Cortisol spikes at night: what the evidence shows

Cortisol is supposed to be low at night. That's not opinion — it's basic circadian biology. But for a meaningful proportion of people, something disrupts that pattern, and the downstream effects on sleep quality are well-documented. Spiegel et al. (1999) showed that even modest sleep restriction measurably altered evening cortisol concentrations in healthy adults. This article is my attempt to lay out what the evidence actually shows, where it gets thin, and what — if anything — is worth doing about it.

What the evidence actually shows

Let's start with the basics. Cortisol follows a diurnal rhythm: it peaks roughly 30 minutes after waking — the so-called cortisol awakening response — and should decline steadily across the day, reaching its nadir around midnight. That much is well-established physiology, documented consistently across decades of endocrine research.

What's less often discussed is how fragile that rhythm turns out to be. Andreadi et al. (2025) examined cortisol circadian rhythm in night-shift workers and found significant flattening and phase-shifting of the normal curve — with evening and nocturnal cortisol concentrations remaining elevated compared to day-shift controls. The clinical implication is that disrupted cortisol rhythm isn't just a curiosity; it's associated with measurable metabolic and cardiovascular consequences in that population.

Juliana et al. (2025) reviewed cortisol detection methods in the context of circadian rhythm disruption and noted that nocturnal cortisol elevation is increasingly recognised as a distinct physiological pattern — separate from generalised hypercortisolaemia — and that it may be more common in the general population than previously appreciated. The human data on exactly how common is still thin, and I'd be overstating it to give you a precise prevalence figure.

What I can say with reasonable confidence: if you're regularly waking between 2am and 4am, lying there with your mind running, and struggling to get back to sleep, elevated nocturnal cortisol is one plausible mechanism worth considering. It's not the only one. But it's not a fringe theory either.

The biology: what's actually happening at night

Cortisol is produced by the adrenal cortex in response to adrenocorticotropic hormone (ACTH), which is itself released from the anterior pituitary. The whole system — the hypothalamic-pituitary-adrenal (HPA) axis — is governed by a master circadian clock in the suprachiasmatic nucleus (SCN) of the hypothalamus. Light exposure is the primary zeitgeber, or time-giver, that keeps this clock synchronised.

At night, under normal conditions, the SCN suppresses ACTH release. Cortisol falls. The body shifts into repair and consolidation mode. Sleep architecture — particularly slow-wave sleep — is partly dependent on this low-cortisol window. Tuladhar et al. (2021) demonstrated, in a study of infants, that higher diurnal cortisol predicted shorter and more disrupted sleep — a relationship that appears to hold across the lifespan, though the mechanisms differ somewhat with age.

When something disrupts this suppression — psychological stress, light exposure, irregular sleep timing, certain medical conditions — ACTH can pulse inappropriately during the night, producing what is recognisable as a cortisol spike. Russell et al. (2024) showed that the normal ultradian pulsatility of cortisol — those small, rhythmic pulses throughout the day — has real neurological consequences when disrupted, altering emotional processing and fatigue in patients with adrenal insufficiency. The implication for healthy people is indirect, but it's a useful reminder that cortisol isn't just a stress hormone. It's a signalling molecule with precise timing requirements.

It's also worth understanding that cortisol spikes at night aren't always pathological. Acute physiological stressors — surgery, illness, intense exercise — can produce dramatic nocturnal cortisol elevations as a normal adaptive response. Raju et al. (2023) characterised cortisol secretion patterns during and after cardiac surgery, finding substantial nocturnal elevations in the post-operative period. The biology is context-dependent. A single bad night doesn't mean your HPA axis is broken.

When nocturnal cortisol elevation signals something clinical

I want to be careful here. Most people reading this don't have Cushing's syndrome. But it's worth knowing what the clinical threshold looks like, because it helps calibrate what "elevated" actually means.

