This study verified the reliability and validity of the Edinburg Postpartum Depression Scale (EPDS) administered by telephone interviews. In a cross-sectional study of a cohort from Brazil (BRISA), the EPDS was administered by telephone to 1,083 women within 12 months postpartum, and 257 (23.7 %) participants had an EPDS score ≥10. At 67 ± 48 days after their telephone interview, 199 (EPDS ≥10 = 96; EPDS <10 = 103) participants were interviewed face-to-face using the Structured Clinical Interview for DSM-IV (SCID) and completed the EPDS again by self-report. In 90 participants, the diagnosis of major depressive episode was confirmed by the SCID (EPDS ≥10 = 65; EPDS <10 = 25). The Cronbach's alpha coefficient was 0.861. The Spearman's correlation between the EPDS administered by telephone and the self-reported EPDS was 0.69 (p < 0.001). The receiver-operating characteristic (ROC) curve for the EPDS administered by telephone was 0.78 (95 % confidence interval (CI) = 0.72 to 0.84). Scores ≥10 showed a sensitivity of 72.2 %, a specificity of 71.6 %, and a positive predictive value of 67.7 %. The application of the EPDS by telephone is a suitable alternative for clinical practice and research and represents a method to optimize the diagnosis of postpartum depression.