legrandangle-lgm

Researchers from Loughborough University have developed an AI tool designed to increase safety in maternity care by identifying the human factors that can influence outcomes.

The I-SIRch solution, developed by Professor Georgina Cosma and Professor Patrick Waterson, analyzes pregnancy incident reports to identify human factors such as communication, teamwork and decision-making, which could benefit from additional support.

Findings from the analysis of 188 real reports of pregnancy incidents, on which the tool has been trained, have been published in the International Journal of Population Data Science on November 20, 2024.

It is hoped that the tool will help inform targeted maternity care interventions that will improve safety for mothers and babies the 2022 Ockenden Review into NHS maternity care, which found that between 2000 and 2019, more than 200 babies died after ‘repeated failures’ in care.

Patrick Waterson, Professor of Human Factors and Complex Systems at Loughborough University, said: “Our work opens new possibilities for understanding the complex interplay between social, technical and organizational factors that influence maternal safety and population health outcomes.

“The need for such research was emphasized in the Ockenden reviewin which maternity care was investigated and safety and quality of care in maternity care was improved.

“By gaining a more comprehensive picture of maternal health care, we can develop targeted interventions to improve maternal outcomes for all mothers and babies.”

The I-SIRch project was jointly funded by the Health Foundation and the NHS Transformation Directorate’s NHS AI Lab, and supported by the National Institute for Health and Care Research.

When adverse incidents occur in maternity care, investigations are conducted that involve time-consuming and labor-intensive manual reviews to gain insights into human factors that may have influenced outcomes, but which depend on individual interpretation and expertise, which can lead to different conclusions.

I-SIRch automates and standardizes this process, analyzing multiple reports to identify recurring factors and pinpoint areas that would most benefit from additional support.

In tests, it successfully identified human factors in each report and provided accurate insights into where additional support could improve results, the researchers said.

Prof. Cosma and Prof. Waterson are now seeking funding to refine I-SIRch using a larger, more diverse data set, allowing them to validate the effectiveness of the instrument and ensure it can address the challenges faced by mothers from ethnic minority groups in maternity care face.

“We are looking to work with hospitals, healthcare organizations and research bodies to further refine our AI tool and apply it to reports.

“These partnerships will help us gather vital information 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 adverse police incident reports, where understanding the human factors involved could help prevent future incidents and improve response strategies,” said Prof. Cosma.

Dr. Jonathan Back, safety insights analyst at the Health Services Safety Investigations Body, said the AI ​​tool “can help analysts working in health and care to identify where inequalities exist, maximizing learning by combining findings from multiple studies to bring”.

Meanwhile, a review of more than 12,000 papers and 87 articles from the University of Birmingham published in October 2024 found that AI software improved the chances that women would have a good pregnancy care with 69%.