Blog · Sleep & Energy

Middle Insomnia: Why You Wake at 3am and Can't Get Back to Sleep

By Vitae Team · 2026-10-09

Waking at 3am and unable to fall asleep again? The science behind middle insomnia, from sleep cycles and cortisol to hidden causes and treatment.

Waking in the night is part of normal sleep. Lying awake for an hour afterwards is something different. Here is what the research says about why it happens, which popular explanations do not hold up, and what the evidence supports as treatment.

TL;DR

Night waking is extremely common. In a general-population survey of 8,937 American adults, 35.5% reported waking at least three nights a week. That is a measure of frequent waking, not of diagnosed insomnia.
The awakening is rarely the problem. Sleep becomes lighter towards morning and brief awakenings are part of its normal architecture. What distinguishes insomnia is what happens next.
Age changes sleep. A meta-analysis of 65 studies found time awake after falling asleep rose by about ten minutes per decade between 30 and 60.
The cortisol and blood-sugar theories are overstated. Cortisol normally rises towards morning, and overnight glucose changes have not been established as a cause of middle insomnia.
Breathing problems are easy to miss. In a small case series, nine in ten awakenings were preceded by a breathing event, in patients whose symptoms had not suggested sleep apnoea. Preceded is not the same as caused.
CBT-I is the strongest established treatment. Sleep hygiene advice alone is not, and the main US guideline recommends against it as a stand-alone therapy.

Night Waking Is Normal, and Very Common

You fall asleep without difficulty, sleep soundly for several hours, then wake at 3.17am. Nothing obvious has disturbed you, the room is quiet, and there is no reason to be awake. An hour later you are still looking at the ceiling. This is the pattern of middle insomnia, or sleep-maintenance insomnia: unlike sleep-onset insomnia, the difficulty is not falling asleep but staying asleep or returning to sleep after waking.

The waking itself is not abnormal. Sleep is not eight uninterrupted hours of unconsciousness but a series of cycles of non-rapid eye movement (NREM) and rapid eye movement (REM) sleep, with shifts in brain activity, muscle tone and responsiveness throughout. Deep slow-wave sleep is concentrated at the start of the night; later, REM occupies more of each cycle while the biological pressure to stay asleep declines. Brief awakenings become more likely as the night goes on, and many are so short they are never remembered.

Nor is it rare. In a telephone survey of 8,937 adults across California, New York and Texas, Maurice Ohayon and colleagues found that 35.5% reported waking at least three nights a week and 23% reported waking every night, rising to 34.6% nightly among those aged 65 and over. What separated an inconvenience from a problem was what followed: 43% of those who woke reported difficulty resuming sleep, amounting to 15.2% of the whole sample, and of that group 77.5% reported some daytime impairment as a result. More than 80% of people reporting any insomnia symptom also reported night waking, which makes sleep-maintenance difficulty the most widely shared feature of insomnia rather than a niche variant.

It is worth being precise about what those numbers describe. The 35.5% figure is self-reported frequent night waking in a general population, and the 15.2% figure is self-reported difficulty getting back to sleep. Neither is a measure of how many people have clinically diagnosed chronic insomnia, which requires a clinical assessment rather than a survey question: persistent difficulty initiating or maintaining sleep despite adequate opportunity, together with distress or impaired daytime functioning, usually at least three nights a week for three months or more. What the survey establishes is the scale of the experience, not the scale of the diagnosis.

Why 3am, and Why the Hour After Is the Hard Part

There is nothing biologically unique about 3am. Its significance depends on when you went to bed, your habitual waking time and the timing of your body clock. Two processes regulate sleep: homeostatic sleep pressure, which builds during wakefulness and dissipates while we sleep, and the circadian system, which helps determine when the body is biologically prepared for sleep and for wakefulness. After four or five hours of sleep, much of the pressure accumulated during the previous day has been discharged and deep sleep is less prominent, so an awakening at that point is easier to sustain than one shortly after bedtime.

