Natural Language Processing to Extract Meaningful Information from Patient Experience Feedback.

Area of Science:

• Health Informatics
• Natural Language Processing
• Patient Experience Research

Background:

• Hospitals collect patient feedback to improve care quality and patient experience, as reimbursement is partly tied to patient perceptions.
• Traditional analysis of patient feedback often relies on quantitative data, potentially missing nuanced insights.

Purpose of the Study:

• To demonstrate the application of Natural Language Processing (NLP) for extracting meaningful information from textual patient feedback.
• To identify specific factors contributing to negative patient experiences using NLP techniques.

Main Methods:

• Standardized textual data using NLP libraries, including spell correction and stop-word removal.
• Analyzed sentiment and care aspects, focusing on categories with more negative comments.
• Utilized NLP to identify frequently occurring words, adjectives, and bigrams, followed by manual review of comments.

Main Results:

• Patient feedback predominantly focused on doctors and nurses.
• "Room," "discharge," and "tests and treatments" emerged as areas with more negative than positive comments.
• Identified specific issues like climate control, noise, discharge delays, and procedural discomfort as key contributors to negative feedback.

Conclusions:

• Natural Language Processing (NLP) is a powerful and efficient tool for gaining actionable insights from raw patient feedback.
• NLP provides valuable information not obtainable through quantitative data alone, aiding targeted interventions for patient care improvement.