What the research actually says

A pilot in dialysis patients, whose insomnia has causes yours probably doesn't. Sixty-three people is small, and sleep was scored by questionnaire rather than measured in a lab. It does show the placement matters — which is the claim most worth testing.

The authors are blunt about the weaknesses: only two of eighteen trials described their randomisation properly, none described allocation concealment, none could blind participants, heterogeneity was extreme (I² = 98%), the funnel plot suggests publication bias, and every trial came from Asian countries. A wide pooled estimate built on weak trials is a signal, not a settled fact.

Post-surgical recovery is a specific, short-lived kind of stress. It doesn't follow that the same points touch chronic everyday anxiety, and cortisol is a crude proxy for how you feel at 4pm on a Tuesday.

IVF is one of the most stressful medical experiences there is, and expectation effects in an unblinded fertility setting are hard to rule out. Read it as evidence about acute anxiety, not about sleep.

There was no control group at all — the authors say so in the title. Sleep in the weeks after birth changes on its own, so improvement here cannot be attributed to the seeds. Included because leaving out the weak studies is how citation lists start lying.

This is auricular acupuncture — needles, placed by a clinician, ten times. Seeds are not needles and your bathroom is not a clinic, so this supports the mechanism, not our product. The comparison was also a booklet rather than a sham.

Nothing. This is a plan, not a finding — we list it because it is the rigorous trial this field still needs, and because a citation list that hides the pending studies is marketing rather than evidence.