Hair vs saliva cortisol test: the moment or the months?

A saliva test measures the free, active cortisol in your body at the moment you collect the sample. A hair test measures the cortisol built up in your hair over roughly the past three months.¹ Saliva is the right tool for seeing how cortisol rises and falls across a day, including the surge after waking, and late-night saliva is one of the standard tests doctors use to look for Cushing's syndrome.² For your long-term level, hair is the most accurate measure from a single sample, because one day of saliva mostly tells you about that day.

What a saliva test measures

Only a small share of the cortisol in blood, about 4 to 6 per cent, circulates free; the rest is bound to carrier proteins.³ Cortisol in saliva is directly proportional to that free fraction, and it passes from blood into saliva in much less than five minutes, so a saliva sample is a close reading of the active hormone at that moment.⁴ It is not affected by changes in the main carrier protein, corticosteroid-binding globulin (CBG).³ That is a real advantage over blood. In a study of women taking oral oestrogen, total blood cortisol was 67 per cent higher than in women taking none, but saliva cortisol was similar in all the groups.⁵

Home saliva kits have you collect saliva into small tubes at set times. A typical UK kit asks for four samples across one day, after waking, at noon, at 4pm and before sleep, with no food or drink for 60 minutes and no tooth-brushing for 30 minutes before each.⁶ The result is a curve showing your cortisol at those four moments. Our guide to the saliva cortisol test covers collection in detail.

One day of saliva is mostly about that day

Saliva's strength, catching the moment, is also its weakness when the question is long term. In a classic study of saliva taken at 1pm on two days six weeks apart, about 75 per cent of the variation in results reflected the situation on the day, about 21 per cent the person's stable level, and about 4 per cent measurement error.⁷ Across three studies of healthy adults, about half the variation in the awakening response, the daily slope and total daily output was day-to-day fluctuation. Over spans longer than a year, intraclass correlations were 0.13 or lower, meaning single measurements barely agreed with each other.⁸

The answer is more days of sampling. Researchers estimate that you need saliva from at least three separate days for a reliable average, four to eight days for total daily output and ten days for the slope from morning to night. They recommend fewer samples per day over more days.⁹ A one-day kit gives a snapshot of one day's curve, which can be useful, but it is not the same as your usual pattern.

Hair does that averaging inside the sample. In repeated measurements months apart, a stable personal component explained 59 to 82 per cent of the variation in hair cortisol.¹⁰ In a study that measured both in the same people, hair correlated at r = 0.84 from one month to the next, against 0.38 to 0.61 week to week for saliva collected three times a day.¹¹

The cortisol awakening response

The cortisol awakening response (CAR) is the sharp rise in cortisol over the first 30 to 45 minutes after you wake.¹² In three studies of 152 people, from children to older adults, free cortisol rose by 50 to 75 per cent within 30 minutes of waking.¹³ A review put the rise in healthy adults at 50 to 160 per cent, about 9 nmol/L on average, and noted that there are no agreed norms for absolute levels.¹⁴ Our guide to cortisol's daily rhythm explains why it happens.

Only saliva can measure this, and it is demanding. Because the rise depends on sampling from the exact moment of waking, an international expert panel has published guidelines on controlling sampling accuracy.¹² The response on any single day is largely shaped by that day, and two to six days of sampling are needed for a reliable measure.¹⁵ Hair cannot show it at all: in two studies, hair cortisol was not associated with the awakening response.¹¹¹⁶ If your question is about the shape of your day, saliva is the tool.

Late-night saliva and Cushing's syndrome

Cortisol should be at its lowest around midnight. In healthy people with regular sleep, salivary cortisol at bedtime or between 11pm and midnight is below 145 ng/dL (4 nmol/L), and one NHS laboratory uses below 3.2 nmol/L for late-night samples.²¹⁷ In Cushing's syndrome that night-time low is lost. That is why late-night saliva is a first-line test, with published sensitivity of 92 to 100 per cent and specificity of 93 to 100 per cent.²

It still needs repeating. The Endocrine Society recommends at least two late-night samples; the 2021 consensus on Cushing's disease suggests two or three and advises against the test in night-shift workers.²¹⁸ In a study that followed people with Cushing's disease for up to 42 nights, extreme fluctuations were common, and every patient with recurrent or persistent disease had at least two normal results.¹⁹ Results can also be pushed up by steroid creams or gels getting into the tube, liquorice, chewing tobacco, smoking on the day, or stress just before collection.²

Hair has been studied for Cushing's too. In one study, a single hair sample identified the condition with accuracy similar to the standard first-line tests,²⁰ but hair is not part of the diagnostic guidelines.¹⁸ A late-night saliva test for Cushing's should be arranged and interpreted by a doctor. See testing for Cushing's syndrome.

How hair compares with a month of saliva

The fair comparison for hair is not one day of saliva but many. In a study of 17 healthy adults who collected saliva three times a day for 30 days, the centimetre of hair nearest the scalp correlated with their 30-day saliva total at r = 0.61, and the authors concluded that hair reliably estimates long-term free cortisol production.¹¹ In 54 female students, hair cortisol was consistently linked with daytime saliva output measured two, four and six weeks before the hair was cut.¹⁶ With fewer saliva samples, agreement is weaker: hair correlated with three weekly samples at r = 0.383, but those saliva samples correlated with each other only at r = 0.514 from one week to the next.²¹

So hair reflects the total amount of cortisol over weeks and months, the area under the daily curve, rather than its shape. These correlations are moderate, and there is no perfect yardstick for cortisol over months. But against the strongest comparisons available, hair holds up well, and it is far simpler: one sample, cut at any time of day, with no effect from the time it is taken.²² See is hair cortisol testing accurate? for the full evidence.

Which test answers your question?

• Has my cortisol been high or low over recent months? Hair. One sample covers roughly three months and is consistent from test to test.
• What does my daily rhythm look like? Saliva, collected at set times, ideally on more than one day.
• Do I have a strong awakening response? Saliva, timed precisely from waking, on two to six days.¹⁵
• Could I have Cushing's syndrome? See a doctor. Late-night saliva is one of the tests they may arrange.
• I take the pill or oral HRT. Saliva and hair both avoid the large carrier-protein effect that distorts blood tests, and hair studies of the pill have found at most a small and inconsistent effect.²³

Saliva tests from major providers are listed below, with prices as of October 2026. For the other comparisons, see hair vs blood and hair vs urine.