# Why do I keep having nightmares?

> You're awake now, and the dream is over. You're back in the room where you fell asleep. The pounding and the sweat are your body finishing the alarm the dream set off, and they ease as the room comes back to you.

**In short:** Nightmares tend to come with stress, some medicines, alcohol wearing off and, for many people, trauma. A dream that keeps returning is thought to be a pattern the sleeping mind has learned, and what was learned can be relearned. You can rewrite the dream while you're awake and practice the new version for ten to twenty minutes a day. Sleep specialists back this above anything else for nightmares, and it often helps even when people do it alone at home.

Maybe you're lying in the dark right now with the phone lit in your hand and some small hour showing on it. You're holding very still. You don't want to close your eyes in case the dream comes back. Or it's the morning after another one, and you're wondering why the same story keeps finding you, and whether it'll ever stop.

## What can I do right now, if a nightmare has woken me?

Your heart may still be pounding and your skin damp, and that alarm is a real one. A nightmare usually comes out of REM sleep, the part of the night we do most of our dreaming in, and it brings the body along with it. The sweating and the short, quick breaths you woke with can go on for a while after the dream has ended[^1], as your body goes on answering a danger that has already passed.

Waking fully, knowing where you are and remembering the dream sharply is what sets a nightmare apart[^2]. The sharpness you'd rather be without right now means you're properly awake, in the room, with the ordinary night going on around you.

**If you’d like to try something now**

You could switch on a lamp (or, if someone is asleep beside you, sit up with your phone's screen turned low), put your feet on the floor and name a few things you can see and hear: the shape of the door, the hum of the fridge, a car somewhere outside. Then let your breathing slow, and give the room a minute to become the room again.

If sleep doesn't come back after that, and you find yourself checking the clock and doing the sum of hours left until morning, it's fine to get up for a while. You could go somewhere else, or to another corner of the room if that's all there is, and do something unhurried until you feel sleepy again. It's a standard way of helping people with insomnia[^3] keep the bed tied to sleep, called stimulus control. It hasn't been tested after nightmares, but it may do the same there, so the pillow doesn't become the place where you lie very still and wait for the dream to pick up where it left off.

In the US about four adults in every hundred have nightmares often enough to count as nightmare disorder[^2]. It's a recognized condition, and it has a well-tested treatment[^2] you can start on your own.

## Why does the same dream keep coming back?

For many people it's the same dream, or the same kind of dream, that returns: being chased, arriving too late, someone gone, the thing that happened, each time in slightly different clothes. One leading explanation[^1] is that a nightmare can become a learned script. The sleeping mind falls into the same fixed pattern whenever something in a dream resembles the original stress, and it can go on playing long after that stress has faded. This is a model researchers use more than a tested fact, though the main treatment is built on it.

Researchers picture a returning dream as a groove the sleeping mind has worn by replaying the same script, and a groove that was worn in by practice can be worn a different way. If you've felt foolish for being so shaken by a dream, or wondered what it says about you, their answer is plain. A nightmare that keeps coming back is a pattern your nights have learned. It isn't a verdict on you or a sign that you're damaged, and specific treatments exist[^1] to change it, one of them something you can begin at home.

Why would pushing a dream away bring it back? It sounds backwards. Yet in small laboratory studies, trying hard not to think about something made it more likely to turn up in dreams[^1], and avoiding reminders goes along with nightmares that recur. So the morning habit of shoving the dream out of your mind may be part of what keeps it coming.

The dread that arrives at bedtime, long before any dream, belongs to the same picture. If you've lain there afraid to let go of the day, the doctors who diagnose nightmare disorder count that fear of sleep as one of the signs of the condition itself[^2]. Researchers also find that the distress nightmares cause matters more for how people feel than how many they have[^1]. So if the dread wears you down as much as the dreams do, it's worth getting help for, however many nightmares you have.

