AI Feedback Tool for Enhancing Communication Skills

(CLEAR2 Trial)

Overseen By William D Rieger

Age: 18 - 65

Sex: Any

Trial Phase: Academic

Sponsor: The University of Texas Health Science Center, Houston

Disqualifiers: Non-MMS residents, Non-PGY 1-5, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: 1

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial tests a new AI tool designed to help medical residents improve their communication skills. The focus is on using AI to provide feedback on how residents interact with patients. The trial will compare an AI feedback tool (LLM-based feedback tool) with no intervention to determine if the AI enhances communication. General surgery residents at McGovern Medical School might be suitable candidates for this study. As an unphased trial, this study offers a unique opportunity to contribute to innovative educational tools that could enhance medical training.

Do I need to stop my current medications for this trial?

The trial information does not specify whether you need to stop taking your current medications.

What prior data suggests that this LLM-based feedback tool is safe for enhancing communication skills?

Research shows that large language models (LLMs), like the one used in this study's feedback tool, are advancing rapidly. However, safety concerns remain. Studies have found that LLMs can produce unsafe statements if not properly managed, especially when human feedback is involved.

Another survey highlights the importance of checking LLM safety, noting it as a growing research field. This research aims to understand and improve how these models interact safely with users. A different article discusses the strengths of LLMs but also notes their limitations, particularly in ensuring safe interactions.

While LLMs are not typically linked to physical safety issues, awareness of these potential problems is crucial when using them for feedback and communication. The goal is to enhance these tools to make them safer and more reliable for users. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the AI Feedback Tool for Enhancing Communication Skills because it leverages a cutting-edge LLM-based feedback system to improve communication skills in a way that traditional methods, like in-person coaching or self-assessment, may not. Unlike standard options that often require human intervention, this tool utilizes artificial intelligence to provide immediate, personalized feedback. This not only makes the learning process more efficient but also widely accessible, potentially transforming how communication skills are developed across various fields.

What evidence suggests that the LLM-based feedback tool is effective for enhancing communication skills?

Research shows that large language models (LLMs) can enhance communication skills. In this trial, participants will use an educational LLM-based feedback tool. Studies have found that this tool aids learning, particularly in school settings where it improves writing skills. One study found that AI tools designed for communication significantly enhance communication effectiveness, which is crucial in healthcare. LLM feedback has also proven beneficial in training scenarios, helping individuals understand and refine their skills. Overall, evidence suggests that LLM feedback is a valuable tool for improving communication abilities. ^{6 7 8 9 10}

Who Is on the Research Team?

Krislynn M Mueck, MD, MS, MPH

Principal Investigator

The University of Texas Health Science Center, Houston

Are You a Good Fit for This Trial?

This trial is for individuals with communication disorders. Participants should be residents in a healthcare setting who are seeking to improve their communication skills. There's no specific mention of exclusion criteria, so additional factors may determine eligibility.

Inclusion Criteria

McGovern Medical School (MMS) general surgery residents

Postgraduate year (PGY) 1-5

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

1-2 weeks

Intervention

Participants receive LLM-based feedback and participate in communication assessments

1 hour

1 visit (in-person)

Follow-up

Participants are monitored for feedback on LLM interface and communication outcomes

1 hour

What Are the Treatments Tested in This Trial?

Interventions

• LLM-based feedback tool

Trial Overview The study is testing an educational tool based on a large language model (LLM) AI to give feedback to residents. It aims to see if this AI feedback can help them communicate better with patients and assess their own communication skills.

How Is the Trial Designed?

Group I: Educational LLM-based feedback tool Experimental Treatment 1 Intervention

Group II: Control Active Control 1 Intervention

Who Is Running the Clinical Trial?

The University of Texas Health Science Center, Houston

Lead Sponsor

Trials

974

Recruited

361,000+

Health Science Education Small Grants Program

Collaborator

Citations

1. sciencedirect.com sciencedirect.com/science/article/pii/S2666920X23000784

Using LLMs to bring evidence-based feedback into the ... The present study is the first to provide experimental evidence of the effectiveness of LLM-generated feedback for student outcomes in the writing domain.

2. withpower.com withpower.com/trial/phase-patient-communication-9-2025-1b5ec

AI Feedback Tool for Enhancing Communication Skills ... The purpose of this study is to refine and test existing enterprise-grade large language model (LLM) based on generative artificial ...

3. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC12582511/

Effectiveness of Communication Competence in AI ... This review highlights communication competence as a critical component in the design of health care CAs, particularly in improving users' ...

4. arxiv.org arxiv.org/html/2506.17006v1

LLM-Generated Feedback Supports Learning If Learners ... This study investigates how on-demand LLM-generated explanatory feedback influences learning in seven scenario-based tutor training lessons.

5. sciencedirect.com sciencedirect.com/science/article/pii/S1877129725002242

Development and evaluation of AI chatbot tool for written ... This study explored the potential of LLM-based chatbots to support pharmacy students in developing written communication skills relevant to self ...

6. aclanthology.org aclanthology.org/2025.emnlp-main.714.pdf

A Comprehensive Dataset for Enhancing Safety in LLM- ... LLMs are evolving into assistants that lever- age tools, significantly expanding their ca- pabilities but also introducing critical safety.

7. arxiv.org arxiv.org/html/2506.11094v1

A Comprehensive Survey on Safety Evaluation of LLMs This survey aims to provide a comprehensive and systematic overview of recent advancements in LLMs safety evaluation, focusing on several key aspects: (1) "Why ...

8. technologyinsights.dnv.com technologyinsights.dnv.com/safe-responsible-and-effective-use-of-llms/

Safe, responsible and effective use of LLMs This article elaborates on the strengths and limitations of large language models (LLMs) based on publicly available research and our experiences working with ...

9. link.springer.com link.springer.com/article/10.1007/s41019-025-00296-9

LLM-Based Agents for Tool Learning: A Survey Specifically, LLMs lack the ability to analyze the safety of tool feedback. In Human Feedback scenarios, if unsafe statements are input as ...

10. nature.com nature.com/articles/s41598-025-92889-7

A framework for mitigating malicious RLHF feedback in ... We propose the COnsensus-Based RewArd framework (COBRA), a consensus-based technique that can effectively negate the malicious noise generated ...