[The Edinburgh Postnatal Depression Scale: the validity of its Quebec version for a population low socioeconomic status mothers].

This paper investigates the construct validity and reliability of a Quebec version of the Edinburgh Postnatal Depression Scale (EPDS) for a population of low-socioeconomic-status mothers. This scale was constructed for the specific purpose of measuring mothers' symptoms of depression during the postnatal period in an effort to alleviate the validity problems that could arise from depression scales intended for the general population. Two hundred and twenty-four mothers participating in a Quebec prevention program, "Naître égaux, grandir en santé" (Martin & Boyer, 1995) filled out the EPDS between the 22nd and the 35th day postpartum. A confirmatory factor analysis, conducted with LISREL, gives a 2-factor structure for the EPDS, the first representing symptoms of depression and the second symptoms of anxiety. This structure differs from the one presented by Cox, Holden, and Sagovsky (1987), authors of the EPDS. It corresponds, however to the results of other authors who looked at the EPDS with confirmatory factor analysis (Pop, Komproe, & van Son, 1992) and indicates a good construct validity. The reliability of the scale also appears satisfactory, with a Cronbach alpha co-efficient of 0.82.