Maternal common mental disorder in Malawi, Africa

Abstract

Background Maternal common mental disorder (CMD), characterised by depression, anxiety and somatic symptoms of distress, is known to be an important health problem in low-and-middle- income countries, but had not been investigated in Malawi, Africa. In preliminary work, we adapted a CMD screening measure and showed that post-partum CMD was common and associated with child stunting. In the research presented here, we expanded the investigation of CMD in Malawi to include pregnant women and mothers of children with severe acute malnutrition (SAM). Using validated measures, we aimed to estimate the prevalence of antenatal CMD/depression and investigate its relationship to social support and intimate partner violence. Secondly, we sought to explore women’s lived experiences of the perinatal period. Finally, we investigated CMD amongst mothers of children admitted with SAM and other life-threatening illness and, in the former, we tested the hypothesis that maternal CMD would be associated with impaired child recovery. Method In an antenatal clinic-based study, we validated and compared Chichewa and Chiyao versions of the Self Reporting Questionnaire (SRQ) and Edinburgh Postnatal Depression Scale (EPDS) using major depressive episode (MDE) as the criterion diagnosis. We adapted and validated the Multi-dimensional Scale of Perceived Social Support (MSPSS). We measured the prevalence of antenatal MDE and identified associated factors. We conducted focus group discussions with women and enquired about perinatal stressors and supports. We administered the SRQ to mothers of children with SAM during admission to a nutritional rehabilitation unit (NRU) and at 1-month post-discharge, and investigated whether CMD was a risk factor for lower child weight gain at follow-up. In a subsequent study, we compared levels of CMD symptoms between mothers of children admitted to a NRU, a high dependency unit and an oncology ward. Results We found that the adapted EPDS and SRQ were both valid screening instruments for antenatal CMD/depression. The weighted prevalence of antenatal MDE was 10.7% (95% CI 6.9% - 14.5%). The adapted MSPSS showed adequate test characteristics and differentiated between sources of social support. MDE was associated with lack of support by a significant other; intimate partner violence moderated this association. We found that women in rural Malawi recognised depressive and anxious states in the perinatal period and identified lack of partner support as a key stressor. During admission with a severely malnourished child to a NRU, mothers had very high SRQ scores that greatly reduced post-discharge. There was no association between SRQ score and child weight gain at follow-up. We found no higher level of CMD symptoms amongst mothers of children admitted for treatment of SAM compared with those admitted to other wards. Conclusions We demonstrated that measures of antenatal CMD and perceived social support can be adapted for use in Malawi. We found that maternal CMD is common and associated with lack of social support, intimate partner violence and child illness. We did not find evidence for a specific association between maternal CMD and child SAM but further prospective studies are required. Our findings suggest that treatment of CMD in mothers in Malawi will require attention to social support and partner behaviour.