How to Lower Cortisol: What the Research Supports — and Why a Single Test Tells You Nothing

Cortisol has become the hormone everyone wants less of, which is a strange fate for a molecule you cannot live without.

The premise behind most “lower your cortisol” advice is that there’s a number, that yours is too high, and that the right supplement or breathing protocol brings it down. Almost none of that survives contact with the endocrinology.

Cortisol isn’t a level. It’s a waveform — pulses riding on a daily rhythm, peaking shortly after you wake and falling through the evening.¹ A single measurement samples one point on that curve and tells you very little about the curve itself.

What the trial evidence supports is narrower and less exciting than the internet version: sleep, moderate aerobic and mind-body exercise, and not training yourself into chronic overreach. Here’s what each actually does, and what you can track without pretending to measure a hormone no consumer device can see.

Question	Short answer
Does exercise lower cortisol?	Over weeks, yes — modestly, and mostly in people who are already distressed. Yoga showed the largest effect in a 44-trial network meta-analysis²
Does working out increase cortisol?	Yes, acutely, and it’s intensity-gated. Thirty minutes at 80% VO₂max raised cortisol 83.1%; 40% raised it 5.7%³
How long does it stay up?	Hours, not days. Elevated through 60 minutes post-session, below baseline by 120–180 minutes, back to baseline at 24 hours⁴
Best exercise to lower cortisol?	Yoga and qigong in the distressed populations studied, at roughly 530 MET-min/week — about 150 minutes of moderate activity²
Can I test it at home?	Not usefully. Valid awakening-response measurement requires objectively time-verified samples across multiple days⁵⁶
What actually moves the curve?	Sleep. One night of partial sleep loss raised the next evening’s cortisol 37%⁷

Why Your Cortisol Number Almost Certainly Means Nothing

There is no single “normal cortisol level” to compare yourself against, because cortisol is not a level — it’s a pulsatile waveform on a daily rhythm, and a single value does not reliably reflect your glucocorticoid status.⁸ A single measurement is a photograph of one intersection, taken at an unknown time, used to judge a city’s traffic.

Cortisol Is a Waveform, Not a Level

Cortisol is secreted in ultradian pulses — bursts every hour or so — layered on top of a circadian rhythm. Stafford Lightman, PhD, FRS, FMedSci, Professor of Medicine at the University of Bristol, has spent his career making the case that this pulsatility is not noise around a “true” level but the signal itself: the hypothalamic-pituitary-adrenal axis produces a dynamic oscillating output that target cells decode at the receptor level.¹ Flatten the pulses — as long-acting synthetic glucocorticoids do — and that dynamic signal is lost, with the potential to affect physiology adversely.¹

A 2024 Endocrine Reviews assessment led by Sophie Clarke and colleagues at Imperial College London says so directly: single cortisol values do not necessarily reflect an individual’s glucocorticoid status. The reasons are inter- and intra-individual variation, pulsatile secretion, differences in binding proteins, local tissue conversion by 11β-hydroxysteroid dehydrogenase, and receptor sensitivity. Current tools spot profound disturbances well and confirm optimal status poorly.⁸

Why At-Home Saliva Kits Can’t Fix This

Now consider the direct-to-consumer saliva kit that ships you four tubes.

Tobias Stalder, PhD, Professor of Clinical Psychology at the University of Siegen, led the International Society of Psychoneuroendocrinology panel that wrote the consensus guidelines for measuring the cortisol awakening response. Their central requirement is objective verification of when you actually woke and when you actually spat, because validity depends on the first sample landing at the true moment of awakening.⁵ When the panel re-audited the field in 2022, researchers who had read the guidelines still largely weren’t doing it.⁶

If funded labs can’t reliably manage this, a mail-order tube on your nightstand isn’t managing it either.

What Cortisol Actually Does All Day

You don’t want less cortisol. You want a curve with a strong morning peak and a clean evening descent.

The rise over the first 30–45 minutes after waking — the cortisol awakening response — is a feature, not a malfunction.⁵ It’s your metabolism switching on for the day. A blunted morning rise is the worrying pattern, not a reassuring one.

The other half of the curve matters more than most people realize. Emma K. Adam, PhD, Professor of Human Development and Social Policy at Northwestern University, led the meta-analysis making the case that what predicts health outcomes is the diurnal slope — the steepness of the fall from morning to night.

