Abstract

A longitudinal follow-up study of postnatal depression was performed in the years 1995–2000 in the University of Oulu. A volunteer, ramdomly selected group of 187 mothers from maternal wards of the University Hospital of Oulu were studied with different questionnaires in the first postpartum week. Depressive symptoms were re-assessed with Edinburgh Postnatal Depression Scale four months postpartum when paternal depression was evaluated with Beck Depression Inventory. 16.2% of the women were immediately after delivery screened as being depressed. Four months postpartum 13.0% of the mothers were depressed whereas 5.1 % of the fathers were having depressive symptoms. The cumulative incidence of maternal postnatal depression within the first four months was 22.2%.

These mothers who immediately showed depressive symptoms were at a higher risk to be depressed later. Maternal age 30 years or less predicted postnatal depression. The occurrence of maternal postnatal depression varied slightly during different seasons; during dark time immediate depressive symptoms increased and the spring seemed to protect from later postnatal depression. Analgesia during vaginal delivery, e.g. nitrous oxide, epidural analgesia or paracervical blockade, protected from postnatal depression as well. Caesarean section, either elective or emergency, did not predict postpartum mental well-being. Scores from the GHQ and the EPDS were strongly interrelated. Seven (5%) fathers were depressed four moths postpartum. They all were men whose partners also scored high in the EPDS.

Those mothers who were depressed interpreted infant facial signals differently, seeing less joy, disgust and anger, but more sadness in the infant facial pictures. Cultural variability was found in complex blended facial features of emotions, e.g. distress, in the Infant Facial Expression from Looking at Picture scale, although remarkable agreement was achieved and reinforced.

In the videotaped early mother-infant interaction small, but essential, changes were observed at 10 months postpartum. Overall, mothers who had had persistent depressive symptoms showed less negative expressions and had less anger and anxiety in their interaction. Their children were slightly less impulsive and seemed to have less visual and communicative contact with their mothers. The dyad was characterised by short periods of uninvolvement between the mother and the infant; e.g. the moments of reciprocity were less frequent than in the non-depressed mothers. At 42 months postpartum, the children of the persistently depressed mothers scored lower on the Expressive language scale than the children of the mothers who were depressed in one measurement or never.