Objective:

To evaluate the contribution of LLM-assisted analysis compared to traditional thematic analysis in chronic illness management.

Approach:

• Study Design and Participants: The study employed thematic analysis within an interpretivist framework, conducting semistructured interviews with individuals managing multiple chronic conditions to identify recurring themes.
• Phased Methodology: The study followed a multiphase qualitative approach, including the development of an interview guide, data collection through interviews, and comparative analysis of manual and AI-assisted thematic analysis.

Key Findings:

• LLMs can identify patterns and generate thematic insights from large textual datasets.
• LLM-assisted analysis may reveal alternative thematic groupings that challenge or complement researchers' assumptions.
• The study's design allowed for a comparative evaluation of manual and LLM-assisted analysis.

Interpretation:

LLMs offer a scalable alternative to traditional qualitative analysis, potentially enhancing the depth and breadth of insights in chronic illness management research.

Limitations:

• Manual thematic analysis is time-intensive and resource-demanding.
• Variations in coding and interpretation may affect consistency in team-based settings.

Conclusion:

The study highlights the comparative evaluation of LLMs in qualitative research, particularly in the context of chronic illness management.

Sources:

• Standards for Reporting Qualitative Research