Arrow
use cases

Forecasting Maternal Mortality and Birth Outcomes: Enhancing Maternal Health

Detect patients that will have increase rates of mortality and negative obstetric outcomes using clinical variables.

Problem

The 2020 U.S. maternal mortality rate ranked last among all industrialized countries, with 17.4 deaths per 100,000 pregnancies (2). For every maternal death measured, 100 women also experience severe obstetric morbidity, resulting in significant health consequences (1). More than 60,000 women suffer from severe maternal morbidity annually, and the rate of maternal morbidity increased by 36% from 2008 to 2018 (1)(3). Severe maternal morbidity is associated with a 111% increase in maternity-related costs for commercially-insured populations and 175% for those covered by Medicaid (4). In 2019, there were 3.75 million births, with Medicaid paying for approximately 50% of them (5)(6). Maternal health data reveals immense racial and ethnic disparities. Black women are three to four times more likely to die from pregnancy-related causes compared to White women, and up to 12 times more likely in some cities (1).

Non-Hispanic Black women also have the highest rates for most severe morbidity indicators and are more likely to suffer from pregnancy-induced and chronic conditions (1). Reducing these disparities is crucial for improving obstetric outcomes. Studies indicate that 46% of maternal deaths among Black women are potentially preventable, compared to 33% among White women (1). Additionally, socioeconomic factors, such as hospital quality, play a significant role in these disparities. Approximately 75% of Black deliveries occur in a quarter of U.S. hospitals, which have higher risk-adjusted maternal morbidity rates, compared to 18% of White deliveries in the same hospitals (1). Black and Hispanic women are nearly three times as likely to report concerns about their treatment due to race, ethnicity, and cultural background (7)(8).

Why it matters

  • More than 60,000 women suffer annually in the U.S.
  • Black women are 3-4 times more likely to die from pregnancy-related causes than White women.
  • 46% of maternal deaths among Black women are potentially preventable.
  • Severe maternal morbidity increases maternity-related costs by 111% for commercially-insured.
  • Medicaid funded about 50% of U.S. births in 2019.

Solution

To address this issue, a predictive AI model has been designed to assess maternal and obstetric risks. By evaluating clinical data, including medical history and current health status, the model equips healthcare providers with actionable information to provide preventative and personalized care to at-risk pregnant women.

Discover more and interact with our AI!

Datasources

The model uses the Maternal Health Risks dataset from the UCI Machine Learning Repository (9). This data set, which includes information on systolic and diastolic blood pressure, blood sugar, body temperature, heart rate and age, was collected through an IoT-based risk monitoring system in several hospitals and community clinics in Bangladesh. The data presents six significant and relevant risk factors for maternal mortality, addressing a key concern of the United Nations Sustainable Development Goals.

Citations

  1. Howell, Elizabeth. “Reducing Disparities in Severe Maternal Morbidity and Mortality”” Clinical Obstetrics and Gynecology, vol. 61, no. 2, Jan. 2018, pp. 387-399, doi: 10.1097/GRF.0000000000000349. Accessed 22 Oct.
  2. Declerq, Eugene, Zephyrin, Laurie. “Maternal Mortality in the United States: A Primer.” The Commonwealth Fund, 16 Dec. 2020, https://doi.org/10.26099/ta1g-mw24. Accessed 22 Oct. 2021.
  3. Premier. “Bundle of Joy: Maternal 8 Infant Health Trends” Premier, 2020, pp. 1-14, https://explore.premierinc.com/MaternalHealth Trends/landing-page-copy-67V7-8638Hhtml?. Accessed 22 Oct. 2021.
  4. Black M., Christopher, Vesco K. Kimberly, et al. “Costs of Severe Maternal Morbidity in U.S. Commercially Insured and Medicaid Populations: An Updated Analysis.” Women's Health Reports, vol. 2, no. 1, 27 Sep. 2021, pp. 443-451. Accessed 20 Oct. 2021.
  5. Alliman, Jill, et al. “Strong Start in Birth Centers: Socio-Demographic Characteristics, Care Processes, and Outcomes for Mothers and Newborns.” Birth, vol. 46, no. 2, 17 May 2019, pp. 234-243, 10.1111/birt.12433. Accessed 20 Oct. 2021.
  6. CDC. “National Vital Statistics System.” Centers for Disease Control and Prevention, last reviewed 27 Sep. 2021. https://www.ede.gov/nchs/nvss/births.htm. Accessed 22 Oct. 2021.
  7. Declerq, Eugene, Sakala, Carol, et al. “Listening to Mothers lll Pregnancy and Birth.” Childbirth Connection, May 2013, pp. 1-75. https://www.nationalpartnership.org/our-work/resources/health-care/maternity/listening-to-mothers-iAccessed 20 Oct. 2021.
  8. Alliman, Jill, Bauer, Kate. “Next Steps for Transforming Maternity Care: What Strong Start Birth Center Outcomes Tell Us Journal of Midwifery 8 Women's Health, vol. 65, no. 4, 11 Apr. 2020, pp. 462-465, doi:10.1111/jmwh.13084.
  9. "Maternal Health Risk - UCI Machine Learning Repository," https://archive.ics.uci.edu/ml/datasets/Maternal+Health+Risk.

Book a Free Consultation

Trusted by the world's top healthcare institutions