nothing here is a reason to change how you take a medication.
In this article, you'll learn:
What the randomized-trial evidence does and does not establish
Why the answer looks different at the end of treatment than at follow-up
Three common misreadings of this question, and what is true instead
Why clinicians ask what the medication is doing in the moment, not just whether you take it
Specific questions to bring to your prescriber and your therapist
The short answer
There is no definitive evidence that taking a prescribed benzodiazepine makes exposure-based anxiety therapy fail. There is also no clean bill of health. The most rigorous review to date looked at 12 randomized clinical trials and concluded that, as of its publication, there was no definitive evidence that benzodiazepines hinder the efficacy of exposure-based interventions in adults with anxiety or post-traumatic stress disorders [1].
Read the results underneath that conclusion, though, and the picture has texture. At the end of treatment, benzodiazepines did not affect outcomes in nine studies, improved them in two, and reduced them in one. Eleven of the twelve trials reported follow-up — after the medication had been discontinued — and there, six showed no effect and five showed reduced benefit [1]. Every trial carried some concern or high risk of bias, and the authors graded the overall level of evidence as B rather than A.
So the honest summary is narrower than either camp usually states it: the blanket warning is not supported, and neither is reassurance. If you want a single sentence to carry out of here, it is that this is an open question being handled by people who know your case, and that evidence-based anxiety therapy is worth starting either way.
Key takeaway: ⚖️ The strongest available review found no definitive evidence of harm — and a follow-up signal that stops it being a clean all-clear.
Why this question keeps coming up
Benzodiazepines and exposure therapy arrived from different directions. Benzodiazepines act relatively quickly — how quickly depends on the specific drug — and they reduce the physical experience of anxiety. Exposure-based therapy works slowly and asks you to stay in contact with that experience long enough for your predictions about it to be revised. One turns the volume down; the other asks you to listen to the sound.
Put like that, an interference story writes itself, which is part of why it spreads so easily. The story is also plausible enough to have been taken seriously by researchers for decades — this is not internet folklore, it is a genuine mechanistic hypothesis that people have been trying to test in controlled trials.
What complicates it is that most people in this situation are not choosing between the two. They are already taking a medication a clinician prescribed, and now they are being asked to add therapy on top. If you are tracking your symptoms with a tool like the GAD-7, you may already be able to see the shape of what the medication does and does not touch.
Three things people get wrong here
"Taking a benzodiazepine cancels out exposure therapy." In reality, the trial evidence does not show that. Nine of twelve randomized trials found no difference in outcome at the end of treatment [1]. If this effect were large and reliable, twelve trials would be a reasonable place to have found it.
"If benzodiazepines aren't first-line, my prescriber made a mistake." In reality, first-line is a statement about what to try first in a typical case, not a verdict on any individual prescription. NICE tells clinicians not to offer a benzodiazepine for generalized anxiety disorder except as a short-term measure during crises, and not to prescribe one for panic disorder at all [3]. A review of 113 guidance documents from around the world found broad agreement that these medicines are not routine first-line agents, with recommended duration usually under four weeks [5], and a Japanese expert consensus panel reached the same conclusion for unspecified anxiety disorder specifically [4]. A short-term measure during a crisis is a real clinical category, and people are in it.
"The safest move is to stop the medication before I start therapy." In reality, this is the most dangerous of the three. The FDA's 2020 boxed-warning update states that physical dependence can occur when benzodiazepines are taken steadily for several days to weeks, even as prescribed, and that stopping abruptly or reducing the dose too quickly can produce withdrawal reactions including seizures, which can be life-threatening [2]. Nothing in this article is a reason to change a dose. That decision belongs to your prescriber, and starting therapy is a good reason to book that conversation rather than to pre-empt it.
Key takeaway: 🚫 Never taper or stop a benzodiazepine on your own initiative. The prescriber who started it is the person who changes it.
What the research actually found
The human trials
The 2020 systematic review screened 1,529 studies and included 12 randomized clinical trials [1]. That is the strongest body of human evidence available on this specific question, and its conclusion — no definitive evidence of hindrance — is the sentence most worth knowing. Its limitations are worth knowing too: bias concerns across every included trial, and a level-of-evidence grade of B rather than A.