Across 179 associations drawn from 80 studies, flatter slopes were significantly associated with poorer health, average effect size r = 0.147. Flatter slopes tracked with worse outcomes in 10 of 12 health subtypes examined, strongest for immune and inflammatory outcomes at r = 0.288.⁹

Note the size of those numbers: real, replicated, and modest — the kind of relationship that shows up reliably in populations and predicts very little about any one person. Adam and colleagues deliberately left open whether a flat slope is a marker of disease or a mechanism driving it.⁹

So the reframe is: stop asking “is my cortisol high?” and start asking “does my day have a shape?” That’s a question about light, sleep timing, and training load — the territory SensAI works in, and answerable with data you can actually collect.

Exercise Raises Cortisol — and That’s the Point

Yes, exercise raises cortisol acutely, and the effect is gated by intensity rather than duration.

The cleanest demonstration remains a controlled study by Emily Hill and colleagues, including A.C. Hackney at UNC Chapel Hill: twelve active men, 30 minutes of exercise at three intensities plus a resting control session, with time of day, prior diet, and training status controlled.

30-minute session	Change in circulating cortisol
Resting control (no exercise)	−6.6 ± 3.5%
40% VO₂max	+5.7 ± 11.0%
60% VO₂max	+39.9 ± 11.8%
80% VO₂max	+83.1 ± 18.5%

The 60% and 80% responses were significantly larger than the others and than each other. ACTH — the pituitary signal driving cortisol release — mirrored the pattern but only reached significance at 80%. Once the authors corrected for plasma volume loss and circadian drift, the 40% session actually reduced circulating cortisol.³

Same person, same 30 minutes, opposite direction. The variable is intensity.

Here’s the part that defuses the whole “exercise is a stressor” anxiety. Flavio Cadegiani and Claudio Kater at the Federal University of São Paulo compared 25 healthy male athletes with 12 non-physically-active controls using gold-standard, exercise-independent provocation tests. Under an insulin tolerance test, the athletes’ cortisol response came earlier and larger: 21.7 ± 3.1 µg/dL versus 16.9 ± 4.1 µg/dL (p < 0.001), a mean increase 52.2% higher than controls. Growth hormone and prolactin followed suit; response to direct adrenal stimulation was similar between groups.¹⁰

The athletes’ axis wasn’t damaged. It was conditioned — the way a trained heart is better at raising stroke volume on demand. A bigger, faster cortisol response to a real challenge is a fitter stress system, not a broken one.

How Long Cortisol Stays Elevated After a Hard Session

A hard workout is a several-hour cortisol event, not a several-day one.

Manuel Dote-Montero and colleagues meta-analysed the acute hormonal response to a single HIIT session across 10 controlled studies (213 participants) and 50 pre-post groups (677 participants). The time course is unusually clean:

Time after session	Cortisol effect size (controlled studies)	Cortisol effect size (pre-post groups)
Immediately after	d = 2.17	d = 0.64
30 minutes	d = 1.62	d = 0.67
60 minutes	d = 1.32	d = 0.27
120 minutes	—	d = −0.95
180 minutes	—	d = −1.08
24 hours	—	d = −0.02

Cortisol climbs, stays up for about an hour, then undershoots baseline for the next couple of hours, then returns to where it started by the next day.⁴

That undershoot is the detail nobody quotes. Three hours after a hard interval session, the average participant’s cortisol sat substantially below where it started — a finding that comes from the uncontrolled pre-post groups rather than the controlled subset, since the later timepoints have no control-group estimate.⁴

Test yourself at that moment and you’d conclude HIIT lowers cortisol. Test two hours earlier and you’d conclude the opposite. Both readings real; both conclusions wrong.

Which is why training decisions get made on multi-day trends. SensAI’s coach reads the direction of your sleep, resting heart rate, and HRV over rolling windows rather than reacting to any single day’s number — because a single day’s number, in this domain, is mostly timing.

What the Trials Say Actually Lowers Cortisol

Of the exercise modalities tested head-to-head, mind-body practice has the strongest evidence for reducing cortisol — and the effect has been demonstrated almost exclusively in people who were already distressed.

