Postpartum depression (PPD) is one of the most common disorders following childbirth. This systematic review and meta-analysis (SR/MA) aimed to assess the effectiveness of psychological interventions in preventing PPD in non-depressed women. PRISMA guidelines were followed. MEDLINE (Ovid and PubMed), PsycINFO, Web of Science, Scopus, CINAHL, CENTRAL, OpenGrey, Australian New Zealand Clinical Trial Registry and clinicaltrial.gov were searched. Randomized controlled trials (RCTs) conducted with pregnant or postpartum (up to 12 months) women who were non-depressed at baseline were selected. The outcomes were the incidence of PPD and/or the reduction of postpartum depressive symptoms. The standardized mean difference (SMD) using random-effect models was calculated. Sensitivity, sub-group and meta-regression analyses were performed. 17 RCTs were included in the SR and 15 in the MA, representing 4958 participants from four continents. The pooled SMD was −0.175 [95% confidence interval (CI) −0.266 to −0.083; p < 0.001] and sensitivity analyses confirmed the robustness of this result. Heterogeneity was low (I2 = 21.20%) and was fully explained by a meta-regression model including one variable (previous deliveries). The meta-regression model and MA stratified by previous deliveries indicated that interventions focused on primiparous women are more effective. There was no evidence of publication bias. Few RCTs had an overall low risk of bias. According to GRADE, the quality of evidence was moderate. Psychological interventions have very little effectiveness in preventing PPD in non-depressed women, although this effectiveness is greater in interventions focused on primiparous women. Further RCTs with a low risk of bias and more effective interventions are needed.