Walk into almost any spa and you'll hear some version of the same story: massage flushes toxins, releases lactic acid, and lowers your cortisol. It sounds authoritative. It's also, for the most part, wrong.
Here's the more interesting truth. Massage therapy does have measurable effects on pain and psychological wellbeing — some of them substantial. But the mechanisms the industry cites are mostly outdated physiology, and the strongest evidence points somewhere unexpected: not at your muscles, but at your nervous system and your mood.
We went through the meta-analyses, systematic reviews, and Cochrane assessments to work out what the research actually supports. Here's what we found.
What the biggest analyses found
The most cited piece of research in this field is a 2004 meta-analysis published in Psychological Bulletin, which pooled results from 37 randomised studies covering more than 1,800 participants. Rather than looking at a single benefit, it examined nine separate outcomes — some measured immediately after one session, others after a course of treatment.
The results split cleanly into two groups.
After a single massage, participants showed reliable drops in state anxiety (the temporary, situational kind), blood pressure, and heart rate. That's a genuine physiological shift, and it happens fast.
After multiple sessions, the effects got more interesting. The largest benefits in the entire analysis were reductions in trait anxiety — a person's general disposition toward anxiousness — and in depression. The authors noted that a full course of massage produced benefits comparable in magnitude to a course of psychotherapy.
That is a striking finding, and it's the one most spa marketing overlooks in favour of talking about muscle knots.
But the same analysis found something the industry mentions even less often: single sessions did not produce significant changes in negative mood, immediately assessed pain, or cortisol levels. Pain relief only showed up in the delayed measurements, after repeated treatment.
In other words: massage looks less like a painkiller and more like a course of treatment.
The cortisol claim doesn't hold up
"Massage lowers your cortisol" is probably the single most repeated scientific claim in the wellness industry. It's also the one with the shakiest support.
A comprehensive quantitative review published in the Journal of Bodywork and Movement Therapies set out specifically to test it, applying conventional meta-analytic methods across the available randomised trials. Its conclusion was that the cortisol reduction effect is far smaller and less consistent than commonly asserted — and that it almost certainly isn't the mechanism responsible for massage's benefits in anxiety, depression, or pain.
The 2004 meta-analysis reached the same place: across the seven studies that measured cortisol, no significant effect emerged.
This doesn't mean massage doesn't reduce stress. The anxiety, heart rate, and blood pressure findings are real. It means the explanation is wrong. Something else is doing the work, and hormone levels in a saliva sample aren't capturing it.