Evaluation of Patient Experience Using Natural Language Processing Algorithms

Sofia Veronica Ortega Haro, University of Texas at El Paso


INTRODUCTION: Healthcare organizations are making extensive efforts to improve the patient experience. Enhancing patient/client experience and outcomes is crucial for patient-centered care and can reveal improvement opportunities. Healthcare settings currently rely on surveys (e.g., HCAHPS) and patient feedback to measure patient experience. Studies have identified that utilizing patient journey mapping can better capture patient experience throughout all stages of the patient's journey and provide quality and process improvement recommendations at specific hotspots. However, these measurement techniques are time-consuming and resource intensive. AIM: This research aims to measure patient experience of breast cancer patients from social media data using natural language processing algorithms. METHODS: This study analyzes data obtained from social media (e.g., Twitter and Reddit) referent to breast cancer. Natural Language Processing (NLP) algorithms were applied to identify latent topics via Latent Dirichlet Allocation (LDA) and sentiments via Sentiment Analysis (SA) associated to specific hotspots. DISCUSSION: The use of AI to capture patient experience during the patient's journey in the healthcare continuum provides valuable insights to improve individualized, empathetic, and respectful care in clinical systems. Such patient experience information is invaluable for healthcare quality improvement efforts and improving patient-centered care.

Subject Area

Biomedical engineering|Information science|Bioinformatics|Applied Mathematics|Public health

Recommended Citation

Ortega Haro, Sofia Veronica, "Evaluation of Patient Experience Using Natural Language Processing Algorithms" (2021). ETD Collection for University of Texas, El Paso. AAI28714757.