Racing Mind at Bedtime: When Anxiety and Insomnia Feed Each Other
Last reviewed: 08/15/2026
Reviewed by: Dr. Kiesa Kelly
You get into bed exhausted. Within about ninety seconds, your mind produces a complete inventory of everything unresolved in your life — the email you should have sent, the conversation that went sideways, the thing your doctor said in March. You are too tired to solve any of it and too activated to stop trying. By 1 a.m. you have added a new worry to the pile: how badly tomorrow is going to go on this little sleep.
If that is your night, you are not dealing with one problem. You are dealing with two that keep handing work to each other — an anxiety problem and chronic insomnia. And the question that actually determines how fast you get out of it is not which one came first — it is which one to treat first.
In this article, you'll learn:
How worry and sleep loss escalate each other, and which direction the evidence is stronger in
Why "fix the anxiety and the sleep will follow" is the most common wrong turn
What decides whether you should start with sleep treatment or anxiety treatment
When the two genuinely need to run at the same time
How we sequence both at our Nashville office and by telehealth
The loop: worry delays sleep, sleep loss amplifies worry
The nighttime half of the loop is familiar. Worry is a cognitive activity, and cognitive activity is arousing. Lying still in a dark, quiet room removes every distraction that kept the worry manageable during the day, which is why 11 p.m. is when your brain finally has the bandwidth to review your finances. Sleep does not arrive on top of that arousal, so you lie there longer. Lying there longer teaches your nervous system that bed is a place where you think hard — and that association does its own damage on subsequent nights, independent of whatever you were worrying about.
The daytime half is less obvious and more important. Short sleep does not just make you tired. It narrows your tolerance, shortens your fuse, and biases how you read ambiguous situations — your coworker's flat reply lands as hostility rather than as a busy person typing fast. You then spend the day generating exactly the material you will process at midnight. It is worth putting a number on that daytime half early — a structured measure such as the GAD-7 gives you a baseline to compare against later [10], though a questionnaire is a screener rather than a diagnosis and cannot answer the sequencing question on its own.
Here is where the evidence gets genuinely interesting, because the two arrows are not the same size. Longitudinal studies that follow people who sleep badly but have no psychiatric diagnosis find that insomnia predicts the later onset of anxiety disorders, with a pooled odds ratio of about 3.2 across six studies [1]. A 2023 replication of that work confirmed the pattern, most robustly for depression [2]. Two honest caveats belong with that number: six studies is a small base, the confidence interval is wide, and the risk of bias across the primary studies was rated moderate. It is not a precise figure. But the direction is consistent, and it is the direction most people do not expect.
💤 Key takeaway: Poor sleep is not only a symptom of anxiety. In prospective studies it is also a risk factor for developing it — which is why leaving the insomnia untreated has a cost of its own.
Which came first matters less than which to treat first
Most people arrive at our Nashville office with a firm theory about causation. "I'm an anxious person, so of course I don't sleep." That theory feels obvious, and it produces a specific plan: deal with the anxiety, and the sleep will sort itself out.
That plan is where a lot of time gets lost.
Misconception one: insomnia is a symptom, so treating it directly is treating a symptom. For decades clinical practice split insomnia into "primary" and "secondary to" another condition, and secondary insomnia was expected to resolve once the primary condition did. Current diagnostic criteria abandoned that split [3]. The reason is empirical: insomnia so frequently outlives the condition it supposedly derived from that treating it as a downstream symptom was leaving people with chronic insomnia and a well-managed anxiety disorder. Insomnia earns its own treatment plan, not a place at the back of someone else's.
Misconception two: you have to be calm enough before sleep treatment will work. This one delays treatment for months. Cognitive behavioral therapy for insomnia is the recommended first-line treatment for chronic insomnia in adults in the American Academy of Sleep Medicine guideline [4], the American College of Physicians guideline [5], and the 2023 European Insomnia Guideline [6] — and none of them make an anxious mind a reason to wait. A meta-analysis of CBT-I delivered specifically to people who also carried a psychiatric diagnosis found meaningful improvement in insomnia severity across every comorbidity group studied, with the co-occurring symptoms improving as well [7].