A 2025 network meta-analysis by Xiongjie Li and colleagues pooled 44 randomized controlled trials of exercise in adults with psychological distress. Yoga produced the largest cortisol reduction (SMD = −0.59; 95% CI −0.90 to −0.28) with a SUCRA ranking of 93%, meaning it ranked first across the network of pairwise comparisons. Qigong and multicomponent exercise followed. High-intensity interval training trended toward increasing cortisol, though not significantly.²

Rank	Modality	Direction of effect
1 (SUCRA 93%)	Yoga	Largest reduction, SMD = −0.59 (95% CI −0.90 to −0.28)
2	Qigong	Reduction
3	Multicomponent exercise	Reduction
—	High-intensity interval training	Trended toward increasing cortisol; not significant

The dose-response finding is the more useful number. Effects followed an inverted U with the optimum at roughly 530 MET-min/week — about 150 minutes a week of moderate activity, which is also the standard public-health target. More was not better. Longer intervention duration did predict greater reductions, and certainty of evidence ranged from very low to high, with yoga-versus-control on the strongest footing.²

Meditation shows a similar shape with an asterisk. Ádám Koncz and colleagues found a significant, medium-sized cortisol reduction across the 10 randomized trials using blood samples — but on inspection that effect existed only in at-risk samples, specifically patients with a somatic illness. Across the 21 saliva-sampled studies, the effect was small and not significant.¹¹

Now the caveat most articles on this topic omit.

Every trial above recruited distressed, stressed, or unwell participants. Lowering cortisol in a dysregulated axis is a plausible therapeutic target. Lowering it in someone whose curve is already normal has never been shown to benefit anything — and given what Adam’s slope data implies about flattened curves, it isn’t obvious you’d want to.

So if you’re sleeping well, training sensibly, and not miserable, the correct cortisol intervention is probably none. The programming questions worth your attention — how much intensity, how often, how much recovery — are the ones SensAI is built to answer, and none of them require a hormone panel.

Cortisol and Sleep: The Lever With the Cleanest Causal Evidence

If you want one intervention with a clean, causal effect on cortisol, it’s sleep — and the classic experiment is nearly thirty years old.

Rachel Leproult, Georges Copinschi, Orfeu Buxton, and Eve Van Cauter at the University of Chicago tracked plasma cortisol continuously across 32 hours in healthy young men under normal sleep, partial sleep deprivation, and total sleep deprivation. The effect showed up in the evening after the disrupted night: cortisol across the 18:00–23:00 window rose 37% after partial deprivation (p = 0.03) and 45% after total deprivation (p = 0.003), with the onset of the evening quiescent period pushed back by at least an hour.⁷

Read that again — the damage wasn’t a morning spike. It was a failure to come down at night. Sleep loss flattens exactly the part of the curve Adam’s meta-analysis flagged as health-relevant. And sleep debt doesn’t clear on the schedule people assume; we’ve covered how long recovery markers actually take to normalize.

Then there’s the light finding, where the “cortisol detox” genre quietly gets it backwards. Katja Petrowski and colleagues exposed healthy men to standardized LED light for one hour during the post-awakening window. Bright white light (414 lux) raised salivary cortisol relative to dim light (under 2 lux); blue light (201 lux) raised it relative to both red and dim light.¹²

Morning light increases cortisol. That’s a sharper awakening response and a steeper slope into the evening — precisely the shape you want. Anyone selling you a protocol to suppress morning cortisol is selling you a flatter curve.

Cortisol and Belly Fat: What the Evidence Supports, and What It Doesn’t

There is a real association between cortisol and central adiposity. It is modest, and in ordinary people the direction of causation remains unresolved.

The study that launched the popular version is Elissa Epel’s, with Bruce McEwen among the co-authors. Fifty-nine healthy premenopausal women — 30 with a high waist-to-hip ratio, 29 with a low one — went through three consecutive laboratory stress sessions plus a rest session across four days. The high-WHR women rated the challenges as more threatening, performed worse, and secreted significantly more cortisol during the first session. The striking subgroup finding: lean women with a high WHR failed to habituate, still over-secreting cortisol on days 2 and 3 when the tasks were already familiar.¹³

Good study. Also cross-sectional, n = 59, and about acute laboratory reactivity — not about whether cortisol built the fat.

