Physician decision-making--evaluation of data used in a computerized ICU

Area of Science:

• Medical Informatics
• Critical Care Medicine
• Health Information Systems

Background:

• Modern intensive care units (ICUs) generate vast amounts of patient data due to advanced instrumentation and computing.
• This data deluge poses a risk of information overload for clinicians, potentially hindering effective decision-making.
• A computerized system was implemented in the Shock-Trauma ICU at LDS Hospital to manage integrated patient data.

Purpose of the Study:

• To evaluate physician data usage in decision-making within a highly computerized Intensive Care Unit (ICU).
• To compare the proportion of data categories in the computerized database with their actual use during clinical decision-making.
• To identify areas for optimization in patient data collection and storage.

Main Methods:

• Physician data usage was assessed in two decision-making environments: ICU rounds and on-site.
• Patient data was categorized into six types: bedside monitor, laboratory, drugs, input/output and IV, blood gas laboratory, and observations.
• The percentage of database occupied by each data category was compared with its frequency of use in clinical decisions.

Main Results:

• Combined laboratory data constituted 38-41% of reviewed data but only 16.3% of the database.
• Observations represented 21-22% of reviewed data and 6.8% of the database.
• Drugs, input/output, and IV data (13-23% usage) occupied 36% of the database, while bedside monitor data (12.5-22% usage) occupied 32.5%.

Conclusions:

• Significant discrepancies exist between patient data collection/storage and actual clinical utility in the ICU.
• Optimization of data collection and storage is necessary to mitigate information overload.
• Implementing tools like the computerized ICU Rounds Report can improve data presentation and support physician decision-making.