That same meta-analysis is also where the honest limit sits. Its comorbidity groups were depression, PTSD, alcohol dependency, bipolar disorder and psychosis — there was no anxiety-disorder subgroup [7]. So the claim we can make cleanly is that CBT-I holds up in the presence of serious co-occurring psychiatric conditions. The claim that it performs identically for a primary anxiety disorder is reasonable but less directly evidenced than most articles imply. We would rather tell you that than round it off.
Misconception three: sleeping badly for a long time means the damage is permanent. Duration predicts very little about how well insomnia responds. People who have slept poorly for fifteen years routinely respond to a course of treatment that runs a couple of months.
🧭 Key takeaway: The practical decision rule: if you meet criteria for chronic insomnia — trouble falling or staying asleep three or more nights a week for three months or more, with daytime consequences — start with the sleep, even when anxiety is obviously present.
How CBT-I handles the racing mind (stimulus control, scheduled worry)
Two tools do most of the work here, and both are already covered in depth elsewhere on this site, so we will point rather than repeat. Stimulus control rebuilds the association between your bed and sleep instead of between your bed and thinking; the mechanics, including why getting out of bed helps when it feels like the opposite, are walked through step by step in our guide to how CBT-I resets a stuck sleep cycle. Scheduled worry time moves the reviewing to a fixed earlier window so it stops colonizing the hour after lights-out.
What is worth adding — because it is routinely overstated — is what scheduled worry actually buys you. A 2025 randomized controlled trial tested worry postponement over fourteen days and found that the version with specific if-then plans meaningfully reduced how long people spent worrying, by roughly fifteen minutes a day, but produced no significant improvement in sleep onset, awakenings, or sleep quality compared with controls [8]. A separate randomized trial found similar benefits for worry itself [9].
Read that carefully, because it reframes the whole tool. Scheduled worry is a good worry intervention. It is not, on its own, a sleep intervention. In a course of CBT-I the change in your nights comes mostly from the timing work — consolidating your sleep window, holding a fixed wake time, and breaking the bed-and-thinking association. Scheduled worry makes those changes tolerable for an anxious mind and keeps rumination from filling the space that sleep restriction opens up. It is the support act, not the treatment. Sessions that lean on it as the treatment are the ones where people conclude, wrongly, that CBT-I did not work for them.
⏱️ Key takeaway: If a sleep plan for anxiety consists mainly of a worry journal and a wind-down routine, it is missing the part that moves the needle.
When anxiety treatment needs to run alongside
Sleep-first is a default, not a rule. There are patterns where running anxiety treatment concurrently — or ahead — is the better clinical call.
Consider a recognizable version of the first pattern. You sleep badly, but the thing that has actually reorganized your life is avoidance. You stopped driving on I-40 after a panic episode. You take the surface streets to work, which adds twenty-five minutes each way, which is time you no longer spend on anything else. You have started declining plans that would put you on the interstate, and you have a story ready for why. Your sleep is genuinely bad, and it is the fourth most disruptive thing in your week. Sequencing sleep first here would be technically defensible and practically wrong — the avoidance is what is shrinking your life, and it needs direct exposure-based work.
Or: your nights are punctuated by waking with your heart already going, gasping, certain something is badly wrong. That is not the ordinary insomnia pattern of lying awake thinking. Nocturnal panic needs to be identified as panic, because the treatment differs and because sleep restriction handed to someone with untreated nocturnal panic tends to backfire.
A third pattern: your worry is not about sleep or about any specific thing, but is a constant background hum across health, money, work, and the people you love, present for years and predating the sleep problem by a decade. That profile suggests generalized anxiety carrying the sleep problem rather than the reverse, and it warrants treatment in its own right rather than being filed under "poor sleep."
