Association between perinatal depressive symptoms and parental cohabitation status in a Nordic high-income country

This data supports findings in a nested study within the population-based prospective Biology, Affect, Stress, Imaging, and Cognition (BASIC) and U-BIRTH cohort studies conducted from September 1st, 2009 to February 24th, 2022 in Uppsala, Sweden (Written, informed consent was obtained from all participants in accordance with the Declaration of Helsinki and the research was approved by the Regional Ethical Review Board in Uppsala [BASIC: Dnr 2009/171; U-BIRTH: Dnr 2012/010, both with amendments]). The primary aim was to examine the relationship between maternal perinatal depressive symptoms (PND) and parental cohabitation status over time. Although PND is a known risk factor for compromised maternal and child well-being, little is known about how the persistence of depressive symptoms during the perinatal period influences long-term family structure and parental cohabitation. Previous studies have typically relied on a single time point to assess depression, overlooking the role of symptom duration and trajectory. In this study, maternal PND symptoms were measured at three time points—during pregnancy, 6 weeks postpartum, and 6 months postpartum—using the Edinburgh Postnatal Depression Scale (EPDS). Parental cohabitation status was assessed at 6 weeks and again at 6 years postpartum. The exposure variable was the number of time points with positive PND screenings, while sociodemographic, psychosocial, and clinical variables were collected as potential confounders. This dataset includes maternal mental health data, cohabitation information, and background characteristics such as age, education, parity, pregnancy planning, psychiatric history, exposure to stressful life events, and experience of violence. The individual data and code associated with this study, are available upon reasonable request and after necessary agreements have been signed by the involved parties. The data cannot be shared openly as they contain sensitive information and due to the characteristics of the sample, deductive disclosure cannot be completely excluded.