Related Experiment Video
Updated: Aug 24, 2026
A Neonatal Imaging Model of Gram-Negative Bacterial Sepsis
Published on: August 12, 2020
Quantifying uncertainty: physicians' estimates of infection in critically ill neonates and children
Joachim E Fischer1, Stephan Harbarth, Alexander G Agthe
1Children's Hospital, University of Zurich, Zurich, Switzerland. fischer@ifv.gess.ethz.ch
Abstract:
To determine the diagnostic accuracy of physicians' prior probability estimates of serious infection in critically ill neonates and children, we conducted a prospective cohort study in 2 intensive care units. Using available clinical, laboratory, and radiographic information, 27 physicians provided 2567 probability estimates for 347 patients (follow-up rate, 92%). The median probability estimate of infection increased from 0% (i.e., no antibiotic treatment or diagnostic work-up for sepsis), to 2% on the day preceding initiation of antibiotic therapy, to 20% at initiation of antibiotic treatment (P<.001). At initiation of treatment, predictions discriminated well between episodes subsequently classified as proven infection and episodes ultimately judged unlikely to be infection (area under the curve, 0.88). Physicians also showed a good ability to predict blood culture-positive sepsis (area under the curve, 0.77). Treatment and testing thresholds were derived from the provided predictions and treatment rates. Physicians' prognoses regarding the presence of serious infection were remarkably precise. Studies investigating the value of new tests for diagnosis of sepsis should establish that they add incremental value to physicians' judgment.
Insights
Physicians accurately estimated serious infection risk in critically ill children. Their predictions showed high precision, guiding treatment decisions and suggesting new diagnostic tests should prove superior to clinical judgment.
Area of Science:
- Pediatric Critical Care Medicine
- Infectious Disease Diagnostics
- Clinical Decision Making
Background:
- Accurate diagnosis of serious infection in critically ill neonates and children is crucial for timely and appropriate treatment.
- Physicians' subjective probability estimates play a significant role in clinical decision-making for sepsis management.
- The diagnostic accuracy of these estimates in pediatric intensive care settings requires thorough evaluation.
Purpose of the Study:
- To assess the diagnostic accuracy of physicians' prior probability estimates for serious infections in critically ill pediatric patients.
- To evaluate how these probability estimates evolve over time relative to the initiation of antibiotic therapy.
- To determine the predictive performance of physicians' judgments in identifying serious infections, including blood culture-positive sepsis.
Main Methods:
- A prospective cohort study was conducted in two pediatric intensive care units.
- Twenty-seven physicians provided 2567 probability estimates for 347 critically ill patients based on available clinical, laboratory, and radiographic data.
- The study analyzed the relationship between probability estimates, treatment initiation, and subsequent infection classification, calculating discrimination metrics like the area under the curve (AUC).
Main Results:
- The median probability estimate of infection significantly increased from 0% to 20% from before to the initiation of antibiotic treatment (P<.001).
- Physicians' predictions demonstrated strong discrimination between proven infections and unlikely cases (AUC = 0.88) at treatment initiation.
- The ability to predict blood culture-positive sepsis was also good (AUC = 0.77).
Conclusions:
- Physicians' prognoses regarding the presence of serious infection in critically ill children were remarkably precise.
- The findings highlight the substantial diagnostic value of experienced clinical judgment in managing potential sepsis.
- Future research on new sepsis diagnostic tests should aim to demonstrate added incremental value beyond existing physician expertise.