Use this URL to cite or link to this record in EThOS:
Title: Perinatal depression in rural Ghana : burden, determinants, consequences, and impact of a community-based intervention
Author: Weobong, Benedict
Awarding Body: London School of Hygiene & Tropical Medicine
Current Institution: London School of Hygiene and Tropical Medicine (University of London)
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
The relative lack of research in mental health in low and middle income countries is symptomatic of the 10/90 gap in general health research where only 10% of the world’s expenditure on health research is dedicated to the poorest 90% of the world’s population. Globally there has been modest declines in both maternal and child deaths but there are still wide disparities between developed and developing countries; as the total number of under 5 deaths has declined, from 11.6 million in 1990 to 7.2 million in 2011, the proportion of deaths occurring in sub-Saharan Africa has increased from 33% in 1990 to 49% in 2011, and the region also bears the biggest burden (>50%) of maternal deaths. Innovations to reducing this burden are urgently needed in parallel with intensified efforts to increase coverage of proven effective maternal and child health interventions. One such innovation might be to include a focus on eliciting contextual determinants, and preventing and/or treating perinatal depression that is depression occurring during pregnancy or after birth, since there is some evidence suggesting that this is associated with adverse effects on infant health and development, and maternal health. This thesis is designed to add to this sparse evidence base by providing data on the burden of antenatal and postnatal depression in rural Ghana, examining determinants of this burden, investigating the links between perinatal depression and maternal and child health outcomes, and evaluating whether a home-visits intervention had reduced this burden. The research was undertaken within seven contiguous districts of the Brong Ahafo region of Ghana between January 2008 and July 2009. All women of reproductive age in these districts were part of a surveillance system supporting two randomised controlled trials that involved 4-weekly visits by resident fieldworkers who collected data on socio-demographics, obstetric histories, pregnancies, births, deaths and infant and maternal health. The research for this PhD involved training the surveillance field workers to also administer the depression module of the Patient Health Questionnaire screening tool (PHQ-9) to pregnant women and recently delivered mothers between 4-12 weeks after birth. 21135 pregnant women and 18356 recently delivered women were screened for depression, 13929 of whom were screened at both time points. The prevalence of postnatal depression (PND) was 3.8% (95% CI 3.5%, 4.1%), comprising 0.1% (95% CI: 0.08%-0.1%) who met criteria for major depression and 3.7% (95% CI: 3.4%-3.9%) for minor depression. The prevalence of antenatal depression (AND) was much higher 9.9% (95% CI: 9.5%-10.3%); 12.5% of these cases persisted into the postnatal period and accounted for 34.4% of postnatal cases. The following determinants were identified for antenatal depression: maternal age 30 years or older, never married, lower wealth status, non-Catholic religion, non-indigenous ethnicity, unplanned pregnancy, and previous pregnancy loss. And the following were identified for postnatal depression: never married, non-indigenous ethnicity, AND, season of delivery, peripartum/postpartum complications, newborn ill-health, still birth or neonatal death. Determinants were similar for ‘new’ cases of postnatal depression and for cases where depression was also detected antenatally. AND was found to be associated with the following consequences: prolonged labour, postpartum complications, peripartum complications, CS/instrumental delivery, severe newborn illness, and bed net non-use during pregnancy. PND was associated with increased risk of infant mortality up to six months (rate ratio [RR], 2·83 (1·56-5·16) and 12 months (RR, 1·79 (1·04-3·09) of age. Postnatal depression was also associated with increased risk of infant morbidity. Home-visits by community volunteers aimed at preventing neonatal deaths had no impact on attenuating prevalence of postnatal depression (relative risk [RR] 0.99 (95% CI 0.65, 1.50; p=0.96). This is the first large cohort study in SSA to provide evidence of determinants and consequences of perinatal depression, rather than studying the more general common mental disorder which include depression. The conclusions reached in this PhD are:1) Most risk factors of postnatal depression relate to adverse birth outcomes of the mother and/or baby, whereas those of antenatal depression are sociodemographic and pregnancy-specific, 2) Both antenatal and postnatal depression may have deleterious effects on the health of the mother and/or on child health and survival, 3) A case for clinical interventions for depression is established both during pregnancy and after birth, 4) Though often self-limiting, tackling antenatal depression could prevent up to a third of the burden of postnatal depression, 5) The timely implementation of such interventions using existing primary care structures may provide an important adjunct to improving maternal health and child health and survival efforts.
Supervisor: Kirkwood, B. R. Sponsor: Department for International Development ; Bill & Melinda Gates Foundation ; World Health Organization ; Psychiatric Research Trust (PALTHPA)
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral