US national, state and county-level trends in fertility and maternal mortality from 1980-2014

Abstract

Importance: Tracking subnational variability in fertility rates, quantifying inequality in maternal mortality risk, and understanding important drivers behind increases in maternal deaths is essential in guiding state- and county-level decision makers to target policies that improve reproductive and maternal health in the United States. Objective: To estimate fertility rates and maternal mortality ratios for all US counties between 1980 and 2014 and decompose changes in the number of maternal deaths due to the differences in population size, population age structure, fertility rates, and maternal mortality ratios between 1980 and 2014. Design, Setting, and Participants: Using deidentified birth records from the National Center for Health Statistics (NCHS) and population counts from the US Census Bureau, NCHS and Human Mortality Database, age-specific fertility rates were estimated with previously validated small area estimation models. Similar methods were previously used to model the maternal mortality rate which was combined with the estimated fertility rates to calculate county-level maternal mortality ratios. Das Gupta decomposition methods were used to decompose changes in the number of maternal deaths due to differences in multiple effects. Exposures: County of residence. Main Outcomes and Measures: Age-specific and total fertility rates, and age-specific, all-ages, and maternal age-standardized maternal mortality ratios (MMR). Results: The mean age at childbirth has increased in every single county in the US from 1980 to 2014 but there is still substantial variation between counties with a gap of 7.85 years between counties with the highest and lowest mean age at childbirth in 2014. US maternal mortality ratio has increased nationally from 14.76 maternal deaths per 100,000 live births (95% UI, 13.56-16.07) in 1980 to 35.3 maternal deaths per 100,000 live births (95% UI, 33.42-37.17) in 2014 and geographic inequality between counties has increased. The increasingly older fertility age distribution and increases in MMR in the 20 to 34 age group were primarily responsible for the increase in maternal deaths from 1980 to 2002 across US counties. From 2002 to 2014, increases in maternal deaths were primarily due to increases in MMR across all ages. Conclusions and Relevance: Across US counties women are having children at older ages which has partly driven the increase in the number of maternal deaths from 1980 to 2014. But the main contributor to the large increases seen in maternal deaths since 2002 can be attributed to increases in MMR. More research is needed to understand how much of this increase in MMR throughout the US is due to improvements in identifying maternal deaths because of changes to the US standard death certificates versus actual increases in MMR. Regardless, the US MMR remains well above other high-income countries and there are large inequalities between counties.