Social factors linked to postpartum depression: A mixed-methods longitudinal study

Abstract

Background: There is a large body of research on the mental well-being of childbearing women focusing on the prevalence rates of antenatal and postpartum depression, and associated risk factors. However, there is a paucity of research on postpartum depression in its social context and in relation to the lived experience of the mother.

Aims: To investigate the association between social factors, the organization of maternity care services, and the prevalence of depression among a purposive sample of new mothers at 3 – 4 weeks and 4 – 6 months post-partum.

Method: Longitudinal, mixed-methods research design.

Results: Our quantitative analyses demonstrate that income and postpartum depression were linked and there was also an association between satisfaction with the birth experience and depression at 3 – 4 weeks postpartum. Thematic analysis of the qualitative interview data revealed two main sources of dissatisfaction: disruption of birth plans and inadequate support from maternity providers.

Discussion & limitations: We discuss some of the benefits and limitations of our mixed-methods longitudinal study and directions for further research.

Declaration of interests: None.

Keywords:

• Social factors
• mixed-methods
• longitudinal design
• satisfaction
• postpartum depression

Notes

1 In the research site, pregnant women can choose either a certified midwife or family physician as their primary birth attendant within the public health care system, but not both. Women must be referred by either a certified midwife or family physician to receive their primary care from an obstetrician.

2 Census Metropolitan Areas in Canada are regions consisting of one or more adjacent municipalities situated around a major urban core. To form a census metropolitan area, the urban core must have a population of at least 100,000. The Victoria CMA population during the time period of the study was 312,000 and was served by a single regional maternity hospital.

3 Our estimate is based on a birthrate of 9.6 per 1000 in the province of BC during the study period.

4 For 21 item original scale, <5 = unusual (“faking good”), 5 – 9 = normal ups and downs, 10 – 18 = mild/moderate depression, 19+ = moderate to severe depression, 30+ = severe. As expected, the majority of participants score in the normal range (<10).  The BDI was measured as a continuous variable except in one case were the sample was dichotomized in order to examine movement into and out of a significant depression symptoms category. When looking at patterns regarding depression symptoms over time by socioeconomic status, the sample was divided into two groups: those with scores higher than or equal to 10 and those who scored under 10.

5 Results at 3 – 6 weeks post-partum indicate that depression was inversely related to birth satisfaction, but ANOVA results indicate the results are not statistically significant (p = .17).