Staying up late, leaving a light on or scrolling until you're too tired to dream can feel like a sensible way to give the dream less night to work with. Putting off sleep is also one of the shapes the dread itself takes, and there's more on that in [What can I do when I dread going to bed?](https://nitez.com/dreading-bedtime) Over time, keeping away from bed night after night can loosen the tie between bed and sleep that insomnia treatment works to keep[^3].

## What makes nightmares more likely?

Nightmares tend to gather around hard stretches of life, and a difficult month can bring more of them. During the first COVID-19 lockdown, a survey of 968 Canadians[^4] found bad dreams most often in people under moderate to severe stress. Their dreams were of being chased or attacked, of death, of trying again and again or arriving late. That was one survey, and stress doesn't map neatly onto any single night. Among people with PTSD who kept a diary, the day's stresses didn't predict that night's nightmares[^5].

If you've been ill, a fever can darken dreams as well. In a small online survey[^6], the dreams people remembered from a fever were stranger and more negative than their usual ones. If your nights are badly broken, as they can be with sleep apnea, that may feed nightmares too. Several studies link disordered breathing in sleep with them[^1], and treating it with a breathing machine has eased nightmares in some people with PTSD.

If you've recently started a new prescription and the nightmares began around then, or changed, you may well be onto something. Of everything medicines can set off in sleep, from sleepwalking to acting out dreams, nightmares are among the most common[^7]. Coming off a medicine can bring them on too, if it's one that cuts down REM sleep[^2].

Which medicines? The names that turn up in reports include the stop-smoking drug varenicline, beta-blockers such as metoprolol and propranolol, and some SSRI antidepressants[^7]. Most of those reports are single cases and small studies, so nobody can yet say these drugs cause the nightmares. With an SSRI, one suggested reason is that it can raise anxiety for the first couple of weeks before easing it.

If you've noticed a change like this, it's worth telling whoever prescribed the medicine, who can go through it with you. Please don't stop or cut down on your own, because some medicines are dangerous to stop quickly: stopping a benzodiazepine suddenly or lowering the dose too fast can cause withdrawal seizures that can be life-threatening[^8].

**A drink before bed helps me sleep through them.**

It can seem to work at first, because alcohol cuts down dreaming sleep in the first half of the night[^9], so the early hours can pass without the dream.

As the alcohol wears off, REM sleep comes back longer than usual in the second half of the night[^9], a REM rebound, which can mean more dreaming, and more intense dreaming, in the hours before morning. In people who drink heavily and stop, the rebound is bigger, and vivid dreams are a common complaint[^10][^11], so nightmares during a week off the wine may come from that change. If you drink heavily, please cut down with a doctor's help rather than all at once, because alcohol withdrawal after heavy drinking can be dangerous[^12]. For someone who has the occasional glass, there's no withdrawal to worry about.

## How do I rewrite a nightmare?

If what you want most is for the dream to stop, there's a treatment that works directly on the groove. It's called imagery rehearsal, and it is the only nightmare treatment the American Academy of Sleep Medicine fully recommends[^2], both for nightmares after trauma and for nightmares with no clear cause. You choose a nightmare, change it while you're awake in any way you like, and rehearse the new version in your imagination for ten to twenty minutes a day[^2][^1]. The original instruction was simply to change the nightmare however you wish, so the new ending is yours to choose. The chase might stop at a door that opens, or the person who was gone might be standing in the kitchen.

**If you’d like to try something this week**

In daylight, you could write down the dream that comes most often, then write it again with whatever changes you like, and spend ten to twenty minutes each day picturing the new version[^2] from start to finish, as clearly as you can. If the dream replays something that happened to you, it may be better to read [what helps when nightmares follow a trauma](#after-trauma) first, and to do the rewriting with a therapist.

The largest self-help trial was done without a therapist at all. In the Netherlands, 399 people who had found their way to a nightmare website[^13] worked through a six-week program at home, on their own. Those who rewrote and rehearsed their dreams went from about five nightmares a week to under three by week eleven, roughly half as many as before. Forty-two weeks on, they were down to two and a half. They were adults from the general public with nightmares of their own, which makes theirs the trial closest to ordinary recurring nightmares.

