Key points are not available for this paper at this time.
BACKGROUND: Cognitive behavioral therapy (CBT) for adult ADHD provides adaptive coping strategies targeting executive dysfunction and emotion regulation, processes underlying functional impairment. However, most research has prioritized symptom reduction over functional outcomes. This meta-analysis quantified CBT effects on global functioning, quality of life (QoL), and domain-specific functioning (social, occupational, educational), while examining moderators and the role of ADHD symptom change. METHOD: We included 70 studies (5152 participants): 38 RCTs, four non-randomized trials, 24 single-arm trials, and four single-case designs. Random-effects meta-analyses estimated standardized mean differences for between-group and within-subject outcomes at post-treatment and follow-up. Risk of bias was formally assessed, and GRADE assessed evidence certainty. RESULTS: CBT produced moderate effects on global functioning at post-treatment (SMD = 0.47) that increased at follow-up (SMD = 0.67). QoL effects were smaller and non-significant at follow-up. Strongest effects were found for occupational functioning, effects on social functioning were moderate, and academic benefits were limited. Individual format and longer duration were associated with better outcomes. ADHD symptom improvement predicted QoL but not functioning. CONCLUSIONS: CBT was associated with improvements in functioning (often the primary treatment concern), with moderate effects observed at post-treatment and some evidence of sustained effects at follow-up, particularly in directly targeted domains. The dissociation between symptom change, QoL, and functioning suggests that functional gains may not be fully explained by symptom reduction alone, although the mechanisms underlying these effects remain unclear. These findings support a more process-oriented perspective on CBT, while highlighting the potential relevance of directly targeting everyday functioning within psychological interventions. Risk of bias, self-report reliance, Western sample concentration, and short follow-ups (typically 3 months) warrant cautious interpretation.
López-Pinar et al. (2026) studied this question.