JMIR Publications

ABSTRACT

Background:

Qualitative health research often focuses on how patients experience and manage chronic illness. With the emergence of large language models (LLMs), such as Claude and GPT, new opportunities are arising to support and scale the thematic analysis of narrative health data. Yet, their role and added value compared to traditional human led approaches remain underexplored.

Objective:

To compare traditional manual thematic analysis with AI analysis using an LLM, focusing on how each method captures patient experiences in managing multiple chronic conditions.

Methods:

Semi structured interviews were conducted with 30 individuals living with two or more chronic illnesses. Transcripts were analyzed using both manual thematic coding and Claude 3.5 Sonnet. A comparative analysis assessed shared and unique themes, as well as differences in depth and contextual framing.

Results:

Both approaches revealed core themes related to the patient experience, such as healthcare navigation and challenges, support systems and family dynamics, and emotional challenges and coping. Manual analysis provided richer contextual interpretations, while the LLM approach identified a broader range of sub themes and showed efficiency in handling larger volumes of qualitative data. Each method also revealed distinct insights: the manual analysis highlighted themes such as faith, caregiving roles, and a proactive mindset, whereas the LLM emphasized themes like future planning and multiple health conditions.

Conclusions:

A hybrid approach that integrates AI assisted and human led thematic analysis can enhance both analytical depth and scalability. This study offers early evidence for the feasibility and added value of using LLMs to augment qualitative health research in complex clinical contexts.