Can you test cortisol in dyed or treated hair?

Yes, in most cases. Dyed, styled and grey hair can all be tested, but some of what happens to hair after it grows can lower the cortisol measured in it. In the Whitehall II study of more than 3,500 older adults, people who dyed their hair had 12.5 per cent lower hair cortisol, and heat treatments, frequent washing and sunlight have been linked with lower values too.¹²³⁴ None of this stops a test being useful. It means telling the laboratory what you do to your hair, sampling the newest growth, and keeping things the same if you test again. This guide takes each factor in turn and ends with practical advice.

Does hair dye affect a cortisol test?

It tends to lower the result, though how much varies between studies:

• In an early study, hair that had been coloured before sampling had lower cortisol, but dyeing hair samples after they had been collected made no difference.⁵
• In a multi-ethnic US sample, dyed hair was linked with lower hair cortisol.⁶
• In Whitehall II, people who dyed their hair had 12.5 per cent lower values, with a 95 per cent confidence interval of 1.9 to 22.0 per cent lower.¹
• A 2026 LC-MS/MS reference-range study found no link with dyeing.⁷

Treatments can also hide real differences between people. In a 2026 study of women's hair cortisol, leaving out women with treated hair revealed differences that were otherwise masked.⁸ That is why the 2017 meta-analysis of hair cortisol studies lists hair treatment, alongside age, sex, washing frequency and contraceptive use, as something researchers should account for.⁹

In practice, a difference of around 12.5 per cent is small next to the differences between people and between laboratory methods.¹ Where it matters most is when you compare your own results over time: if you started or stopped colouring your hair between two tests, part of any change could come from that rather than from your cortisol.

Heat treatments, bleach and perms

Heat-based hair treatments were linked with lower hair cortisol in a South London community study.² Laboratory work fits with that: hair segments soaked in hot water lost cortisol over time.¹⁰

Bleaching and perms (permanent waving) have not been studied separately in people, so their effect on hair cortisol is not known. The sensible approach is to treat them like dye and heat treatments: mention them to the laboratory, and if you can, take your sample before the next treatment rather than just after it.

Washing and hair products

Washing matters mainly for hair further from the scalp. In a Dutch study, people who washed their hair more often had lower hair cortisol and cortisone even after allowing for other factors, and the Lifelines cohort found the same pattern.³¹¹ But another study found that washing frequency made no difference to the segment nearest the scalp, and lowered cortisol only in the third segment out.¹² In children, recent washing, but not how often hair was washed, was linked with lower values, and a systematic review concluded that washing frequency and hair products matter little when the segments nearest the scalp are used.¹³¹⁴

Laboratory work on animal hair suggests why: in pooled monkey hair, cortisol fell as the number of washes rose, mainly because of the water.¹⁵ Each wash takes a little more, so older hair has lost more, which is one reason laboratories test the 3 cm nearest the scalp. There is no need to change your routine before a test. Washing as usual keeps one test comparable with the next.

Sunlight and the time of year

Ultraviolet light lowers cortisol in hair. In an experiment with hair from nine volunteers, 40 hours of natural sunlight cut hair cortisol by a median of 54 per cent and artificial UV light by 75 per cent; cortisone, a related hormone, fell by 32 and 52 per cent.⁴ Hair that has spent a summer in the sun will have lost more than hair that has not, another reason to use the newest growth.

Season matters in a less predictable way. A 2025 systematic review of 29 studies with 10,520 people found that 76 per cent reported seasonal differences, and most of the studies that compared different people found lower values in winter and spring than in summer and autumn (10 of 15).¹⁶ That is the opposite of what sunlight alone would suggest, and the reasons are not settled. If you are comparing tests, take them at a similar time of year.

Storage and posting

Cut hair can be posted without refrigeration, but its cortisol slowly declines. In a study of samples from 81 people, cortisol fell steadily over two years of storage, and agreement with samples analysed straight away weakened over time.¹⁷ The authors recommend analysis ideally within three to six months of sampling.¹⁷ Post your sample promptly rather than cutting it weeks in advance.

Grey hair, hair colour and body hair

Grey hair is fine to test. A study that compared pigmented and grey hairs from the same people found no difference in cortisol.¹⁸ Findings on natural hair colour are mixed: darker hair was linked with higher values in a Dutch adult study, while another study found no effect of natural colour.³¹² If you test more than once, comparing your results with your own earlier ones takes care of this.

Body hair is a different matter. The same study compared scalp hair with body hair from the same people, found that most steroid hormones correlated only weakly between them, and concluded that body hair is not a suitable substitute when scalp hair is unavailable. Hair and nail cortisol did not correlate either.¹⁸ If your head hair is very short or sparse, ask the laboratory before ordering.

How to get a fair sample

• Sample before you treat. If you are about to colour, bleach, perm or heat-treat your hair, take the sample first. It is treatment before sampling that has been linked with lower values; dyeing hair after it had been cut made no difference in the laboratory.⁵
• Cut close to the scalp, from the back of the head just below the crown. The newest hair has had the least washing, sun and treatment, and that area gives the most consistent results: cortisol varied by 15.6 per cent between samples from there, against 30.5 per cent across the whole head.⁵¹⁴
• Tell the laboratory about dye, heat and chemical treatments and when you last had them, and about any steroid tablets, inhalers, nasal sprays, creams or injections, which can raise or lower the result.¹¹¹
• Keep your usual routine. Wash and style as normal. If you test again, keep treatments, time of year and laboratory the same so the results are comparable.¹⁶
• Post promptly. Analysis within three to six months of cutting is recommended.¹⁷

Even with treated hair, the result is a useful measure of your long-term cortisol, provided you read it with these factors in mind. See how to take a hair sample for the cutting itself and hair cortisol results explained for reading the result.