TY - JOUR
T1 - Longitudinally investigating patterns of maternal psychological distress in a South African birth cohort
AU - MacGinty, Rae
AU - Hoffman, Nadia
AU - Koen, Nastassja
AU - Zar, Heather J.
AU - Stein, Dan J.
AU - Pellowski, Jennifer
N1 - Publisher Copyright:
© The Author(s) 2025.
PY - 2025/12
Y1 - 2025/12
N2 - Background: Psychological distress, a broad construct that encompasses of a range of emotional difficulties—including depression and anxiety—is prevalent during pregnancy and postnatally with up to 25% of women experiencing psychological distress globally. However, this is mostly described in high income countries (HIC), with little data from low- and middle-income countries (LMICs). The aim of this study was to describe and determine latent trajectory classes of maternal psychological distress from pregnancy through five years postpartum in a LMIC setting. Methods: Data were used from women enrolled in the Drakenstein Child Health Study (DCHS), a birth cohort in South Africa. Women who had completed the Self-Report Questionnaire-20 items (SRQ-20) at two or more timepoints from pregnancy to 5 years postpartum, were included in the analysis. Latent class mixed modelling (LCMM) was used to generate latent trajectory classes of maternal psychological distress. Predictors, including trauma exposure, socio-economic indicators, substance use and pregnancy complications, of the latent classes were investigated using multinomial logistic regression. Results: In 973 women, four trajectory classes of maternal psychological distress were derived from pregnancy through 5 years postpartum, 1) persistent psychological distress symptoms, 2) antenatal symptoms only, 3) late postnatal onset of symptoms (post perinatal stage) and 4) low symptoms of psychological distress. Predictors of the persistent symptom class included early and recent trauma exposure, smoking during pregnancy, and gestational diabetes, whereas partner support was protective. Trauma exposure prior or during pregnancy was a predictor for the antenatal symptom class, while postnatal trauma exposure was associated with the late onset symptom class. Conclusion: Trajectories and predictors of maternal psychological distress were found to be similar to those seen in depression only and in HIC, which suggests there could be a degree of universality with regards to predictors of psychological distress. Trauma experienced by women was found to be a critical risk factor for psychological distress, as was maternal smoking or lack of partner support. There is an urgent need for social transformation surrounding gender-based violence and prevention and smoking cessation programs targeting women of child bearing age.
AB - Background: Psychological distress, a broad construct that encompasses of a range of emotional difficulties—including depression and anxiety—is prevalent during pregnancy and postnatally with up to 25% of women experiencing psychological distress globally. However, this is mostly described in high income countries (HIC), with little data from low- and middle-income countries (LMICs). The aim of this study was to describe and determine latent trajectory classes of maternal psychological distress from pregnancy through five years postpartum in a LMIC setting. Methods: Data were used from women enrolled in the Drakenstein Child Health Study (DCHS), a birth cohort in South Africa. Women who had completed the Self-Report Questionnaire-20 items (SRQ-20) at two or more timepoints from pregnancy to 5 years postpartum, were included in the analysis. Latent class mixed modelling (LCMM) was used to generate latent trajectory classes of maternal psychological distress. Predictors, including trauma exposure, socio-economic indicators, substance use and pregnancy complications, of the latent classes were investigated using multinomial logistic regression. Results: In 973 women, four trajectory classes of maternal psychological distress were derived from pregnancy through 5 years postpartum, 1) persistent psychological distress symptoms, 2) antenatal symptoms only, 3) late postnatal onset of symptoms (post perinatal stage) and 4) low symptoms of psychological distress. Predictors of the persistent symptom class included early and recent trauma exposure, smoking during pregnancy, and gestational diabetes, whereas partner support was protective. Trauma exposure prior or during pregnancy was a predictor for the antenatal symptom class, while postnatal trauma exposure was associated with the late onset symptom class. Conclusion: Trajectories and predictors of maternal psychological distress were found to be similar to those seen in depression only and in HIC, which suggests there could be a degree of universality with regards to predictors of psychological distress. Trauma experienced by women was found to be a critical risk factor for psychological distress, as was maternal smoking or lack of partner support. There is an urgent need for social transformation surrounding gender-based violence and prevention and smoking cessation programs targeting women of child bearing age.
KW - Anxiety
KW - Depression
KW - Longitudinal
KW - Low and middle income
KW - Maternal psychological distress
UR - https://www.scopus.com/pages/publications/105018259645
U2 - 10.1186/s12889-025-24445-x
DO - 10.1186/s12889-025-24445-x
M3 - Article
C2 - 41063014
AN - SCOPUS:105018259645
SN - 1472-698X
VL - 25
JO - BMC Public Health
JF - BMC Public Health
IS - 1
M1 - 3409
ER -