Some people in that trial didn't rewrite anything. They only kept a nightmare diary. Why would a plain notebook help at all? They wrote each dream down and changed nothing, yet still ended up with about a quarter fewer nightmares. In other trials, people who faced the dream without changing it did about as well as those who rewrote it. A pared-down version of rewriting and rehearsing worked as well. Researchers don't yet agree on which part does the work[^1][^13], whether it's the writing down, the facing or the new ending.

**Nightmares a week, counted out, in the self-help trial**

Each mark is one nightmare a week, as people counted them themselves: adults from the general public who followed a six-week program at home[^13]. Both groups fell, the diary alone included. The people who rewrote and rehearsed fell further. They were still down forty-two weeks on[^2]. The diary group wasn't counted that late.

- **At the start** 5.2

- **Week 11** 2.8

- **Week 42** 2.5

- **At the start** 5.6

- **Week 11** 4.0

- **Week 42** -

## What if my nightmares began after something that happened to me?

Up to eight in ten people with PTSD have nightmares, and they can go on even after the PTSD itself has eased[^2]. If yours replay what happened, or circle close to it, researchers have tested help for exactly this.

Imagery rehearsal was first tried with trauma survivors. In New Mexico, researchers took 168 women with PTSD, nearly all of them survivors of sexual assault[^14], and gave them either three sessions of imagery rehearsal or a place on a waiting list. Six months later, the women who'd had the sessions were down from nearly four nights a week with nightmares to just over one. The waiting list had barely moved. Across thirteen studies of people with nightmares after trauma[^15][^16], the gains were large and lasted six to twelve months. Most of those studies only compared people with themselves before and after, though, which tends to flatter a treatment, and in the one study that compared it with another active treatment, its edge was small[^15][^16].

It has helped less for veterans who have lived with severe combat PTSD for many years. In trials with them it did no better than standard therapy, or left many still having nightmares every week[^2]. For a few people, working on the dream stirs up images that hurt. In the trial of assault survivors some women who had this left the study[^2], while in another trial one man changed how he did it and was free of nightmares within three months. If rewriting brings up pain, the rewrite itself can be changed, and finding a version you can bear to rehearse is part of what a therapist helps with.

If something happened to you, there is also help that works on what happened itself, and it tends to ease the nights as well. The therapies are prolonged exposure, cognitive processing therapy and EMDR, and a major US guideline recommends them for PTSD ahead of medication[^17]. In trials, treating PTSD this way brought large drops in trauma nightmares[^5][^2]. It runs the other way too, since treating the nightmares directly tends to ease PTSD symptoms and depression by a moderate amount[^5].

Maybe you're wondering whether there's simply a pill for this, or someone has mentioned prazosin. It's a blood-pressure medicine some doctors prescribe for trauma nightmares, and its trials don't agree. The early ones were small and encouraging. Then came the biggest: 304 US veterans with chronic PTSD took prazosin or a placebo for twenty-six weeks, and those on prazosin had no fewer nightmares and slept no better[^18]. After that, sleep doctors recommended it less firmly[^2].

When researchers pooled the smaller trials, though, they found a moderate benefit[^19][^20], with no clear difference from imagery rehearsal[^19]. Even while downgrading it, the same sleep doctors still called it the first-choice medicine, saying many patients respond very well[^2]. So one large trial says no, and the smaller ones, pooled together, say yes, moderately. Every one of them was in people with PTSD, many of them military. That makes prazosin a reasonable thing to ask about if trauma nightmares are wearing you down. Because it lowers blood pressure[^18], whether it suits you is a decision for a doctor who knows your health.

