OBJECTIVE: To validate the Edinburgh Postnatal Depression Scale (EPDS) according to DSM-IV, for detection of psychiatric morbidity at six weeks postpartum.
METHOD: A two-phase study was conducted in the population of women attended at the Obstetric Ward for postpartum check-up over one year. In the first phase, 1453 women completed the EPDS and a sociodemographic questionnaire. During the second phase, a stratified randomized sample (N=404) according to EPDS score and working status during pregnancy, was administered the Structured Clinical Interview for DSM-IV (SCID) in order to establish Axis-I psychiatric diagnosis. Reverse weighting was used to calculate sensitivity and specificity. Receiver Operating Characteristic (ROC) curve was also constructed. Positive and negative predictive values were calculated for the optimal cut-off point. Stata Release 7.0 and SPSS 10.0 were used, with P=0.05 and a confidence interval of 95% (CI 95%).
RESULTS: For the cut-off point of 7, sensitivity and specificity values were 88.2% (CI 95%: 73.2-95.3) and 84.0% (CI 95%: 80.2- 87.4) respectively, and positive and negative predictive values were 55.4% and 96.9% respectively. For the cut-off point of 8, sensitivity and specificity values were 83.3% (CI 95%: 70.1-91.4) and 87.7% (CI 95%: 84.3-90.5) respectively, and positive and negative predictive values were 60.3% and 95.9% respectively. The area under curve ROC was 0.953 (CI 95%: 0.938-0.969).
CONCLUSIONS: The EPDS is a valid screening instrument for identifying cases of the most common postpartum psychiatric morbidity with an optimal cut-off point of 7/8.
