Does massage lower cortisol?

Cortisol is the number the wellness industry reaches for when it wants a stress claim to sound biological. Somebody tested whether massage actually moves it, across the whole literature, and the answer is awkward.

It is also the myth with the best consolation prize, because what massage does do for stress is better documented than almost anything else it is claimed to do.

The short version

• The cortisol reduction is very small and, in most cases, not statistically distinguishable from zero.¹
• It cannot be the cause of massage's benefits. The review that measured it says so directly.¹
• Single sessions did not reduce cortisol in the earlier meta-analysis either, though they did reduce state anxiety, blood pressure and heart rate.²
• The effect on anxiety and depression is the largest in the literature, and a course of treatment produced benefits similar in size to psychotherapy.²
• Across 137 touch studies covering 12,966 people, touch reduced depression and state anxiety, with clinical groups benefiting most.³

What the review found

In 2011 Christopher Moyer and colleagues ran a comprehensive quantitative review of every study that had measured cortisol before and after massage therapy. The conclusion is unusually blunt for a research paper: the reduction is generally very small and, in most cases, not statistically distinguishable from zero, and it cannot be the cause of massage therapy's well-established beneficial effects.¹

Note the phrasing carefully. "Well-established beneficial effects" is not a hedge. The same author group had already established those effects. What this paper removes is the explanation, not the outcome.

The earlier 2004 meta-analysis of 37 studies pointed the same way. Single applications of massage reduced state anxiety, blood pressure and heart rate. They did not reduce negative mood, immediate pain or cortisol.²

Why cortisol was an attractive explanation

It is measurable, it has a number, and it can be taken from saliva. A therapist who says "this lowers your cortisol" sounds like they are describing a physiological process rather than a feeling.

The problem is that cortisol is a poor proxy for how stressed someone actually is. It follows a strong daily rhythm, peaking in the morning and falling through the day, so a sample taken after an afternoon appointment will be lower than one taken before it regardless of what happened in between. It responds to food, sleep, caffeine and the anticipation of being measured. Detecting a small treatment effect through all of that requires study designs that most of the studies in the review did not have.

What massage does do for stress

Here the evidence is genuinely strong, and the industry undersells it.

The 2004 meta-analysis found that reductions in trait anxiety and depression were massage therapy's largest effects. Trait anxiety is the disposition rather than the moment, and a course of treatment produced benefits similar in magnitude to those of psychotherapy.²

A 2010 meta-analysis specifically in depressed people found massage therapy was associated with a significant reduction in depressive symptoms.⁴

And in 2024 Nature Human Behaviour published a meta-analysis of 137 touch studies covering 12,966 individuals. Touch reduced pain, depression and state anxiety. Clinical cohorts benefited more than healthy ones, and, notably, the benefit did not differ between touch from a familiar person and touch from a health professional.³

That last finding is worth being honest about, because it cuts both ways. It says the effect is real and substantial. It also says the effect belongs to touch itself rather than to any proprietary technique.

The honest summary

Massage does not work by lowering your stress hormone. It works, for anxiety and low mood in particular, better than it works for most of the muscular complaints it is usually booked for, and better than the industry claims. The mechanism is a nervous system one, set out on the how massage works page, rather than an endocrine one.

Frequently asked questions

Does massage lower cortisol?

Barely, if at all. The comprehensive review of the literature found the reduction is generally very small and in most cases not statistically distinguishable from zero, and that it cannot be the cause of massage's beneficial effects.

Does massage help with stress and anxiety then?

Yes, and this is the strongest part of the evidence base. Reductions in trait anxiety and depression are massage therapy's largest measured effects, with a course of treatment producing benefits similar in magnitude to psychotherapy.

Why is cortisol a bad measure of stress?

It follows a strong daily rhythm and responds to food, sleep, caffeine and the anticipation of being measured. Detecting a small treatment effect through all of that noise takes a study design most of the published trials did not have.

Is one massage enough to help my mood?

Single sessions reduced state anxiety, blood pressure and heart rate but not negative mood. The larger effects on trait anxiety and depression came from a course of treatment rather than a one-off.

Does it matter who gives the massage?

For the mental health benefit, less than you would expect. A meta-analysis of 137 touch studies found the benefit did not differ between touch from a familiar person and touch from a health professional.