Sleep and Anxiety
Poor sleep makes anxiety worse. Anxiety makes sleep worse. The cycle runs in both directions, and once it’s established, each side maintains the other. Here’s the science behind it — and what actually breaks the loop.
The bidirectional relationship
Most people know that anxiety disrupts sleep. Fewer realize how reliably the reverse is also true: poor sleep generates anxiety. The relationship runs in both directions, creating a self-reinforcing cycle that can be difficult to break from either end.
The statistics are stark. People with insomnia are 10 times more likely to have depression and 17 times more likely to have anxiety than the general population, according to research from Stanford Medicine. A large cohort study found that sleep disturbances increase the risk of anxiety by 1.9-fold — a larger effect than anxiety’s 1.2-fold increase in sleep disturbance risk. The cycle exists, and it’s not symmetric: poor sleep appears to generate anxiety somewhat more reliably than anxiety generates poor sleep.
Andrea Goldstein-Piekarski at Stanford’s Computational Psychiatry, Neuroscience, and Sleep Laboratory has spent years studying this using fMRI. Her work shows that sleep deprivation changes biological function in the brain regions that process emotions — making emotional regulation harder at a neurological level, not just a behavioral one.
Sleep deprivation impairs the prefrontal cortex — the brain region responsible for regulating the amygdala’s threat-detection response. A sleep-deprived brain is less able to put the brakes on anxiety responses, which means the same situation generates more anxiety on less sleep. The overthinking that follows is partly a direct consequence of impaired emotional regulation.
What poor sleep does to anxiety and overthinking
Sleep deprivation doesn’t just make you tired. It specifically impairs the cognitive and emotional systems that manage anxiety and overthinking.
Emotional regulation deteriorates
Sleep-deprived people show heightened emotional reactivity to mild stressors — situations that would barely register on a full night’s sleep become genuinely upsetting. Research shows sleep-deprived participants reported significantly greater stress, anger, and anxiety in response to low-level stressors compared to rested controls. The amygdala’s threat response amplifies; the prefrontal cortex’s ability to dampen it reduces.
Negative cognitive bias increases
Inadequate sleep skews perception toward negative interpretations. Ambiguous situations — a tone of voice, an unexplained delay, a neutral expression — are more likely to be read as threatening or negative after poor sleep. The cognitive distortions that fuel overthinking become more accessible and harder to challenge. The same event that would pass without rumination on a rested day generates a loop on a sleep-deprived one.
Rumination increases the following day
Poor sleep doesn’t just affect the night — it affects the next day’s thinking. Research shows that insufficient sleep is associated with elevated negative affect during waking hours, which increases the threshold for ruminative processing. The default mode network — the brain’s overthinking engine — runs more readily when you’re under-slept. The loops that might have stayed manageable after good sleep become harder to interrupt.
Cortisol patterns shift
Chronic sleep deprivation disrupts cortisol rhythms. Cortisol should peak in the early morning and drop through the day, reaching its lowest point in the evening to allow sleep onset. Disrupted sleep keeps cortisol elevated at the wrong times — maintaining the nervous system in a state of mild physiological arousal that lowers the threshold for anxious thinking throughout the day and makes it harder to wind down at night.
Breaking the cycle: where to start
Because the relationship is bidirectional, the cycle can be broken from either direction. Improving sleep reduces anxiety; reducing anxiety improves sleep. In practice, most people find sleep hygiene changes easier to start with — they don’t require confronting anxious thoughts directly, and the effects on anxiety follow relatively quickly.
-
1Consistent sleep and wake times
The most reliable single sleep intervention in the research. Going to bed and waking at the same time every day — including weekends — anchors the circadian rhythm and cortisol pattern. The consistency matters more than the specific time. A study of nearly 75,000 people by Jamie Zeitzer at Stanford found that going to bed earlier and waking earlier is better for mental health even for natural night owls, though the timing matters less than the consistency.
-
2Pre-bed worry dump
30–60 minutes before bed: write every worry, to-do, and unresolved thought. Not to solve them — just to capture them. The brain’s tendency to keep generating thoughts it’s afraid of losing is part of what keeps people awake. Writing them down signals that the material is captured and doesn’t need active processing. Research on structured worry time shows this specifically reduces pre-sleep cognitive arousal — the racing-thoughts-at-bedtime pattern that maintains the sleep-anxiety cycle.
