Neurofeedback for Attention-Deficit/Hyperactivity Disorder: A Systematic Review and Meta-Analysis.

Importance: Neurofeedback has been proposed for the treatment of attention-deficit/hyperactivity disorder (ADHD) but the efficacy of this intervention remains unclear.

Objective: To conduct a meta-analysis of randomized clinical trials (RCTs) using probably blinded (ie, rated by individuals probably or certainly unaware of treatment allocation) or neuropsychological outcomes to test the efficacy of neurofeedback as a treatment for ADHD in terms of core symptom reduction and improved neuropsychological outcomes.

Data sources: PubMed (MEDLINE), Ovid (PsycInfo, MEDLINE, Embase + Embase Classic), and Web of Science, as well as the reference lists of eligible records and relevant systematic reviews, were searched until July 25, 2023, with no language limits.

Study selection: Parallel-arm RCTs investigating neurofeedback in participants of any age with a clinical ADHD or hyperkinetic syndrome diagnosis were included.

Data extraction and synthesis: Standardized mean differences (SMDs) with Hedges g correction were pooled in random effects meta-analyses for all eligible outcomes.

Main outcomes and measures: The primary outcome was ADHD total symptom severity assessed at the first postintervention time point, focusing on reports by individuals judged probably or certainly unaware of treatment allocation (probably blinded). Secondary outcomes were inattention and/or hyperactivity-impulsivity symptoms and neuropsychological outcomes postintervention and at a longer-term follow-up (ie, after the last follow-up time point). RCTs were assessed with the Cochrane risk of bias tool version 2.0.

Results: A total of 38 RCTs (2472 participants aged 5 to 40 years) were included. Probably blinded reports of ADHD total symptoms showed no significant improvement with neurofeedback (k = 20; n = 1214; SMD, 0.04; 95% CI, -0.10 to 0.18). A small significant improvement was seen when analyses were restricted to RCTs using established standard protocols (k = 9; n = 681; SMD, 0.21; 95% CI, 0.02 to 0.40). Results remained similar with adults excluded or when analyses were restricted to RCTs where cortical learning or self-regulation was established. Of the 5 neuropsychological outcomes analyzed, a significant but small improvement was observed only for processing speed (k = 15; n = 909; SMD, 0.35; 95% CI, 0.01 to 0.69). Heterogeneity was generally low to moderate.

Conclusions and relevance: Overall, neurofeedback did not appear to meaningfully benefit individuals with ADHD, clinically or neuropsychologically, at the group level. Future studies seeking to identify individuals with ADHD who may benefit from neurofeedback could focus on using standard neurofeedback protocols, measuring processing speed, and leveraging advances in precision medicine, including neuroimaging technology.