What animal research adds, and what it cannot settle
Laboratory work in rats has produced the most specific mechanistic finding in this area. In a 2014 study, midazolam disrupted extinction learning when it was given before the first extinction session, but did not disrupt it when given before a second session after some drug-free extinction had already happened [6].
That result is genuinely interesting, because it suggests timing rather than presence may be what matters. It is also a study of rats and a benzodiazepine used as an anesthetic agent, and it cannot tell you what your prescription will do during your therapy. Treating it as human evidence would be a mistake. For what the human trials show, the 2020 review [1] remains the best available summary, and it did not find definitive evidence of hindrance.
Where the honest uncertainty sits
The gap between "no definitive evidence of harm" and "no harm" is where clinicians actually work. Nobody has run the trial that would settle it: a large, low-bias study of people with diagnosed anxiety disorders, on stable prescribed doses, randomized by dosing pattern, followed long enough to see whether early gains hold. Until that exists, anyone who tells you this question is closed is overstating what the literature supports in one direction or the other.
Key takeaway: 🔍 The end-of-treatment and follow-up results point different ways. That difference is the actual finding, not a footnote to it.
A question clinicians find more useful
There is a more useful question than does this drug interfere: what is the medication doing in the moment you take it?
In anxiety treatment, a safety behavior is an action taken to prevent, escape, or reduce the severity of a feared outcome [7]. These have traditionally been eliminated during exposure work, on the theory that they prevent the learning the exposure is meant to produce. That theory has been challenged, and the findings are genuinely mixed [7]. A randomized trial of the "judicious use" of safety behaviors during exposure found no significant differences in outcome or in how tolerable people found treatment [8], and a 2025 randomized trial found that keeping a safety behavior in place actually helped participants who were highly intolerant of distress, while removing it undermined exposure when it blocked what the person expected to happen [9]. Both of those trials studied spider fear, not medication, so the read-across is a suggestion rather than a result — no trial has compared pattern-of-use against drug presence.
Here is what that looks like in a life rather than a study.
You have a presentation on Thursday. On Tuesday you take your medication before the practice run with your therapist, and it goes well enough that you agree to try the real thing without taking it first. By Thursday morning the tablet is in your pocket "just in case," and you get through the presentation without it, but you touch the pocket four times to check. The exposure happened. What you learned from it is partly about the room and partly about the pocket, and only you and your therapist can work out the ratio.
Or: your prescription is scheduled rather than as-needed — the same dose every morning, whether or not anything is happening that day. You take it before therapy because you take it before everything. Nothing about that session is organized around the medication; it is simply background. That is a different situation from the pocket, and the fact that both are described as "taking a benzodiazepine before therapy" is most of why this question generates such confused advice.
This is also why guidance is thinner than you would hope. The 2024 review of international guidance documents found agreement on keeping use short, but inconsistency across documents on whether to use regular or as-needed dosing, and for which condition [5]. The distinction most relevant to your therapy is precisely the one the guidelines have not resolved.
Key takeaway: ⏱️ "What is the medication doing in the moment you take it?" is a question your therapist can work with. It has not been tested against the drug question in a trial.
How to think about your own situation
You are not going to resolve a contested literature before your next appointment. You can, though, sort yourself into a rough position:
If your dose is scheduled and stable and your therapy is progressing, the trial evidence gives you no specific reason to act on your own. Tell your therapist what you take and when, mention it at your next prescriber appointment, and get on with the work — any change is their call, not therapy's.
If you use it as needed and you notice you are timing doses around sessions or feared situations, that is the pattern worth naming out loud. Not because it proves interference, but because it is the version of this question your therapist can actually do something with.
If you are already thinking about coming off it, that is a prescriber conversation with its own timeline, and it does not have to happen before therapy starts. Doing both at once is a choice some people make well and others find overwhelming.
If therapy has stalled and you are wondering whether medication is the reason, there are several other well-recognized reasons anxiety therapy stops working, and they are worth ruling out alongside this one.
Questions worth asking your prescriber
Is my prescription intended as short-term or ongoing, and what would tell us it is time to revisit that?
Is as-needed or scheduled dosing a better fit while I am doing exposure-based therapy?
If I want to reduce it eventually, what would that process look like and how long would it take?
Are there interactions or side effects that could affect my ability to concentrate during sessions?
Would you want to coordinate directly with my therapist, and do you need me to sign anything for that?
