Cortisol is released in a circadian rhythm. It climbs sharply in the period around waking, reaches its peak within the first hour of the day, then declines across the afternoon and evening to a low point near midnight.¹ That curve is the signal. The amount of cortisol present at any single moment is only one point on it.

This is why the question “what is my cortisol level?” is incomplete. A value of 400 nmol/L means one thing at 7am and something entirely different at 9pm. Read without a time attached, a cortisol number cannot tell you whether the rhythm is intact.

The shape of the curve carries clinical weight. A systematic review and meta-analysis of 80 studies found that a flatter cortisol slope across the day — less of a fall from morning to night — was consistently associated with poorer mental and physical health outcomes, with the strongest links in immune and inflammatory measures.²

Layered on top of the daily curve is a separate, sharper event: the cortisol awakening response, or CAR. In healthy adults, free cortisol rises by roughly 50% or more in the first 30 minutes after waking.¹

The CAR is not simply the start of the morning peak. It is regulated independently of the rest of the diurnal cycle and is treated as a distinct feature of HPA-axis function.³ A blunted or absent awakening response is one of the patterns that prompts closer attention to the stress axis.

Because the CAR unfolds in a narrow window, measuring it accurately depends on tight timing — samples taken at the moment of waking and at set intervals afterward. International consensus guidelines set out exactly how the collection should be done for the result to be valid.⁴ A single sample, taken at no fixed point relative to waking, cannot capture it.

Each collection method answers a different question, and they are not equal in what they resolve.

Serum (blood). A blood draw measures total cortisol — both the protein-bound fraction and the small free fraction — at the moment the sample is taken. It is precise for that time-point and is used in diagnostic protocols such as ACTH-stimulation testing.

Dried urine. A dried-urine collection captures five timed samples across the day — one more than a four-point salivary curve — so it resolves the daily rhythm, including the post-waking rise, at higher resolution. It also measures cortisol metabolites: markers of how the body clears cortisol once it has done its work. The five-point curve plus the metabolite picture make it the most complete of the four methods, and it is the method used within the program.

Saliva. Salivary cortisol reflects the unbound, biologically active fraction of the hormone, and is an established, reliable proxy for free cortisol in blood.⁵ Saliva is collected without a needle and can be sampled repeatedly through the day — the basis of four-point curve testing and CAR measurement. It resolves the curve well; it does not capture cortisol clearance.

Hair. As hair grows, cortisol is incorporated into the shaft. A hair segment provides a retrospective estimate of average cortisol exposure over the preceding months,⁶ and is used as a marker of longer-term, chronic stress load rather than day-to-day variation.⁷

Multi-point cortisol testing has become practical to do at home, which matters — the curve and the CAR can only be measured where the day actually happens: on waking, in your own environment, on an ordinary schedule.

An at-home kit for dried-urine or salivary testing provides the collection materials and a fixed schedule: a sample on waking, then at set intervals through the day and one before bed — five collections for a dried-urine curve, four for a salivary curve. The samples are stable at room temperature and posted to the laboratory.

The single thing that determines whether an at-home result is trustworthy is adherence to the timing. The validity of curve and CAR measurement rests on samples being taken at the scheduled moments, beginning at the instant of waking.⁴ Following the schedule precisely is the part of the test that cannot be outsourced.

A cortisol result is read as a shape, not a score. Several patterns recur.

A flattened curve — cortisol that neither rises much in the morning nor falls much at night. Chronic, uncontrollable stress has been shown to produce exactly this high, flat diurnal profile,⁸ and flatter slopes track with poorer health outcomes.²

A blunted awakening response — little or no rise in the 30 minutes after waking, often reported alongside fatigue that a full night of sleep does not resolve.

Elevated night-time cortisol — cortisol still high when it should be at its daily low, a common correlate of difficulty falling asleep.

Distinct from these functional patterns, cortisol testing also has a defined diagnostic role. Where genuine cortisol excess is suspected, late-night salivary cortisol has greater than 90% sensitivity and specificity for endogenous Cushing’s syndrome,⁹ and clinical practice guidelines list late-night salivary cortisol and 24-hour urinary free cortisol among the recommended initial tests.¹⁰ A result that falls well outside the expected range is a reason to involve a medical practitioner.