Social withdrawal is an early manifestation of several neuropsychiatric disorders, and it is characterised by a gradual disengagement from social interactions, potentially leading to complete isolation. This study investigated the association between social withdrawal at baseline and short-term symptom remission in five independent cohorts, including patients with major depressive disorder (MDD), bipolar spectrum disorders, and schizophrenia. Measures of social withdrawal were derived in each study, and clinical remission was estimated based on the psychopathological severity assessed after short-term psychopharmacological treatment (12 weeks). Logistic regression was performed in each sample, adjusting for age and baseline psychopathological severity residualised for social withdrawal. Results were then meta-analysed across samples within a random-effect framework. A total of 4461 patients were included in the analyses (3195 patients with MDD, 655 with bipolar spectrum disorders and 611 with schizophrenia). The meta-analysis showed that higher baseline levels of social withdrawal were associated with a decreased likelihood of short-term remission (ORadj = 0.67, 95% CI, 0.58-0.79, P = 5.28 × 10-7), with the strongest effect in patients with schizophrenia. Overall, our study highlighted the need to address social withdrawal in the early phases of the disease to promote symptom remission in patients with major psychiatric disorders. Understanding the neurobiology underlying social withdrawal may aid the development of medications that can specifically reverse social impairment, thereby fostering clinical remission.