Sleep and cortisol: What the evidence shows

Sleep and cortisol are locked in a bidirectional relationship that most people underestimate. Cortisol follows a strict 24-hour rhythm — peaking roughly 30 minutes after waking and falling through the day — and disrupted sleep measurably distorts that pattern. A 2024 meta-analysis found that acute sleep deprivation may significantly alter evening cortisol levels, with effect sizes varying by timing and duration of deprivation. The relationship runs both ways: elevated cortisol at night appears to fragment sleep architecture in return.

What the evidence actually shows

The clearest signal in the literature is this: sleep and cortisol regulate each other, and when one goes wrong, the other usually follows. Chen et al. (2024) conducted a systematic review and meta-analysis specifically on acute sleep deprivation and cortisol, and found that evening cortisol levels may be particularly sensitive to sleep loss — though the authors note considerable heterogeneity across studies, which makes sweeping conclusions difficult. I'd be overstating it to say we have a clean dose-response curve here. We don't.

What we do have is a reasonably consistent picture from multiple angles. Liu (2024) published a detailed review arguing that cortisol rhythms actively mediate the relationship between sleep architecture and broader health outcomes — not just as a downstream marker, but as a mechanistic driver. That framing matters. It means cortisol isn't simply a readout of poor sleep. It's part of the causal chain.

Zhao et al. (2021) reviewed the evidence on how sleep quality affects the salivary cortisol response to acute stressors. Their conclusion was measured: sleep appears to modulate cortisol reactivity, but the direction and magnitude depend heavily on sleep stage, timing, and individual differences. That kind of nuance rarely makes it into supplement marketing. It should.

If you want to go deeper on the specific effects of chronic sleep restriction on cortisol output, I've written a more focused piece on lack of sleep cortisol that covers the longer-term data.

The mechanism: what's biologically happening

Cortisol is produced by the adrenal cortex in response to adrenocorticotropic hormone (ACTH), which is itself released from the pituitary in response to corticotropin-releasing hormone (CRH) from the hypothalamus. That HPA axis — hypothalamic-pituitary-adrenal — is the core circuit here.

Under normal conditions, cortisol secretion follows a circadian rhythm driven by the suprachiasmatic nucleus (SCN), the brain's primary circadian clock. Cortisol is typically lowest around midnight, begins rising in the early hours of the morning, and peaks 20–40 minutes after waking — a phenomenon called the cortisol awakening response (CAR). Elder et al. (2014) reviewed the CAR extensively and noted it represents a distinct, active process — not just a continuation of the overnight rise — and that its magnitude appears sensitive to sleep quality, psychosocial stress, and HPA axis reactivity.

Sleep itself is not a passive state for cortisol regulation. During slow-wave sleep (SWS), HPA activity is actively suppressed. Growth hormone secretion peaks during SWS, and there's a reciprocal inhibitory relationship between GH and cortisol. Disrupting SWS — through fragmentation, shortened duration, or poor sleep efficiency — may reduce that suppression and allow cortisol to remain relatively elevated into the night.

REM sleep appears to play a different role. The transition into REM is associated with rising cortisol, and the proportion of REM sleep in the latter half of the night tracks the pre-waking cortisol rise. Liu (2024) describes this as a functional coupling — cortisol rhythm and sleep architecture are co-regulated rather than one simply causing the other.

One clinical example that illustrates this clearly: in Cushing's syndrome, where cortisol is chronically and pathologically elevated, sleep architecture is severely disrupted. Ferrari et al. (2025) found that chronotherapy targeting cortisol rhythm restoration may improve sleep quality in Cushing's patients — which suggests the relationship is mechanistically reversible, at least in that context. Cushing's is an extreme case, but it makes the pathway legible.

Sex differences and oral contraceptive use

This is an area I don't see discussed enough. The sleep-cortisol relationship is not uniform across sexes, and hormonal context matters considerably. Morssinkhof et al. (2025) examined cortisol dynamics and sleep quality with specific attention to sex and oral contraceptive (OC) use. Their data suggest that OC users may show altered cortisol dynamics compared to naturally cycling women, with potential downstream effects on sleep quality. The sample sizes in this space are still relatively modest, and I'd be cautious about overgeneralising — but it's a meaningful signal that warrants more attention.

For men, the picture is somewhat different. Liu et al. (2022) reviewed the interplay between sleep, testosterone, and cortisol in ageing men, noting that the testosterone-to-cortisol ratio may be a more informative marker of physiological stress than cortisol alone. Sleep restriction appears to shift that ratio in a direction associated with poorer recovery. The human data here is reasonably consistent, though most studies are observational rather than interventional.

Cortisol, sleep, and cognitive function

The cognitive consequences of disrupted sleep-cortisol dynamics are worth taking seriously. Henry et al. (2022) explored the relationship between sleep, cognition, and cortisol in the context of Addison's disease — a condition of cortisol insufficiency — and found evidence for a mechanistic link between abnormal cortisol patterns and impaired cognitive performance. Again, this is a pathological model, but it illuminates the normal physiology by showing what happens when the system fails.

