When people ask whether craniosacral therapy works, the honest answer depends on which condition you mean and what kind of evidence you'd accept. The CST research base isn't one block. Some conditions have been studied more carefully than others. Some trials were well-powered and sham-controlled. The results vary.
What follows is what the published trial data actually shows for the conditions where CST has been studied most. The aim is accuracy, not advocacy. Plenty of people report real benefit, and some of that experience now shows up in the trial record, even where the picture isn't tidy.
Migraine and headache
Migraine is one of the better-studied areas. A 2022 RCT found positive effects on migraine frequency and severity. A 2023 crossover study added more support. These results are encouraging, partly because migraine research is hard to blind in the first place. It's tough to hide from a patient whether they got hands-on treatment.
Headache-related disability, measured with tools like the HIT-6 (Headache Impact Test), also improved in some trials. If you have chronic migraine and conventional approaches haven't done enough, these results suggest CST may be worth trying, while the evidence base remains small.
Infant colic
A 2019 RCT on infant colic enrolled 132 participants, which is sizeable for this literature. It found significant reductions in crying hours and colic severity in the CST group compared with controls. It was reasonably well conducted and is one of the larger infant studies in the field.
Colic is a condition where small improvements matter a lot to families, and parents often want options beyond waiting it out. In terms of sample size and clarity of result, this trial is probably the strongest in the CST literature.
Neck pain
A 2015 sham-controlled RCT with 54 participants found statistically significant and clinically relevant effects of CST on neck pain intensity and disability over eight weeks, compared with sham treatment. The sham control matters here. The comparison wasn't against no treatment, but against something designed to feel like treatment without the elements thought to be active.
Results held up on both pain intensity and disability. For a small trial, that's a fairly clean finding. Whether it would replicate at scale is the obvious next question, but the 2015 trial is one of the stronger individual studies in the CST literature.
Fibromyalgia
Fibromyalgia has been studied in at least two relevant RCTs. A 2010 trial with 92 patients found significant reductions in tender point count and pain measures. A 2011 RCT with 84 patients found significant improvements in anxiety, depression, and quality of life at six months. Six-month follow-up is longer than many trials in this area, which adds to its weight.
Fibromyalgia patients often cycle through treatments with partial results, so these findings are worth attention. Neither trial settles the question, but they point in the same direction.
Low back pain and pelvic girdle pain
Low back pain is more mixed. A 2016 RCT had a primary endpoint of pain reduction that didn't quite hit statistical significance, with a p-value of 0.060. Close, but outside the conventional 0.05 threshold. Some secondary outcomes looked better, but the primary endpoint is what the trial was designed to test, and it wasn't met.
Pregnancy-related pelvic girdle pain was studied in a 2013 RCT. The researchers themselves described the treatment effects as small and clinically questionable. There were statistical trends, but the practical significance for patients was unclear.
Taken together, the picture is genuinely varied. Some conditions have promising data. Others don't. That variation is itself useful. It suggests that any benefit from CST, where it exists, is condition-specific, not a general placebo or attention effect.
The research is incomplete, and many of these trials will be superseded by larger, better-powered work. But the existing data shows a reasonably honest picture: stronger evidence for neck pain, fibromyalgia, and infant colic; weaker for low back pain and pelvic girdle pain. Reading it carefully, condition by condition, gives the most accurate view.