Maternal Mortality in New Jersey: An Interactive Social Determinant of Health Tool
Authors: Ru-fong Cheng, MD, FACOG, Mariana Coelho, Douglas Friedman
While the COVID-19 pandemic has dominated every aspect of our lives these past few months, another public health crisis has been burning in the background. The U.S. is in the middle of a maternal mortality crisis. The maternal mortality rate in the U.S. is the highest of all developed nations and is rising.[i] A woman today is twice as likely to die in childbirth than her mother was a generation ago.[ii] Healthcare professionals, public health officials, and policy makers are all working to understand the drivers of maternal mortality. In addition, there are stark racial disparities. Black women die during pregnancy and childbirth at a rate 3–4 times above the rate of white women.[iii] The root cause for this difference is not well understood, but healthcare inequities, structural racism, social determinants of health and bias all play a role. Combined with recent events highlighting structural racism in the U.S., this crisis has taken on new urgency.
According to the CDC, “700 women die each year in the United States as a result of pregnancy or delivery complications.” [iv] This figure has been on the rise overall since the Pregnancy Mortality Surveillance System was implemented in 1986.[v] However, these figures have been notoriously difficult to calculate as pregnancy-related mortality data has been historically poor and/or inconsistent.[vi]
A team of experts in maternal health and advanced analytics partnered to develop a data visualization tool for maternal health related information in New Jersey, a state with the 4th highest maternal mortality in the nation, with almost double the national maternal mortality rate.
The New Jersey Maternal Mortality Dashboard was built using public data sources, brought together in a unique and interactive way. Data from the state and federal government about population demography and social variables are integrated into a geographic map to provide a comprehensive high-level picture of the relationships of these factors and maternal mortality, at a state and county level. Rather than rely on a single measure of maternal mortality, the application provides multiple measures allowing the user to have a broader view of maternal mortality based on different data points. The application allows users to easily explore maternal mortality rates and ratios by year, considering key demographic groups (e.g. race/ethnicity, age) and social and public health factors such as hospitals, insurance, location of Ob-Gyn healthcare providers, midwives, and cesarean section rate.
There is limited access to maternal mortality figures (much less detailed ones) for the general public. When access to these figures is available, it is often not clear how they are calculated or what data may be involved. This application seeks to fill that gap by providing detailed measures of maternal mortality with full transparency in calculation and the associated data. Additionally, users can quickly see how public health factors, rather than just women specific health issues, play a role in maternal mortality which would otherwise be difficult to align to the same timeframe/population without technical or subject matter expertise. Trends and patterns may be observed which would be helpful to understand for development of public health interventions. Although correlations may be noted, causation cannot be determined by the tool.
Maternal mortality is a complex issue; multiple angles such as policy, economics, education, integrated clinical support including both physical as well as mental health, among others, need to be considered in the architecture of successful solutions. Integrating public feedback and the voice of women in understanding the underlying causes and developing impactful solutions are critical for sustaining change.
The dashboard is built using the R/Shiny web application framework and hosted through the Shinyapps service. We encourage you to explore our code, data, and/or methodology to apply to your own state’s maternal mortality data so you too can be a part of raising awareness about such an important issue.
You can find the Github repository with all of our code and data here — https://github.com/johnsonandjohnson/njmm.
You can find the New Jersey Maternal Mortality Dashboard here — https://johnsonandjohnson.shinyapps.io/njmm/.
[i] Slomski A. Why Do Hundreds of US Women Die Annually in Childbirth? JAMA. 2019;321(13):1239–1241.
[ii] Associated Press. An American mom today is twice as likely to die in childbirth than her own mother was. Available at: https://www.marketwatch.com/story/heres-why-us-pregnancy-deaths-are-three-times-more-likely-among-some-minority-groups-2019-05-09. Retrieved August 28, 2020.
[iii]https://www.cdc.gov/nchs/maternal-mortality/index.htm#:~:text=For%202018%2C%20the%20maternal%20mortality,births%20in%20the%20United%20States. accessed August 28, 2020.
[iv] https://www.cdc.gov/reproductivehealth/maternal-mortality/index.html
[v]https://www.cdc.gov/reproductivehealth/maternal-mortality/pregnancy-mortality-surveillance-system.htm#trends
[vi] https://www.cdc.gov/nchs/data/nvsr/nvsr69/nvsr69_01-508.pdf