Where sleep and a co-occurring condition genuinely interact and both need adapting, our guide to keeping sleep on track alongside trauma, OCD, or strong anxiety covers the condition-by-condition adjustments in detail.
🚦 Key takeaway: Start with anxiety treatment when avoidance, panic, or trauma symptoms are doing more damage than the sleep loss is. Otherwise, start with sleep.
Doing both in Nashville (sequencing)
In practice, sequencing looks less like choosing a lane and more like staggering two starts.
A common shape: weeks one through six are CBT-I, with anxiety addressed only where it directly interferes with the sleep plan. Around week four we look at what the daytime anxiety is doing now that sleep has consolidated. For a meaningful number of people it has dropped enough that no separate course is needed. For others, what remains is clearer and more targeted than what they walked in with — and that residual is what anxiety treatment then works on, whether that is exposure-based CBT or an acceptance and commitment therapy approach to anxiety when the problem is that fighting the anxiety has become its own full-time job.
Two local logistics matter more than people expect. First, CBT-I is unusually well suited to telehealth — the work is a weekly review of your sleep diary and adjustments to your schedule, none of which needs a room. Across Tennessee, that means the Franklin-to-downtown commute is not a reason to postpone starting. Second, when you do come to the office, morning appointments are worth requesting during the sleep-restriction phase. That phase is temporarily tiring by design, and a 4:30 p.m. slot on Elm Hill Pike after a short night, in Nashville traffic, is a needless additional cost. Our local overview of CBT-I in Nashville covers what a first month looks like.
One more thing worth naming: if you have been tracking your sleep on a wearable and the score has become part of the bedtime anxiety, say so early. Chasing a sleep score reliably makes the loop worse, and it is common enough to have its own name and its own literature.
📋 Key takeaway: Ask for morning appointments during sleep restriction, and bring two weeks of sleep diary data to the first session — it changes what can be decided on day one.
Getting assessed
Before any of this is a plan, someone needs to establish which pattern you are actually in. A good assessment sorts a few things quickly: whether your insomnia meets criteria on its own, whether your daytime anxiety is at a level that needs its own treatment, whether something else is fragmenting your nights — sleep apnea, a delayed body clock, a medication side effect — and which of those to move on first.
Questions worth asking whoever you consult:
1. Scope — will you assess the insomnia in its own right, or only as a symptom of my anxiety?
2. Methodology — do you use CBT-I with sleep scheduling and stimulus control, or a sleep hygiene and relaxation approach? These are not the same treatment and they do not perform the same.
3. Sequencing — what is your reasoning for treating one before the other in my case, and what would change your mind?
4. Output — at the end, will I have a written plan I can follow independently, or does maintaining the gains depend on staying in weekly sessions?
If the answers to those come back thin, that is useful information about fit. Our clinical team works with both sides of this pattern, and the first conversation is largely about establishing which one is currently driving.
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
If I treat my insomnia, will my anxiety get better on its own?
Often it improves, but rarely all the way. Trials of CBT-I in people who also have a mental health condition show the sleep problem improving substantially and the co-occurring symptoms improving alongside it, though usually by less. Plan for sleep treatment to take a real bite out of daytime anxiety, and plan to look again at what is left over once you are sleeping. What remains after your sleep stabilizes is a much clearer target.
Should I start CBT-I or anxiety treatment first if I have both?
If you meet criteria for chronic insomnia, starting with CBT-I is usually the better first move. It is the first-line treatment for chronic insomnia in current guidelines whether or not another condition is present, it works on a shorter timeline than most anxiety treatment, and better sleep makes anxiety work easier. The main exception is when panic, avoidance, or trauma symptoms are the larger source of impairment.
Can I do CBT-I while I am taking medication for anxiety?
Yes. CBT-I is routinely delivered alongside psychiatric medication, and current guidelines position it as first-line treatment rather than something you earn after medication. We do not make medication decisions here, and any change to a prescription belongs with the prescriber who wrote it. What we do ask for is a stable dose during the early weeks, because a changing dose and a changing sleep schedule at once make it impossible to read what is working.
