Variations in faculty assessment of NICU flowsheet data: implications for electronic data display

Area of Science:

• Medical Informatics
• Clinical Decision Making
• Neonatology

Background:

• Physician interpretation of patient data is crucial for clinical decision-making.
• Effective presentation of patient data impacts diagnostic reasoning and resident education.
• Flowsheets in the neonatal intensive care unit (NICU) are complex data visualization tools.

Purpose of the Study:

• To investigate how physicians reason through patient data presented on NICU flowsheets.
• To understand the process of constructing clinically meaningful information from raw patient data.

Main Methods:

• A cohort of 20 neonatologists participated in the study.
• Participants diagnosed 5 hypothetical NICU case scenarios using standardized flowsheets.
• Diagnostic choices and the reasoning behind them were recorded.

Main Results:

• High agreement was observed among neonatologists regarding diagnostic choices.
• There was minimal consensus on which specific data points most influenced their diagnostic reasoning.
• On average, only 6% of data items per case achieved 75% agreement among participants.

Conclusions:

• Neonatologists frequently agree on diagnoses but diverge significantly on the key data elements driving those conclusions.
• Further research is needed to elucidate how physicians prioritize information from flowsheets.
• Optimizing patient data presentation in the NICU may improve clinical reasoning and data utilization.