Computational method of identifying medical complications based on Hospital Information System data

Area of Science:

• Medical Informatics
• Radiology
• Health Services Research

Background:

• Medical complications following diagnostic procedures pose risks to patient safety.
• Accurate identification of adverse events is crucial for quality improvement in healthcare.
• Hospital Information Systems (HIS) contain vast amounts of data that can be leveraged for patient safety research.

Purpose of the Study:

• To develop and validate an objective method for identifying medical complication cases using only HIS data.
• To assess the feasibility of using routinely collected HIS data for adverse event detection.
• To specifically identify cases of shock associated with contrast media administration during CT examinations.

Main Methods:

• An objective rule-based system was designed to identify specific types of medical complications.
• Hospital Information System (HIS) data from The University of Tokyo Hospital was prepared and utilized.
• The developed identifying rule was applied to one year of HIS data (2003/4/1 to 2004/3/31).

Main Results:

• The study successfully identified three cases of shock.
• These shock cases were specifically linked to the intravenous administration of contrast media during CT examinations.
• The method demonstrated the ability to detect adverse events using solely HIS data.

Conclusions:

• An objective, data-driven method for identifying medical complications from HIS data is feasible.
• This approach can aid in the proactive detection of adverse events, such as shock from contrast media.
• Leveraging HIS data offers a valuable tool for enhancing patient safety and quality of care in hospitals.