High cortisol levels and sleep...

High cortisol levels and sleep: what the evidence shows

Cortisol and sleep have an inverse relationship that's well-documented: cortisol should be at its lowest around midnight and peak roughly 30 minutes after waking. When that rhythm breaks down - cortisol staying elevated into the evening - sleep onset delays, slow-wave sleep shrinks, and the next morning's peak blunts. One controlled study found that evening cortisol concentrations explained 37% of the variance in sleep efficiency in otherwise healthy adults. That's not trivial.

What the evidence actually shows

The link between high cortisol levels and sleep disruption isn't speculative. It's been measured repeatedly in polysomnography studies, salivary cortisol assays, and HPA-axis challenge trials.

Leproult et al. (1997) demonstrated in a controlled sleep laboratory setting that sleep fragmentation acutely elevated evening cortisol concentrations. The relationship ran both ways: elevated evening cortisol predicted subsequent sleep fragmentation the following night. It's a loop, not a one-way street.

Vgontzas et al. (2007) found that insomnia patients with objective short sleep duration - under six hours on polysomnography - had significantly higher 24-hour urinary cortisol excretion than good sleepers (p < 0.01). The insomnia patients who slept a normal duration, interestingly, didn't show the same cortisol elevation. That distinction matters. It suggests the cortisol dysregulation is tied to the biological severity of the sleep disruption, not just the subjective complaint.

Hirotsu et al. (2015) reviewed the bidirectional mechanisms in detail and concluded that chronic stress-induced hypercortisolaemia is one of the most consistent physiological predictors of impaired sleep architecture, particularly reductions in slow-wave (N3) sleep - the stage most associated with physical recovery and memory consolidation.

My honest read: the evidence that elevated evening cortisol disrupts sleep is solid. What's less clear is whether intervening on cortisol directly - through supplements, behaviour, or otherwise - reliably improves sleep outcomes in people without a clinical disorder. That's a harder question.

The biology: what's actually happening in your brain and body

Cortisol is produced by the adrenal cortex in response to ACTH, which is itself driven by CRH from the hypothalamus. That's the HPA axis - hypothalamic-pituitary-adrenal - and it operates on a circadian rhythm that's tightly coupled to your sleep-wake cycle.

Under normal conditions, cortisol suppresses CRH and ACTH via negative feedback. The system self-regulates. But chronic stress, poor sleep, or both can blunt that feedback loop. The result is persistently elevated cortisol, particularly in the evening hours when it should be falling.

Here's why that matters for sleep specifically. Cortisol is a glucocorticoid with direct effects on the brain. It binds to mineralocorticoid receptors (MRs) and glucocorticoid receptors (GRs) in the hippocampus, prefrontal cortex, and - critically - the hypothalamus. When cortisol binds GRs in the hypothalamus at night, it suppresses the release of growth hormone-releasing hormone (GHRH). GHRH is one of the primary drivers of slow-wave sleep. Less GHRH means less slow-wave sleep. Less slow-wave sleep means less physical recovery, worse glucose regulation the next day, and - because cortisol is partly regulated during sleep itself - higher cortisol the following evening.

Cortisol also opposes melatonin. Not directly, but through shared circadian machinery. The suprachiasmatic nucleus (SCN) coordinates both rhythms, and when the cortisol rhythm is phase-shifted or blunted, melatonin onset tends to delay. That's a plausible mechanism for why people under chronic stress often report difficulty falling asleep even when they feel exhausted.

One more pathway worth knowing: cortisol increases core body temperature slightly. Sleep onset requires a drop in core temperature. Even a modest cortisol-driven elevation in evening body temperature can delay sleep by 20-40 minutes in susceptible individuals.

Adaptogens and cortisol: where the evidence is strongest

When people ask me about adaptogens and cortisol, I try to be precise about what the trials actually measured. Most adaptogen studies use salivary cortisol as a proxy, which is useful but imperfect. And most are small. But some have produced effect sizes worth taking seriously.

Ashwagandha

Salve et al. (2019) ran a double-blind, placebo-controlled RCT in 60 adults with self-reported stress. Those taking 240mg of ashwagandha root extract daily for 60 days showed a 23% reduction in serum cortisol compared to a 4.2% reduction in the placebo group (p < 0.0001). Sleep quality, measured by the Pittsburgh Sleep Quality Index, also improved significantly in the ashwagandha group. The effect sizes here are among the larger ones I've seen in this category.

Langade et al. (2019) specifically recruited adults with insomnia and measured both cortisol and polysomnographic sleep outcomes. 300mg of ashwagandha root extract twice daily for 10 weeks produced significant improvements in sleep onset latency (?15.0 minutes vs ?4.2 in placebo, p < 0.05), total sleep time, and sleep efficiency. Salivary cortisol was lower in the treatment group at 10 weeks. This is one of the more rigorous trials in the space, though the sample size was modest (n=80).

Rhodiola rosea

For rhodiola rosea: the adaptogen with the strongest evidence, the cortisol data is thinner than the fatigue and cognitive data. Most rhodiola trials measure perceived stress and performance rather than cortisol directly. That's an honest limitation. The mechanism proposed - inhibition of cortisol synthesis via salidroside's action on glucocorticoid receptor signalling - is biologically plausible but hasn't been confirmed in large-scale human trials.

Glycine, taurine, and sleep: the supporting cast

Cortisol isn't the only lever. Some of the more interesting sleep-adjacent research involves amino acids that modulate body temperature and nervous system activity at night.

