LLM-Powered Healthcare UX: Smart Interfaces for Telemedicine and Respiratory Care

Abstract

The rapid expansion of telemedicine has transformed healthcare delivery, yet user experience (UX) challenges continue to hinder patient engagement and clinical effectiveness. This study investigates the integration of Large Language Model (LLM)-powered smart interfaces into telemedicine platforms, with a specific focus on respiratory care for conditions such as asthma, chronic obstructive pulmonary disease (COPD), and post-COVID complications. A mixed-methods research design involving 382 patients and 112 healthcare professionals evaluated UX outcomes, interface performance, and clinical impacts. Results showed significant improvements in usability, satisfaction, accessibility, and cognitive load reduction, alongside high natural language understanding accuracy and personalization of responses. Clinical outcomes also improved, with higher medication adherence, enhanced symptom reporting compliance, and reduced hospitalization rates. Multivariate analysis confirmed that UX parameters collectively influenced patient adherence, symptom management, and quality of life, while cluster analysis revealed distinct patient engagement groups. Importantly, perceptions of trust, inclusivity, and data security improved substantially, highlighting the ethical potential of LLM-powered systems. These findings suggest that LLM-driven telemedicine platforms represent a scalable and equitable solution for advancing patient-centered digital healthcare, particularly in managing chronic respiratory conditions.