Certainty and mortality prediction in critically ill children.

Area of Science:

• Pediatric Critical Care Medicine
• Clinical Decision-Making
• Medical Prognostication

Background:

• Physician's subjective mortality predictions are crucial for patient management in pediatric intensive care.
• Understanding the interplay between prediction accuracy and confidence is vital for refining prognostic assessments.

Purpose of the Study:

• To investigate the relationship between physician's subjective mortality predictions and their confidence levels in pediatric intensive care unit (ICU) patients.
• To determine if confidence in prognostication varies with the predicted mortality risk.

Main Methods:

• Prospective cohort study of pediatric ICU patients (<18 years).
• Physicians provided mortality risk predictions and confidence levels within 24 hours of admission.
• Median confidence levels were compared across different mortality risk prediction ranges.

Main Results:

• Physicians expressed significantly less confidence in mortality predictions between 5% and 95%.
• Predictions <5% and >95% were associated with higher confidence levels.
• Physician experience correlated positively with increased confidence in prognostication.

Conclusions:

• Physician confidence significantly impacts subjective mortality predictions, with lower confidence leading to intermediate risk assessments.
• Assessing physician confidence alongside mortality predictions is essential for understanding potential biases in therapeutic planning.
• Confidence levels in prognostic assessments may influence clinical decision-making and future patient outlooks.