Cortisol, anxiety and depression: what the research shows

There is no simple rule that depression or anxiety means high cortisol. When researchers reviewed studies of hair cortisol, which reflects average cortisol over months, most found no difference between people with major depression and people without.¹ Anxiety disorders, including post-traumatic stress disorder (PTSD), were linked on average with 17 per cent lower hair cortisol, not higher.² Some groups, such as people with both depression and anxiety, do show higher levels,³ and results depend on the condition, how long it has lasted and what the person is still going through. So no cortisol test, including a hair test, can diagnose or rule out a mental health condition. If you are struggling, the place to start is a GP or, in England, NHS talking therapies, which you can usually contact yourself.⁴

Why cortisol and mood are connected

Cortisol is controlled by the hypothalamic-pituitary-adrenal (HPA) axis. Nerve signals prompt the hypothalamus in the brain to release corticotrophin-releasing hormone, the pituitary gland responds with adrenocorticotrophic hormone (ACTH), and ACTH tells the adrenal glands how much cortisol to make.⁵⁶ Cortisol then feeds back on the brain, including limbic areas involved in emotion such as the hippocampus, to switch the response off again.⁵

Long-term stress does not change this system in one fixed way. Depending on how long, how intense, how frequent and what type the stress is, and on a person's genes, early life, environment, sex and age, it can lead to constant over-production of cortisol or to stress responses that become sensitised.⁵ That is one reason studies of mood and cortisol give mixed results.

Hormone disorders can also affect mood directly. Mood changes are a recognised symptom of Cushing's syndrome,⁷ and anxiety and depression are listed among the symptoms of Addison's disease.⁸ In the other direction, MedlinePlus notes that depression and anxiety are among the conditions linked with cortisol that is high some of the time.⁶

Depression: no consistent difference

A systematic review of 16 studies found that most showed no difference in hair cortisol between people with major depression and healthy controls. Pooling seven of them gave a standardised mean difference of -0.02 (95% confidence interval -0.36 to 0.32), effectively zero, with considerable disagreement between studies.¹ A broader meta-analysis of hair cortisol likewise found no consistent link with mood disorders or with how depressed people said they felt.²

Individual studies do find differences in particular groups:

• In 23 people taking medication for depression, hair cortisol was higher than in 64 controls in both the most recent 3 cm of hair (26.7 against 18.7 pg/mg) and the 3 cm before that (21.9 against 13.4 pg/mg).⁹
• In the Netherlands Study of Depression and Anxiety, with 1,166 participants, depressive or anxiety disorders on their own were not linked with hair cortisol, but having both was linked with slightly higher levels.³
• In the Whitehall II study of older adults, depressive symptoms were associated with 20 per cent higher hair cortisol.¹⁰

So raised long-term cortisol is found in some people with depression, particularly in some clinical groups, but many people with depression have levels like anyone else's, and a raised level on its own says little about mood.

Anxiety: often lower, not higher

Anxiety disorders tend to go the other way. In a small study of 15 people with generalised anxiety disorder and 15 controls, hair cortisol was 50 to 60 per cent lower in the anxiety group in both of the 3 cm segments tested, covering roughly six months, even though their daily saliva profiles showed no difference.¹¹ Pooled across studies, anxiety disorders including PTSD were linked with 17 per cent lower hair cortisol.² In the large Dutch study, anxiety disorders on their own showed no difference.³

Lower long-term cortisol can seem surprising, but it fits the wider finding that long-term stress can alter the stress system in more than one direction.⁵ A similar pattern has been reported with some kinds of work stress: in the Dresden Burnout Study, an imbalance between effort and reward at work predicted lower long-term cortisol later on.¹² See burnout and cortisol and cortisol and stress.

PTSD and trauma: timing matters

PTSD shows both patterns. In a study of 28 people with PTSD, 27 people who had been through trauma without developing PTSD and 32 who had not been traumatised, the PTSD and trauma groups had 59 and 51 per cent lower hair cortisol respectively. Saliva samples did not show the difference.¹³

But among severely traumatised people still living in a conflict area of Uganda, the 10 with PTSD had higher hair cortisol than the 17 without.¹⁴ A systematic review of 31 studies largely confirmed that hair cortisol is raised during recent or ongoing traumatic stress, while the evidence for lower levels long after trauma was less clear.¹⁵ Whether the danger has passed, and how long ago the trauma was, seem to matter.

Why a cortisol test cannot diagnose a mental health condition

Put together, the research shows averages that differ between some groups, in different directions, with a lot of overlap between individuals. Hair cortisol does not track self-reported stress consistently,² and in pregnancy the overall link between psychological distress and hair cortisol was small and not significant.¹⁶ No cortisol result can tell you whether you have depression, anxiety or PTSD, or that you do not. Those are diagnosed by talking through your symptoms and history with a GP or mental health professional.

Where hair cortisol is useful is in research, and as a measure of long-term cortisol in its own right. Because it averages months and is far more consistent from test to test than saliva,¹⁷ it is the measure researchers use to see long-term patterns: in PTSD, for example, it revealed a difference that saliva missed.¹³ Studies have also used it to see whether psychological treatments change cortisol, with mixed results so far.¹⁸ A hair test measures your long-term cortisol; it does not diagnose a mental health condition.

When to check for a physical cause

Low mood or anxiety that comes with physical signs is worth raising with a GP, who can check for hormonal causes. Cushing's syndrome can cause mood changes alongside more fat on the tummy and upper back, a red, round face, easy bruising and stretch marks.⁷ Addison's disease can cause anxiety and depression alongside extreme tiredness, weight loss, dizziness on standing, salt craving and darker patches of skin.⁸ Both are rare, and they are diagnosed with blood, urine and saliva tests arranged by a doctor, not with a hair test. See high cortisol symptoms and low cortisol symptoms.

Where to get help

If you are struggling with low mood, worry or the after-effects of trauma, help is available whatever your cortisol level.

• Your GP can talk through your symptoms, advise on treatments and help you reach mental health services.¹⁹
• NHS talking therapies offer psychological treatments for problems including persistent low mood, panic attacks, phobias, OCD and PTSD. In England you can usually refer yourself without seeing a GP if you are 18 or over, or 16 or 17 in some areas, and you do not need a diagnosis.⁴
• Urgent help. Call NHS 111 and choose the mental health option, or use 111 online. If someone's life is at risk, or you feel you cannot keep yourself or someone else safe, call 999 or go to A&E.¹⁹
• Someone to talk to. Samaritans can be reached on 116 123, and people under 19 can call Childline on 0800 1111.¹⁹

Outside the UK, your doctor or local health service can point you to similar help. Looking after sleep, activity and alcohol can support recovery, and some of these habits have evidence for lowering cortisol (see how to lower cortisol), but they are not a replacement for treatment.