LLM-Assisted Patient Consultation Study with 38 Participants

38 female participants with dysmenorrhea were randomized in a mixed-methods study investigating an LLM-driven knowledge translation module during virtual gynecological consultations. Quantitative analysis indicates the module significantly enhances knowledge acquisition (β=6.471, p=0.007) and positively influences user satisfaction. The study, authored by Fei Du and published on figshare in 2026, provides preliminary empirical support for patient-centered AI medical services.

Use Cases

• Evaluating the impact of AI modules on patient knowledge acquisition based on the reported quantitative metrics.
• Studying mechanisms of patient empowerment and autonomy based on the constructed 'Patient Empowerment Model'.
• Identifying technical limitations for LLM-assisted consultation modules based on the findings regarding interpretation depth and personalization.
• Designing patient-centered AI medical services based on the empirical support for cognitive scaffolding.

Strengths

• Includes quantitative results with specific statistical metrics (β=6.471, p=0.007).
• Constructs a qualitative 'Patient Empowerment Model' from participant data.
• Focuses on a specific medical condition (dysmenorrhea) and consultation context (virtual gynecology).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small (59.0 KB), indicating limited scope and likely a research paper rather than a large-scale data collection.

Provenance

Source

figshare

Collection Method

Mixed-methods research study involving randomization of 38 participants.

Freshness

Last updated 2026-05-29 11:43:09; freshness should be verified.