Cushing's syndrome — caused by chronic cortisol excess, most commonly from a pituitary adenoma or exogenous glucocorticoids — is diagnosed partly through late-night salivary cortisol measurements. Ceccato et al. (2017) reviewed the screening and diagnostic criteria, noting that late-night salivary cortisol has a sensitivity of around 92–100% for Cushing's, making it one of the most reliable first-line tests. The normal late-night salivary cortisol is typically below 4 nmol/L in most laboratory reference ranges.

If you're experiencing significant weight gain around the abdomen, persistent high blood pressure, easy bruising, or stretch marks appearing on previously unaffected skin, those are symptoms worth discussing with a GP — not something to self-manage with supplements. This article is about the more common, subclinical pattern of functional nocturnal cortisol elevation. That's a different thing.

For a deeper look at what cortisol high at night actually looks like in practice — including how to test for it — I've written a more detailed piece on that specifically.

The sleep-cortisol feedback loop

This is where it gets genuinely interesting, and genuinely complicated. The relationship between cortisol and sleep isn't one-directional. Poor sleep may raise nocturnal cortisol. Elevated nocturnal cortisol may worsen sleep. Both things appear to be true simultaneously, which makes causality difficult to establish and intervention difficult to target.

Spiegel et al. (1999) restricted healthy young men to four hours of sleep per night for six nights. By the end of the restriction period, evening cortisol concentrations were significantly higher than in the fully-rested condition — suggesting that sleep loss itself may drive the very cortisol pattern that then makes sleep harder. It's a feedback loop, and it's worth naming it clearly rather than pretending there's a simple upstream cause.

The relationship between high cortisol levels and sleep is something I've written about in more detail elsewhere. The short version: the evidence for bidirectional interference is reasonably solid; the evidence for any specific intervention breaking that loop is considerably weaker.

Measuring cortisol at night: what's actually useful

One thing that frustrates me about the wellness conversation around cortisol is that people talk about it as though it's easy to measure and straightforward to interpret. It isn't.

Balasamy et al. (2024) reviewed cortisol biosensing strategies and noted that while salivary cortisol is a reasonable proxy for free serum cortisol, there are significant inter-individual differences in salivary flow rate, collection technique, and diurnal timing that can produce misleading results if not carefully controlled. Consumer-facing cortisol tests — the ones you order online and spit into a tube at home — can be useful, but they need to be interpreted with appropriate scepticism.

Serum cortisol, measured at a specific time, gives you a snapshot. Urinary free cortisol over 24 hours gives you a total output figure. Late-night salivary cortisol gives you a trough reading. Each measures something slightly different. None of them, alone, gives you the full picture. If you're genuinely concerned about your cortisol pattern, a GP referral to an endocrinologist — with properly timed, laboratory-grade testing — is the right move, not a home test kit interpreted in isolation.

What the research says about lifestyle factors and nocturnal cortisol

Light exposure and sleep timing

The evidence here is probably the strongest of any lifestyle intervention. Evening light exposure — particularly blue-spectrum light from screens — may delay the circadian signal that suppresses ACTH at night. The mechanism is well-characterised, even if the precise magnitude of the cortisol effect in free-living humans is harder to quantify. Keeping the bedroom dark and avoiding bright light in the two hours before sleep is one of the few interventions with a credible mechanistic rationale and essentially no downside.

Grounding and other emerging areas

Ghaly et al. (2005) investigated the effects of grounding — physically connecting the body to the earth's electrical field during sleep — on cortisol levels and subjective sleep quality. The study found that grounding appeared to normalise the cortisol profile in a small sample of participants, with reported improvements in sleep. I'll be honest: this is a small, preliminary study, and I wouldn't hang a protocol on it. But I find it interesting that the intervention was specifically targeted at nocturnal cortisol rhythm rather than overall output. It's an area where the human data is thin and I'd be overstating it to claim grounding is an evidence-based cortisol intervention.

