Cortisol in athletes: training load, overtraining and how to measure it

Exercise raises cortisol, and the harder the session the bigger the rise. Over weeks and months, heavy endurance training shows up as higher long-term cortisol, while a regular moderate amount of exercise appears to bring it down.¹²³ For an athlete, that makes the timing of a cortisol test awkward: a blood or saliva sample mostly reflects what you did that morning. A hair test sidesteps the problem. It is the most accurate way to measure long-term cortisol from a single sample, it has no timing rules, and a sweaty session does not change it.⁴ What no cortisol test can do, hair included, is diagnose overtraining.⁵

What a training session does to cortisol

How much cortisol a session releases depends on its intensity. In a study of 30-minute sessions, cortisol fell by 6.6 per cent at rest and rose by 5.7 per cent at 40 per cent of VO2max (your maximum aerobic capacity), 39.9 per cent at 60 per cent and 83.1 per cent at 80 per cent.¹ Once the researchers allowed for changes in blood volume and time of day, low-intensity exercise actually lowered cortisol.¹

All of this sits on top of the normal daily rhythm, in which blood cortisol peaks at about 8.30 in the morning and falls to its lowest point around midnight.⁶ A blood test taken after an early hard session, or a saliva sample after an evening race, reflects the training as much as your underlying level. For exercise in general, see cortisol and exercise.

Heavy training and long-term cortisol

Repeated hard sessions add up. In a study of 304 amateur endurance athletes, including long-distance runners, triathletes and cyclists, and 70 non-athletes, the athletes had higher hair cortisol in all three 3 cm segments tested. Cortisol in the segment nearest the scalp correlated with every measure of training volume the researchers recorded.²

A study of female university footballers, measured in August and again in December before a national tournament, found a similar pattern: self-reported training load correlated with hair cortisol, even though it was unrelated to fatigue measured by questionnaire.⁷ The authors suggested hair cortisol could help guard against overtraining, though their study did not test that.

Higher long-term cortisol during heavy training is not in itself a sign of harm. It is what you would expect from repeated physical stress, and whether it matters for athletes' health is still being studied.²

Regular moderate exercise brings it down

The picture is different for ordinary amounts of exercise. In a 12-month trial, 130 healthy adults aged 26 to 58 were randomly assigned either to 150 minutes a week of moderate-to-vigorous aerobic exercise or to a health-information control group. Hair cortisol fell in the exercise group relative to the controls, although other stress and cardiovascular markers did not change consistently.³

So the dose matters. Regular moderate activity is linked with lower long-term cortisol, and sustained heavy volume with higher. Our guide to how to lower cortisol covers the other approaches that have been tested.

Can a cortisol test detect overtraining?

Overtraining syndrome is a prolonged fall in performance, with fatigue and mood changes, caused by too much training load and too little recovery. A joint consensus statement from the European College of Sport Science and the American College of Sports Medicine says it is very hard to tell apart from non-functional overreaching, a shorter-lived version. The diagnosis rests on how things develop and on ruling out other causes, such as illness or infection, too few calories, too little carbohydrate or protein, and iron deficiency.⁵ Hormones, performance tests, psychological questionnaires and blood markers are all used, but none meets the criteria to be generally accepted.⁵

Cortisol in particular has given mixed results. A systematic review of 38 studies found that resting hormone levels were mostly normal in athletes with overtraining or overreaching, and that cortisol responses to exercise tests were conflicting. The authors concluded that resting hormone levels are not good predictors.⁸ So a single blood or saliva cortisol result cannot tell you whether you are overtrained, and neither can a hair result. If your performance has dropped for weeks despite rest, with persistent fatigue or low mood, see a doctor, ideally one with sports medicine experience, to look for other causes.

What hair can add is a measure of how much cortisol you have been exposed to over a training block, set against your own earlier results. Alongside training logs, performance and how you feel, a rising trend is one more piece of information, not a diagnosis.

Why hair suits monitoring a training block

Blood and saliva tests capture a moment, and an athlete's moments vary more than most. Serum cortisol swings across the day, and having blood taken can itself trigger a cortisol rise in about 30 to 47 per cent of people.⁶⁹ To get a reliable average from saliva, researchers estimate you need samples on at least three separate days.¹⁰

Hair avoids most of this. In two experiments, a treadmill session that made 42 people sweat and a sauna session with 52 people did not change hair cortisol, which was also unrelated to the time of day the sample was taken.⁴ It is also 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.¹¹ In repeated measurements over months, a stable personal component explained 59 to 82 per cent of the variation.¹² That consistency is what makes a change between two hair tests easier to interpret than a change between two saliva tests.

Timing a hair test around your training

A standard test uses the 3 cm of hair nearest the scalp, which covers roughly three months at the average growth rate of 1 cm a month, although real growth ranges from 0.6 to 1.4 cm a month.¹³ There is also a delay. The centimetre nearest the scalp reflects hair formed roughly 1.3 to 2.2 months before the cut, because new hair takes about two weeks to reach the surface and scissors leave a short stub.¹⁴

In practice, a sample cut on the last day of a training block misses its final few weeks. Cutting it six to eight weeks after the block ends lines the window up better; that is a rough calculation from the growth figures above, so treat it as approximate. A sample taken in a quieter part of the season gives a baseline to compare with. Longer hair can be cut into month-by-month segments to show how cortisol moved through a season. See segmental hair cortisol analysis and how often to test cortisol.

What can skew an athlete's result

• Sunlight. In an experiment, 40 hours of natural sunlight lowered hair cortisol by a median of 54 per cent, so long hours training outdoors may lower a result.¹⁵
• Season. In a review of 29 studies, 76 per cent reported seasonal differences in hair cortisol.¹⁶ Compare like with like, such as the same point in successive seasons.
• Frequent washing. Washing more often has been linked with lower hair cortisol, an effect clearest in hair further from the scalp.¹⁷¹⁸
• Steroid medicines. In the large Lifelines cohort, people who used steroid inhalers, nasal sprays or creams had lower levels of hair cortisone, cortisol's close relative.¹⁹ Tell the laboratory about any inhalers, creams or steroid injections.
• Laboratory method. Healthy ranges differ more than tenfold between methods, from 0.84 to 8.07 pg/mg by LC-MS/MS to 17.7 to 153.2 pg/mg by an early immunoassay, so compare results only from the same laboratory.²⁰²¹

A hair result measures exposure; it is not a verdict on your training. Read it against your own earlier results and alongside how you are sleeping and recovering. See hair cortisol results explained and cortisol and sleep.