Peritoneal Dialysis (PD) Specialized LLM for PD Management (PIONEER-PD)

Physician Implementation of an NLP-Empowered Electronic Resource for Peritoneal Dialysis Management: A Randomized Controlled Trial (The PIONEER-PD Trial)

This study conducted a randomized controlled trial using clinical vignettes to evaluate differences in the quality of PD management among two large language model-assisted workflows and a physician-only decision-making process, and to identify potential risks (e.g., generating clearly erroneous or even harmful suggestions).

Study Overview

• Status: Enrolling by invitation
• Conditions: Peritoneal Dialysis (PD); Large Language Models
• Intervention / Treatment: Other: Generalized LLM tools such as deepseek; Other: Peritoneal dialysis-specialized large language model.

• Study Type: Interventional
• Enrollment (Estimated): 150
• Phase: Not Applicable

Contacts and Locations

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510080
      • Sixiu Chen

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

1. Internal medicine or nephrology standardized training residents, licensed residents, or attending physicians.
2. Independent experience in PD management ≤ 3 years.
3. Provided signed informed consent and agreed to comply with the trial procedures.

Exclusion Criteria:

1. Direct involvement in the development or training of the specialized PD large language model, or in the construction of the clinical scenarios/ scoring criteria used in this trial.
2. Participation in the pilot testing of all clinical scenarios used in this trial.
3. Inability or unwillingness to access the study platform or use online resources during the study period.
4. Experienced PD experts.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
No Intervention: Phys Group; Participants could search for relevant knowledge, but access to large language model (LLM)-related websites was prohibited.
Placebo Comparator: Phys+DS Group; Participants had access to both the web and the DeepSeek interface.	Other: Generalized LLM tools such as deepseek; Participants were permitted to ask the LLM any questions related to the clinical scenarios. However, they were explicitly instructed to critically evaluate the model's suggestions and to take full responsibility for the final clinical plans.
Experimental: Phys+PD-LLM Group; Participants had access to the web and a specialized PD-LLM provided by the First Affiliated Hospital of Sun Yat-sen University.	Other: Peritoneal dialysis-specialized large language model.; The peritoneal dialysis-specialized large language model (PD-LLM) used in this study was jointly developed by the Department of Nephrology at the First Affiliated Hospital of Sun Yat-sen University and Digital Health China (DHC).

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Management Reasoning	Percent correct (range: 0 to 100) for each case.	Within 90-min study

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Domain-Specific Scores	Percent correct (range: 0 to 100) for each case.	within 90-min study.
Severity of Potential Harm	The severity of potential harm will be classified as none, mild-to-moderate, or severe.	Within 90-min study.
Time Spent on Management	The time participants spend per case.	Within 90-min study.
Self-Reported Confidence per Case	Scale 1-10. 10 represents being very confident.	Within 90-min study.

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Total vignette performance score of LLM generated answers	Percent correct (range: 0 to 100) for each case.	Within 90-min study.
Severity of Potential Harm of LLM-generated answers	The severity of potential harm will be classified as none, mild-to-moderate, or severe.	Within 90-min study.

Collaborators and Investigators

• Sponsor: First Affiliated Hospital, Sun Yat-Sen University
• Collaborators: Vantive Health LLC

Study record dates

Study Major Dates

• Study Start (Actual): April 5, 2026
• Primary Completion (Estimated): May 30, 2026
• Study Completion (Estimated): May 30, 2026

Study Registration Dates

• First Submitted: April 5, 2026
• First Submitted That Met QC Criteria: April 13, 2026
• First Posted (Actual): April 20, 2026

Study Record Updates

• Last Update Posted (Actual): April 20, 2026
• Last Update Submitted That Met QC Criteria: April 13, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Other Study ID Numbers: IIT-2025-965

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No