Large Language Model-Driven Analysis of Acoustic and Other Environmental Factors in Palliative Care Environments: Constructing an Acoustic Environment Evaluation Framework

Palliative care plays a crucial role in enhancing quality of life during end-of-life care, with the environment significantly influencing the emotional and psychological well-being of environment users. Constructing an environmental evaluation framework typically relies on qualitative data from interviews and other sources, but qualitative data collection and analysis in this context is high in time and resource requirements, with potentially causing negative impacts on individuals in these sensitive settings and even involving ethical reviews. This study uses a large language model (LLM) to analyze the impact of acoustic and other environmental factors on the emotional experiences of family members and healthcare providers in palliative care environments. We conducted automated qualitative analysis on offline semi-structured interview data from 26 relevant personnel in palliative care environments and online reviews from palliative care institutions across different countries, extracting key themes and emotions related to these environmental factors. The results indicate that noise levels and odors significantly affect the emotional and psychological well-being of both family members and healthcare providers. Family members prioritize privacy and quietness, while healthcare providers focus on space comfort and efficiency. This study demonstrates the potential of AI-assisted qualitative analysis, providing valuable insights for future palliative care environment design strategies.