Anxiety insomnia: why you can't sleep and how to break the cycle

There are nights when the body is exhausted but the mind will not stop. And mornings when, after hours in bed, you get up more tired than when you lay down. If that sounds familiar, you are probably caught in one of the most studied, and at the same time most underestimated, biological cycles in modern medicine: the feedback loop between chronic stress and light sleep.

What many people describe as anxiety insomnia is neither a failure of willpower nor two separate problems that happen to coincide through bad luck. Stress and light sleep are two sides of the same physiological process: stress activates the sympathetic nervous system and raises cortisol, which fragments sleep; and fragmented sleep, in turn, drives up stress markers and anxiety the following day. The loop closes in on itself.

Which of the two is best tackled first? The answer is not as simple as it seems, but science offers clear guidance. Drawing on the available evidence, this article examines exactly how that feedback loop works and what happens in the brain and in the body during the process.

What type of insomnia you have: why it matters to distinguish shallow sleep from fragmented sleep

Sleep is not a uniform state. Every night, the brain moves through four to six cycles of roughly 90 minutes, each made up of distinct stages with very different biological functions.

Sleep stages fall into two broad categories: NREM sleep (without rapid eye movements) and REM sleep (with rapid eye movements). Within NREM there are three substages: N1 (the transition from wakefulness to sleep), N2 (consolidated light sleep) and N3 (deep or slow-wave sleep). REM sleep, which predominates in the second half of the night, is critical for the consolidation of emotional memory and for affective regulation.

When we talk about 'light sleep' in a clinical context, we are referring to two distinct but related phenomena:

1. A predominance of stages N1 and N2 over N3 and REM. Someone with light sleep spends more time in the shallow stages and less in the restorative ones. This can happen without the person formally waking up, yet sleep does not fulfil its restorative functions.

2. Sleep fragmentation. Frequent awakenings —sometimes so brief that they are not consciously remembered— that interrupt the normal architecture of sleep. Polysomnography can detect micro-arousals of 3-15 seconds that the sleeper does not register subjectively but which significantly degrade sleep quality.

The available epidemiological literature puts the prevalence of subjectively poor sleep quality at between 15% and 35% of the adult population, with higher rates in women and in middle-aged people — precisely the group with the greatest exposure to chronic stressors.

The distinction matters because the interventions are not the same. Shortening sleep latency (taking less time to fall asleep) does not resolve fragmentation. And improving sleep continuity does not guarantee a higher proportion of N3. Knowing which type of light sleep a person has is the first step towards choosing the right strategy.

The HPA axis: the central mechanism linking stress and sleep

To understand why stress degrades sleep, you need to know about the hypothalamic-pituitary-adrenal (HPA) axis, the neuroendocrine system that regulates the stress response in mammals.

When the brain perceives a threat —real or imagined— the hypothalamus releases corticotropin-releasing hormone (CRH), which prompts the pituitary gland to secrete ACTH, which in turn stimulates the adrenal glands to produce cortisol. The process takes minutes and has a clear evolutionary purpose: to prepare the body to fight or flee.

The problem arises when that axis is activated chronically, with no real physical threat to resolve. Work stress, relationship conflict, financial uncertainty or information overload activate the HPA axis in the same way a predator would. Cortisol rises. And elevated cortisol has direct, documented effects on the architecture of sleep.

Cortisol and sleep: the evidence

A study published in Journal of Psychosomatic Research documented that people with chronic insomnia had significantly higher cortisol levels than healthy controls, and that this elevation was associated with greater sleep fragmentation and reduced slow-wave sleep (N3). The relationship is dose-dependent: the greater the activation of the HPA axis, the worse the architecture of sleep.

Night-time cortisol: the rhythm that should not be disrupted

Cortisol follows a natural circadian rhythm: it peaks in the early hours of the morning and falls progressively through the day, reaching its lowest levels between 10 pm and 2 am. In the first 30-45 minutes after waking there is also a sharp rise known as the 'cortisol awakening response' (CAR), which prepares the body for the day. That night-time fall is necessary for deep sleep to become established.

When chronic stress keeps cortisol elevated through the afternoon and evening —a pattern documented in people with burnout and anxiety disorders— the threshold for entering N3 rises. The brain stays in a state of activation that is biologically incompatible with deep sleep. This is not insomnia of the will: it is chemistry.

Besides cortisol, stress activates the sympathetic nervous system, raising heart rate, body temperature and high-frequency electroencephalographic activity (beta and gamma waves) — all of them signals the brain reads as 'this is not the moment to sleep'.

A review published in Journal of Clinical Endocrinology & Metabolism (Buckley & Schatzberg, 2005) concluded that hyperactivity of the HPA axis is the mechanism most consistently associated with primary insomnia and with non-restorative sleep in adults.

