Acupuncture for children with autism spectrum disorders: a systematic review of randomized clinical trials.

Soo and her team hunted for every randomized trial that poked needles into kids with autism. They found 11 studies done in China, Hong Kong, and Egypt between 2000 and 2011.

The kids were 2–12 years old. Most trials compared true acupuncture against sham needles or no extra care. The reviewers graded study quality and looked at autism scores, language, and behavior.

Six trials claimed acupuncture helped; five found no difference. When the team tried to pool the numbers, the studies were too different in age, dose, and outcome tools.

Biggest red flags: tiny samples (most under the kids), no blinding of parents, and different needle sites every time. Bottom line: the evidence is mixed and too shaky to trust.

Hudson et al. (2012) looked at 33 drug trials the same year. Like Soo, they saw lots of weak studies and only a few meds with solid proof. Both reviews warn that ‘more RCTs’ must be bigger and tighter.

Wang et al. (2025) later found clear, medium-size gains for computerized cognitive training. Their 15 newer studies used similar kids but better designs. The contrast shows the field has moved past tiny, sloppy trials.

Pickard et al. (2019) also reported mixed news: group sports helped social skills a bit, yet communication did not budge. Together these papers paint the same picture—small gains are real only when methods are strong.

Parents still ask about acupuncture. You can now say, ‘Eleven small studies give a shrug, not a green light.’ Use the paper to shift talk toward options with firmer footing—like CBT for anxiety (A et al., 2014) or computerized training for thinking skills. If families insist on trying needles, set single-case data rules: pick one target, track daily, and review after four weeks so you can pivot fast if nothing changes.