Screening for common perinatal mental disorders in South Africa

Background: In South Africa the burden of common perinatal mental disorders, is approximately three times higher than in high-income countries, with negative consequences for maternal and child health. The high prevalence and 80% treatment gap may be attributable to multiple contextual risk factors that women encounter as well as health care system barriers. This thesis sought to establish the diagnostic prevalence and psychosocial predictors for Common Perinatal Mental Disorders (CPMD) amongst women living in adversity and to develop a brief screening tool to detect such disorders in low-resource settings. Methods: This cross-sectional study was conducted at a primary level, midwife-run obstetric unit in Hanover Park, Cape Town. The data from 376 pregnant women attending their antenatal, “booking” visit were included. These data were collected using a diagnostic interview and self-reporting screening questionnaires. Logistic regression models explored demographic and socio-economic characteristics, psychosocial risk factors and psychiatric comorbidity as predictors for CPMD. The Area Under the Curve (AUC) from Receiver Operator Characteristic (ROC) curve analysis compared screening tool performance. Bestperforming items from screening tools were analysed against diagnostic data using multiple logistic regression and ROC analysis to develop a novel screening tool. Findings: Prevalence of diagnosable major depressive episode (MDE) and anxiety disorders were 22% and 23% respectively, 50% of depressed women expressed suicidality. CPMD diagnosis was significantly associated with multiple risk factors, including a history of mental health problems, food insecurity, traumatic life events and lower perceived social support. Multigravidity, unintended and unwanted pregnancy and pregnancy loss were strongly associated with antenatal anxiety. The Edinburgh Postnatal Depression Scale (EPDS) and Whooley questions were the best performing screening instruments. Multiple logistic regression identified four items independently predictive of CPMD. At a cut-off of 2, the combined items yielded an AUC of 0.83, with a sensitivity of 78% and specificity of 82%, comparable to the EPDS. Conclusions: These findings confirm the high prevalence of antenatal CPMD in low- and middle-income country (LMIC) settings, where women experience multiple risk factors. Ultrashort screening tools adequately detect CPMD, and a novel, four-item screening tool may be useful for the early identification of mental health symptoms in pregnancy, as part of collaborative stepped care. This is the first measure of its kind developed in South Africa, further novelty is that it is a combined depression and anxiety tool, and includes screening for suicidal ideation