For chronic exposure you need hair. Clemens Kirschbaum, PhD, Professor of Biopsychology at TU Dresden, pioneered hair-cortisol analysis precisely because blood, saliva, and urine can’t capture months of secretion; hair incorporates cortisol as it grows, giving a retrospective integrated measure.¹⁴ With Sarah Jackson and Andrew Steptoe at UCL, he applied it to 2,527 adults aged 54 and older in the English Longitudinal Study of Ageing.

The correlations: weight r = 0.102, BMI r = 0.101, waist circumference r = 0.082. Hair cortisol was significantly elevated in participants with obesity and raised waist circumference, and was associated with the persistence of obesity over the preceding four years.¹⁵

An r near 0.10 means cortisol explains roughly one percent of the variance in body weight. Real signal in a large sample. Not the reason for your waistline.

The contrast case is instructive. In Cushing’s syndrome — genuine, prolonged cortisol excess — visceral fat accumulation is a defining feature, alongside moon face, proximal myopathy, easy bruising, and hirsutism in women.¹⁶ That’s what pathological hypercortisolism does to body composition, and it looks nothing like ordinary stubborn abdominal fat. For what actually drives that, see our guide to losing belly fat.

Signs of High Cortisol: Which Ones a Doctor Actually Cares About

The signs that actually point to high cortisol are the ones hard to explain any other way: easy bruising, proximal muscle weakness (trouble rising from a chair or climbing stairs), facial plethora, and wide purple striae. Fatigue, abdominal weight gain, and facial puffiness are not discriminating — they are three of the most common complaints in general practice and have dozens of causes.

Sign	Discriminating?	Why
Wide purple striae	Yes	Few other conditions produce them
Easy bruising without trauma	Yes	Reflects capillary fragility from cortisol excess
Proximal muscle weakness	Yes	Cortisol-driven myopathy hits proximal muscles first
Facial plethora	Yes	Distinct from generic puffiness
Fatigue	No	Near-universal presenting complaint
Abdominal weight gain	No	Dozens of causes, most of them not endocrine
Facial puffiness alone	No	Sodium, alcohol, sleep, allergies, cycle phase

The Endocrine Society’s clinical practice guideline, led by Lynnette Nieman, points to exactly this discriminating category — not the generic complaints — as the trigger for testing.¹⁷

Why “Cortisol Face” Isn’t a Sign

The trend’s premise is that facial puffiness is a readable sign of elevated cortisol. It isn’t. Facial fullness has a long differential — sodium, alcohol, sleep position, allergies, cycle phase, weight change, dental and sinus issues — and in genuine Cushing’s it arrives bundled with the discriminating features above, not on its own.¹⁷

A puffy face plus a stressful job is not a diagnosis. It’s a Tuesday.

How Doctors Actually Test

The real workup is physician-ordered, and it’s good. Brandon Galm and colleagues at Massachusetts General Hospital meta-analysed 139 studies covering 14,140 participants to compare the first-line tests:

Test	Sensitivity (95% CI)	Specificity (95% CI)
Overnight 1-mg dexamethasone suppression	98.6% (96.9–99.4)	90.6% (86.4–93.6)
Late-night salivary cortisol	95.8% (93–97.2)	93.4% (90.7–95.4)
24-hour urinary free cortisol	94.0% (91.6–95.7)	93.0% (89.0–95.5)
Midnight serum cortisol	96.1% (93.5–97.6)	93.2% (88.1–96.3)

All four are highly sensitive and specific; the dexamethasone suppression test was the most sensitive, urinary free cortisol the least.¹⁸

Notice what every one has in common. They’re either dynamic (give a drug, see if the axis suppresses) or integrated across a defined window (24-hour urine, late-night sample at a controlled time). None is “measure cortisol once and compare to a range.” The profession abandoned that approach because it doesn’t work.

What Doesn’t Work: Adaptogens, “Cortisol Detoxes,” and At-Home Kits

Of the three, only ashwagandha has any randomized evidence at all — “cortisol detoxes,” “adrenal fatigue” protocols, and at-home saliva kits have none that answers the question they claim to answer. Ashwagandha deserves reporting honestly before it’s caveated.

