One of the foundational claims of craniosacral therapy is that practitioners can feel a distinct rhythm in the body — slower than heartbeat and breath — that reflects the movement of cerebrospinal fluid and the flexion and extension of cranial structures. If that's true, trained practitioners assessing the same patient independently should arrive at similar findings. That's the question of interrater reliability.
A 1994 study tested it directly. Three physical therapists with CST training each independently examined 12 patients and recorded the craniosacral rate — the number of pulses per minute they perceived. They didn't communicate, allowing for a real test of whether their perceptions agreed.
The result was striking. It's become one of the most frequently cited pieces of evidence in the scientific critique of CST: the interrater reliability was essentially zero.
What the study found
The statistical measure was the intraclass correlation coefficient (ICC), which runs from -1 to 1. A score of 1 would mean perfect agreement — the three examiners found exactly the same rate in every patient. A score of 0 would mean agreement at the level of random chance. The score the study found was -0.02, which means the examiners weren't agreeing even to the level you'd expect by coincidence.
The examiners also showed no meaningful agreement on the timing or character of the rhythm beyond the rate measurement. So it wasn't only that they disagreed on how many pulses per minute they felt. They also disagreed on when pulses occurred and on the qualitative character of what they were perceiving.
Three examiners with CST training, assessing the same 12 patients, essentially perceived different things. That's a significant finding for anyone trying to understand CST theoretically. If the rhythm exists as an objective feature of physiology, you'd expect trained practitioners to perceive it consistently. This study found they don't.
Why this challenges CST theory
The interrater reliability failure matters for two reasons. First, it raises the question of whether the craniosacral rhythm, as typically described in CST theory, is a real and consistently detectable phenomenon. If three trained practitioners can't agree on what they're feeling, that suggests what each is perceiving may be at least partly internally generated — shaped by expectation, attention and the practitioner's own body rhythms — rather than something objectively present in the patient.
Second, it has implications for the diagnostic logic of CST. If practitioners can't reliably detect the same rhythm in the same patient, treatment decisions based on that rhythm rest on an unreliable foundation. That doesn't mean the treatment can't help. Plenty of helpful interventions are built on theoretical frameworks that don't hold up under scrutiny. But it does complicate claims about precise assessment and targeted intervention.
CST practitioners often respond in several ways. By noting that no study has found zero reliability across all available research. By suggesting the 1994 study had methodological limitations. Or by shifting the theoretical emphasis from the rhythm itself to the quality of contact and the nervous system response to touch.
Mechanism uncertainty versus session value
The interrater reliability question is about the theoretical basis of CST — specifically, whether the craniosacral rhythm is a reliably detectable phenomenon. It doesn't directly answer the question of whether CST sessions help people feel better.
Many therapeutic approaches help without their practitioners fully understanding why. The quality of attention, the therapeutic relationship, the effect of sustained gentle touch on the nervous system, the time and space to be still — all real, all able to contribute to someone feeling better, regardless of whether the specific theoretical claims about cranial bone movement and cerebrospinal fluid hold up.
The honest position: the reliability problem is a real challenge to the theoretical framework of CST, not an argument the field has answered. It calls for some humility from practitioners about what they're actually detecting and why their interventions help when they do. For clients, it's worth knowing — not as a reason to dismiss CST, but as a reason to hold the explanatory framework loosely while staying open to your own experience of sessions.
The 1994 interrater reliability study remains one of the most cited challenges to CST's theoretical foundation. Its findings don't determine whether sessions are valuable, but they do ask serious questions about what practitioners are actually perceiving — questions the field has not yet fully resolved.