Practitioner Review: Cognitive bias modification for mental health problems in children and adolescents: a meta-analysis

• Published in: Journal of child psychology and psychiatry 2015-07, Vol.56 (7), p.723-734
• Main Authors: Cristea, Ioana A., Mogoașe, Cristina, David, Daniel, Cuijpers, Pim
• Format: Article
• Language: English
• Subjects: Adolescent; Adolescents; Anxiety Disorders - diagnosis; Anxiety Disorders - psychology; Anxiety Disorders - therapy; Attention; Behavior Therapy; Child; Child & adolescent psychiatry; children; clinical efficacy; Cognition & reasoning; Cognition Disorders - diagnosis; Cognition Disorders - psychology; Cognition Disorders - therapy; Cognitive bias modification; Cognitive Therapy; Depressive Disorder - diagnosis; Depressive Disorder - psychology; Depressive Disorder - therapy; Effect Size; Humans; Judgment; Mental disorders; Mental Health; Meta Analysis; Outcome and Process Assessment (Health Care); Psychopathology; Randomized Controlled Trials; Randomized Controlled Trials as Topic; RCT interventions; Social Stigma; Systematic review
• ISSN: 0021-9630; 1469-7610
• DOI: 10.1111/jcpp.12383

Background Despite accumulating research and bold claims about the efficacy of cognitive bias modification (CBM) for young populations, no meta‐analysis has attempted to synthesize the research literature so far. We examined whether there was empirical evidence for the clinical efficacy of CBM interventions in youths, while also considering the methodological quality of this evidence. Methods Studies were identified through systematic searches in bibliographical databases (PubMed, PsychInfo, Cochrane Library and EMBASE to June 2014). We included randomized controlled trials of CBM interventions, and considered both clinical outcomes and targeted biases. We examined the quality of the trials, as well as potential publication bias and possible moderators. Results We identified 23 trials that reported on four types of outcomes: mental health, anxiety, depression and bias. Effect sizes were small and nonsignificant for all symptom outcomes considered. We found a moderate significant effect size for bias outcomes (Hedges' g of 0.53), with significant heterogeneity. There were no differences between types of CBM interventions, or between one versus multiple‐session applications. A small but significant effect size for mental health problems arose when the intervention was delivered in schools. The quality of almost all of the included studies was suboptimal and the vast majority did not include information needed for allowing quality assessment. Conclusions We conducted the first meta‐analysis of CBM interventions for children and adolescents and found no effects for mental health outcomes, but we did find moderate and significant effects on the targeted biases. Our results cast serious doubts on CBM interventions having any clinical utility for nonadult populations. Demand characteristics might play an important part in CBM research. Read the Commentary on this article at doi: 10.1111/jcpp.12436