-
3CBT-I: stimulus control
From Cognitive Behavioral Therapy for Insomnia — the most evidence-based sleep treatment available, with better long-term outcomes than sleep medication. Stimulus control means using the bed only for sleep — not for lying awake, worrying, scrolling, or watching TV. If you’ve been awake and anxious for 20 minutes, get up until you’re sleepy. It builds the association between bed and sleep rather than between bed and wakefulness and anxiety. It takes a week or two to work. It works.
-
4Physiological wind-down
Core body temperature has to drop to initiate sleep. Anxiety keeps it elevated. A warm bath or shower 1–2 hours before bed triggers the temperature drop afterward, which promotes sleep onset. Dimming lights 90 minutes before bed starts the melatonin signal. Avoiding screens for 60 minutes before bed reduces blue light suppression of melatonin. These aren’t marginal interventions — the temperature drop mechanism is well-established in sleep physiology.
-
5Aerobic exercise — but not too late
30 minutes of moderate aerobic exercise, three to five times a week, reduces both anxiety and insomnia symptoms with consistent evidence across multiple studies. The mechanism involves cortisol regulation, reduced baseline nervous system arousal, and improved sleep architecture. The timing matters: exercise within three hours of bed can delay sleep onset by raising core temperature and cortisol. Morning or afternoon exercise produces the sleep benefits without the timing downside.
When sleep problems are the primary issue
Sometimes anxiety is primarily maintaining the sleep problem, and treating the anxiety is the priority. Sometimes the sleep problem is primary — driven by learned behavioral patterns rather than anxiety — and CBT-I alone produces dramatic improvement in both sleep and anxiety.
Goldstein-Piekarski’s research found that CBT for insomnia led to sleep improvements that, in turn, led to lower depression levels. The causal direction went from sleep to mood, not mood to sleep. This supports starting with sleep treatment even when anxiety and depression are prominent — because improving sleep directly reduces the physiological substrate for those conditions.
On melatonin: Melatonin helps with sleep timing — shifting the circadian rhythm earlier or later — but doesn’t address the anxiety component of sleep disruption. It’s useful for jet lag and delayed sleep phase. For anxiety-driven insomnia, it’s most effective as an adjunct to behavioral interventions rather than a standalone treatment.
Common questions
Does poor sleep cause anxiety, or does anxiety cause poor sleep?
Both. The relationship is bidirectional and well-established. Research suggests sleep disturbances increase anxiety risk more strongly than anxiety increases sleep disturbance risk — sleep deprivation generates anxiety somewhat more reliably than anxiety generates sleep problems. But once the cycle is established, both sides maintain each other. Breaking either side tends to improve both.
How much sleep do I actually need to manage anxiety?
Most adults need seven to nine hours. Chronic restriction to six hours or less — even if you feel adapted to it — shows measurable impairment in emotional regulation and increased anxiety reactivity. The “I function fine on six hours” finding doesn’t hold up under objective testing; people consistently overestimate their functioning on restricted sleep. Seven hours is a reasonable minimum for most people managing anxiety.
Why is overthinking worse when I’m tired?
Sleep deprivation impairs the prefrontal cortex — the brain region that regulates the amygdala’s threat response. A sleep-deprived brain is less able to put the brakes on anxious reactivity, and more likely to interpret ambiguous situations negatively. The cognitive distortions that fuel overthinking become more accessible, and the ability to challenge them reduces. The same event that would pass unremarkably on a rested day can generate a full loop after poor sleep.
Is CBT-I better than sleep medication for anxiety-related insomnia?
For long-term outcomes, yes. CBT-I consistently produces more durable improvements in sleep than medication and addresses the underlying behavioral patterns that maintain insomnia. Medication provides faster initial relief but doesn’t change the maintaining mechanisms. Many sleep specialists now recommend CBT-I as the first-line treatment, with medication as a short-term bridge if needed. Several digital CBT-I programs also have solid evidence behind them for people without access to a specialist.