Maternal mortality ratios in low- and middle-income countries: a comparison of estimation methods and relationships with sociodemographic covariates

Abstract

BACKGROUND

Maternal mortality is most prevalent in low- and middle-income countries, especially those from sub-Saharan Africa and South Asia. The Sustainable Development Goal 3.1 aims to reduce global maternal mortality by 2030 to <70 per 100,000 live births globally and <140 per 100,000 live births at the national level. For maternal mortality ratio estimations, the World Health Organization recommends a census in low- and middle-income countries that lack civil registration and vital statistics; however, other methods have also been used.

OBJECTIVE

This study aimed to compare maternal mortality ratio estimating methods and maternal mortality ratio trends over time. Associations between sociodemographic variables in low- and middle-income countries and maternal mortality ratios are described and compared between countries projected to meet or fall short of Sustainable Development Goal 3.1.

STUDY DESIGN

Publications from the World Health Organization and the Maternal Mortality Estimation Inter-Agency Group were used to identify countries that reported maternal mortality ratio estimates at least twice since 2004 using census, noncensus, or both approaches. Maternal mortality ratios were extracted from the Maternal Mortality Estimation Inter-Agency Group, and covariates associated with maternal mortality ratios were obtained from the Our World in Data and the Fragile States Index web pages. Group comparisons were performed using paired t tests, and correlations between variations among maternal mortality ratio estimates and population demographic covariates were analyzed using linear mixed-effect models. Projected maternal mortality ratio estimates for 2030 were calculated using the exponential growth/decay method used by the World Health Organization.

RESULTS

Data were available for 45 countries for comparison; 21 countries had data from different maternal mortality ratio estimation methods, and 42 countries reported maternal mortality ratios using the same estimation method over time. Census maternal mortality ratio estimates were 83.2 per 100,000 live births higher than the estimates from noncensus methods, although this difference was statistically nonsignificant (P=.19). Of the 45 countries assessed, 30.1% were projected to meet the Sustainable Development Goal 3.1 maternal mortality ratio target of <140 per 100,000 births by 2030. National maternal mortality ratio estimates were significantly influenced by total fertility rate, skilled birth attendance rate, gross domestic product per capita, female and male literacy rates, female rate of access to modern contraceptives, and the Fragile States Index.

CONCLUSION

Maternal mortality ratio estimates are reproducible using different estimation methods in low- and middle-income countries. Only 30% of the low- and middle-income countries for which serial data are available are projected to meet the Sustainable Development Goal 3.1. Maternal mortality ratios are significantly lower in countries with sociodemographic characteristics that indicate a higher status for women.