- **Would a sleeping pill get me through the night?** Researchers who've reviewed the options find that no medicine has earned a recommendation without reservation for nightmares[^1]. Sleep doctors advise against clonazepam, a benzodiazepine, for them[^2], and guidance for PTSD advises against benzodiazepines as well[^17]. There's more on this in [Are sleeping pills safe to take?](https://nitez.com/sleeping-pills)

- **Could I learn to know I'm dreaming, and change it from inside?** Some people try lucid dreaming, noticing from inside a dream that it is one, and sleep doctors say it may be used[^2]. Not everyone can learn it. In one small study only six of sixteen people became lucid in a nightmare, though their nightmares eased whether or not they did[^1][^2]. Rewriting the dream in daylight asks less of you, and it has firmer backing[^2].

Much of the strongest evidence comes from people with PTSD, and results are weaker in long-standing combat PTSD. Nightmares in these trials were counted by people themselves, week by week.

- **Imagery rehearsal is the one nightmare treatment the American Academy of Sleep Medicine fully recommends, for nightmares after trauma and without a clear cause.** well-established

- **Done at home without a therapist, imagery rehearsal cut nightmares by about half in adults from the general public, against about a quarter for a diary alone, and the drop lasted.** good-evidence

- **In survivors of sexual assault with PTSD, three sessions cut nights with nightmares from about four a week to about one.** good-evidence

- **In three trials, trauma-focused therapy for PTSD brought large drops in trauma nightmares too.** good-evidence

- **A nightmare that keeps returning may be a learned script the sleeping mind replays.** early-evidence

- **Researchers don't agree on which part of imagery rehearsal does the work, since a nightmare diary and facing the dream also helped.** still-debated

- **Researchers disagree about prazosin: its largest trial found no benefit, while pooling the smaller trials suggests a moderate one.** still-debated

- **Early reports, mostly single cases and small studies, link varenicline, some beta-blockers and some SSRIs with nightmares, without showing they cause them.** early-evidence

If nightmares come with thoughts of ending your life, please talk to someone today. Frequent nightmares are linked with a higher later risk of suicidal thoughts and acts[^5], and that is reason enough to get help now.

It's also worth seeing a doctor, or a therapist who treats nightmares, if:

- nightmares come most nights, or the fear of them has you putting off sleep
- they're spilling into your days, as worry, low mood or exhaustion
- they began after something that happened to you
- they started or changed after a new medicine, so your prescriber can look at it with you
- you, or someone beside you, notice you act out your dreams, shouting, punching, kicking or falling out of bed. That is a different condition, REM sleep behavior disorder[^21], and it needs a doctor, partly because it can cause injuries.

## Can the nights really change?

When a dream has been coming for months or years, it can feel like part of the furniture of the night, as fixed as the wardrobe in the dark. Yet in the trials, among ordinary adults at home and among women who had survived assault, it changed. People counted their nightmares before they started rewriting, then again weeks or months later, and by then the counts had fallen[^13][^14]. In the self-help trial they were still down forty-two weeks on[^13][^2]. Tomorrow morning, if you'd like, the dream can go into a notebook on the kitchen table, with a line through the old ending and a new one written underneath. The next time the night reaches for that story it may find the version you've practiced waiting there.

## Related questions

- [What can I do when I dread going to bed?](https://nitez.com/dreading-bedtime)
- [Should I get up if I can’t sleep?](https://nitez.com/getting-up-at-night)
- [Does alcohol help you sleep?](https://nitez.com/alcohol-and-sleep)
- [What are night terrors, and what should I do when my child has one?](https://nitez.com/night-terrors)
- [What is REM sleep?](https://nitez.com/rem-sleep)
- [Why do we dream?](https://nitez.com/why-we-dream)