There's also the question of what happens to decision-making under sleep-disrupted, high-cortisol conditions. Law et al. (2024) found that sleep quality, cortisol reactivity, and risk-reward decision-making appear to interact in ways that compound impairment — each variable amplifying the effect of the others. That's a sobering finding, and the human data on this particular interaction is still developing.

There's a longer-horizon concern here too. Pistollato et al. (2017) reviewed associations between sleep, cortisol regulation, and diet in the context of Alzheimer's disease risk, noting that chronically dysregulated cortisol — particularly when combined with poor sleep — may be associated with increased amyloid burden over time. The evidence here is largely observational and mechanistic; I wouldn't overstate the clinical implications. But it's another data point suggesting this isn't a trivial relationship.

What the evidence supports for dietary and supplement approaches

I want to be direct here: there is no supplement with strong, replicated RCT evidence for directly normalising the sleep-cortisol axis in healthy adults. Anyone telling you otherwise is selling something. That said, a few ingredients have been studied in relevant contexts, and it's worth being honest about what the data does and doesn't show.

Creatine and sleep deprivation

Creatine is the most interesting case in this context. McMorris et al. (2006) found that creatine supplementation may attenuate some cognitive and psychomotor performance decrements associated with sleep deprivation combined with mild exercise — though the study was small (n=10) and the cortisol data specifically was not the primary outcome. Creatine increases physical performance in successive bursts of short-term, high intensity exercise — that's the registered claim, and it's well-supported. Whether it has a meaningful role in sleep-deprivation resilience is plausible but not yet conclusively established in large trials. The KōJō Daily Formula includes 5,000mg of micronised creatine monohydrate — the dose range used in most performance research.

Glycine

Glycine has been studied for its potential role in sleep quality — some small trials suggest it may help with subjective sleep satisfaction and morning alertness — but the large-scale human trial data is limited, and I wouldn't overstate what we know. Research is ongoing. The formula includes 2,000mg.

Taurine

Taurine has been studied in the context of the nervous system and stress response in animal models, with some preliminary human data. Large-scale RCTs in healthy humans are limited, and the evidence base is still developing. The formula includes 2,000mg.

Polyphenol extracts

Aged Garlic Extract, Olive Leaf Extract, Grape Seed Extract, and Pine Bark Extract are included in the formula for their studied antioxidant properties — Vitamin C contributes to the protection of cells from oxidative stress, which is the registered claim in that space. For the polyphenol extracts specifically, research into their relationship with the HPA axis and cortisol is preliminary, and large-scale human trials are limited. I include them because the mechanistic rationale is interesting, not because I can point to definitive RCT evidence for cortisol modulation.

If you're interested in the adaptogen literature — specifically ashwagandha, which has more RCT data on cortisol than most ingredients — I've covered that in detail in the piece on adaptogens ashwagandha.

The cortisol awakening response as a practical marker

The cortisol awakening response (CAR) deserves its own section because it's genuinely useful as a window into HPA axis function. It's measurable via salivary samples — typically collected at waking, then 15, 30, and 45 minutes post-waking — and it gives you a snapshot of HPA reactivity that evening blood draws can't capture.

Elder et al. (2014) reviewed the CAR's applications in sleep medicine and noted that blunted or exaggerated CARs are associated with different pathological profiles — burnout, PTSD, and shift work disorder all show characteristic CAR patterns. A blunted CAR isn't always a sign of low cortisol overall; it may reflect HPA axis dysregulation rather than simple deficiency.

For most people, the practical takeaway is that consistent wake times matter more than most people realise. The CAR is partly anchored to anticipated wake time — disrupting that anchor through irregular sleep schedules appears to attenuate the response. If you want to understand your own cortisol pattern better, I'd suggest reading more about high cortisol symptoms sleep before reaching for any supplement.

Frequently asked questions

My honest take

I started paying attention to the sleep-cortisol literature because I noticed my own sleep was worse during high-stress periods — which, as a founder, is most of the time. I assumed it was psychological. The research suggests it's also physiological, and that the causal arrows run in both directions simultaneously. That's both reassuring and slightly annoying, because it means there's no single lever to pull.

What I've taken from this personally: consistent wake times matter more than I used to think. The CAR data is compelling enough that I treat irregular sleep schedules as a genuine stressor rather than a minor inconvenience. I've also stopped assuming that feeling tired means my cortisol is low — the relationship is more complicated than that, and a blunted CAR can coexist with elevated evening cortisol in ways that feel confusing from the inside.

On supplements: I'm genuinely cautious about overclaiming here. The ingredients in the KōJō Daily Formula are included because the mechanistic rationale is sound and the safety data is good — not because I can point to large RCTs showing they normalise cortisol in healthy adults. Vitamin C contributes to the reduction of tiredness and fatigue, and that's a registered claim I'm comfortable making. For the rest, I think the honest position is: promising, worth including, but the large-scale human trial evidence isn't there yet.

The thing I keep coming back to is that sleep hygiene — consistent timing, dark environment, avoiding bright light in the hour before bed — has more evidence behind it than any supplement I'm aware of for this specific application. I include that not to be preachy, but because it's true and I'd rather say it than pretend the pill does the work the behaviour should do.

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