Navigating geographical disparities: access to obstetric hospitals in maternity care deserts and across the United States.

BACKGROUND: Access to maternity care in the U.S. remains inequitable, impacting over two million women in maternity care "deserts." Living in these areas, exacerbated by hospital closures and workforce shortages, heightens the risks of pregnancy-related complications, particularly in rural regions. This study investigates travel distances and time to obstetric hospitals, emphasizing disparities faced by those in maternity care deserts and rural areas, while also exploring variances across races and ethnicities. METHODS: The research adopted a retrospective secondary data analysis, utilizing the American Hospital Association and Centers for Medicaid and Medicare Provider of Services Files to classify obstetric hospitals. The study population included census tract estimates of birthing individuals sourced from the U.S. Census Bureau's 2017-2021 American Community Survey. Using ArcGIS Pro Network Analyst, drive time and distance calculations to the nearest obstetric hospital were conducted. Furthermore, Hot Spot Analysis was employed to identify areas displaying significant spatial clusters of high and low travel distances. RESULTS: The mean travel distance and time to the nearest obstetric facility was 8.3 miles and 14.1 minutes. The mean travel distance for maternity care deserts and rural counties was 28.1 and 17.3 miles, respectively. While birthing people living in rural maternity care deserts had the highest average travel distance overall (33.4 miles), those living in urban maternity care deserts also experienced inequities in travel distance (25.0 miles). States with hotspots indicating significantly higher travel distances included: Montana, North Dakota, South Dakota, and Nebraska. Census tracts where the predominant race is American Indian/Alaska Native (AIAN) had the highest travel distance and time compared to those of all other predominant races/ethnicities. CONCLUSIONS: Our study revealed significant disparities in obstetric hospital access, especially affecting birthing individuals in maternity care deserts, rural counties, and communities predominantly composed of AIAN individuals, resulting in extended travel distances and times. To rectify these inequities, sustained investment in the obstetric workforce and implementation of innovative programs are imperative, specifically targeting improved access in maternity care deserts as a priority area within healthcare policy and practice.

Deep Neural Networks for Fine-Grained Surveillance of Overdose Mortality.

Surveillance of drug overdose deaths relies on death certificates for identification of the substances that caused death. Drugs and drug classes can be identified through the International Classification of Diseases, Tenth Revision (ICD-10), codes present on death certificates. However, ICD-10 codes do not always provide high levels of specificity in drug identification. To achieve more fine-grained identification of substances on death certificate, the free-text cause-of-death section, completed by the medical certifier, must be analyzed. Current methods for analyzing free-text death certificates rely solely on lookup tables for identifying specific substances, which must be frequently updated and maintained. To improve identification of drugs on death certificates, a deep-learning named-entity recognition model was developed, utilizing data from the Kentucky Drug Overdose Fatality Surveillance System (2014-2019), which achieved an F1-score of 99.13%. This model can identify new drug misspellings and novel substances that are not present on current surveillance lookup tables, enhancing the surveillance of drug overdose deaths.