How poor sleep amplifies anxiety: the loop closes

If stress degrades sleep, the reverse relationship is equally powerful: fragmented or shallow sleep raises stress markers the following day. This is where the problem perpetuates itself without intervention. Sustained poor sleep is, in fact, one of the routes through which anxiety ends up expressing itself in the body (for a wider picture, see the effects of anxiety on your body).

Sleep deprivation and stress reactivity

A study published in Current Biology (Yoo, Gujar, Hu, Jolesz & Walker, 2007) used functional magnetic resonance imaging to show that sleep deprivation increased the magnitude of the amygdala response —the fear and threat processing centre— to emotionally negative stimuli by around 60%, compared with people who had slept. Sleep-deprived people tend to report more anxiety, irritability and a sense of losing control in situations that under normal conditions would not cause them significant distress.

The role of REM sleep in emotional regulation

REM sleep has a specific role in emotional processing that goes beyond memory consolidation. During REM the amygdala —the fear and threat processing centre— reactivates, but it does so in a context of low noradrenaline, the neurotransmitter associated with the alarm response. That combination allows the brain to 'reprocess' the emotionally charged experiences of the previous day, reducing their affective load.

Research published in Psychological Bulletin (Walker & van der Helm, 2009) proposed the 'Sleep to Forget, Sleep to Remember' model: REM sleep allows the content of stressful experiences to be remembered while their emotional component is attenuated. When REM sleep is fragmented or reduced —as happens with chronic stress— that processing is not completed, and the previous day's experiences keep their full emotional charge on waking.

Sleep deprivation and cortisol

A clinical trial published in The Lancet (Spiegel, Leproult & Van Cauter, 1999) in 11 healthy young men showed that six days of sleep restriction (4 hours per night) significantly raised evening cortisol relative to baseline. That rise in evening cortisol —precisely at the point when it should be falling— created the conditions for another night of fragmented sleep.

The loop is thus completely closed:

1. Stress raises night-time cortisol → sleep becomes fragmented
2. Fragmented sleep fails to process the day's emotions → stress reactivity increases
3. Greater stress reactivity → more night-time cortisol → more fragmentation

This cycle can persist for weeks or months without intervention and, over time, produces documented structural changes: reduced hippocampal volume (associated with chronic exposure to cortisol), altered regulation of the HPA axis and greater vulnerability to mood disorders.

What should you tackle first: stress or sleep?

This is the question most frequently asked by the people affected and by health professionals alike. The honest answer: it depends on the clinical profile. But the evidence offers clear guidance on the general principles.

The case for tackling sleep first

The logic is physiological: if fragmented sleep amplifies stress reactivity, improving sleep turns down the 'gain' of the alarm system. With more restorative sleep, the brain has more cognitive and emotional resources to manage stress during the day.

A randomised controlled trial published in Biological Psychiatry (Irwin et al., 2015) in 123 older adults with insomnia compared three interventions: cognitive behavioural therapy for insomnia (CBT-I), Tai Chi and an educational seminar on sleep. The CBT-I group showed not only improvements in sleep but also significant reductions in the stress-related inflammatory markers (CRP, IL-6) at 16 months of follow-up. Improving sleep had systemic effects on biological stress.

CBT-I is currently the first-line intervention recommended by the American Academy of Sleep Medicine (AASM) for chronic insomnia, with efficacy rates higher than those of hypnotics in long-term studies.

The case for tackling stress first

When stress is the primary trigger —for example, in situations of intense acute stress such as bereavement, a crisis at work or an ongoing relationship conflict— working on sleep without addressing the cause may not be enough. Cortisol will remain elevated regardless of sleep hygiene techniques.

The available evidence suggests that stress-focused interventions —mindfulness, cognitive therapy, emotional regulation techniques— can improve sleep secondarily when stress is the main trigger. Harvey et al. (2011, Clinical Psychology Review) argue that sleep acts as a bidirectional transdiagnostic factor alongside emotional disorders, which supports the logic of addressing the cause when it can be identified.

The evidence-based recommendation: sequential intervention with sleep as the priority

The position best supported by the current literature is this: in most adults with an established stress-sleep cycle —that is, when there is no longer an identifiable acute stressor but a chronic pattern— sleep should be the initial priority, complemented at the same time by stress regulation techniques.

The reason is pragmatic: changes in sleep are measurable more quickly and produce improvements in subjective well-being within 2-4 weeks, which increases adherence to the stress interventions. Someone who sleeps better has more resources to practise mindfulness, take exercise or apply cognitive techniques.

This sequence —sleep first, stress management integrated from the outset— is the one applied by third-generation CBT-I protocols, which build in emotional regulation and mindfulness components from the first session.

