Ai advice can improve sepsis treatment decisions with human judgment
Indirect Influences of Artificial Intelligence Advice on Complex Sepsis Treatment Decisions
Human-Computer InteractionArtificial Intelligence
Summary
Sepsis is a dangerous condition that is hard to treat because decisions about care are complicated and uncertain. The authors looked at how an AI system’s advice affected doctors’ choices for treating sepsis. They found that doctors sometimes made better decisions when they used AI advice but still relied on their own judgment, especially in complex cases where the AI was less accurate. This suggests that AI can help doctors think through tough problems, not just give them direct answers.
What this means in practice
- •For critical care physicians: Use AI recommendations to prompt reevaluation of treatment decisions in complex sepsis cases, improving patient care outcomes.
- •For hospital quality improvement teams: Develop training programs that teach clinicians how to interact with AI advice to enhance decision quality in complex treatments.
Authors
Venkatesh Sivaraman, Joel Levin, Eric P. Mason, Victor Talisa, Suchi Saria, Jean Feng, Andrew J. King, Jeremy M. Kahn, Adam Perer
Abstract
Sepsis is a major public health concern, and many sepsis patients die due to ineffective hemodynamic management. Artificial intelligence could support hemodynamic decisions, particularly in complex cases where conflicting information and comorbidities drive uncertainty and practice variation. However, these systems' real-world value depends on not just accuracy, but how they influence the quality of clinician decisions. We first retrospectively evaluated an AI system's recommendations in 6,928 sepsis patients, finding that they improved upon the original providers' decisions overall but had significantly worse performance in complex cases. We then conducted a pre-registered experiment with 228 US critical care physicians to test their use of this imperfect advice in six complex treatment decisions, using one of three advice types designed to vary reasoning autonomy. Participants given simple model recommendations overcame poor advice, choosing optimal treatments in 56.3% of decisions compared to 31.8% if they had always followed the recommendations. They also made significantly better decisions than those receiving a less prescriptive advice type (42.9%). Physicians selectively adapted and improved weak advice, likely because it prompted them to reconsider their assessment. Our findings suggest that improving sepsis care decisions with AI may require accounting for these indirect influences resulting from human-AI collaborative reasoning.