AI recommends more expensive treatment than physicians, analysis suggests (www.bmj.com)

A recent analysis published in the BMJ reveals that AI systems propose medical tests and treatment plans that can be over three times more expensive than those recommended by physicians. The study conducted by researchers from Israel and the US examined 24 different AI systems using 200 real-world outpatient and emergency department clinical notes sourced from MTSamples, a public collection of deidentified clinical documentation. The evaluation highlighted significant discrepancies in cost between AI recommendations and traditional physician recommendations. This finding is significant for the AI/ML community as it raises questions about the decision-making processes of AI in healthcare and its implications for cost-effectiveness and patient care. The results suggest that while AI can enhance clinical decision-making, there is a risk of prioritizing more expensive interventions, which could strain healthcare resources. Understanding the factors that drive these cost differences will be crucial for integrating AI tools into clinical settings while ensuring that they support economically sustainable practices in patient care.