Toward a human-centered hyperlipidemia management system: the interaction between internal and external information...

Area of Science:

• Human-Computer Interaction
• Information Science
• Health Informatics

Background:

• User search performance is influenced by data representation and cognitive distribution.
• Existing frameworks lack specific guidance for designing relational data search interfaces in healthcare.

Purpose of the Study:

• To propose a human-centered search model (UFuRT) and task taxonomy for relational data search.
• To investigate the impact of data representation and task type on search performance in a healthcare context.
• To provide design guidelines for effective search interfaces in electronic medical records.

Main Methods:

• Developed a human-centered framework (User, Function, Representation, Task).
• Created a taxonomy for relational data search tasks.
• Designed and tested experimental prototypes of hyperlipidemia data displays.
• Conducted two within-subject experiments with 24 participants.

Main Results:

• Search performance (response time, accuracy) is significantly affected by representation dimensions, data scales, and search task types.
• Increased external representations on the interface correlate with better user search task performance.
• Optimal search efficiency is achieved when question types align with their corresponding data scale representations.

Conclusions:

• The proposed model and taxonomy effectively predict search performance.
• Interface design elements, particularly data representation and matching question types to data scales, are crucial for optimizing search efficiency.
• Findings inform the design of more effective search interfaces for laboratory results within electronic medical records.