Corticosteroids such as prednisone are widely listed as causing insomnia: sleep disturbance is a commonly cited adverse effect, and a review describes three proposed pathways, including changes in the stress-hormone axis and in natural sleep-wake cycles. Antidepressants vary: activating ones such as fluoxetine may disrupt sleep in short-term treatment. In a UK Biobank study of type 2 diabetes patients, metformin alone was not linked to more sleep difficulty, and a US survey linked low dietary creatine, not high, to more trouble sleeping. A dated record of your nights is the most useful thing to bring to the prescriber.
Every package leaflet lists insomnia somewhere, usually between nausea and dry mouth, with no sense of whether it hits one patient in a hundred or one in three. The leaflet is not wrong. It just does not help when you are lying awake on day nine of a new medication and wondering whether to blame it.
This page takes four of the common suspects, tells you what the research says about each, and shows how to document a suspicion for the prescriber. One rule runs through all of it: if a prescribed drug is involved, the decision about it belongs to the person who prescribed it.
Medications that cause insomnia: corticosteroids carry the clearest warning
A word on terms. This section means corticosteroids such as prednisone, the anti-inflammatory kind prescribed for asthma flares, rashes or autoimmune disease. Anabolic steroids are a different topic and are not covered here.
For corticosteroids the answer is a qualified yes: psychiatric and sleep effects are a standard part of the warning list. A practical guide to monitoring systemic corticosteroid therapy names psychiatric disturbances among the more serious side effects, particularly at high doses for long periods. A focused review of steroid-induced sleep disturbance and delirium goes further and describes three pathways proposed in the literature: changes in the hypothalamic-pituitary-adrenal axis (the stress hormone system), changes in natural sleep-wake cycles, and hyperarousal from effects on neuroinhibitory pathways.
Cole JL. Steroid-Induced Sleep Disturbance and Delirium: A Focused Review for Critically Ill Patients. Fed Pract 2020. PMID 32669778
Liu D, Ahmet A, Ward L, Krishnamoorthy P, Mandelcorn ED, Leigh R, Brown JP, Cohen A, Kim H. A practical guide to the monitoring and management of the complications of systemic corticosteroid therapy. Allergy Asthma Clin Immunol 2013. PMID 23947590
The focused review rests on 8 articles, written for critically ill patients. Its practical conclusion is that minimizing dose and duration are the main ways clinicians reduce these effects. So the mechanism is credible and the warning is standard, but a good estimate of how often a short course of prednisone disturbs a night at home is not documented. Individual experiences are real, but they are anecdotes and not a rate.
Antidepressants can disrupt sleep early in treatment
Fluoxetine is on the list of antidepressants with an activating effect. A review of how antidepressants act on sleep concluded that the effect depends on the kind of drug, the dose, the time of day it is taken and the duration of treatment. Activating antidepressants such as fluoxetine and venlafaxine may disrupt sleep, at least in short-term treatment, while sedating ones such as mirtazapine or trazodone improve sleep quickly but can cause oversedation in the long run. The same review notes that mid-night insomnia is the most frequent residual symptom of depression itself, so the drug is not the only suspect.
Wichniak A, Wierzbicka A, Walęcka M, Jernajczyk W. Effects of Antidepressants on Sleep. Curr Psychiatry Rep 2017. PMID 28791566
An older systematic review of randomized trials found insomnia commonly with monoamine oxidase inhibitors, most tricyclic antidepressants and all the SSRIs, as well as venlafaxine and moclobemide.
Mayers AG, Baldwin DS. Antidepressants and their effect on sleep. Hum Psychopharmacol 2005. PMID 16229049
For a look at what happens in the sleeping brain, there is a small lab study: nine depressed patients on fluoxetine and six unmedicated depressed patients were recorded on two consecutive nights. The fluoxetine group had lower sleep efficiency and significantly more arousals and eye movements during non-REM sleep, and 44% of them (four of nine) showed clinically significant periodic limb movements, versus none of the controls. Fifteen people is a small basis for anything, but it shows how a drug can fragment sleep without the patient being able to say how.
Dorsey CM, Lukas SE, Cunningham SL. Fluoxetine-induced sleep disturbance in depressed patients. Neuropsychopharmacology 1996. PMID 8726754
Prozac is not a sleep medicine, and the idea of Prozac for insomnia runs against the evidence above: at the start of treatment it belongs with the alerting drugs, not the sedating ones. That does not mean anyone should stop taking an antidepressant that is working. Untreated depression disrupts sleep too, and stopping suddenly has its own problems. Report the change and let the prescriber weigh it.
A general review of psychotropic drugs and sleep explains why this list is long: antidepressants, antipsychotics, stimulants and benzodiazepines act on monoamine systems and on histamine and muscarinic receptors, and those same receptors help regulate sleep, so effects on sleep are built in rather than rare accidents.
Doghramji K, Jangro WC. Adverse Effects of Psychotropic Medications on Sleep. Psychiatr Clin North Am 2016. PMID 27514301
Metformin alone shows no sleep signal in the best data
A UK Biobank analysis of 11,806 people with type 2 diabetes compared sleep by medication group. Patients on non-metformin therapy (815 people, 86% of them on sulphonylureas) had 1.24-fold odds of reporting regular difficulty falling and staying asleep, compared with untreated patients and with patients on metformin alone. No significant differences were reported between untreated patients and metformin patients. So the study finds no signal for metformin as a cause of insomnia, and some signal for the other group.
