Researchers at Loughborough University have unveiled a groundbreaking artificial intelligence (AI) tool designed to identify critical human factors affecting maternity care outcomes, aiming to enhance safety for mothers and their babies.
Developed by Professor Georgina Cosma, an AI and data scientist, and Professor Patrick Waterson, an expert in human factors and complex systems, the innovative tool analyzes maternity incident reports to pinpoint key elements—such as communication, teamwork, and decision-making—that may have influenced care outcomes. This analysis provides valuable insights into areas requiring additional support.
In England, when adverse maternity incidents occur, detailed investigation reports are generated to uncover opportunities for learning and improving safety. While these reports offer insights into clinical factors such as health conditions and medical procedures, identifying the human factors involved can be more challenging due to their complexity.
Traditionally, experts have relied on manual reviews of incident reports to extract insights related to human factors. This labor-intensive process can be time-consuming and subjective, often leading to inconsistent conclusions. The new AI tool addresses these challenges by swiftly and consistently categorizing human factors within reports. Its standardized methodology enables the analysis of multiple reports, revealing recurring issues that may benefit from targeted intervention.
The AI model was trained and validated using data from 188 real maternity incident reports, successfully identifying and analyzing human factors across all cases. “AI has transformed our analysis of maternity safety reports,” said Professor Cosma. “We’ve uncovered crucial insights far quicker than manual methods, allowing us to gain a comprehensive understanding of areas needing improvement in maternity care, ultimately enhancing patient safety and outcomes for mothers and babies.”
The analysis revealed that teamwork and communication were the most frequently identified human factors, highlighting the critical role of effective collaboration and clear communication among healthcare professionals and patients in ensuring safety and quality in maternity care.
Additionally, the research underscored the importance of comprehensive patient evaluations, including thorough assessments and screenings, as well as the impact of individual patient characteristics—such as birth history and conditions like pre-eclampsia—on care outcomes.
The AI tool also identified challenges related to the use of medical technology and staff performance, suggesting that ongoing training and support could further improve care outcomes. The analysis noted the effects of the COVID-19 pandemic on maternity services, emphasizing the need for adaptability in healthcare practices.
Furthermore, the tool indicated that certain human factors may disproportionately affect mothers from ethnic minority groups, although the limited availability of ethnicity data in the reports necessitates further research to draw definitive conclusions.
The Loughborough research team aims to secure funding to refine the AI model using a larger and more diverse dataset, as expanded testing is crucial for validating the tool’s effectiveness and understanding the challenges faced by mothers from ethnic minority backgrounds.
“We are seeking collaboration with hospitals, healthcare organizations, and investigative bodies to further refine and apply our AI tool to reports,” stated Professor Cosma. “These partnerships will help us extract vital intelligence to prevent adverse incidents and ensure the safety of all mothers and babies. We also hope to adapt the tool for use with other types of reports, such as those related to police incidents, to enhance understanding of human factors and improve response strategies.”
Professor Waterson emphasized the importance of the research for improving maternity care: “Our work opens up new possibilities for understanding the complex interplay between social, technical, and organizational factors influencing maternal safety and health outcomes. The need for such research was underscored in the Ockenden Review, which aimed to enhance safety and care quality in maternity services.”
Dr. Jonathan Back, a safety insights analyst at the Health Services Safety Investigations Body (HSSIB), remarked that this research “could help analysts in health and care identify inequalities and maximize learning by synthesizing findings from multiple investigations.”
The findings from the analysis of the 188 reports, along with information on the AI tool, have been published in the International Journal of Population Data Science.
Related Topics:
-
USF Study Tracks The Long-Term Impact Of Digital Media On Youth Well-Being
-
Pregnancy Antiepileptic Drugs Are Associated With Neurodevelopmental Risks In Children
