Maternal mortality estimation methodologies: a scoping review and evaluation of suitability for use in humanitarian settings

Around the world, a maternal death occurs approximately every two minutes-most of these deaths are preventable. The maternal mortality ratio is a key indicator for the Sustainable Development Goals, yet we have no reliable way to estimate maternal deaths in refugee or internally displaced persons (IDP) camps and settlements. The goal of this study was to understand the methodologies most suited for adaptation for use to estimate the proportion of maternal mortality due to abortion complications in these settings. We conducted a scoping review of methodologies to estimate maternal mortality and evaluated them using a predetermined set of criteria. We evaluated nine original methodologies using eleven categories related to implementation in refugee or IDP camps and settlements: data sources, definitions, sample size, timing of point estimate relative to data collection, bias, human resources, time needed for implementation, data collection training, statistical training, digitalization, and cost. Each category could be assigned zero to four points, for a total score of 44 points. After evaluating each original methodology, we reviewed the original publication's citations or searched for other implementations through October 2022. We revised the original scores and developed a rank-order list of the methodologies according to their suitability for implementation in refugee and IDP camps. We identified 124 publications that estimated maternal mortality. The Maternal Deaths from Informants/Maternal Death Follow on Review (MADE-IN/MADE-FOR) (33.5), hospital- or facility-based (33.5), and community informant-based (32.5) methodologies ranked highest due to low costs, short time interval needed for implementation, small sample sizes and close timing of point estimate relative to data collection, easy digitalization, and the need for no statistical training. Similar to the lack of a "perfect" methodology to estimate maternal mortality in stable settings, there are compromises to consider when applying these methodologies to humanitarian settings. The most promising methodologies are adaptable to practical constraints in refugee and IDP camps and settlements. New methodologies that adapt and strengthen the MADE-IN/MADE-FOR, hospital- or facility-based, and community informant-based methodology show promise and must be further developed.