How to Check Cortisol Levels (And How to Read the Results)

You are exhausted but wired at 11 p.m. You wake at 3 a.m. with your heart going. You crash mid-afternoon, crave salt and sugar, and carry stubborn weight around your middle no matter how clean you eat. Somewhere along the way someone told you it might be "your cortisol" — so now you want an actual number.

Here is the problem: a single cortisol reading is one of the most misunderstood lab values in all of medicine. Cortisol is not a thermostat you check once. It is a rhythm — a wave that is supposed to rise sharply in the morning and fall to almost nothing by bedtime. Measure it at the wrong moment, with the wrong test, and you can look perfectly "normal" while feeling anything but.

This guide walks you through every legitimate way to check cortisol levels — blood, saliva, urine, and hair — what each one actually measures, when to do it, and most importantly how to read the results in the context of how you feel. The goal is not just a number. It is the pattern.

Why checking cortisol is different for women — and why timing beats everything

Cortisol is your primary stress hormone, released by your adrenal glands under the direction of a feedback loop called the HPA (hypothalamic-pituitary-adrenal) axis. Its single most important feature is that it follows a strong daily curve. In a healthy rhythm, cortisol surges 50–75% in the first 30–45 minutes after you wake — the cortisol awakening response — then tapers across the day to a low point around midnight (Psychoneuroendocrinology 2022).

That rhythm is exactly why timing matters more than the raw value. A morning cortisol of 18 µg/dL can be completely healthy at 8 a.m. and alarmingly high at 10 p.m. If a test ignores the clock, the result is close to meaningless.

For women specifically, two things complicate the picture. First, daily and chronic psychological stressors flatten and shift the diurnal cortisol curve differently than they do in men, so a "normal range" built largely on mixed populations can hide a meaningful pattern in you (Pharmacology Biochemistry and Behavior 2026). Second, estrogen raises cortisol-binding globulin — the protein that ferries cortisol through the blood — so a blood test can read high in total cortisol while your free, biologically active cortisol is normal. That is why the question is never simply "what is my cortisol" but "what is my cortisol, measured how, at what time, in what fraction."

If you want the diet and lifestyle side of this, our companion guide to foods that lower cortisol pairs naturally with testing — but test first so you know what you are actually working with.

1. Morning serum (blood) cortisol — the screening standard

The most common way to check cortisol is a morning blood draw, typically between 7 and 9 a.m. when your level should be near its daily peak. This measures total cortisol — both the protein-bound, inactive portion and the small free fraction.

Mechanistically, a morning serum cortisol is best at answering one specific question: are your adrenals producing enough cortisol at all? A clearly low morning value (commonly below ~3 µg/dL) raises concern for adrenal insufficiency, while a robust morning value largely rules it out (Diagnostics 2023). What it is not good at is diagnosing "high cortisol," because a single elevated morning draw can simply reflect the white-coat stress of the needle itself.

2. Late-night salivary cortisol — the rhythm test that catches what morning blood misses

If morning blood answers "is there enough," late-night salivary cortisol answers "is the rhythm switching off when it should." Saliva is collected at home, usually around 11 p.m. to midnight, the moment cortisol should be at rock bottom.

The elegant part: saliva contains only free, unbound cortisol — the fraction that actually acts on your tissues — so it sidesteps the binding-globulin problem that distorts blood tests in women on estrogen or birth control. An elevated late-night salivary cortisol is one of the most reliable screening tools for cortisol excess (Cushing's syndrome), because it captures a curve that fails to descend (Singapore Medical Journal 2019). It is also the test most likely to validate what you already suspect when you feel "tired but wired" at night.

3. The cortisol awakening response (CAR) — a window into HPA-axis health

The sharpest, most information-dense moment in your whole day is the half hour after you open your eyes. Collecting saliva immediately on waking, then again at 30 and sometimes 45–60 minutes, captures the cortisol awakening response — the size of that morning surge.

A blunted or absent CAR is associated with burnout, chronic stress, and HPA-axis dysregulation, while an exaggerated one can track with anxiety and anticipatory stress. Because the CAR is exquisitely sensitive to when you sample and to that first hour of behavior, expert consensus guidelines insist on strict timing, no snoozing, and no eating or brushing teeth before the first sample (Psychoneuroendocrinology 2022). Done sloppily it is noise; done correctly it is one of the most useful functional markers you can run.

4. The four-point salivary cortisol curve — mapping the whole wave

Rather than a single timepoint, a four-point (sometimes six-point) salivary curve samples cortisol on waking, around noon, late afternoon, and at bedtime. The result is an actual graph of your day.

This is the test that distinguishes patterns clinical ranges often blur: a high-then-crash curve, a flat-line low curve, or an inverted curve that is too low in the morning and too high at night. Because chronic psychological stress reshapes the shape of the curve — not just its average — a multi-point profile reveals dysregulation that any single draw would miss (Pharmacology Biochemistry and Behavior 2026).

5. 24-hour urinary free cortisol — your daily total output

A 24-hour urine collection measures how much free cortisol your body excretes across a full day. Because it integrates every peak and trough into one number, it cancels out the minute-to-minute swings that make single blood draws so jumpy.

Mechanistically, urinary free cortisol reflects the portion of cortisol that overflowed past binding proteins and was filtered by the kidney — a reasonable proxy for total daily exposure. It is a long-standing screening test for cortisol excess and one of the recognized first-line options for evaluating suspected Cushing's syndrome (Journal of Clinical Endocrinology and Metabolism 2008). Its weakness is the opposite of saliva's: it tells you the daily total but loses the rhythm entirely, and an incomplete collection skews the result.

