Redundancy within hospital-related harm prediction tools

Area of Science:

• Healthcare Quality and Safety
• Clinical Informatics
• Patient Outcomes

Background:

• Inpatient harm screening tools are used to identify at-risk patients.
• The predictive accuracy and efficiency of these tools require further investigation.
• Current tools may have limitations in identifying diverse patient harms.

Purpose of the Study:

• To evaluate the predictive ability and efficiency of inpatient harm screening tools.
• To identify factors contributing to the performance of these screening instruments.
• To inform the development of more effective patient safety strategies.

Main Methods:

• Retrospective analysis of approximately 25,000 patient admissions from 2019.
• Assessment of the discriminatory ability of existing harm screening tools.
• Examination of question overlap and their impact on predicted harm.

Main Results:

• The overall discriminatory ability of the evaluated harm screening tools was found to be moderate.
• A significant portion of the predictive power could be attributed to one or two specific questions.
• These key questions demonstrated overlap in the types of harm they predicted.

Conclusions:

• Current inpatient harm screening tools exhibit moderate predictive capabilities.
• The effectiveness of these tools may be influenced by redundant or overlapping questions.
• Optimizing screening tools requires a focus on distinct and high-yield predictive elements for improved patient safety.