Skip to content

Evaluating ChatGPT Responses to patient inquiries in Urology: A Comparative Study between the Large Language Model ChatGPT and treating Urologists

Evaluating ChatGPT Responses to patient inquiries in Urology: A Comparative Study between the Large Language Model ChatGPT and treating Urologists - ECURIOUS

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
DRKS
Registry ID
DRKS00033968
Enrollment
40
Registered
2026-02-02
Start date
2026-02-28
Completion date
Unknown
Last updated
2026-03-02

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

N00-N99 R32 E29.1 I86.1 C00-D48

Interventions

Group 1: Study procedure The study comprises four phases: (1) enrolment, (2) generation and preparation of responses, (3) evaluation, and (4) data analysis. Patients are enrolled during routine outpat
n = 40
based on the sample size calculation), relevant contextual information, and the treating urologists’ verbal answers are captured via audio recording, subsequently transcribed, and anonymised prior to
to ensure blinding, only explicit source cues and non-content elements are removed for both sources (without paraphrasing or altering informational content). The resulting response pairs are randomise

Sponsors

Klinik für Urologie, Universitätsklinikum Freiburg
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: • Diverse patients with medical conditions in urology that require a consultation by a board certified urologist • sufficient German language skills • Consent to participate

Exclusion criteria

Exclusion criteria: • Without consent to participate • insufficient German language skills

Design outcomes

Primary

MeasureTime frame
The primary endpoint of the study is the criterion medical accuracy. This score was selected due to its relevance to clinical practice, its low variability, and its high relevance to patient care. This endpoint will be operationalised as a composite score based on the first three medical accuracy items (items 1–3) rated on a 0–10 Likert scale. The primary endpoint will be assessed exclusively by expert raters.

Secondary

MeasureTime frame
All outcomes other than the primary medical accuracy composite (items 1–3) will be analysed as secondary or exploratory. These will include all additional expert-rated scales and items beyond the primary endpoint, as well as all outcomes assessed in the lay-rater group.

Countries

Germany

Contacts

Public ContactArkadiusz Miernik

Klinik für Urologie, Universitätsklinikum Freiburg

arkadiusz.miernik@uniklinik-freiburg.de0761 270-28930, -25821

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Mar 14, 2026