Ashwagandha: Real Evidence, Narrow Scope

A 2026 systematic review and meta-analysis by Fornalik and colleagues pooled 23 randomized placebo-controlled trials covering 1,706 patients. Oral ashwagandha significantly reduced circulating cortisol (SMD = −1.18, p < 0.04) and increased serotonin. Testosterone rose in men but not women. Subgroup analysis suggested a dose-response relationship keyed to withanolide content.¹⁹

That’s a real result. Here’s the frame it belongs in.

The trials are small, short, and heterogeneous — the authors flag heterogeneity, limited data on several endpoints, and the need for standardized trials. The samples skew stressed or unwell, matching the pattern in the exercise and meditation literature.

And most fundamentally: reducing a hormone value is not the same as improving an outcome. A supplement that lowers cortisol in a stressed cohort has demonstrated that it lowers cortisol in a stressed cohort. Whether that translates into better sleep, body composition, or health in someone with a normal curve is a question the trials didn’t ask.

”Cortisol Detoxes” and “Adrenal Fatigue”

“Cortisol detox” protocols are a different animal. They have no diagnostic construct — no defined state they claim to correct — and no measurable endpoint. Neither does “adrenal fatigue,” which isn’t a recognized diagnosis and has no validated test. You can’t evaluate a protocol whose success condition is unstated.

At-Home Saliva and Urine Kits

At-home saliva and urine kits sit in between: real assays applied to a question they can’t answer. The kit returns a number sampled from a pulsatile waveform with no verification of when you woke, no control for the previous night’s sleep, and no repeat days — each of which the CAR consensus guidelines call essential.⁵⁶ You’ll get a number back. It will feel like information.

For which recovery supplements clear the evidence bar and which don’t, see our review of supplements and wearable recovery metrics.

The Proxies You Can Actually Track

No consumer wearable measures cortisol. Not the Apple Watch, not Oura, not WHOOP, not Garmin. Nothing shipping to consumers reads this hormone.

What you can track are downstream signals that move when recovery and stress load move. They’re non-specific — that’s the entire caveat — but observable daily, and they respond to the same interventions.

Proxy	What to watch	What a problem looks like
Sleep duration and continuity	Total sleep time and awakenings, 7-day rolling average	A sustained drop of an hour or more, or fragmentation that persists past a bad week
Resting heart rate	Multi-day drift from your own baseline	A steady climb of several beats over 5–7 days with no obvious explanation
HRV trend	7-day average versus your 30-day normal	A sustained decline that doesn’t rebound after a rest day
Training-load tolerance	Whether prescribed sessions feel proportionate to their difficulty	Familiar work suddenly feeling hard, or performance falling at unchanged load

Every one of these has confounders. A rising resting heart rate could be a cold, a late meal, three drinks, a hot bedroom, or a flight. Hence the rule: trends over 7–14 days, never a single reading — the same discipline that makes the validation data on wrist HRV worth understanding before you act on it.

SensAI’s role here should be stated precisely, because the temptation to overclaim in this category is enormous. SensAI does not measure, estimate, or predict cortisol. No app does, and any that claims to is telling you a story.

What it does is read the proxies in aggregate — sleep, resting heart rate, HRV, and your actual completed training — and adjust programming only when several drift together, because any one of them alone has too many confounders to act on. Data flows through Apple HealthKit: Apple Watch directly, and Garmin, Oura, and WHOOP through HealthKit as well. For the deeper version of how those signals combine to flag genuine overreaching, we covered it in distinguishing overtraining from overreaching with wearable data.

Training Load: The One Cortisol Story Where the Dose Genuinely Matters

There is one context where the cortisol conversation has real training consequences, and it isn’t stress management. It’s chronic overreach.

Sports science distinguishes three states. Functional overreaching is short-term performance decrement that resolves into improvement after recovery — the intended outcome of a hard block. Non-functional overreaching is what happens when load and recovery fall out of balance and performance stays suppressed longer. Overtraining syndrome is prolonged maladaptation.

The honest distinction between the last two is duration of impairment, not a hormone panel.

