Computerized cognitive training in attention-deficit/hyperactivity disorder (ADHD): a meta-analysis of randomized controlled trials with blinded and objective outcomes

This meta-analysis investigated the effects of computerized cognitive training (CCT) on clinical, neuropsychological and academic outcomes in individuals with attention-deficit/hyperactivity disorder (ADHD). The authors searched PubMed, Ovid, and Web of Science until 19th January 2022 for parallel-arm randomized controlled trials (RCTs) using CCT in individuals with ADHD. Random-effects meta-analyses pooled standardized mean differences (SMD) between CCT and comparator arms. RCT quality was assessed with the Cochrane Risk of Bias 2.0 tool (PROSPERO: CRD42021229279). Thirty-six RCTs were meta-analysed, 17 of which evaluated working memory training (WMT). Analysis of outcomes measured immediately post-treatment and judged to be “probably blinded” (PBLIND; trial n = 14) showed no effect on ADHD total (SMD = 0.12, 95%CI[−0.01 to −0.25]) or hyperactivity/impulsivity symptoms (SMD = 0.12, 95%[−0.03 to−0.28]). These findings remained when analyses were restricted to trials (n: 5–13) with children/adolescents, low medication exposure, semi-active controls, or WMT or multiple process training. There was a small improvement in inattention symptoms (SMD = 0.17, 95%CI[0.02–0.31]), which remained when trials were restricted to semi-active controls (SMD = 0.20, 95%CI[0.04–0.37]), and doubled in size when assessed in the intervention delivery setting (n = 5, SMD = 0.40, 95%CI[0.09–0.71]), suggesting a setting-specific effect. CCT improved WM (verbal: n = 15, SMD = 0.38, 95%CI[0.24–0.53]; visual-spatial: n = 9, SMD = 0.49, 95%CI[0.31–0.67]), but not other neuropsychological (e.g., attention, inhibition) or academic outcomes (e.g., reading, arithmetic; analysed n: 5–15). Longer-term improvement (at ~6-months) in verbal WM, reading comprehension, and ratings of executive functions were observed but relevant trials were limited in number (n: 5–7). There was no evidence that multi-process training was superior to working memory training. In sum, CCT led to shorter-term improvements in WM, with some evidence that verbal WM effects ...