Xue P, Wu J, Tang X, Tan X, Benedict C. Oral Antidiabetics and Sleep Among Type 2 Diabetes Patients: Data From the UK Biobank. Front Endocrinol (Lausanne) 2021. PMID 34803924
Keep the limits in view. Sleep was self-reported at one point in time, the authors call for replication with objective measures, and the comparison tells you nothing about an individual who started metformin and began waking at 3 a.m. Diabetes itself, with night-time urination and blood sugar swings, is a plausible reason to wake, and that is a guess, not a finding of this study.
A creatine survey points the other way
The evidence on creatine consists of two odd pieces. In rats, four weeks of oral creatine lowered total sleep time and non-REM sleep during the inactive (light) period, and blunted the rise in a sleep-pressure chemical after sleep deprivation. That is a plausible mechanism and a rat.
Dworak M, Kim T, Mccarley RW, Basheer R. Creatine supplementation reduces sleep need and homeostatic sleep pressure in rats. J Sleep Res 2017. PMID 28397310
In humans there is a US survey of dietary creatine, the amount people get from food, mostly meat and fish. Among 5,988 people, those with intake under 1 g a day reported trouble sleeping more often than those at or above 1 g a day: 23.7% versus 19.3% (odds ratio 1.30). More severe sleep disorders did not differ. The direction is the opposite of what the rumor predicts, and it is a cross-sectional survey of food, not a test of supplements.
Baltic S, Grasaas E, Ostojic SM. Creatine and sleep habits and disorders in the general population aged 16 years and over: NHANES 2007-2008. Nutr Health 2025. PMID 39569425
People who take creatine for training are therefore in an evidence gap: neither a warning nor a clearance. If you suspect it, mention it to a doctor or pharmacist, and bring the kind of dated record described in the next section.
Whether it is the pill: what a dated record can show
What most people want to know is whether the drug is doing it to them. Writing things down helps the person who prescribes it answer that, and none of it involves changing anything yourself.
- Write down the dates: when the medication or supplement started, when the dose or timing changed, and when your sleep changed. Memory blurs these quickly, and the sequence is the whole argument.
- Note anything else that changed in the same weeks: caffeine, alcohol, bedtime, illness, stress. Otherwise nobody can tell the pill from the coffee, and the prescriber needs that picture.
- Record the nights. With a recording app such as SleepTrace you get a timeline of when the room went quiet and any sounds in it, so the week before a change can be set beside the week after, instead of relying on how the nights felt.
A recording works from sound, not brain activity, and it cannot say whether a drug caused anything. It can show a pattern that stays still or moves with the dates, which is a better thing to bring to an appointment than a feeling. The app has to stay open with the phone plugged in; the page on how to record yourself sleeping covers the setup.
Do not stop or change a prescribed medication because of this page. Stopping some of these drugs suddenly causes problems of its own. Tell the prescriber what you notice and when it began. If you have thoughts of harming yourself, seek urgent help in your country now (in the US, call or text 988).
What a medication change leaves behind
Two practical thoughts to finish. First, a short course of a drug that disturbs sleep is not the same as chronic insomnia, but a bad stretch can set off habits that outlast the drug, as described in what causes insomnia. If sleep has not recovered a few weeks after a medication ended, the habits may be what is keeping it going. Second, supplements are a separate question for a doctor or pharmacist, since some interact with prescribed drugs. The evidence on them is thinner than the shelves suggest; melatonin for sleep goes through the one with the most data.
Frequently asked questions
Corticosteroids such as prednisone are widely listed as causing sleep disturbance. A focused review proposes three pathways: changes in the hypothalamic-pituitary-adrenal stress axis, changes in natural sleep-wake cycles, and hyperarousal through effects on neuroinhibitory pathways. The review is based on 8 relevant articles and written for critically ill patients, so it tells you the mechanism is plausible, not how likely it is for you. Ask the prescriber whether anything about dose timing can change; do not stop the drug yourself.
It can, especially early in treatment. A review of antidepressants and sleep lists fluoxetine among the activating antidepressants that may disrupt sleep in short-term use, and a small sleep-lab study of nine fluoxetine-treated depressed patients found lower sleep efficiency and more arousals than in six untreated depressed patients. Depression itself disturbs sleep, so a doctor has to sort out which is which.
The best available data say probably not on its own. In 11,806 UK Biobank patients with type 2 diabetes, metformin monotherapy was not linked to more self-reported difficulty sleeping than no medication, while non-metformin therapy (86% sulphonylureas) was, with an odds ratio of 1.24. It was self-reported data from one point in time. Diabetes itself, including blood sugar swings and night-time urination, can also wake people.
A human trial of creatine supplements and insomnia is not documented. A rat study found that four weeks of oral creatine reduced total sleep time. In a US survey of 5,988 people, a low dietary creatine intake (under 1 g a day from food) was linked to more trouble sleeping (23.7% versus 19.3%), not less. Survey data cannot say what supplements do.
SleepTrace is a wellness app, not a medical device. This article is general information, not medical advice. If your symptoms are frequent, severe or worrying, please talk to a doctor.
Do not stop a medication on your own because of sleep. SleepTrace cannot see which pill does what, but it records your estimated sleep phases and sounds from the first night, so your doctor sees a timeline. Note the date of every change in a diary; it is a record, not a diagnosis, and it does not measure side effects.
Hear your own night. SleepTrace turns a night of audio into your sleep phases, the sounds you made, and how it all trends. Works without a wearable, better with one. Your phone on the nightstand is enough to start. Get it for iPhone or Android →