The research picture for craniosacral therapy is genuinely mixed. Not a clean verdict in either direction, but a developing body of evidence with some positive findings, some negative ones, and a lot of methodological work still to do. Reading it honestly is more useful than dismissing it or cherry-picking the positive studies.
What follows is where the evidence currently sits, what the main trials have found, and why careful researchers can read the same literature and reach different conclusions.
The broad reviews and their limits
The most thorough recent evaluations are not encouraging. Two major 2024 systematic reviews, Ceballos-Laita et al. in Healthcare and Amendolara et al. in Frontiers in Medicine, both concluded that CST produced no significant effects across the conditions studied. The Amendolara review covered 24 RCTs and 1,613 participants. These are the largest reviews in the field so far, and their conclusions deserve serious weight.
But systematic reviews aggregate across studies, and the quality of what they aggregate determines what you get. Most included trials in both reviews were rated as having high or unclear risk of bias. Practitioner training varied widely, sham conditions were hard to standardise, and most trials used short intervention periods (often four to eight sessions) that experienced practitioners would consider inadequate for complex or chronic conditions. The 2023 headache systematic review was more specific. It found statistically significant changes in some outcomes but rated the evidence as 'very low certainty' and questioned the clinical significance. Statistically significant and clinically meaningful aren't the same thing.
Where individual trials look more promising
At the individual trial level, certain conditions have shown more encouraging results. A 2015 neck pain RCT (PMID 26340656) with 54 patients found significant and clinically relevant effects on pain intensity compared with sham CST. A 2010 fibromyalgia RCT with 92 patients showed significant pain reduction at 13 of 18 tender points assessed. A 2019 infant colic RCT with 132 participants found significant reductions in crying hours (3.2 fewer hours of crying per day by day 24) and improved sleep, compared with usual care.
For migraine, a 2022 RCT and a 2023 crossover study both showed significant reductions in pain intensity, headache frequency, disability scores, and medication use. These aren't small effects in clinically relevant populations, and they don't disappear because the broader systematic reviews were less favourable. A specific trial for a specific condition carries different information than a pooled analysis across all conditions.
The 2019 chronic pain meta-analysis
One systematic review stands out as notably more positive: the 2019 meta-analysis by Jäkel and von Hauenschild. It focused on chronic pain and found significant and robust effects of CST on pain and function lasting up to six months. This review is cited often by practitioners and supporters of CST, and fairly so. It represents the most careful positive analysis of the pain evidence available. That later reviews have been less favourable reflects partly a wider literature base and partly different methodological choices about which studies to include.
The contrast between the 2019 results and the 2024 reviews shows a real challenge in evidence synthesis for manual therapy: conclusions are genuinely sensitive to which trials are included, how quality is rated, and how outcomes are pooled. There isn't yet a clean, final answer. Claiming there is, in either direction, would overstate what the current evidence supports.
What this means in practice
For most conditions, CST shouldn't be the first or only intervention considered. For conditions with established conventional treatments (acute infection, fractures, serious medical illness) CST isn't a substitute. What the evidence does support: CST is safe, a meaningful number of people experience genuine benefit, and for conditions involving nervous system dysregulation, chronic stress, or where conventional approaches have plateaued, it's a reasonable option to explore.
Practitioners who are honest about the evidence make better partners in care. The most trustworthy ones don't dismiss the sceptical reviews or pretend the positive trials prove everything. The research is evolving, the methodology is improving, and the picture will sharpen over the next decade of better-designed trials. In the meantime, individual experience, approached with clear expectations and inside a broader care plan, is still a meaningful data point.
Mixed evidence, genuine safety, meaningful individual benefits for many people, and an improving research trajectory. For clients considering CST, that's enough to make it worth trying with a well-trained practitioner, without expecting it to replace other forms of care.