Cortisol and stress: what happens in the short and the long term

When you face a stressor, your brain sets off a chain of hormones that raises cortisol, and the same system then switches itself off again.¹ That short-term response is normal and useful. The harder question is what months of stress do, and there is no single answer. Chronic stress can keep cortisol high in some people and make it overreact in others, depending on the person and the kind of stress.¹ What researchers measuring cortisol in hair have found is that, across studies, groups living with chronic stress had 22 per cent higher long-term cortisol, rising to 43 per cent when the stress was still going on.²

What happens to cortisol when you are stressed

The stress response starts in the brain. Nerve signals activate cells in the hypothalamus that release corticotrophin-releasing hormone (CRH).¹ The pituitary gland then releases adrenocorticotrophic hormone (ACTH), which tells the adrenal glands how much cortisol to make.³ Cortisol makes glucose available, breaks down protein and fat for fuel, and adjusts immune and inflammatory activity.⁴ Finally, it acts back on the brain, including the hippocampus and the CRH-producing cells themselves, to switch the response off. This negative feedback is what normally returns cortisol to its usual level once the stress has passed.¹ For the full picture of what cortisol does, see what is cortisol.

Which kinds of stress raise cortisol most

Not all stress is equal. A meta-analysis of 208 laboratory studies found that the biggest cortisol and ACTH responses, and the slowest recovery, came from tasks that were uncontrollable and involved social-evaluative threat: being judged by other people while you perform.⁵ Feeling out of control while others assess you seems to be the combination the HPA axis responds to most.

Physical stressors count too. Even having blood taken triggers a cortisol response in roughly 30 to 47 per cent of people.⁶ Exercise raises cortisol once it is hard enough: 30 minutes at 60 per cent of maximum aerobic capacity raised it by 39.9 per cent, and at 80 per cent by 83.1 per cent.⁷ Serious illness has a bigger effect still. In intensive care patients, blood cortisol was 3.5 times higher than in healthy controls.⁸

These are all short-term changes, and they are why a single blood or saliva sample can reflect the morning you had rather than your usual level. In one classic study, about 75 per cent of the variation in a saliva sample reflected the situation on the day.⁹

When stress becomes chronic

With long-running stress the picture gets more complicated. The HPA axis can settle into different patterns, such as producing too much cortisol all the time or responding more strongly to each new stressor. Which pattern appears depends on how long the stress lasts, how intense and frequent it is and what kind it is. It also depends on the person's genes, early-life experience, environment, sex and age.¹

Some forms of long-term stress are linked with lower, not higher, cortisol. In the main meta-analysis of hair cortisol, people with anxiety disorders, including post-traumatic stress disorder (PTSD), had 17 per cent lower levels.² In the Dresden Burnout Study, an imbalance between the effort people put in at work and the reward they got back predicted lower long-term cortisol later on.¹⁰ So "stressed" does not always mean "high cortisol". See burnout and cortisol and cortisol, anxiety and depression.

What hair cortisol shows about chronic stress

Because a single blood or saliva sample mostly reflects the day, researchers studying long-term stress have turned to hair. One sample averages roughly the last three months,¹¹ and results are consistent: the same people's hair cortisol correlated at r = 0.84 from one month to the next, against 0.38 to 0.61 week to week for saliva.¹² That makes hair the most accurate way to measure long-term cortisol from a single sample, and a natural tool for asking whether a stressful period has shown up in your body.

The main meta-analysis pooled 124 groups from 66 studies, 10,289 people in all. Groups exposed to chronic stress had 22 per cent higher hair cortisol on average, rising to 43 per cent where the stress was still going on. Where the stress was in the past, there was no significant difference.² Individual studies show the same pattern.

These are group averages from different laboratories, so the pg/mg figures cannot be compared across rows or with your own result. See hair cortisol results explained.

Why feeling stressed and cortisol do not always match

You might expect the people who feel most stressed to have the highest cortisol. Often they do not. The hair meta-analysis found no consistent link between hair cortisol and self-reported perceived stress, depressive symptoms or social support.² In the unemployment study, long-term unemployed people had higher hair cortisol, but how stressed they said they felt was not related to it.¹³ And in a randomised trial of meditation-based mental training, hair cortisol fell over the first three to six months, yet the fall was unrelated to changes in self-reported stress.¹⁶

Part of the gap is that the two measure different things. A questionnaire asks how you feel about recent weeks, while hair records what your body produced over months. And, as described above, the HPA axis can respond to chronic stress in more than one direction.¹ How you feel and what your cortisol is doing are two separate measurements, and neither stands in for the other.

That also means a hair test cannot diagnose "stress", anxiety, depression or burnout in an individual. A meta-analysis found no consistent difference in hair cortisol between people with major depression and people without,¹⁷ and a 2026 review of burnout research found no biological marker good enough for routine diagnosis.¹⁸

Social support and other buffers

Other people seem to soften the cortisol response. In a laboratory study, people who had support from their closest friend showed a smaller rise in saliva cortisol during a standard public-speaking stress test, the Trier Social Stress Test.¹⁹ Over the longer term the evidence is thinner: across hair studies, self-reported social support showed no consistent link with hair cortisol.²

The meditation trial described above is one of a small number of controlled trials that have used hair cortisol to test whether a sustained change in daily life brings long-term cortisol down. Those trials, and the evidence on what else may help, are covered in how to lower cortisol and cortisol and sleep.

When stress symptoms need a doctor

Stress and a hormone disorder can feel alike, but some signs point to something else and need a doctor rather than a home test. Easy bruising, stretch marks, a red, round face, and weakness in the muscles at the top of the arms and legs can be signs of Cushing's syndrome.²⁰ Extreme tiredness, weight loss, dizziness on standing and darker patches of skin can be signs of Addison's disease.²¹

Those conditions are diagnosed with tests arranged by a doctor: late-night saliva, 24-hour urine and dexamethasone suppression tests for Cushing's syndrome,²² and the Synacthen blood test for adrenal insufficiency.²³ A hair test shows your long-term cortisol, but it does not diagnose or rule out either condition. See high cortisol symptoms and low cortisol symptoms.