What to tell your therapist
Tell them the name, the dose pattern, and — most usefully — whether you have taken it before or during a session, and what you were hoping it would do. Therapists doing exposure-based work with physical sensations are used to this conversation and will not treat it as a confession. A good one will fold the answer into how sessions are structured rather than asking you to change anything on your own.
Key takeaway: 🤝 Your therapist needs the dosing pattern and the timing, not just the prescription name.
Where this fits in anxiety treatment
Exposure-based interventions are the first-line treatment for anxiety disorders and post-traumatic stress disorder [1], and the international treatment guidelines that cover panic disorder, generalized anxiety disorder, social anxiety disorder and specific phobias are built on a large randomized-trial base [10]. Whether a benzodiazepine sits alongside that work is a genuine clinical judgment, and one where our position is that the prescriber owns the prescription and the therapy is planned around what is actually happening.
In our own practice, we ask about medication early — including timing and dosing pattern — and we plan CBT-based anxiety work with that information rather than around it. If you are still deciding what kind of help you want, it can be worth understanding which type of anxiety therapist fits your situation before you book.
It is also worth knowing that anxiety rarely arrives alone. A broader mental health screening can surface things a single anxiety measure will miss, and depression in particular changes what a treatment plan should prioritize — the PHQ-9 is the usual starting point there.
Key takeaway: 🧭 The medication question is worth asking. It is one of several things to check, not the automatic explanation.
If you came here worried that a prescription has been quietly undoing your therapy, the evidence does not support that fear as stated — and it does not dismiss it either. What it supports is a specific conversation with two people: the one who prescribed the medication, and the one doing the therapy with you. Neither of those conversations requires you to change anything first.
Anxiety running the show?
Evidence-based therapy can turn the volume down on anxiety — a clinician can help you find the approach that fits your life rather than a one-size-fits-all plan.
Frequently Asked Questions
Can I take a benzodiazepine before an exposure therapy session?
This article cannot answer that for your case, and moving or skipping a prescribed dose is not a decision to make from a web page. If your dose is scheduled, keep taking it as prescribed and raise the question at your next prescriber appointment. If it is as-needed, tell your prescriber and your therapist when you take it and what you hope it will do, and let them plan around that. A 2020 review of 12 randomized trials found no definitive evidence that benzodiazepines hinder exposure-based therapy.
Does anxiety medication make therapy less effective?
For benzodiazepines specifically, the evidence is mixed and unsettled. Across 12 randomized trials reviewed in 2020, they did not affect the outcome of exposure-based therapy at the end of treatment in nine studies, improved it in two, and reduced it in one. At follow-up, five showed reduced benefit and six showed none. Every trial carried bias concerns and the authors graded the evidence B rather than A. That review covers benzodiazepines only, not anxiety medication in general.
Why are benzodiazepines no longer a first-line treatment for anxiety?
Because guidelines favor treatments with better long-term profiles. NICE tells clinicians not to offer a benzodiazepine for generalized anxiety disorder except as a short-term measure during crises, and not to prescribe one for panic disorder at all. A review of 113 international guidance documents found broad agreement that they are not routine first-line agents, with recommended duration usually under four weeks. That describes the typical case, not a verdict on your prescription.
What is the difference between as-needed and scheduled dosing during therapy?
Scheduled dosing means taking medication at set times regardless of symptoms; as-needed means taking it when anxiety rises. Guidance is genuinely inconsistent here: a 2024 review of international guidance documents found agreement on short duration but not on which dosing pattern suits which condition. One theoretical concern about as-needed use is that it may pair relief with avoidance, but that idea comes from studies of physical safety aids in specific phobia, not medication. Your prescriber sets the pattern.
Should I stop my benzodiazepine before starting anxiety therapy?
No, not on your own. Stopping a benzodiazepine abruptly or reducing the dose too quickly can cause withdrawal reactions including seizures, which the FDA describes as potentially life-threatening. Any change to how much you take, or how often, belongs with the clinician who prescribed it. Starting therapy is a good reason to open that conversation, not a reason to make a change before you have had it.
About the Author
Dr. Kiesa Kelly is a licensed clinical psychologist and the founder of ScienceWorks Behavioral Healthcare. Her graduate cognitive-behavioral therapy practicum was completed at The Chicago Medical School Anxiety Disorders Clinic, where her work included exposure and response-prevention CBT for adults and children with anxiety disorders — panic disorder among them — along with assessment, treatment planning, and the development and facilitation of exposure hierarchies. She also completed adult individual psychotherapy training at the University of Wisconsin–Madison Psychiatric Institute and Clinics, working with outpatients presenting with generalized anxiety disorder, major depression, and adjustment disorders.
