Abstract Objective The aim of this systematic review and meta-analysis was to study the association between specific environmental risk factors (ERF) and later development of Bipolar disorder and Psychotic depression. Methods A systematic search of prospective studies was conducted in MEDLINE, EMBASE and PsycINFO databases, and supplemented by hand searching, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (registration number: CRD42018092253). Selected ERF included: pre-/peri-natal factors—paternal age at birth, maternal infection, obstetric complications, perinatal stress; early childhood factors—urbanicity at birth, childhood infection, childhood adversity; later life factors—substance misuse, ethnic minority and migration, urbanicity later in life, stressful life events, and traumatic head injury. Pooled effect sizes of the association between these ERF and affective psychoses were calculated from systematically selected studies. When studies examining each ERF were insufficient for meta-analysis, results were presented narratively. Results Forty-six studies were included for quantitative analyses among selected ERF for affective psychosis, with significant association found for paternal age >40 years (OR 1.17, 95%CI 1.12–1.23), early (OR 1.52, 95%CI 1.07–2.17) and late (OR 1.32, 95%CI 1.05–1.67) gestational age, childhood adversity (OR 1.33, 95%CI 1.18–1.50), substance misuse (OR 2.87, 95%CI 1.63–5.50), and being from an ethnic minority (OR 1.99, 95%CI 1.39–2.84). Conclusions These results suggest some shared environmental load between non-affective and affective psychosis, implying generalized risks for psychosis rather than for specific diagnostic categories. Nonetheless, published studies for some ERF in the affective psychoses are scarce, and further longitudinal studies are needed.