Efficacy and safety of acupuncture therapy for neuropsychiatric symptoms among patients with Parkinson's disease: A systematic review and meta-analysis.

Only Chinese- and English-language papers were included, which the authors say can bias the picture. Just 28 trials, and several questions rested on very few of them: six trials for anxiety, three for fatigue, three for quality of life, and one for impulse control disorders. Twenty-five of the 28 trials were done in China, so the results may not carry over to patients elsewhere. Twenty-six of 28 trials did not say how they concealed who got which treatment, and only five reported blinding anyone. Acupuncture is hard to blind, which the authors acknowledge. The trials disagreed strongly with each other on depression (I2 = 86%) and anxiety (I2 = 91%), so the pooled averages hide a lot of variation. Most trials did not follow people up, so nothing is known about whether any benefit lasts. Different trials used different scales for the same symptom, so results are pooled as standardised differences rather than directly comparable point scores. Acupuncture technique varied widely: manual or electroacupuncture, frequencies from 2 to 100 Hz, 10 to 60 minutes a session, 4 to 12 weeks.

The authors declare no commercial or financial relationships that could be a conflict of interest. The work was funded by the National Natural Science Foundation of China, the Natural Science Foundation of Guangdong Province, and the Sanming Project Medicine in Shenzhen Nanshan, with grant numbers given.

Do not conclude from this that acupuncture works for these symptoms on its own. When acupuncture was compared with sham acupuncture or no treatment, with nothing added, it made no difference to depression or anxiety; the pooled benefits in this review came from trials where acupuncture sat on top of medication or brain stimulation. The positive headline is about an add-on therapy, not a stand-alone one.