Romain Meeusen, PhD, Professor at Vrije Universiteit Brussel, led the joint European College of Sport Science and American College of Sports Medicine consensus statement on this. Their biomarker conclusion is unambiguous: several markers are in use — hormones, performance tests, psychological questionnaires, biochemical and immune measures — but none meet all the criteria for general acceptance. Overtraining syndrome remains a diagnosis of exclusion, reached by ruling out infection, caloric restriction, inadequate carbohydrate or protein intake, and iron or magnesium deficiency. The consensus also notes that athletes in non-functional overreaching and overtraining syndrome often show the same clinical and hormonal signs.²⁰

So no, a cortisol test won’t tell you whether you’re overtrained. The lead author of the field’s consensus statement says so directly.

What you can manage is the input. Count your genuinely hard sessions per week and hold that count steady instead of letting it creep. Deload on a schedule rather than waiting to feel broken — we’ve written a data-driven approach to timing deload weeks using sleep and HRV trends instead of guesswork. SensAI regenerates your weekly program from what you actually completed and how you actually recovered, which is how the hard-session count stops creeping without you auditing it yourself.

A Four-Week Plan That Targets the Curve, Not the Number

Every intervention below targets the shape of the day — a real morning peak, a real evening descent — rather than pushing a number down.

Week 1 — Baseline. Fix your wake time to within 30 minutes, seven days a week. Log sleep duration. Record morning resting heart rate and HRV for seven consecutive days. Change nothing else. You’re establishing what your normal actually is, which is the only thing that makes weeks 2–4 interpretable.

Week 2 — Light and sleep. Get 10+ minutes of outdoor light within an hour of waking. Protect a 7.5-hour sleep opportunity — time in bed, not time asleep. Know what you’re doing: morning light raises the post-awakening peak,¹² and that’s the goal. Adequate sleep is what lets the evening end of the curve come down.⁷

Week 3 — Aerobic dose. Build toward roughly 150 minutes of moderate aerobic activity across the week, matching the ~530 MET-min/week optimum.² Cap genuinely high-intensity sessions at two per week — the inverted-U means more is not better.

Week 4 — Mind-body block. Add two 30-minute yoga sessions (or qigong, or an equivalent breath-led practice), keeping weeks 2 and 3 in place. This is the modality with the largest measured effect in the trial evidence.²

Day-28 checkpoint	Target
Wake-time consistency	Within 30 minutes, 6 of 7 days
Average sleep duration	Up versus your Week 1 baseline
Morning light exposure	5+ days per week
Resting heart rate, 7-day average	Flat or below Week 1
HRV, 7-day average	Flat or above Week 1
Hard sessions per week	≤ 2, consistently
Moderate aerobic minutes	~150 per week

Look at that table and notice what’s missing.

Not one checkpoint is a cortisol measurement. That’s deliberate, and it’s the whole argument. You cannot measure your cortisol curve at home in a way any researcher would accept,⁵⁶ and the number you’d get wouldn’t tell you what to do differently. The behaviors that shape the curve — sleep timing, light exposure, aerobic dose, intensity discipline — are all observable, all modifiable, and all worth doing regardless of what the hormone is doing.

Frequently Asked Questions

Does exercise lower cortisol? Over weeks, moderate exercise modestly lowers cortisol in people who are psychologically distressed — yoga showed the largest effect across 44 randomized trials (SMD −0.59).² Acutely, exercise raises cortisol in proportion to intensity.³ The two findings aren’t in conflict: chronic adaptation and acute response go in opposite directions.

What are the signs of high cortisol? The signs that discriminate are wide purple striae, easy bruising, proximal muscle weakness, and facial plethora — features hard to explain any other way. Fatigue, abdominal weight gain, and facial puffiness are not discriminating; they have dozens of causes. The Endocrine Society guideline treats the discriminating features, not the generic ones, as the trigger for testing.¹⁷

How do you reduce cortisol naturally? Three things have randomized evidence, all of it in already-distressed populations: adequate sleep (one night of partial sleep loss raised the next evening’s cortisol 37%⁷), roughly 150 minutes a week of moderate activity (~530 MET-min/week, the inverted-U optimum²), and mind-body practice — yoga produced the largest reduction across 44 trials, SMD = −0.59.² Morning bright light raises cortisol,¹² and that is the goal: a steeper curve, not a flatter one. No trial has shown a benefit from lowering cortisol in someone whose curve is already normal.

