Frequent Nightmares: Causes, Medications and How to Stop Them

Nightmare disorder is estimated to affect about 4% of adults, but occasional nightmares occur in a far larger share of the population. What separates an ordinary bad dream from a clinical problem is frequency, intensity and daytime consequences: fear of falling asleep, deliberately cutting sleep short, irritability, poor concentration. This article sorts out where nightmares come from, how to tell them from other nocturnal events, and what the research says about treatment — with particular attention to prazosin, around which many misunderstandings have grown.

This text is educational and does not replace a diagnosis. If nightmares are linked to past trauma, suicidal thoughts or violent movements during sleep, see a specialist — the warning signs are listed in a separate section below.

A nightmare is an intensely unpleasant, remembered dream that usually wakes the sleeper with fear, anxiety or disgust. After waking the person quickly knows where they are and can describe the dream. Nightmares occur mostly in the second half of the night, when REM periods are longer and richer in imagery; sleep structure and the role of REM are covered in our entry on sleep stages.

A nightmare is easily confused with a night terror, which occurs in deep NREM sleep, usually in the first half of the night. A person with a night terror screams and may be agitated, but afterwards often remembers nothing, or only a vague sense of threat, and is hard to calm. Sleepwalking belongs to the same family. The distinction matters because management can differ — see our entry on sleepwalking and night terrors.

REM sleep is the time of the most vivid dreaming and also of particular activity in emotion-related structures, including the amygdala. In typical REM, noradrenaline levels in the brain are very low, which lets emotional memories be processed without the accompanying stress response. A leading hypothesis holds that in nightmares, and especially post-traumatic ones, this mechanism fails: noradrenergic activity stays elevated and the memory of the trauma is replayed with a full fear reaction instead of gradually losing its emotional charge.

This hypothesis was the basis for attempts to treat nightmares with alpha-1 adrenergic blockers, above all prazosin. We return to those results below because they are one of the best examples of a promising mechanism and early studies not translating into a result in a large, well-designed trial.

The second large group of causes is substances that change sleep architecture. A 2025 systematic review of drug-induced parasomnias and sleep-related movement disorders named nightmares among the most common drug-induced parasomnias. The authors point out that drugs inhibiting noradrenergic, serotonergic or orexin transmission can promote REM sleep and thereby nightmares. Parasomnias are, however, rarely listed in package leaflets, so patients and doctors are often surprised by the timing.

The practical key: nightmares that began soon after starting, changing the dose of, or stopping a drug should be reported to the prescriber — without stopping the drug yourself. Abruptly stopping a beta-blocker or an antidepressant can be more dangerous than the nightmare itself. A useful reference point is also how alcohol changes the structure of the night, described in our separate article on alcohol before bed and sleep architecture.

Nightmares are one of the most characteristic symptoms of post-traumatic stress disorder (PTSD), but they also occur in depression, anxiety disorders, chronic stress, grief and personality disorders. The relationship runs both ways: worse mental health aggravates nightmares, while chronic nightmares and sleep loss worsen mood and resilience to stress. We describe this two-way relationship in more detail in our entry on sleep and mental health.

For people after trauma, nightmares are often not only distressing but also sustain the whole symptom picture — if sleep is interrupted every night, recovery and effective psychotherapy are harder. For this reason modern practice treats nightmares as a treatment target in their own right, not just a 'side symptom' of PTSD that will disappear once the trauma is treated.

Factors that promote nightmares also include an irregular sleep rhythm, sleep deprivation (followed by increased REM), heavy meals and alcohol shortly before bed, and watching disturbing content in the evening. These are modifiable and worth putting in order before any pharmacotherapy is considered.

Prazosin is an alpha-1 blocker originally used for hypertension. Its use for post-traumatic nightmares followed from the noradrenergic hypothesis above and from promising smaller studies. In 2018, however, the largest randomized trial of the drug to date was published — and its result surprised many clinicians.

The American Academy of Sleep Medicine (AASM), in a 2018 position paper, reviewed the evidence on treating nightmare disorder in adults. According to the summary of that publication, the only method labelled as recommended for PTSD-associated nightmares and for nightmare disorder is imagery rehearsal therapy (IRT). Other psychological and pharmacological methods were classified as 'may be used', and two drugs — clonazepam and venlafaxine — as not recommended for nightmare disorder.

What is IRT? In simplified form: the patient picks a recurrent nightmare, deliberately changes its ending or content to something less threatening, and for several minutes a day imagines the new version while awake. The aim is to 'rewrite' the script so the brain stops replaying it in its original form. The method is usually delivered by a psychologist or sleep therapist in a short series of sessions — a major advantage over long-term drug treatment.

Another option the AASM lists as possible is CBT for insomnia (CBT-I). If nightmares coexist with insomnia and an irregular sleep rhythm, tidying up the schedule, limiting time in bed awake and stabilizing the wake-up time often lowers the overall 'density' of nightmares. This is worth remembering because many people with nightmares cut sleep short out of fear of falling asleep, which paradoxically makes them worse.

In unclear cases a sleep specialist may order a sleep study to exclude other causes of night awakenings. This article does not cover treatment in children or nightmares in the course of neurological disease — those situations need a separate assessment.

Nightmares are a good example of a phenomenon where intuition says 'the right drug will fix it'. The data say otherwise: the largest randomized trial of prazosin showed no advantage over placebo, and the only method the AASM labels as recommended is a short behavioural therapy. A sensible order is: check medications, alcohol and sleep rhythm, introduce simple techniques, and for post-traumatic or chronic nightmares look for a therapist. Keep drugs for situations in which a doctor sees a clear rationale.