## Sources

[^1]: Gieselmann A, Ait Aoudia M, Carr M, Germain A, Gorzka R, Holzinger B, et al. Aetiology and treatment of nightmare disorder: State of the art and future perspectives. Journal of sleep research. 2019;28(4):e12820. [PubMed](https://pubmed.ncbi.nlm.nih.gov/30697860/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6850667/) · [DOI](https://doi.org/10.1111/jsr.12820)
[^2]: Morgenthaler TI, Auerbach S, Casey KR, Kristo D, Maganti R, Ramar K, et al. Position Paper for the Treatment of Nightmare Disorder in Adults: An American Academy of Sleep Medicine Position Paper. Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine. 2018;14(6):1041–1055. [PubMed](https://pubmed.ncbi.nlm.nih.gov/29852917/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC5991964/) · [DOI](https://doi.org/10.5664/jcsm.7178)
[^3]: Edinger JD, Arnedt JT, Bertisch SM, Carney CE, Harrington JJ, Lichstein KL, et al. Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline. Journal of Clinical Sleep Medicine. 2021;17(2):255–262. [PubMed](https://pubmed.ncbi.nlm.nih.gov/33164742/) · [DOI](https://doi.org/10.5664/jcsm.8986) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC7853203/)
[^4]: Solomonova E, Picard-Deland C, Rapoport IL, Pennestri MH, Saad M, Kendzerska T, et al. Stuck in a lockdown: Dreams, bad dreams, nightmares, and their relationship to stress, depression and anxiety during the COVID-19 pandemic. PloS one. 2021;16(11):e0259040. [PubMed](https://pubmed.ncbi.nlm.nih.gov/34818346/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC8612516/) · [DOI](https://doi.org/10.1371/journal.pone.0259040)
[^5]: Sheaves B, Rek S, Freeman D. Nightmares and psychiatric symptoms: A systematic review of longitudinal, experimental, and clinical trial studies. Clinical psychology review. 2023;100:102241. [PubMed](https://pubmed.ncbi.nlm.nih.gov/36566699/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10933816/) · [DOI](https://doi.org/10.1016/j.cpr.2022.102241)
[^6]: Schredl M, Erlacher D. Fever Dreams: An Online Study. Frontiers in psychology. 2020;11:53. [PubMed](https://pubmed.ncbi.nlm.nih.gov/32047464/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6997236/) · [DOI](https://doi.org/10.3389/fpsyg.2020.00053)
[^7]: Dumont S, Bloch V, Lillo-Lelouet A, Le Beller C, Geoffroy PA, Veyrier M. Parasomnias and sleep-related movement disorders induced by drugs in the adult population: a review about iatrogenic medication effects. Journal of sleep research. 2025;34(2):e14306. [PubMed](https://pubmed.ncbi.nlm.nih.gov/39243188/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC11911033/) · [DOI](https://doi.org/10.1111/jsr.14306)
[^8]: US Food and Drug Administration. FDA requiring Boxed Warning updated to improve safe use of benzodiazepine drug class. Drug Safety Communication, September 23, 2020. [www.fda.gov](https://www.fda.gov/drugs/drug-safety-and-availability/fda-requiring-boxed-warning-updated-improve-safe-use-benzodiazepine-drug-class)
[^9]: Roehrs T, Roth T. Sleep, sleepiness, and alcohol use. Alcohol Research & Health. 2001;25(2):101–109. [PubMed](https://pubmed.ncbi.nlm.nih.gov/11584549/)
[^10]: Brower KJ. Alcohol’s effects on sleep in alcoholics. Alcohol Research & Health. 2001;25(2):110–125. [PubMed](https://pubmed.ncbi.nlm.nih.gov/11584550/)
[^11]: Colrain IM, Nicholas CL, Baker FC. Alcohol and the sleeping brain. Handbook of Clinical Neurology. 2014;125:415–431. [PubMed](https://pubmed.ncbi.nlm.nih.gov/25307588/) · [DOI](https://doi.org/10.1016/B978-0-444-62619-6.00024-0)