What happens next is where insomnia is made. For someone sleeping well, waking at 3am may be followed by an almost effortless return to sleep. For someone with insomnia, the same awakening triggers a sequence of thoughts — how long have I been awake, how much is left, how will I manage tomorrow — and the resulting alertness makes sleep progressively harder even after the original cause has passed. Researchers use the term hyperarousal for this heightened cognitive and physiological activation: worry, close attention to bodily sensations, anticipation of a poor night, or a learned association between the bed and wakefulness. Repeated nights of struggling to return to sleep strengthen that association until the bedroom, once a reliable cue for sleep, becomes a place where someone expects to lie awake. This does not make insomnia imaginary, and a medical or environmental disturbance may well be causing the initial awakening, but it does explain why trying harder to sleep so reliably produces the opposite result.

Age matters too. In 2004 Ohayon and colleagues published a meta-analysis of 65 studies involving 3,577 healthy participants aged between five and 102. Wake after sleep onset — the accumulated time spent awake after initially falling asleep — showed the most pronounced relationship with age of any variable assessed, rising by around ten minutes per decade between roughly 30 and 60 before largely levelling off. That is useful context for anyone who slept uninterrupted in their twenties and now wakes more often in their forties, but the analysis examined healthy sleepers rather than people seeking treatment. A tendency towards lighter sleep is not the same thing as lying awake for hours night after night, and normal age-related change should not be used to dismiss clinically significant insomnia.

What Might Be Waking You

Middle insomnia has no single universal cause. Obstructive sleep apnoea is among the most treatable: the upper airway repeatedly narrows or closes during sleep, and the resulting breathing disturbances can trigger brief arousals without the sleeper becoming consciously aware of them. Snoring, witnessed pauses in breathing, gasping, morning headaches and excessive daytime sleepiness are useful warning signs, but not everyone with sleep-disordered breathing fits that picture.

A small 2012 study by Barry Krakow and colleagues enrolled 20 adults with chronic insomnia, none of whom reported classic sleep-disordered-breathing symptoms and none of whom had previously seen a sleep specialist. Across overnight sleep studies the researchers recorded 531 awakenings, of which 478, or roughly 90%, were preceded by a respiratory event: an apnoea, a hypopnoea or a respiratory effort-related arousal. Eleven of the 20 met criteria for obstructive sleep apnoea, and 17 of the 20 were ultimately diagnosed with either obstructive sleep apnoea or upper airway resistance syndrome.

The finding is striking and its limitations are equally important. A respiratory event preceding an awakening is a sequence in time, not a demonstrated cause: in a recording dense with breathing irregularities, some events will precede awakenings that had other triggers, and the study design cannot separate the two. Beyond that, this was a small, uncontrolled case series in a selected group of patients, and the editorial published alongside it in the same issue of Sleep described the work as preliminary. Twenty people cannot establish that 90% of awakenings across the wider insomnia population are caused by breathing problems. What the study does suggest is that sleep-disordered breathing is easy to overlook when insomnia is the presenting complaint, and that persistent awakenings with snoring, gasping or unexplained daytime sleepiness deserve proper assessment.

Alcohol works in the opposite direction to the way most people assume, making sleep easier to start and harder to maintain: its sedative effects wear off as the body metabolises it, and it also alters sleep architecture, can worsen snoring and apnoea, and may increase the need to urinate overnight. In perimenopause and menopause, fluctuating and declining reproductive hormones affect sleep through several routes, and not every menopausal disturbance involves a noticeable hot flush. Nocturia becomes more common with age, though the relationship runs both ways: sometimes bladder discomfort causes the awakening, sometimes a person wakes for another reason and then notices they need the toilet. Chronic pain, reflux, restless legs syndrome and certain medications can all fragment sleep, and stress, anxiety and depression are associated with insomnia, usually bidirectionally. Identifying a real trigger matters, because treating insomnia-related behaviours alone will not resolve awakenings driven by an untreated condition.

Is It a Cortisol Spike or a Blood-Sugar Crash?

Search for an explanation for waking at 3am and two theories appear repeatedly. Both borrow from real physiology. Neither provides a convincing general explanation for middle insomnia.

Cortisol secretion follows a circadian rhythm, reaching lower levels during the early part of the night before increasing towards morning, and there is a separate cortisol awakening response that occurs after waking. These are normal processes, not signs of dysfunction. It is plausible that altered stress-system regulation contributes to insomnia in some people, and research has explored associations between insomnia, hyperarousal and hypothalamic-pituitary-adrenal axis activity, though findings vary across populations and methods. What has not been established is that waking at a particular hour reliably indicates a pathological cortisol spike; establishing abnormal cortisol regulation requires clinical context and validated testing rather than an interpretation of the bedside clock. The related claim that night waking demonstrates adrenal fatigue is more misleading still, since adrenal fatigue is not a recognised medical diagnosis and genuine adrenal disorders cannot be inferred from insomnia alone.

