Examining the Impact of Interface Design and Nurse Characteristics on Satisfaction With Machine Learning Decision...

Area of Science:

• Clinical Informatics
• Human-Computer Interaction
• Artificial Intelligence in Healthcare

Background:

• Machine learning (ML) offers potential for clinical decision support (CDS) systems.
• Explanatory information is crucial for clinician trust and adoption of ML-CDS.
• Limited research exists on nurse satisfaction with ML explanatory information displays.

Purpose of the Study:

• To describe nurse satisfaction with ML-CDS explanatory information displays.
• To examine the impact of display design (format, complexity) on satisfaction.
• To investigate how nurse characteristics (age, numeracy, AI training) influence satisfaction.

Main Methods:

• Survey assessing nurse satisfaction with ML-CDS explanatory information.
• Analysis of associations between display characteristics and satisfaction.
• Investigation of nurse demographics and literacy on satisfaction.

Main Results:

• Local feature-based explanations may not meet nurse information needs.
• Nurse age, AI training level, and numeracy significantly influenced satisfaction.
• Display format and complexity did not significantly affect satisfaction.

Conclusions:

• Nurse characteristics, not display design, appear to be key drivers of satisfaction with ML-CDS explanations.
• Findings highlight the need to tailor ML-CDS design to individual nurse attributes.
• Informing the design of more effective and usable ML-CDS for nursing practice.