Perinatal depressive symptoms among low-income South African women at risk of depression: trajectories and predictors

BACKGROUND: The aim of the study was to identify trajectories of perinatal depressive symptoms and their predictors among women living in a low-resource setting in South Africa, and who present with a risk of depression during pregnancy. METHODS: This is a secondary analysis of a randomised controlled trial among 384 women living in Khayelitsha, a low income setting in South Africa, recruited at their first antenatal visit if they scored 13 or above on the Edinburgh Postnatal Depression Scale, were at least 18 years of age, less than 29 weeks pregnant and spoke isiXhosa. Participants were followed up at 8 months gestation, 3 and 12 months postpartum. Latent trajectories of depressive symptoms were identified using growth mixture modelling, based on the Hamilton Depression Rating Scale (HDRS). There were no differences in HDRS scores between the control and intervention arms, so all participants were assessed together. Health, social and economic predictors of trajectories were investigated to identify high-risk groups with greater or more chronic depressive symptoms, using univariate logistic regression. RESULTS: Two trajectories were identified: antenatal only (91.4%), with moderate to severe symptoms at baseline which later subside; and antenatal and postnatal (8.6%), with severe depressive symptoms during pregnancy and later in the postpartum period, which subside temporarily to moderate levels at 3 months postpartum. Predictors for the antenatal and postnatal trajectory include severe food insecurity, intimate partner violence, lower social support, greater functional impairment, problematic drinking and suicide risk. CONCLUSIONS: A small proportion of women who are at risk for depression antenatally remain at risk throughout the perinatal period, and can be differentiated from those who show a natural remission. Identification and referral strategies should be developed with these findings in mind, especially given the limited mental health resources in low-income settings. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12884-019-2355-y) contains supplementary material, which is available to authorized users.