Is it still insomnia if anxiety is the obvious cause?
Yes. Current diagnostic criteria dropped the older split between primary insomnia and insomnia considered secondary to another condition, precisely because insomnia so often persists after the other condition improves. If you have trouble falling or staying asleep at least three nights a week for three months or more, with real daytime consequences, that meets the definition regardless of what started it.
How long does sleep usually take to improve when anxiety is in the picture?
Most courses of CBT-I run four to eight sessions, and many people notice their nights consolidating within the first two to four weeks. Anxiety that has been present for years typically moves on a slower timeline than that. Expect the sleep curve and the anxiety curve to separate, and treat that as information rather than failure — it is often the clearest signal of how much of your daytime anxiety was sleep debt.
About the Author
Dr. Kiesa Kelly is a licensed clinical psychologist and the founder of ScienceWorks Behavioral Healthcare. Her background includes more than 20 years of experience in psychological assessment and evidence-based treatment, with clinical training at the University of Chicago, Vanderbilt University, and the University of Wisconsin, and NIH-funded research training earlier in her career.
Much of her clinical work sits exactly where this article does — with adults whose presenting problem turns out to involve two conditions maintaining each other rather than one. She practices in Nashville and by telehealth across Tennessee, and reviews clinical content on this site for accuracy before publication.
References
1. Hertenstein E, Feige B, Gmeiner T, et al. Insomnia as a predictor of mental disorders: A systematic review and meta-analysis. Sleep Medicine Reviews. 2019;43:96-105. https://doi.org/10.1016/j.smrv.2018.10.006
2. Hertenstein E, Benz F, Schneider CL, Baglioni C. Insomnia—A risk factor for mental disorders. Journal of Sleep Research. 2023;32(6):e13930. https://doi.org/10.1111/jsr.13930
3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR). American Psychiatric Association Publishing; 2022. https://doi.org/10.1176/appi.books.9780890425787
4. Edinger JD, Arnedt JT, Bertisch SM, 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. https://doi.org/10.5664/jcsm.8986
5. Qaseem A, Kansagara D, Forciea MA, Cooke M, Denberg TD. Management of Chronic Insomnia Disorder in Adults: A Clinical Practice Guideline From the American College of Physicians. Annals of Internal Medicine. 2016;165(2):125-133. https://doi.org/10.7326/M15-2175
6. Riemann D, Espie CA, Altena E, et al. The European Insomnia Guideline: An update on the diagnosis and treatment of insomnia 2023. Journal of Sleep Research. 2023;32(6):e14035. https://doi.org/10.1111/jsr.14035
7. Hertenstein E, Trinca E, Wunderlin M, et al. Cognitive behavioral therapy for insomnia in patients with mental disorders and comorbid insomnia: A systematic review and meta-analysis. Sleep Medicine Reviews. 2022;62:101597. https://doi.org/10.1016/j.smrv.2022.101597
8. McCarrick D, Prestwich A, Ferguson E, O'Connor DB. Effects of worry postponement on daily worry and sleep: a randomised controlled trial. Psychology & Health. 2025:1-21. https://doi.org/10.1080/08870446.2025.2590072
9. Krzikalla C, Buhlmann U, Schug J, et al. Worry Postponement From the Metacognitive Perspective: A Randomized Waitlist-Controlled Trial. Clinical Psychology in Europe. 2024;6(2):e12741. https://doi.org/10.32872/cpe.12741
10. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine. 2006;166(10):1092-1097. https://doi.org/10.1001/archinte.166.10.1092
Disclaimer
This article is for informational purposes only and is not a substitute for individualized clinical assessment, diagnosis, or treatment. Reading it does not create a clinician-patient relationship. Decisions about medication belong with the prescriber who wrote it. If you are in crisis or concerned about your immediate safety, contact 988 (Suicide & Crisis Lifeline) or your local emergency services.