Does HIIT raise cortisol, and is that bad? HIIT raises cortisol sharply and briefly, and there’s no evidence that’s harmful in a well-recovered person.⁴ In the network meta-analysis of distressed populations, HIIT trended toward raising cortisol rather than lowering it — so it’s a poor choice if stress reduction is your specific goal, and a fine one if fitness is.²

How long does cortisol stay elevated after exercise? Roughly an hour. Meta-analytic data show cortisol elevated immediately, at 30 minutes, and at 60 minutes after a HIIT session, then falling below baseline at 120 and 180 minutes, and returning to baseline by 24 hours.⁴

What is a normal cortisol level? There isn’t a single one, which is why the question misleads. Cortisol varies by time of day, assay, sample medium, and individual, and single values don’t reliably reflect glucocorticoid status.⁸ Diagnostic testing uses dynamic or time-integrated protocols instead — dexamethasone suppression, 24-hour urine, or a controlled late-night sample.¹⁸

Can I test cortisol at home? Not in a way that answers a useful question. Valid awakening-response measurement requires objectively verified waking and sampling times across multiple days, a standard most published research still fails to meet.⁵⁶ A consumer kit returns a number without any of those controls.

Does lowering cortisol reduce belly fat? Not demonstrably, in ordinary people. Hair cortisol correlates with waist circumference at r = 0.082 in a sample of 2,527 adults — real, and well under one percent of the variance.¹⁵ Pathological cortisol excess does drive visceral fat,¹⁶ but that is a diagnosed disease, not the everyday situation the marketing implies.

