Data source and search strategy

A comprehensive search on PubMed and EMBASE was made for any studies reporting data on maternal PCOS and neuropsychiatric symptoms in children prior to March 30, 2021. In addition, a google scholar search was also made up to five pages to determine any relevant study for this meta-analysis. The following search terms: (“maternal polycystic ovary syndrome” OR “polycystic ovarian syndrome” OR “PCOS” OR “fetal testosterone”) AND (“neuropsychiatric disorder” OR “neurodevelopmental disorder” OR “mental disease” OR “autism” OR “ASD” OR “attention deficit disorder” OR “ADHD” OR “eating disorder” OR “depression” OR “anxiety” OR “schizophrenia” OR “bipolar disorder”) AND (“offspring” OR “children” OR “child”) were used to screen the eligible articles. Full articles and abstracts were independently reviewed for the eligibility criteria by three authors (PD, SR, and JC). References from the review articles were also cross-checked to screen any pertinent studies for this study. We included any articles evaluating the association between maternal PCOS with NPD in children regardless of the direction of the study. Study selection was also restricted to articles written in English only. Descriptive studies (case series, case reports), clinical trials, non-human studies, abstract, review, and duplicated studies were excluded from this study.