An approach to predicting patient experience through machine learning and social network analysis

Area of Science:

• Healthcare analytics
• Machine learning in medicine
• Patient experience research

Background:

• Patient experience is crucial for healthcare system performance.
• The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey measures patient satisfaction.
• Doctor communication is a key domain influencing overall patient experience.

Purpose of the Study:

• To develop a machine learning model predicting patient responses to HCAHPS
• Doctor Communications
• questions.
• To identify healthcare providers most influential in shaping patient experience.

Main Methods:

• Observational study using electronic health record data (2016-2020).
• Machine learning (random forest) to predict patient survey responses.
• Social network analysis to identify influential providers.

Main Results:

• The model accurately predicted patient responses to doctor communication questions (AUCs 0.876-0.819).
• Doctors with higher network centrality showed a greater impact on patient experience.
• Identified patients at risk for negative feedback.

Conclusions:

• Machine learning can identify patients likely to report negative experiences.
• Social network analysis provides metrics to pinpoint influential providers.
• This approach aids in targeted interventions to enhance patient care and communication.