Research process
The Exemplars in Neonatal and Maternal Mortality study aims to systematically and comprehensively evaluate factors associated with rapid reductions in neonatal and maternal mortality. Selected countries demonstrated success at reducing neonatal and maternal mortality beyond what would be expected based on economic progress alone—as measured by changes in the gross national income per capita from 2000 to 2017. Data availability, regional representation, and existing in-country partnerships were also considered in the country selection process. The countries identified as Exemplars were Bangladesh, Ethiopia, India, Morocco, Nepal, Niger, and Senegal.
Figure 41: NMR/MMR Exemplars Country Selection
A mixed-methods approach was used to assess key drivers of progress and findings have been synthesized across five methods of inquiry: literature review, quantitative analysis of country-level data, qualitative analysis, policy and program review, and financing analysis. The research in Bangladesh has been completed in collaboration with a consortium of research partners, including Countdown to 2030 and the Institute for International Programs at Johns Hopkins Bloomberg School of Public Health.
Summary of Methods
| Method | Data sources | Short description of methods |
|---|---|---|
|
Literature search |
Peer-reviewed literature databases (PubMed, Scopus) | Comprehensive search of policy and program literature focused on Bangladesh and MNH. |
| Mortality rates: national and stratified | DHS (1993–2018) BMMS (2001, 2010, 2016) IGME IHME MMEIG |
Trends analysis using modeled estimates from IGME, MMEIG, and IHME. Analysis of pooled survey data for trends and inequalities. |
| Cause of death |
BMMS (2001, 2010, 2016) WHO/MCEE, IHME |
Causes of maternal death were obtained from the reports of the three BMMS as well as IHME Causes of neonate death came from recent estimates from the WHO/MCEE. |
|
Coverage and equity |
DHS (1993 – 2017) BMMS (2001, 2010, 2016) |
National, subnational, and equity-stratified data were reanalyzed for consistency in coverage indicator definition across surveys and absolute and relative changes were calculated. |
| Hierarchical decomposition |
DHS 1999 & 2017, SPA 1999 & 2017 |
Hierarchical regression of neonatal mortality on proximate, intermediate, and distal determinants; extraction of effects of change in each factor on neonatal mortality change, controlling for upper-level factors. |
| Fertility decomposition |
IGME and MMEIG estimates of maternal and neonatal mortality; UN Population Division’s estimate of population, total births, and crude birth rates |
Use of Jain’s decomposition of maternal lives saved between 2000 and 2017 into components due to decline in birth rates, changes in birth risk composition, and maternal health programs; Adapt the method to decompose neonatal mortality reduction on the same period. |
| Lives Saved Tool |
IGME and MMEIG estimates of maternal and neonatal mortality; DHS and national surveys for coverage estimate; SPA for quality of care |
Lives saved analysis to determine the contribution of coverage change of health interventions to decline in maternal and neonatal mortality between 2000 and 2019. |
| Policy and program |
Country reports; annual statistical yearbook |
Desk review of health systems organization; programs and policies; development of a policy timeline; human resources for health; analysis of qualitative interviews. |
| Qualitative interviews |
Key informant interviews |
In-depth key informant interviews on health system context and organization, policy and program development, success and challenges. |
Abbreviations: ANC, antenatal care; BMMS, Bangladesh Maternal Mortality and Health Care Surveys; DHS, Demographic and Health Surveys; IGME, United Nations Inter-Agency Group for Child Mortality Estimation; IHME, Institute for Health Metrics and Evaluation; MICS, Multiple Indicator Cluster Surveys; MMEIG, United Nations Maternal Mortality Estimation Inter-Agency Group; MNH, maternal and newborn health; SPA, Service Provision Assessment; WHO/MCEE, World Health Organization Maternal Child Epidemiology Estimation Group.