Abstract
The majority of maternal and newborn death occur in low and middle-income countries around the time of birth and are preventable with good quality care. In this thesis, we used mixed methodological approaches to analyse quality of care, synthesise and generate evidence for context-specific strategies to improve the quality of
... read more
care at birth in high-volume, resource-limited labour wards in LMICs, such as Mnazi Mmoja Hospital (MMH), Zanzibar, Tanzania. At this hospital, we found high stillbirth rate (59 per 1000 total births) half of whom died intrahospital, and in general, suboptimal quality of care. We identified a number of inadequate areas to target intervention: 1) admission assessment of maternal foetal characteristics and wellbeing 2) routine surveillance of maternal and foetal wellbeing and progress of labour 3) diagnosis and management of pregnancy and labour complications, 4) clinical documentation and health information systems; 5) respectful maternity care and bereavement care.
We also identified daily challenges to providing good care including insufficient number of well-trained skilled birth attendants, lack of essential supplies and unconducive environment. Given the persistent health system’s constraints in providing adequate resources, efforts are needed to find alternative strategies to optimise care.
We identified several strategies for care improvement: 1) task-shifting of foetal heart rate monitoring to trained lay workers; however, this solution was not acceptable to local stakeholders who raised many concerns. 2) The use of partograph with management guidelines improved perinatal outcomes. However, these top-down generic guidelines are often a mismatch as they do not take context into account and are unachievable in these settings. Thus, through co-creation with skilled birth attendants at MMH, we adapted international guidelines of clinical care to the local context and implemented them through repeated training. Improvements were found in knowledge, partograph skills and clinical practice including foetal surveillance and treatment of complication during labour. These were associated with a reduction in stillbirths by 33% to 39 per 1000 total births, halving of the number of newborns with an Apgar score of 1–5 to 28 per 1000 live births. 3) We developed a high-performance (c-static: 0.8) model consisting of 15 easily-available clinical predictors to assist skilled birth attendants to identify and triage women at risk of intrapartum-related perinatal deaths for proper follow-up during labour.
This thesis sets an example of how to contextualise evidence and interventions using bottom-up approaches and interdisciplinary collaboration of international and local researchers, frontline healthcare workers and service users.
show less
