Abstract
BackgroundMore than two million babies are stillborn each year, and an additional 2.4 million babies die in their first month of life. Nearly all these deaths occur in low- and low-middle-income countries, notably in South Asia and sub-Saharan Africa and the majority could be preventable with high coverage of current evidenced-based interventions. Measurement of coverage and quality of care around the time of birth and in the first two days of life can track progress towards achieving the Sustainable Development Goal for neonatal mortality reduction and Every Newborn Action Plan goals for stillbirth reduction.
Aim
The overarching research question I address in this thesis is: what is the level of coverage, degree of quality of care, and measurement accuracy for newborn care in the first two days after birth.
Methods
For this PhD, three datasets (Demographic and Health Surveys, Service Provision Assessments, and the Every Newborn Birth Indicators Tracking in Hospitals Study) spanning 20 low- and middle-income countries, and including over 100,000 births, were used to examine coverage, quality and measurement of newborn care and outcomes. Descriptive analyses were conducted to assess coverage and quality of care. Regression models, including survey- weighted multi-level models, were fitted to identify factors associated with high coverage of practices and interventions. Criterion validity testing methods were used to analyse the accuracy of survey-reported and register-recorded care and outcomes compared to gold- standard observation data.
Results
The first minutes after birth
Neonatal resuscitation with bag-mask ventilation coverage was high among newborns not breathing at birth, yet quality was very low with almost no newborn receiving timely ventilation within one minute. Women's survey-report underestimated observed coverage of bag-mask ventilation (0.9% compared to 4.0%). Routine hospital registers were highly complete for birth outcomes but accuracy varied for both birth outcome and bag-mask ventilation. Quality gaps were identified in respectful person-centred care for families with stillbirths, including lower rates of birthweight measurement for stillbirths compared with livebirths. Early breastfeeding support was associated with higher rates of early initiation of breastfeeding but not with prelacteal feeding or exclusive breastfeeding.
The first two days of life
Coverage of key newborn care interventions varied widely both between and within countries. Differences between countries were large. Among facility births in Zimbabwe, coverage of all interventions was greater than 60%, yet in Burundi there was no more than 12% coverage for any intervention. Within countries, large differences were found between wealth groups. Coverage of breastfeeding support in Benin was five times higher among the richest compared with the poorest wealth group. Proximity to high-quality facilities and social determinants of health at multiple levels (individual, family, community, structural, contextual) were associated with improved quality of care.
Conclusion
Gaps in coverage and quality of key evidence-based interventions for newborns exist even following facility birth and postnatal care contacts. Accurate measurement of newborn care and outcomes is crucial to track coverage and quality of care and improve progress towards global goals to end preventable neonatal mortality and stillbirth. Population-based surveys remain essential to tracking contact with services and outcomes; however, challenges include accuracy of reporting interventions around the time of birth and frequency of survey implementation. As facility births increase, routine measurement and facility registers hold the potential to track clinical care and outcomes in the critical time around birth. Improvements in data quality would increase confidence to use these data to inform progress towards meeting global newborn survival goals.
| Date of Award | 1 Jan 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Debra Bick (Supervisor), Cath Taylor (Supervisor), Edward Purssell (Supervisor), Louise T. Day (Supervisor) & Jane Sandall (Supervisor) |