A study from the University of Minnesota’s School of Public Health has revealed widespread integration of artificial intelligence (AI) in U.S. hospitals for routine operations, yet inconsistent bias evaluations may undermine fairness in patient care.
Surveying 2,425 hospitals via the 2023 American Hospital Association Annual Survey, the research identified that 65% of hospitals utilize AI tools for forecasting patient health outcomes, pinpointing at-risk individuals, and optimizing scheduling systems.
However, while 61% of these facilities verify their predictive models for accuracy, only 44% assess them for potential bias. This discrepancy poses challenges for equitable healthcare delivery.
Well-funded hospitals have an advantage, with greater technical expertise allowing them to customize AI models for their unique patient demographics. Conversely, underfunded hospitals often purchase standardized AI solutions, which may fail to address the specific needs of their communities.
“The widening digital gap among healthcare facilities threatens patient safety and fair treatment,” emphasized Paige Nong, assistant professor and lead researcher.
To address these issues, future investigations will delve into the prevalence of advanced AI applications, including chatbots and ambient scribes, while collaborating with diverse organizations to guide policy and practical reforms.
The research received funding from the Office of the Assistant Secretary for Technology Policy under the U.S. Department of Health and Human Services.
AI is reshaping India’s workforce, empowering millions
St Aloysius University to host international conference on sustainable development
#HealthcareInnovation #EquityInAI #DigitalDivide #HospitalTechnology