When to reverse the order

There are situations in which stress should be addressed first, or as the priority:

Situation	Why stress comes first
Active acute stress with an identifiable cause (recent bereavement, an ongoing crisis at work)	Cortisol will remain elevated regardless of sleep techniques
Diagnosed generalised anxiety disorder	Insomnia is a symptom, not the primary cause
Symptoms of severe burnout with predominant emotional exhaustion	The capacity to implement sleep routines is compromised
Documented hyperactivation of the HPA axis (elevated salivary cortisol across several measurements)	Behavioural sleep intervention is less effective without first reducing arousal

In these cases, referral to a mental health professional or to a medical specialist is the appropriate step before implementing sleep protocols.

Evidence-based strategies for breaking the cycle

With the mechanism and the order of intervention understood, the practical question is: what works? The strategies that follow are those with the strongest scientific backing, organised by where they act on the cycle.

Sleep-focused interventions

Cognitive behavioural therapy for insomnia (CBT-I)

This is the intervention with the highest level of available evidence. A meta-analysis published in Annals of Internal Medicine (Trauer et al., 2015) that analysed 20 clinical trials with 1,162 participants found that CBT-I improved sleep efficiency by 9.9% and reduced wake time after sleep onset (WASO) by around 26 minutes on average, with effects sustained at follow-up. Its main components are:

• Sleep restriction: temporarily reducing time in bed in order to consolidate sleep and increase homeostatic pressure.
• Stimulus control: associating the bed exclusively with sleep (not with work, screens or worries).
• Cognitive restructuring: identifying and changing dysfunctional beliefs about sleep ('if I don't sleep eight hours, I won't be able to function tomorrow').

Sleep hygiene: what really matters

Sleep hygiene on its own has a modest effect once insomnia is established, but it is a necessary component of any intervention. The measures with the strongest empirical backing are:

• Keeping regular sleep and waking times, including at weekends (this regulates the circadian rhythm).
• A bedroom temperature between 16°C and 19°C (the fall in body temperature is a signal that sleep is beginning).
• Complete darkness or the use of a sleep mask (light suppresses melatonin even through closed eyelids).
• Avoiding blue-light screens in the hour before sleep (Gooley et al., 2011, Journal of Clinical Endocrinology & Metabolism).

Stress-focused interventions with a documented effect on sleep

Mindfulness and meditation

A randomised controlled trial published in JAMA Internal Medicine in 49 older adults with disturbed sleep found that a 6-week mindfulness programme (an adapted MBSR protocol) significantly improved sleep quality (measured with the Pittsburgh Sleep Quality Index, PSQI) compared with a sleep hygiene education programme. The proposed mechanism: mindfulness practice reduces night-time cognitive rumination, one of the main things that keeps insomnia going.

Regular physical exercise

A systematic review published in Mental Health and Physical Activity that analysed 66 studies concluded that regular physical exercise improved subjective sleep quality, reduced sleep onset latency and increased total sleep time. The effect was greater with moderate-intensity aerobic exercise (150 minutes a week) than with high-intensity exercise. The mechanism includes a reduction in baseline cortisol, an increase in adenosine (a marker of homeostatic sleep pressure) and the regulation of body temperature.

Important: intense exercise in the 2-3 hours before sleep can have the opposite effect, raising body temperature and sympathetic activation.

Relaxation techniques

Jacobson's progressive muscle relaxation and slow diaphragmatic breathing (6 breaths per minute) have level B evidence for the reduction of night-time arousal. A clinical trial (Nicassio & Bootzin, 1974, Journal of Abnormal Psychology, replicated many times) showed that progressive muscle relaxation reduced sleep onset latency by an average of 14 minutes in people with sleep onset insomnia.

Slow breathing acts on the vagus nerve, increasing parasympathetic tone and lowering heart rate — signals the nervous system reads as 'it is safe to rest'.

The role of nutrition and micronutrients

Some micronutrients have been investigated for their possible involvement in the regulation of the HPA axis and in the synthesis of neurotransmitters related to sleep:

• Magnesium: a double-blind clinical trial published in Journal of Research in Medical Sciences in older adults with primary insomnia observed, against placebo, differences in subjective sleep quality and in parameters such as serum cortisol and melatonin; this is a single study with a small sample whose results require confirmation. Magnesium acts as a cofactor in more than 300 enzymatic reactions, including processes involved in the normal functioning of the nervous system. It is worth being precise here: magnesium contributes to the normal functioning of the nervous system and helps to reduce tiredness and fatigue, while its specific relationship with sleep quality remains under investigation rather than established.
• L-theanine: an amino acid present in green tea that has been investigated for its possible interaction with the neurotransmission systems involved in the stress response. A study published in Biological Psychology explored its effect on physiological reactivity to stress in an experimental setting, observing differences in markers such as heart rate and salivary IgA. Separate preliminary studies have also explored its possible effect on subjective sleep quality without a direct sedative effect (Hidese et al., 2019, Nutrients). The available evidence is preliminary and its role in sleep quality has not been established.
• Ashwagandha (Withania somnifera): a double-blind clinical trial published in Cureus with 60 participants explored the effects of supplementation with ashwagandha extract (KSM-66, 300 mg twice a day) over 10 weeks on sleep quality and morning salivary cortisol levels. The results showed statistically significant differences against the placebo group in both variables. The authors note that studies with larger samples are needed to confirm these findings. As with L-theanine, the role of ashwagandha here remains an open research question rather than an established effect.

