Background: Autism is a condition of developmental abnormalities in social, communication and behavior aspects. Polycystic ovary syndrome and maternal diabetes during pregnancy contribute more than 50% to the risk of ASD offspring. This study aims to analyze and estimate the influence of polycystic ovarian syndrome and maternal diabetes on autism in children. Subjects and Method: The meta-analysis was carried out according to the PRISMA flow chart and the PICO model. Q: Child, I: Mother with PCOS and maternal diabetes, C: T Mother without PCOS and maternal diabetes, O: Autism. Search for articles in this study through databases that include Google Scholar, Pubmed, ScienceDirect and Sage Journal. With keywords including: Polycystic Ovarian Syndrome” AND “Maternal Diabetes” OR “Gestational Diabetes Mellitus” AND “Autism” OR “Autism Spectrum Disorder” AND “Cohort”. A full paper article with an observational cohort study, the research subject was a child, the size of the relationship used was the adjusted odds ratio, the research outcome was autism. Analysis was performed with Revman 5.3. Results: There were 13 articles with cohort designs originating from America, Denmark, England, Sweden, Israel, China with a total of 4,641,483 research samples. A meta-analysis of 13 cohort studies concluded that children of pregnant women with PCOS had a 1.36 times greater risk of developing autism than those without PCOS, and the effect was statistically significant (aOR= 1.36; 95% CI= 1.24 to 1.49; p<0.001). In addition, pregnant women with maternal diabetes have a 1.24 times higher risk of having a child with autism than those without maternal diabetes, and this effect is statistically significant (aOR= 1.24; 95% CI= 1.08 to 1.43; p=0.002). Conclusion: Pregnant women who have PCOS and maternal diabetes increase the risk of autism in children. Keywords: polycystic ovarian syndrome, maternal diabetes, autism.