Nutritional factors worth knowing about

Micronutrient status intersects with HPA axis function in ways that are often overlooked. Zinc deficiency, for example, has been studied in relation to HPA axis dysregulation, though the direction of causality in humans isn't fully established. Magnesium is another area of active research. Neither addresses cortisol in any straightforward way — that framing doesn't hold up — but addressing genuine nutritional gaps is a reasonable first step before reaching for more targeted interventions.

Supplements that have been studied in this context

I want to be precise here about what "studied" means. There's a meaningful difference between "there is mechanistic or animal data suggesting a plausible effect" and "there are well-powered human RCTs demonstrating a clinically meaningful outcome." Most of what follows sits closer to the first category than the second.

Glycine

Glycine is an inhibitory neurotransmitter and a conditionally essential amino acid. Research suggests it may play a role in sleep onset and core body temperature regulation, both of which interact with nocturnal cortisol dynamics. Large-scale human trials are limited, and the research is ongoing. The KōJō Daily Formula includes 2,000mg of crystalline glycine — a dose consistent with the amounts used in the available human studies — though I'd be the first to say the field needs more data before making strong claims about its effects on nocturnal cortisol specifically.

Taurine

Taurine has been studied for its potential role in GABAergic signalling and autonomic nervous system regulation. Some preliminary animal and in vitro data suggests it may influence stress-related pathways, but large-scale human trials examining its effect on nocturnal cortisol are limited, and the research is ongoing.

Aged Garlic Extract

Aged Garlic Extract has been studied for its antioxidant properties and potential effects on cardiovascular markers. Its relationship to cortisol rhythm in humans is not well-established, large-scale human trials are limited, and the research is ongoing. I include it in this context only because oxidative stress and HPA axis activity are related — not because there's direct evidence linking garlic extract to nocturnal cortisol specifically.

Olive Leaf Extract, Grape Seed Extract, Pine Bark Extract

These polyphenol-rich extracts have been studied primarily in the context of oxidative stress and vascular function. Vitamin C contributes to the protection of cells from oxidative stress — that's a registered claim with solid mechanistic backing. The polyphenol extracts are in a different category: the human data on their direct effects on cortisol rhythm is thin, large-scale human trials are limited, and the research is ongoing. I'd be cautious about anyone selling you these as specific nocturnal cortisol interventions.

Frequently asked questions

My honest take

I started looking into this because I was waking up at 3am with some regularity and couldn't work out why. I wasn't under unusual stress. I wasn't drinking heavily. I was exercising. On paper, nothing was obviously wrong. But something was clearly off with my sleep architecture.

What I found when I read the primary literature was both reassuring and frustrating. Reassuring because the biology is well-understood — cortisol rhythm is a real thing, nocturnal disruption is a real phenomenon, and it has measurable consequences. Frustrating because the intervention evidence is genuinely weak. There's no supplement with a solid human RCT showing it reliably attenuates cortisol spikes at night in otherwise healthy adults. Anyone telling you otherwise is either selling something or hasn't read the papers carefully.

What actually helped me, for what it's worth: consistent sleep and wake times (more important than I'd appreciated), cutting screen use after 9pm, and not eating large meals within two hours of bed. Boring answers. Not sellable. But the data supports them more than any supplement I've found.

I do take glycine before bed — 2,000mg, the same dose as in the KōJō Daily Formula — because the preliminary human data on sleep onset is interesting enough to warrant it, and the safety profile is excellent. But I hold that lightly. It's a plausible intervention, not a proven one.

If you're regularly waking at night and feel like your stress response is dysregulated, please talk to your GP before self-experimenting. Juliana et al. (2025) make the point that proper cortisol measurement requires careful methodology — and that's doubly true when you're trying to track changes over time. A proper test, interpreted by someone qualified, is worth more than a stack of supplements chosen based on a blog post. Including this one.

This article is for informational purposes only and does not constitute medical advice. Consult your healthcare provider before starting any supplement regimen.