Cortisol and exercise: why a hard session raises it and regular activity can lower it

Exercise affects cortisol in opposite ways depending on the timescale. During a session, cortisol rises once the effort is hard enough: in one study, 30 minutes at 60 per cent of maximal aerobic capacity raised it by about 40 per cent, and 30 minutes at 80 per cent by 83 per cent, while light exercise lowered it.¹ Over months the picture changes. In a year-long randomised trial, regular moderate-to-vigorous aerobic exercise lowered hair cortisol, a measure of average cortisol over months.² Very heavy training is the exception: amateur endurance athletes had higher hair cortisol the more they trained.³ The evidence supports regular moderate activity, not overtraining.

Why cortisol rises during exercise

Cortisol helps the body meet demands for energy. It raises the glucose available in the blood and promotes the breakdown of fat and protein for fuel.⁴ A hard session is a physical stress, so a rise in cortisol is part of the normal response, and exercise is one of the everyday things that can raise a cortisol test result.⁵

How much it rises depends on intensity. In a study of 30-minute sessions at different intensities, set against each person's maximal aerobic capacity (VO2max, the most oxygen the body can use, a standard measure of fitness), cortisol changed as follows.¹

Once the researchers corrected for changes in blood volume (blood becomes more concentrated as you sweat) and for the time of day, low-intensity exercise actually lowered cortisol.¹ So there is a threshold somewhere between light and moderate effort: below it, exercise barely moves cortisol or lowers it; above it, the harder you go, the bigger the rise. These short-term rises are normal and are not a reason to avoid hard exercise. The correction for time of day matters because cortisol is naturally highest in the morning and lowest around midnight,⁶ so the same session can start from a very different baseline depending on when you do it.

What regular exercise does to long-term cortisol

The question most people care about is what happens over months. The clearest evidence comes from a 12-month randomised trial in 130 adults aged 26 to 58. One group was assigned to 150 minutes a week of moderate-to-vigorous aerobic exercise, and the other to health information only. Hair cortisol fell in the exercise group compared with the control group.²

That is encouraging, but it is one trial, and other stress-related and heart-related markers measured in it did not change consistently.² It suggests that regular exercise can bring long-term cortisol down a little, not that exercise resets the stress system. The trial measured the effect in hair rather than in blood or saliva. Hair is far more consistent from test to test: a stable personal component explains 59 to 82 per cent of the variation in repeated hair samples,⁷ which makes a real change easier to see.

Gentler exercise: walking and yoga

You do not need to train hard to help your cortisol. Light exercise did not raise it in the intensity study above, and lowered it once other influences were accounted for.¹ In a meta-analysis of 42 randomised trials comparing yoga-based programmes, which combine movement, breathing and meditation, with active control activities, yoga was linked with lower evening and waking cortisol.⁸

These findings fit a simple picture: activity you can keep up week after week, much of it at an easy or moderate effort, is what the evidence on long-term cortisol supports. For the mental side of the same picture, such as meditation training, see how to lower cortisol.

Heavy training and endurance athletes

More is not always better. In a study of 304 amateur endurance athletes and 70 non-athletes, the athletes had higher hair cortisol in all three 3 cm segments tested,³ which together reflect roughly the past nine months.⁹ Cortisol in the segment nearest the scalp rose with every measure of training volume.³

The study does not show that this is harmful. It may reflect the repeated rises of hard sessions adding up over months. The safest reading is that very high training loads go with higher long-term cortisol, so rest and recovery deserve as much attention as the training itself. For more on sport and competition, see cortisol in athletes.

How much exercise is about right?

Putting the evidence together, the pattern it supports is regular, moderate activity rather than as much as possible. The year-long trial that lowered hair cortisol used 150 minutes a week of moderate-to-vigorous aerobic exercise,² which works out at 30 minutes on five days. Moderate effort means you can still talk but not sing; vigorous effort means you can only say a few words before pausing for breath. Hard sessions still have a place, and the short-term rise they cause is normal. What the athlete study suggests is that very high training volumes, kept up for months, go with higher long-term cortisol.³

Sleep matters too: in a laboratory study, a short night raised cortisol the following evening by 37 per cent.¹⁰ The Endocrine Society's advice for people with unexplained tiredness puts exercise alongside a healthy diet and a regular sleep routine.¹¹ If you feel persistently drained despite rest, ease back and talk to a GP, because tiredness has many causes and not all of them are about training. See cortisol and sleep.

Exercise and cortisol tests

Because exercise raises cortisol for a while, it can affect blood and saliva tests. MedlinePlus advises that you may need to rest before a cortisol test, since stress and exercise can raise the result.⁵ Follow the instructions for your test, and avoid a hard session beforehand unless you are told otherwise. Blood cortisol also follows a strong daily rhythm, which is covered in the daily cortisol rhythm.

Hair is different. In two experiments, a treadmill session in 42 people and a sauna in 52, heavy sweating did not change hair cortisol, and hair cortisol was unrelated to sweat cortisol or to the time of day.¹² So a workout on the day you cut a hair sample makes no difference. Hair is also far more consistent from one test to the next, correlating at r = 0.84 from one month to the next,¹³ which makes it the most accurate way to see from a single sample whether a change in training has shifted your long-term cortisol. Researchers do treat physical activity as one of the factors to take into account when they interpret hair results.⁶