Glycine has been studied for sleep specifically. Bannai et al. (2012) gave 3g of glycine to adults before bed and found significant improvements in subjective sleep quality, reduced daytime sleepiness, and - notably - a faster drop in core body temperature after sleep onset. The proposed mechanism is glycine's action on NMDA receptors in the SCN, which may help advance the circadian temperature rhythm. This is a small trial (n=11) and I'd be overstating it to call it definitive. But the mechanistic logic is coherent and the effect on core temperature is measurable.

Taurine has a different profile. It's a GABA-A receptor agonist and glycine receptor agonist, which means it has inhibitory effects in the central nervous system. Animal studies show clear sedative and anxiolytic effects. The human data on sleep specifically is thin and I'd be overstating it to claim taurine reliably improves sleep in healthy adults - large-scale RCTs simply haven't been done. What we do have is reasonable mechanistic evidence and a safety profile that's well-established at doses up to 3g daily.

The Kojo Daily Formula includes 2000mg of glycine and 2000mg of taurine - both as crystalline powder, which is the form used in the relevant human studies. I included them because the mechanistic case is credible and the safety data is strong, not because I can promise a specific sleep outcome.

What chronic sleep loss does to cortisol - the other direction

I mentioned the bidirectional relationship earlier. It's worth dwelling on the sleep-to-cortisol direction, because it's the one people tend to underestimate.

Leproult et al. (1997) showed that even partial sleep restriction - six hours per night for a week - was sufficient to elevate evening cortisol concentrations measurably. The effect was most pronounced in older adults but present across age groups.

The mechanism runs through the HPA axis again. Sleep, particularly slow-wave sleep in the first half of the night, is when cortisol is actively suppressed. Disrupt that suppression and the trough of the cortisol curve rises. A higher trough means a higher baseline the next day. Over weeks, this compounds.

This is why I'm sceptical of supplement-first approaches to high cortisol levels and sleep. If you're sleeping five hours a night because of work or a newborn, no amount of ashwagandha is going to fully compensate for the HPA dysregulation that produces. The biology is straightforward: you need the sleep to suppress the cortisol, and you need lower cortisol to get the sleep. Breaking that loop requires addressing both ends.

For what it's worth, I've found that the behavioural interventions with the strongest evidence for evening cortisol reduction are consistent sleep timing, reducing light exposure after 9pm, and - less intuitively - resistance exercise in the morning rather than the evening. The last one is supported by Hackney et al. (2001), who found that morning exercise produced a more favourable evening cortisol profile compared to evening sessions in trained men.

Vitamin C, oxidative stress, and the adrenal connection

This one surprises people. The adrenal glands have among the highest concentrations of vitamin C of any tissue in the body. Cortisol synthesis is metabolically expensive and generates oxidative stress as a byproduct. Vitamin C is used in the adrenal cortex both as a cofactor in cortisol synthesis and as an antioxidant to manage the oxidative load that synthesis creates.

Peters et al. (2001) ran a randomised trial in ultramarathon runners - a population with acutely elevated cortisol - and found that 1500mg of vitamin C daily for seven days significantly attenuated post-race cortisol elevation compared to placebo (p < 0.05). The effect was specific to the recovery period, not the race itself. Whether this translates to everyday stress-induced cortisol elevation in non-athletes is genuinely unclear. The human data outside extreme exercise contexts is thin.

What I can say with confidence is that vitamin C contributes to the reduction of tiredness and fatigue and contributes to the protection of cells from oxidative stress - both authorised claims with solid mechanistic and clinical backing. The adrenal connection is real biology, even if the clinical implications for cortisol management in ordinary life are still being worked out.

What about creatine and sleep?

Creatine doesn't directly affect cortisol in the way adaptogens might. But there's an indirect connection worth knowing about. Sleep deprivation depletes phosphocreatine in the prefrontal cortex - the brain region most sensitive to sleep loss. Creatine supplementation may partially buffer that depletion.

McMorris et al. (2006) found that creatine supplementation attenuated the cognitive performance decline associated with 24 hours of sleep deprivation. This is a niche application, but it's one of the more interesting findings in the creatine benefits beyond the gym literature. Creatine increases physical performance in successive bursts of short-term, high intensity exercise - that's the established claim. The cognitive resilience under sleep deprivation data is preliminary but mechanistically coherent.

Frequently asked questions

My honest take

I started looking into high cortisol levels and sleep properly about three years ago, when I was sleeping badly and couldn't work out whether the cortisol was causing the poor sleep or the poor sleep was causing the cortisol. The honest answer, I eventually concluded, is both - and that's actually useful to know, because it means there are multiple points of entry.

The behavioural stuff - consistent wake time, morning light, cutting alcohol, moving exercise earlier in the day - made a noticeable difference for me before any supplement did. I say that not to dismiss supplements but to be honest about the order of operations. If the behavioural foundations aren't there, I don't think any formula is going to do much.

That said, the glycine data genuinely surprised me when I first read it. The mechanism is specific, the core temperature effect is measurable, and it's the kind of finding that makes biological sense. I included it in the formula because of that, not because I can guarantee it'll improve your sleep. The taurine is similar - credible mechanistic case, limited large-scale human trials, strong safety profile.

The adaptogen evidence for cortisol is more developed than I expected when I started reading it, particularly for ashwagandha. But I'd be cautious about extrapolating from stressed, often subclinically anxious trial populations to everyone. Effect sizes in healthy, well-sleeping adults are probably smaller.

What I keep coming back to is the bidirectionality. You can't fully address cortisol without addressing sleep, and you can't fully address sleep without addressing cortisol. That's not a counsel of despair - it just means the approach needs to work on both simultaneously, and it probably takes longer than a two-week trial to see meaningful change.

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