References

Lightman SL, Birnie MT, Conway-Campbell BL. “Dynamics of ACTH and Cortisol Secretion and Implications for Disease.” Endocrine Reviews, 2020;41(3):bnaa002. https://pubmed.ncbi.nlm.nih.gov/32060528/ ↩ ↩² ↩³
Li X, Huang J, Zhu F. “The Optimal Exercise Modality and Dose for Cortisol Reduction in Psychological Distress: A Systematic Review and Network Meta-Analysis.” Sports, 2025;13(12):415. https://pubmed.ncbi.nlm.nih.gov/41441399/ ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰
Hill EE, Zack E, Battaglini C, Viru M, Viru A, Hackney AC. “Exercise and circulating cortisol levels: the intensity threshold effect.” Journal of Endocrinological Investigation, 2008;31(7):587-591. https://pubmed.ncbi.nlm.nih.gov/18787373/ ↩ ↩² ↩³
Dote-Montero M, Carneiro-Barrera A, Martinez-Vizcaino V, Ruiz JR, Amaro-Gahete FJ. “Acute effect of HIIT on testosterone and cortisol levels in healthy individuals: A systematic review and meta-analysis.” Scandinavian Journal of Medicine & Science in Sports, 2021;31(9):1722-1744. https://pubmed.ncbi.nlm.nih.gov/34022085/ ↩ ↩² ↩³ ↩⁴ ↩⁵
Stalder T, Kirschbaum C, Kudielka BM, Adam EK, Pruessner JC, Wüst S, Dockray S, Smyth N, Evans P, Hellhammer DH, Miller R, Wetherell MA, Lupien SJ, Clow A. “Assessment of the cortisol awakening response: Expert consensus guidelines.” Psychoneuroendocrinology, 2016;63:414-432. https://pubmed.ncbi.nlm.nih.gov/26563991/ ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶
Stalder T, Lupien SJ, Kudielka BM, Adam EK, Pruessner JC, Wüst S, Dockray S, Smyth N, Evans P, Kirschbaum C, Miller R, Wetherell MA, Finke JB, Klucken T, Clow A. “Evaluation and update of the expert consensus guidelines for the assessment of the cortisol awakening response (CAR).” Psychoneuroendocrinology, 2022;146:105946. https://pubmed.ncbi.nlm.nih.gov/36252387/ ↩ ↩² ↩³ ↩⁴ ↩⁵
Leproult R, Copinschi G, Buxton O, Van Cauter E. “Sleep loss results in an elevation of cortisol levels the next evening.” Sleep, 1997;20(10):865-870. https://pubmed.ncbi.nlm.nih.gov/9415946/ ↩ ↩² ↩³ ↩⁴
Clarke SA, Eng PC, Comninos AN, Lazarus K, Choudhury S, Tsang C, Meeran K, Tan TM, Dhillo WS, Abbara A. “Current Challenges and Future Directions in the Assessment of Glucocorticoid Status.” Endocrine Reviews, 2024;45(6):795-817. https://pubmed.ncbi.nlm.nih.gov/38795365/ ↩ ↩² ↩³
Adam EK, Quinn ME, Tavernier R, McQuillan MT, Dahlke KA, Gilbert KE. “Diurnal cortisol slopes and mental and physical health outcomes: A systematic review and meta-analysis.” Psychoneuroendocrinology, 2017;83:25-41. https://pubmed.ncbi.nlm.nih.gov/28578301/ ↩ ↩²
Cadegiani FA, Kater CE. “Enhancement of hypothalamic-pituitary activity in male athletes: evidence of a novel hormonal mechanism of physical conditioning.” BMC Endocrine Disorders, 2019;19(1):117. https://pubmed.ncbi.nlm.nih.gov/31675953/ ↩
Koncz A, Demetrovics Z, Takacs ZK. “Meditation interventions efficiently reduce cortisol levels of at-risk samples: a meta-analysis.” Health Psychology Review, 2021;15(1):56-84. https://pubmed.ncbi.nlm.nih.gov/32635830/ ↩
Petrowski K, Bührer S, Albus C, Schmalbach B. “Increase in cortisol concentration due to standardized bright and blue light exposure on saliva cortisol in the morning following sleep laboratory.” Stress, 2021;24(3):331-337. https://pubmed.ncbi.nlm.nih.gov/32723201/ ↩ ↩² ↩³
Epel ES, McEwen B, Seeman T, Matthews K, Castellazzo G, Brownell KD, Bell J, Ickovics JR. “Stress and body shape: stress-induced cortisol secretion is consistently greater among women with central fat.” Psychosomatic Medicine, 2000;62(5):623-632. https://pubmed.ncbi.nlm.nih.gov/11020091/ ↩
Stalder T, Kirschbaum C. “Analysis of cortisol in hair — state of the art and future directions.” Brain, Behavior, and Immunity, 2012;26(7):1019-1029. https://pubmed.ncbi.nlm.nih.gov/22366690/ ↩
Jackson SE, Kirschbaum C, Steptoe A. “Hair cortisol and adiposity in a population-based sample of 2,527 men and women aged 54 to 87 years.” Obesity, 2017;25(3):539-544. https://pubmed.ncbi.nlm.nih.gov/28229550/ ↩ ↩²
Bavaresco A, Mazzeo P, Lazzara M, Barbot M. “Adipose tissue in cortisol excess: What Cushing’s syndrome can teach us?” Biochemical Pharmacology, 2024;223:116137. https://pubmed.ncbi.nlm.nih.gov/38494065/ ↩ ↩²
Nieman LK, Biller BM, Findling JW, Newell-Price J, Savage MO, Stewart PM, Montori VM. “The diagnosis of Cushing’s syndrome: an Endocrine Society Clinical Practice Guideline.” The Journal of Clinical Endocrinology and Metabolism, 2008;93(5):1526-1540. https://pubmed.ncbi.nlm.nih.gov/18334580/ ↩ ↩² ↩³
Galm BP, Qiao N, Klibanski A, Biller BMK, Tritos NA. “Accuracy of Laboratory Tests for the Diagnosis of Cushing Syndrome.” The Journal of Clinical Endocrinology and Metabolism, 2020;105(6):dgaa105. https://pubmed.ncbi.nlm.nih.gov/32133504/ ↩ ↩²
Fornalik M, Malkiewicz A, Adamczak D, Nawrocki F, Zielinska A, Mondal SA. “Hormonal Modulation with Withania somnifera: Systematic Review and Meta-Analysis of Randomized-controlled Trials.” Planta Medica, 2026;92(8):790-805. https://pubmed.ncbi.nlm.nih.gov/41740946/ ↩
Meeusen R, Duclos M, Foster C, Fry A, Gleeson M, Nieman D, Raglin J, Rietjens G, Steinacker J, Urhausen A. “Prevention, diagnosis, and treatment of the overtraining syndrome: joint consensus statement of the European College of Sport Science and the American College of Sports Medicine.” Medicine and Science in Sports and Exercise, 2013;45(1):186-205. https://pubmed.ncbi.nlm.nih.gov/23247672/ ↩