A survey conducted by the Ann and Robert H. Lurie Children's Hospital of Chicago found more than 75% of parents are generally receptive to the use of artificial intelligence (AI) tools in the management of children with respiratory illnesses in the emergency department (ED). However, some demographic subgroups, including non-Hispanic black and younger age parents, had greater reservations about the use of these technologies.
These findings point to the importance of involving a broad representation of parents from the earliest stages of development of AI systems for pediatric healthcare. The study was published in the journal Academic Pediatrics.[1]
“Our results suggest that development of AI tools to improve the care of children in an acute care setting needs to involve a diverse set of patient and parent stakeholders early on in the process to ensure that they are comfortable with the technology and that the new tools do not contain unintentional bias,” said lead author Sriram Ramgopal, MD, pediatric emergency medicine physician at Lurie Children’s and assistant professor of pediatrics at Northwestern University Feinberg School of Medicine, in a release about the study.
Currently, AI is rarely used in pediatric acute care settings, but there is an increasing research interest in its potential. Family engagement is seen as critical for the implementation of AI-based clinical decision support tools in pediatrics, which the authors of this study say will play an increasing role in healthcare in the near future.
“In pediatric research, an increasing number of AI or machine learning-based models have been reported,” Ramgopal said. “These models promise to provide greater diagnostic accuracy, identify patients at risk of severe outcomes, or detect patients in need of diagnostic testing or treatment.”
The surveys were completed by 1,620 parents. Most respondents said they were comfortable with the use of computer programs to determine the need for antibiotics (77%) or bloodwork (76%), and to interpret X-ray radiographs (77%). The greatest perceived benefits of computer programs were finding something a human would miss and obtaining a more rapid diagnosis. Areas of greatest concern were diagnostic errors and recommending incorrect treatment.
“It is inevitable that AI will make it into routine pediatric practice,” Ramgopal explained. “In the ED, we already use computer-based decision supports systems, which are precursors to AI. For example, we use a sepsis prediction model that alerts physicians to patients with higher risk. As with all such tools, these systems don’t dictate a particular course of action, but rather inform a physician’s approach to care in situations where a human might easily miss an important pattern in how illness presents itself. Our goal is to provide the most accurate and timely management for children in our care.”
Related AI content:
VIDEO: Getting cardiologist buy-in on artificial intelligence
How machine learning models can predict if patients will show up for appointments
AI-based system affords physicians more flexible schedules, minimizes burnout triggers
AI-based mammo screening protocol reduces radiologist workload by 62%
Explainable AI model accurately auto-labels chest X-rays from open access datasets
6 features of the ideal healthcare artificial intelligence algorithm
Reference:
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: dfornell@innovatehealthcare.com