The blood-sugar version is similarly attractive: glucose supposedly falls overnight, the body releases stress hormones to compensate, and the resulting surge wakes you. But in healthy individuals glucose regulation continues during sleep, with the liver releasing glucose and counterregulatory mechanisms helping to prevent clinically significant hypoglycaemia, and the sequence has not been established as an explanation for middle insomnia in people without diabetes. The 2003 study by Salomon Banarer and Philip Cryer often cited in its support found close to the reverse. Studying eight adults with type 1 diabetes and eight matched participants without diabetes under experimentally induced hypoglycaemia, the researchers found that autonomic responses were reduced during sleep in the participants with diabetes, who were also substantially less likely to wake up. Its finding concerned impaired defences against low blood glucose in type 1 diabetes, not a mechanism by which falling glucose wakes healthy sleepers. Nocturnal hypoglycaemia remains a real concern for anyone using insulin or certain glucose-lowering medications; it is not a reason to recommend bedtime snacks or glucose monitoring to everyone who wakes at 3am.

What Treats Persistent Middle Insomnia

For chronic insomnia, the strongest established treatment is cognitive behavioural therapy for insomnia, usually shortened to CBT-I. The American Academy of Sleep Medicine's 2021 clinical practice guideline gives multicomponent CBT-I its only strong recommendation, with conditional recommendations for brief multicomponent therapies and for stimulus control, sleep restriction and relaxation therapy as single components. The same guideline recommends against using sleep hygiene as a stand-alone treatment for chronic insomnia in adults, while noting it can form part of a multicomponent programme. That is an unusually direct statement from a guideline body, and worth holding onto if the advice you have been given so far amounts to a list of bedroom rules. CBT-I itself commonly combines stimulus control, sleep scheduling, education about sleep and work on unhelpful expectations. Sleep scheduling matters most to people who spend long periods awake in bed, since temporarily matching time in bed more closely to actual sleep aims to consolidate it before gradually extending the window again — a structured intervention rather than a recommendation to sleep less, and one that needs adapting where excessive daytime sleepiness, bipolar disorder or seizure disorders are present.

The UK position is consistent with this. NICE describes CBT-I as the standard first treatment for people with long-term insomnia, offered after sleep hygiene advice has been given, and its appraisal of the medication daridorexant recommends that drug only where CBT-I has been tried without success or is unavailable or unsuitable. Sleep hygiene therefore sits in UK practice where the AASM would put it: as initial advice that precedes treatment rather than as the treatment itself. The practical implication for a reader is simply that the advice stage is not the end of the pathway. If you have worked through the standard recommendations without improvement, the next step is CBT-I rather than more advice, and NICE itself notes that access to CBT-I varies across the UK, which is worth knowing before you ask. Medication may have a role in selected circumstances, but sleeping tablets do not address the processes that sustain chronic insomnia, and where a condition such as sleep apnoea, pain or menopausal symptoms is contributing, that needs attention too.

In the moment, the useful changes are small. Repeatedly checking the time turns an ordinary awakening into a calculation about how little sleep remains, which raises the pressure to fall asleep and tends to heighten alertness, so keeping the clock out of view removes one source of stimulation at no cost. Stimulus-control therapy encourages leaving the bed when you are awake and frustrated and returning when sleepy — not to enforce an arbitrary 20-minute rule, but to stop the bed becoming associated with prolonged, effortful wakefulness. Choose a quiet activity in dim light, and avoid work, stimulating entertainment and repeated tests of whether you feel sleepy yet. Relaxation techniques reduce tension but are not switches that force the brain into sleep, and treating meditation as another performance test recreates the pressure you are trying to escape.

What This Actually Means

If you wake at 3am most nights, the most useful reframing is that waking and lying awake are two questions rather than one, and they do not always have the same answer. The waking itself usually belongs to physiology: your body clock, the natural lightening of sleep towards morning and, in some cases, a specific and treatable disturbance. The hour that follows is often sustained by arousal and learned association, which is the part CBT-I is built to address.

