Memory deficits in children and adolescents with a psychotic disorder: a systematic review and meta-analysis

Early-onset psychosis (EOP) is a severe disorder which takes place before 18 years. It entails diverse clinical and functional implications, and it may lead to critical impairments in neurocognitive functions. Although deficits in memory are well described in adult populations and they appear to be clinically related with psychosis, impairments in memory in EOP show inconsistencies between studies. This study aimed to gain insight into the relationship between EOP and memory impairments, studying the potential contribution of moderators (storage source and memory content) on the observed memory deficits. This systematic review and meta-analysis was conducted following the PRISMA-2020 guidelines. Search was conducted in English and Spanish in five databases. Case-control studies which met all requirements were selected. Overall effect size was calculated under the random-effects model and Z-based tests were used. Heterogeneity was analysed by the I² statistic. Mixed-effects meta-regression analysis was used to study the influence of methodological quality of studies, mean age, proportion of female participants within sample, mean diagnosis, memory storage type, memory content as moderators on individual effect size variability. As a result, 32 articles were finally selected, pooling data from 2636 participants (49.29% EOP participants). Overall effect size was Hedges' g =  - 1.01, CI₉₅ = [ - 1.35,  - 0.67], p < .01, indicating lower memory performance in the EOP group in comparison to healthy controls. Diagnosis and memory storage were found as significant moderators in the memory performance variance: larger deficits were found in children with psychosis and in working memory tasks.