Barbara Geller, MD reviewing Cortese S et al. J Am Acad Child Adolesc Psychiatry 2016 Apr 1.

A meta-analysis shows no significant effects in studies using blinded raters or in those with sham or active control treatments.

Evidence for efficacy of electroencephalographic neurofeedback training for attention-deficit/hyperactivity disorder (ADHD) is controversial. To address this issue, investigators with pharmaceutical-company funding conducted a meta-analysis of neurofeedback studies that used either frequency band training, which targets ratios of theta to beta frequencies, or slow cortical potential training, which uses cued-wave information.

Of 247 publications, 13 met inclusion criteria of randomized, controlled trials in children aged 3 to 18 years (N=520; 20–40 neurofeedback sessions). Only studies using likely nonblinded raters to determine outcomes showed significant effects with neurofeedback. No significant effects on ADHD were found in studies using probably blinded raters or in studies using sham or active control treatments. No studies reported baseline inclusion criteria for theta/beta ratios or training-related changes in participants’ brains. Results were similar for studies using only frequency band training; there were too few studies of slow cortical potential training for separate analyses.


Findings from randomized, controlled studies with blinded raters are negative and do not support the use of electroencephalography-based neurofeedback training for ADHD. Families seeking nonpharmacological interventions — because of family preference or the children’s intolerance of ADHD medications — need to be informed of these results. Of note, in 2013, the FDA approved a device that uses the frequency band method to diagnose ADHD, which may have encouraged families to seek out neurofeedback as treatment. Future studies might show that delayed effects occur; that a larger number of treatments, booster sessions, or both are effective; or that treatment works well for participants with particular baseline theta/beta ratios.

Note to readers: At the time that NEJM Journal Watch reviewed this paper, its publisher noted that it was not in final form and that subsequent changes might be made.

See Original Article here: