Abstract
Introduction: Poor mental health during pregnancy and postpartum is common but remains an under-researched topic. By mapping the relationships between risk factors, protective factors, and symptoms, or between individual symptoms of mental disorders, network analyses provide novel insights into mental health. Few previous studies have used this holistic approach within the field of maternal mental health. Using network analyses, the present study aimed to explore (1) risk factors, protective factors, and symptoms involved in mental health, (2) differences between parous and nulliparous samples, (3) connections between depressive and anxiety symptoms, and (4) longitudinal differences in depressive symptoms before conception and during pregnancy, and during pregnancy and postpartum. Methods: Questionnaire data from N = 333 women (Age: M = 32.04 years, SD = 3.48) collected within a pregnancy study with assessment points before conception (baseline), during pregnancy, and postpartum were analyzed. A network (n = 269) of risk factors, protective factors, and symptoms was conducted, with a strength centrality analysis to identify the most strongly connected variables. A network comparison test (NCT) compared parous (n = 90) and nulliparous (n = 179) subsamples. Next, a comorbidity network (n = 269) of depressive and anxiety symptoms was conducted, with a bridge strength centrality analysis to identify connecting symptoms. Paired NCTs were used for longitudinal comparisons of depressive symptoms networks before conception and during pregnancy (n = 119), and during pregnancy and postpartum (n = 115). Results: Neuroticism, perceived stress, depressive symptoms, and anxiety symptoms showed the highest strength centrality. The parous and nulliparous networks did not differ in their overall structure or global strength. Four bridge symptoms (“I have been anxious or worried for no good reason”, “I have felt scared or panicky for no very good reason”, “Feeling nervous, anxious, or on edge”, and “Not being able to stop or control worrying”) connected depressive and anxiety symptoms. No significant differences between the pre-conception and pregnancy networks were found in overall structure or global strength. The pregnancy and postpartum networks did not differ significantly in overall structure; however, the pregnancy network was significantly more strongly connected than the postpartum network (p < .05). Conclusions: The present findings highlight risk factors and bridge symptoms that could be targeted to improve mental health. The longitudinal results show a dynamic nature of maternal mental health and indicate that ideal time points for interventions may exist. Future research is needed to explore the effect of interventions on networks of maternal mental health. In general, mental health research can benefit from the use of holistic approaches.