This longitudinal study evaluated the validity of earlier established predictors for postpartum posttraumatic stress (postpartum PTS) in a community sample of women (n = 248) in the Netherlands. Data was collected at 32 weeks gestation and at 3, 6 and 12 months postpartum by means of structured interviews. On the basis of the literature, an etiological model was defined with two pathways for the prediction of postpartum PTS. The model encompassed prepartum factors (e.g. depression in family, depression during life and depression during gestation), peripartum factors (e.g. type of delivery, experienced pain, social support by medical staff, supply of information and perinatal dissociation), and postpartum factors (e.g. postpartum depression). The model was evaluated by LISREL. Two pathways for postpartum PTS were confirmed: (1) delivery-related stressors predict postpartum PTS and (2) previous depression predicts postpartum PTS. Implications for mental health care during and after delivery are discussed. Copyright © 2005 John Wiley & Sons, Ltd.