6. Hair cortisol — the long-exposure photograph

If the other tests are snapshots, hair cortisol is a long exposure. As hair grows roughly one centimeter per month, it traps cortisol, so a 3 cm segment closest to the scalp reflects your average exposure over about three months.

This makes hair uniquely suited to chronic stress, where the question is not "what is my cortisol right now" but "what has my baseline been all season." A large meta-analysis confirmed hair cortisol reliably tracks chronic stress states and certain endocrine conditions, while also flagging the practical caveats — hair treatments, washing frequency, and very dark or gray hair can shift values (Psychoneuroendocrinology 2017). It is increasingly used in research and functional settings but is not yet a standard diagnostic test on its own.

7. The dexamethasone suppression test — the confirmatory follow-up

This one is not a baseline measurement; it is a provocation. You take 1 mg of dexamethasone (a synthetic steroid) at 11 p.m., then check blood cortisol at 8 a.m. In a healthy HPA axis, the dexamethasone tricks the brain into thinking there is plenty of cortisol, so morning cortisol should be strongly suppressed.

Failure to suppress suggests the adrenal feedback loop is stuck "on" — a hallmark of true cortisol excess. It is a confirmatory test ordered by a clinician, typically after a saliva or urine screen comes back abnormal, and is one of the recognized first-line diagnostic tools for Cushing's syndrome (Journal of Clinical Endocrinology and Metabolism 2008). You will not do this one at home, but it helps to understand why your clinician orders it.

8. Provoked testing (ACTH stimulation & stress challenges) — for the low end

When the suspicion is too little cortisol rather than too much, a clinician may run an ACTH stimulation test: they inject synthetic ACTH and measure whether your adrenals can rise to the occasion. A flat response points to adrenal insufficiency (Diagnostics 2023).

In research settings, standardized stressors such as the Trier Social Stress Test reliably provoke a measurable cortisol spike, which is how scientists study reactivity. You do not need a lab challenge to understand the principle: a healthy system should both turn cortisol up under stress and turn it back down afterward.

How to actually test cortisol (most people do it wrong)

Here is where most people — and frankly many quick lab orders — go sideways. They run one mid-day blood draw, get a number inside the reference range, and conclude their cortisol is "fine," never knowing whether the rhythm itself is broken.

The root-cause approach is different. We are not chasing a single value; we are reconstructing the shape of your day:

• Match the test to the question. Worried about too much cortisol or a rhythm that won't switch off at night? Late-night saliva or a four-point salivary curve. Worried about exhaustion and too little? Morning serum, possibly an ACTH stimulation test. Worried about months of grinding stress? Hair cortisol.
• Honor the clock. A cortisol result without an exact collection time is uninterpretable. Always record the minute you sampled.
• Measure free, not just total — especially for women. Because estrogen and birth control raise binding globulin, blood total cortisol can mislead. Saliva and urine measure the free, active fraction and sidestep that distortion.
• Capture the curve, not the dot. A single point cannot show a high-morning/crashing-afternoon pattern or an inverted night-high curve. Multiple points across the day reveal the dysregulation you actually feel.
• Read results against your symptoms. A "normal" number that doesn't fit a 3 a.m. wake-up and afternoon crash deserves a better test, not a dismissal.

That last point is the brand wedge of root-cause medicine: the lab range tells you whether you resemble a statistical population; your symptom pattern tells you whether your system is regulating well. Both matter, and the most useful answers live where they intersect.

Evidence-based first steps

Before and around testing, a few low-risk moves make your results cleaner and your rhythm healthier:

• Test before you intervene. Run your saliva curve or hair test before overhauling supplements, so you have a real baseline to compare against.
• Protect the morning surge. Get bright light within 30–60 minutes of waking; a robust morning cortisol peak is a feature of a healthy HPA axis, not a problem (Psychoneuroendocrinology 2022).
• Wind cortisol down at night. Dim lights, limit late screens and late intense exercise, and keep alcohol modest — each helps the curve descend to its proper midnight low.
• Address the chronic-stress load. Because hair cortisol tracks months of accumulated stress, durable changes — sleep, boundaries, breath-based downregulation — matter more than any single hack (Psychoneuroendocrinology 2017).
• Eat to steady the curve. Anchor blood sugar with protein and fiber and lean on the patterns in our foods that lower cortisol guide; blood-sugar swings are a direct cortisol trigger.
• Standardize your collection. No snoozing, eating, or brushing before a waking sample; complete a full 24-hour urine collection; record exact times. Sloppy collection is the number-one reason cortisol tests get misread.

The Bottom Line

Checking cortisol is less about getting one number and more about reading a rhythm. Morning blood tells you whether there is enough; late-night saliva and a four-point curve tell you whether the wave rises and falls when it should; 24-hour urine gives the daily total; hair gives the seasonal average; and provocation tests like the dexamethasone suppression or ACTH stimulation confirm what the screens suggest. Match the test to your question, honor the clock, measure the free fraction, and always read the result against how you actually feel.

The hardest part is rarely the test — it is the interpretation. A flattened curve, a high-then-crash pattern, or a night-high rhythm can each produce overlapping symptoms, and they call for different responses. This is exactly the kind of pattern a naturopathic or functional-medicine practitioner can help you interpret together, connecting your timing, your free-cortisol fractions, and your lived symptoms into a single coherent picture rather than a lone data point. If you'd like help mapping which tests fit your situation, our care coordinator can point you toward a sensible starting blueprint.

This article is for educational purposes and is not medical advice. Cortisol disorders can be serious: seek prompt in-person care for symptoms such as severe or sudden weakness, unexplained rapid weight changes, fainting, very high blood pressure, new purple stretch marks with easy bruising, or signs of an adrenal crisis (severe vomiting, abdominal pain, confusion, or collapse). Always interpret lab results with a qualified clinician.