That is a useful distinction rather than a rule, and it has an important exception. Some causes do not just trigger an awakening, they keep you awake once it has happened: pain that is worse lying down, reflux, restless legs, a breathing disturbance that recurs every time you drift off, or hot flushes that arrive in sequence. In those cases the prolonged wakefulness is being maintained by the condition rather than by what you believe about sleep, and no amount of behavioural work on the bedroom will resolve it until the condition is treated. Equally, the two often run together: a treatable cause starts the pattern and arousal keeps it going long after the original cause has been dealt with, which is why treating one and ignoring the other so often produces partial results.

What this does explain is why so much popular advice disappoints. Cortisol and blood-sugar theories aim at the moment of waking, often the part least worth explaining, and do so with mechanisms the evidence does not support in healthy sleepers. Sleep hygiene aims at the right target but is not strong enough on its own, which is why the main US guideline declines to recommend it as a stand-alone treatment and why UK practice treats it as the step before treatment rather than the treatment. The two questions worth putting to a clinician are therefore whether something identifiable is disturbing your sleep, and how to access CBT-I. For many people the honest answer involves both.

Frequently Asked Questions

Why do I wake up at 3am every night?

Sleep pressure declines as the night progresses, which makes awakenings easier to sustain in the second half of the night. Repeated waking at a similar time can also reflect environmental triggers, an underlying condition or an established insomnia pattern. The timing on its own cannot identify the cause.

Does waking at 3am mean my cortisol is too high?

No. Cortisol follows a daily rhythm and normally begins rising towards morning. Waking at a particular time does not establish abnormal cortisol levels, which require clinical assessment and validated testing.

Should I eat something if I wake during the night?

There is no good evidence that eating during night-time awakenings improves middle insomnia in people without a glucose-regulation problem. Anyone taking diabetes medication that can cause hypoglycaemia needs individualised advice.

Is sleep hygiene advice worth following?

As part of a wider programme, yes. As a treatment on its own for chronic insomnia the American Academy of Sleep Medicine recommends against it, and if you have followed the standard advice without improvement that is a reason to ask about CBT-I.

Could my night waking be sleep apnoea even if I do not snore?

It is possible. In Krakow's small case series, patients without classic breathing symptoms were nonetheless found to have sleep-disordered breathing. That was 20 selected patients rather than a representative sample, but it is grounds for assessment if you have persistent awakenings with unexplained daytime sleepiness.

Can middle insomnia be cured?

Chronic sleep-maintenance insomnia is treatable. CBT-I has an established evidence base, though individual responses vary, and treatment may also require identifying and addressing an underlying medical condition.

The Bottom Line

Waking at 3am is not in itself a sign that something is wrong, and the popular cortisol and blood-sugar explanations are not supported as general causes. If the waking is persistent and affecting your days, the two things with evidence behind them are finding and treating any underlying disturbance and accessing CBT-I.

This article is for information only. If night waking is persistent or affecting your daytime functioning, speak to your GP, who can review your sleep history, medications and symptoms and assess whether another condition is contributing.

Sources

Ohayon MM, Krystal A, Roehrs TA, Roth T, Vitiello MV. Using difficulty resuming sleep to define nocturnal awakenings. Sleep Medicine. 2010;11(3):236-241. doi:10.1016/j.sleep.2009.11.004
Ohayon MM, Carskadon MA, Guilleminault C, Vitiello MV. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals. Sleep. 2004;27(7):1255-1273. doi:10.1093/sleep/27.7.1255
Krakow B, Romero E, Ulibarri VA, Kikta S. Prospective assessment of nocturnal awakenings in a case series of treatment-seeking chronic insomnia patients. Sleep. 2012;35(12):1685-1692. doi:10.5665/sleep.2244
Pigeon WR, Sateia MJ. Is insomnia a breathing disorder? Sleep. 2012;35(12):1589-1590. doi:10.5665/sleep.2222 (editorial commentary)
Banarer S, Cryer PE. Sleep-related hypoglycemia-associated autonomic failure in type 1 diabetes. Diabetes. 2003;52(5):1195-1203. doi:10.2337/diabetes.52.5.1195
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. doi:10.5664/jcsm.8986
National Institute for Health and Care Excellence. Daridorexant for treating long-term insomnia. Technology appraisal guidance TA922, 2023. nice.org.uk/guidance/ta922