Dr. Kelly earned her PhD in Clinical Psychology with a concentration in Neuropsychology from Rosalind Franklin University of Medicine and Science, after an A.B. in Psychology and Neuroscience from Bowdoin College. She completed practica, internship, and an NIH-funded National Research Service Award postdoctoral fellowship at the University of Chicago, the University of Wisconsin, the University of Florida, and Vanderbilt University, and has more than 20 years of experience with psychological assessment. She practices by secure telehealth across Tennessee and in person at the Nashville office.
References
1. Melani MS, Paiva JM, Silva MC, Mendlowicz MV, Figueira I, Marques-Portella C, et al. Absence of definitive scientific evidence that benzodiazepines could hinder the efficacy of exposure-based interventions in adults with anxiety or posttraumatic stress disorders: a systematic review of randomized clinical trials. Depress Anxiety. 2020;37(12):1231-1242. https://pubmed.ncbi.nlm.nih.gov/33241637/
2. U.S. Food and Drug Administration. FDA requiring Boxed Warning updated to improve safe use of benzodiazepine drug class. September 23, 2020. https://www.fda.gov/drugs/drug-safety-and-availability/fda-requiring-boxed-warning-updated-improve-safe-use-benzodiazepine-drug-class
3. National Institute for Health and Care Excellence. Generalised anxiety disorder and panic disorder in adults: management (CG113), recommendations 1.2.26 and 1.3.20. https://www.nice.org.uk/guidance/cg113/chapter/Recommendations
4. Sakurai H, Inada K, Aoki Y, Takeshima M, Ie K, Kise M, et al. Management of unspecified anxiety disorder: expert consensus. Neuropsychopharmacol Rep. 2023;43(2):188-194. https://pubmed.ncbi.nlm.nih.gov/36811273/
5. Brandt J, Bressi J, Lê ML, Neal D, Cadogan C, Witt-Doerring J, et al. Prescribing and deprescribing guidance for benzodiazepine and benzodiazepine receptor agonist use in adults with depression, anxiety, and insomnia: an international scoping review. eClinicalMedicine. 2024;70:102507. https://pubmed.ncbi.nlm.nih.gov/38516102/
6. Hart G, Panayi MC, Harris JA, Westbrook RF. Benzodiazepine treatment can impair or spare extinction, depending on when it is given. Behav Res Ther. 2014;56:22-29. https://pubmed.ncbi.nlm.nih.gov/24755207/
7. Blakey SM, Abramowitz JS. The effects of safety behaviors during exposure therapy for anxiety: critical analysis from an inhibitory learning perspective. Clin Psychol Rev. 2016;49:1-15. https://pubmed.ncbi.nlm.nih.gov/27475477/
8. Blakey SM, Abramowitz JS, Buchholz JL, Jessup SC, Jacoby RJ, Reuman L, et al. A randomized controlled trial of the judicious use of safety behaviors during exposure therapy. Behav Res Ther. 2019;112:28-35. https://pubmed.ncbi.nlm.nih.gov/30481681/
9. Meckes SJ, Cole AC, Gee K, Lancaster CL. Testing judicious use of safety behaviors during exposure: a randomized controlled trial examining when and for whom safety behaviors improve fear reduction. Cognit Ther Res. 2025;50(2):341-359. https://pubmed.ncbi.nlm.nih.gov/42267056/
10. Bandelow B, Allgulander C, Baldwin DS, Costa DLDC, Denys D, Dilbaz N, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for treatment of anxiety, obsessive-compulsive and posttraumatic stress disorders - version 3. Part I: anxiety disorders. World J Biol Psychiatry. 2023;24(2):79-117. https://pubmed.ncbi.nlm.nih.gov/35900161/
Disclaimer
This article is for informational purposes only and is not medical advice, a diagnosis, or a treatment recommendation. It does not describe how any benzodiazepine should be taken, started, adjusted, or discontinued. Decisions about prescription medication belong to you and the clinician who prescribes it. If you are experiencing a mental health emergency, contact your local emergency services or call or text 988 in the United States.