[^12]: MedlinePlus Medical Encyclopedia. Alcohol withdrawal. US National Library of Medicine. Reviewed 1 January 2025. [medlineplus.gov](https://medlineplus.gov/ency/article/000764.htm)
[^13]: Lancee J, Spoormaker VI, van den Bout J. Cognitive-behavioral self-help treatment for nightmares: a randomized controlled trial. Psychotherapy and psychosomatics. 2010;79(6):371–7. [PubMed](https://pubmed.ncbi.nlm.nih.gov/20829648/) · [DOI](https://doi.org/10.1159/000320894)
[^14]: Krakow B, Hollifield M, Johnston L, Koss M, Schrader R, Warner TD, et al. Imagery rehearsal therapy for chronic nightmares in sexual assault survivors with posttraumatic stress disorder: a randomized controlled trial. JAMA. 2001;286(5):537–45. [PubMed](https://pubmed.ncbi.nlm.nih.gov/11476655/) · [DOI](https://doi.org/10.1001/jama.286.5.537)
[^15]: Casement MD, Swanson LM. A meta-analysis of imagery rehearsal for post-trauma nightmares: effects on nightmare frequency, sleep quality, and posttraumatic stress. Clinical psychology review. 2012;32(6):566–74. [PubMed](https://pubmed.ncbi.nlm.nih.gov/22819998/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC4120639/) · [DOI](https://doi.org/10.1016/j.cpr.2012.06.002)
[^16]: Hansen K, Höfling V, Kröner-Borowik T, Stangier U, Steil R. Efficacy of psychological interventions aiming to reduce chronic nightmares: a meta-analysis. Clinical psychology review. 2013;33(1):146–55. [PubMed](https://pubmed.ncbi.nlm.nih.gov/23186732/) · [DOI](https://doi.org/10.1016/j.cpr.2012.10.012)
[^17]: Schnurr PP, Hamblen JL, Wolf J, Coller R, Collie C, Fuller MA, et al. The Management of Posttraumatic Stress Disorder and Acute Stress Disorder: Synopsis of the 2023 U.S. Department of Veterans Affairs and U.S. Department of Defense Clinical Practice Guideline. Annals of internal medicine. 2024;177(3):363–374. [PubMed](https://pubmed.ncbi.nlm.nih.gov/38408360/) · [DOI](https://doi.org/10.7326/M23-2757)
[^18]: Raskind MA, Peskind ER, Chow B, Harris C, Davis-Karim A, Holmes HA, et al. Trial of Prazosin for Post-Traumatic Stress Disorder in Military Veterans. The New England journal of medicine. 2018;378(6):507–517. [PubMed](https://pubmed.ncbi.nlm.nih.gov/29414272/) · [DOI](https://doi.org/10.1056/NEJMoa1507598)
[^19]: Yücel DE, van Emmerik AAP, Souama C, Lancee J. Comparative efficacy of imagery rehearsal therapy and prazosin in the treatment of trauma-related nightmares in adults: A meta-analysis of randomized controlled trials. Sleep medicine reviews. 2020;50:101248. [PubMed](https://pubmed.ncbi.nlm.nih.gov/31855732/) · [DOI](https://doi.org/10.1016/j.smrv.2019.101248)
[^20]: Mendes TP, Pereira BG, Coutinho ESF, Melani MS, Neylan TC, Berger W. Factors impacting prazosin efficacy for nightmares and insomnia in PTSD patients - a systematic review and meta-regression analysis. Progress in neuro-psychopharmacology & biological psychiatry. 2025;136:111253. [PubMed](https://pubmed.ncbi.nlm.nih.gov/39828080/) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC12985405/) · [DOI](https://doi.org/10.1016/j.pnpbp.2025.111253)
[^21]: Matar E, Lewis SJ. REM sleep behaviour disorder: not just a bad dream. The Medical Journal of Australia. 2017;207(6):262–268. [PubMed](https://pubmed.ncbi.nlm.nih.gov/28899330/) · [DOI](https://doi.org/10.5694/mja17.00321)

---
https://nitez.com/nightmares · Conditions · updated 2026-09-23