Perinatal depression among women in pregnancy and postpartum periods in Gaborone, Botswana

Perinatal depression is an important cause of maternal and infant morbidity. Rates of perinatal depression amongst pregnant and postpartum women in Botswana are not well-documented. We examined the prevalence of symptoms of depression and potential associated factors among pregnant and postpartum women participating in a sexually transmitted infection screening study. We conducted a repeated time-point cross-sectional analysis of the Maduo study, a non-randomized clinic-controlled trial evaluating the benefit of antenatal Chlamydia trachomatis and Neisseria gonorrhoeae screening in Gaborone, Botswana. The Edinburgh Postnatal Depression Scale (EPDS) was used to screen for symptoms of depression at the first antenatal care and 6–8 week postnatal care visit. Prevalence of depression was assessed using EPDS cut-off scores of ≥11 and ≥13. Participants with an EPDS score of ≥13 were referred to local mental health services. We used multivariate logistic regression to identify potential associations with symptoms of perinatal depression (EPDS score ≥13) at first antenatal care visit, 6–8 week postnatal visit and overall, including socioeconomic, pregnancy-related and relationship factors. We also conducted a sensitivity analysis with EPDS score specified as a continuous variable. All 500 enrolled women completed at least one EPDS questionnaire during the study: 495 (99%) at first antenatal care and 428 (86%) at the postnatal visit. Median EPDS score was 3 (IQR 1–6) at the first antenatal visit and 2 (IQR 0–5) at the postnatal visit. Similar proportions of women had EPDS scores suggestive of depression at first antenatal (EPDS ≥13, 5% (95% CI 3–7%); EPDS ≥11, 7% (95% CI 5–10%)) and postnatal visits (EPDS ≥13, 3% (95% CI 1–5%); EPDS ≥11, 7% (95% CI 5–10%)), however there was minimal overlap in women with antenatal and postnatal symptoms of depression. Symptoms of depression were strongly associated with experience of intimate partner violence both antenatally (aOR 5.41, 95% CI 1.94–15.10, p = 0.001) and postnatally (aOR 5.18, 95% CI 3.35–8.02, p = < 0.001); whilst symptoms of depression at the first antenatal care visit were also associated with unintended pregnancy (aOR 29.44, 95% CI 2.75–314.6, p = 0.005). Rates of perinatal depression in this study were lower than estimated rates in Southern Africa. Nonetheless, screening pregnant women for depression as part of routine antenatal care in Botswana provides an opportunity to identify those most in need of support and specialist services.