No supplement replaces behavioural and cognitive interventions as the first line. Their role is complementary and should be assessed individually with a health professional. Always consult your doctor or pharmacist before starting any supplement, particularly if you take medication or have pre-existing health conditions.

When the stress-sleep cycle calls for professional attention

Not all light sleep and not all stress require specialist intervention. But there are signs that the cycle has passed the threshold of what self-care strategies can resolve.

Warning signs that justify a medical consultation

• Insomnia present on more than 3 nights a week for more than 3 months (the diagnostic criterion for chronic insomnia in DSM-5 and ICSD-3).
• Excessive daytime sleepiness that interferes with functioning at work or socially.
• Episodes of sleep apnoea (loud snoring, breathing pauses noticed by a partner, waking with a sensation of choking) — obstructive sleep apnoea is a common and treatable cause of non-restorative sleep that requires diagnosis by polysomnography.
• Symptoms of anxiety or depression accompanying the insomnia (intrusive thoughts, persistent sadness, loss of interest in usual activities).
• Using alcohol to bring on sleep — alcohol fragments REM sleep and creates dependence.
• Chronic use of hypnotics without medical supervision.

Professionals who can help

• Primary care doctor: the first point of contact to rule out organic causes (hypothyroidism, sleep apnoea, restless legs syndrome) and to consider referral.
• Clinical psychologist or psychotherapist: for CBT-I and emotional regulation techniques.
• Sleep medicine specialist: for complex cases or when a primary sleep disorder is suspected.
• Pharmacist: for guidance on supplements, interactions and over-the-counter options.

Always consult your doctor or pharmacist before starting any supplement, particularly if you take medication or have pre-existing health conditions.

Stress and light sleep are not two problems that add up: they are a system that feeds itself. Elevated cortisol fragments sleep; fragmented sleep raises cortisol. Understanding that mechanism —not merely living with it— is already the first step out of the loop.

The available scientific evidence indicates that, in most adults with this established chronic pattern, working on sleep first —with behavioural techniques such as CBT-I— produces faster and more sustained improvements than addressing stress in isolation. But the most effective intervention is the one that integrates both dimensions from the start: improving sleep while working on emotional regulation and on reducing arousal.

There is no single, immediate solution. There is a clear path, backed by decades of research, that begins with understanding what is happening in your body and choosing the right tools in the right order.

If you have spent weeks or months caught in this cycle, speak to your doctor or pharmacist. Restorative sleep is not a luxury — it is a biological condition needed for everything else to work.

---

References

The statements in this article are based on the available scientific literature. The verified key references that support the main statements in the Calm and Balance cluster are listed below.

• Lupien SJ, McEwen BS, Gunnar MR, Heim C. Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat Rev Neurosci. 2009;10(6):434-45. PMID: 19401723.
• Pickering G, Mazur A, Trousselard M, et al. Magnesium Status and Stress: The Vicious Circle Concept Revisited. Nutrients. 2020;12(12):3672. PMID: 33260549.
• Boyle NB, Lawton C, Dye L. The Effects of Magnesium Supplementation on Subjective Anxiety and Stress—A Systematic Review. Nutrients. 2017;9(5):429. PMID: 28445426.
• Chandrasekhar K, Kapoor J, Anishetty S. A prospective, randomized double-blind, placebo-controlled study of safety and efficacy of a high-concentration full-spectrum extract of ashwagandha root in reducing stress and anxiety in adults. Indian J Psychol Med. 2012;34(3):255-62. PMID: 23439798.
• Abbasi B, Kimiagar M, Sadeghniiat K, Shirazi MM, Hedayati M, Rashidkhani B. The effect of magnesium supplementation on primary insomnia in elderly: A double-blind placebo-controlled clinical trial. J Res Med Sci. 2012;17(12):1161-9. PMID: 23853635.
• Hidese S, Ogawa S, Ota M, et al. Effects of L-Theanine Administration on Stress-Related Symptoms and Cognitive Functions in Healthy Adults: A Randomized Controlled Trial. Nutrients. 2019;11(10):2362. PMID: 31623400.
• Kimura K, Ozeki M, Juneja LR, Ohira H. L-Theanine reduces psychological and physiological stress responses. Biol Psychol. 2007;74(1):39-45. PMID: 16930802.