Cortisol is widely offered as a way to measure stress, in saliva, blood and hair samples. The hormone is genuinely central to the stress response, but a single number is close to uninterpretable.
The hormone follows a steep daily curve
Cortisol is not steady across the day. It rises sharply in the first half hour after waking, then declines across the day to a low point during the night.
The size of that morning rise and the slope of the decline are the features researchers pay attention to, not the height at any single moment.
A reading taken without knowing when the person woke is therefore missing the information needed to interpret it.
It responds to far more than stress
Cortisol rises with exercise, with eating, with waking, with illness and with anticipating something demanding. All of those are ordinary functioning rather than pathology.
Its role is broader than the popular framing suggests: it participates in regulating blood glucose, inflammation, blood pressure and the sleep-wake rhythm.
Calling it the stress hormone captures one of its jobs and obscures the rest, which is part of why a raised reading is so easily over-read.
Chronic strain does not simply raise it
Intuition suggests that long-term pressure means persistently high cortisol. The research picture is less tidy, with prolonged strain associated in different studies with elevated, blunted or flattened patterns.
A flattened curve — less difference between morning and night — appears in some contexts to be more informative than the overall level.
Because the direction of change is not consistent, a person's result cannot be mapped onto how stressed they are.
What the different sample types actually capture
Saliva captures the fraction available to tissues at that moment, which makes timing critical and makes repeated sampling across a day the only sensible use of it.
Blood captures both bound and free hormone, and the act of taking blood can itself raise the reading in someone who dislikes needles.
Hair captures accumulated exposure over weeks, which avoids the timing problem but is affected by hair treatment, growth rate and washing.
Where testing genuinely belongs
There are clinical conditions of clear cortisol excess or deficiency, and these are diagnosed with structured protocols involving timed samples and stimulation or suppression steps.
Those protocols exist precisely because single readings are unreliable. The complexity is not bureaucratic caution; it is what the hormone's behaviour requires.
For someone who feels persistently depleted, symptoms, sleep and circumstances give a clinician more to work with than a one-off number, and unexplained symptoms warrant that conversation directly.