AI-Generated Nutritional Messaging for Colorectal Cancer Survivors

(NLM Trial)

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: University of Minnesota

Disqualifiers: Non-English speakers, conflicts of interest

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

Compensation: Varies

Number of Visits: Unknown

Duration: 8-12 months

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial evaluates how well AI tools, such as ChatGPT, generate nutrition advice for colorectal cancer survivors compared to experienced dietitians. The goal is to determine if AI can create effective and understandable dietary guidance. Participants must be dietitians with at least five years of experience in cancer patient care. This research explores whether AI, including the Nutritional Language Model, can serve as a helpful tool in colorectal cancer care, focusing on the quality and usefulness of its advice. As an unphased trial, this study offers a unique opportunity to contribute to innovative research that could enhance dietary support for cancer survivors.

Will I have to stop taking my current medications?

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

What prior data suggests that AI-generated nutritional messaging is safe for colorectal cancer survivors?

Research has shown that large language models (LLMs) like ChatGPT and Gemini can provide personalized diet advice. These models have made nutrition information more accessible, benefiting colorectal cancer survivors in particular. Studies have explored how LLMs can offer insights into nutrition and health management.

There are no safety concerns with using LLMs, as they are not medical treatments or drugs. They serve as tools to create nutrition advice. The main question is whether their advice matches the helpfulness and accuracy of human experts. Early findings suggest that AI-generated advice can complement human advice, making nutrition guidance more accessible.

In summary, using LLMs for nutrition advice appears safe, as they are simply a source of information. The goal is to determine if they can effectively support individuals in their health journey, especially when expert advice is not readily available. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the AI-Generated Nutritional Messaging treatment for colorectal cancer survivors because it introduces a personalized way to support dietary changes. Unlike traditional methods that rely on face-to-face meetings with dietitians, this approach uses advanced AI to tailor nutritional advice based on individual needs and preferences, potentially increasing adherence and effectiveness. This innovative method could simplify ongoing dietary management, making it more accessible and engaging for patients recovering from colorectal cancer.

What evidence suggests that AI-generated nutritional messaging is effective for colorectal cancer survivors?

Research has shown that AI tools like ChatGPT and Claude can create personalized nutrition advice for people recovering from colorectal cancer. One study found that these AI tools effectively customize educational content to meet the needs of cancer survivors, offering a new way to aid recovery. These tools provide easy-to-read and relevant information, crucial for personalizing care. In this trial, participants will receive nutritional guidance either from AI-generated messaging or from a dietician. Researchers continue to compare the effectiveness of AI-generated messages with advice from human experts. Early results suggest that AI could complement traditional nutrition guidance from humans. ^{1 6 7 8 9}

Who Is on the Research Team?

Annie Lin, RD, PhD

Principal Investigator

Hormel Institute

Glen Morris, PhD

Principal Investigator

Hormel Institute

Are You a Good Fit for This Trial?

This trial is for colorectal cancer survivors needing nutritional guidance. Participants should be open to receiving dietary advice from both AI language models and human experts. Specific eligibility criteria are not provided, so it's assumed that there are no major restrictions.

Inclusion Criteria

Must have access to computer and internet access

I am a dietitian with over 5 years of experience in cancer care.

Exclusion Criteria

Experts with conflicts of interest related to any of the LLMs that are being evaluated

I am fluent in English.

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Treatment

Participants receive nutrition-related content generated by LLMs and human experts for comparison

8-12 months

Follow-up

Participants are monitored for the effectiveness and implementation of nutrition messages

4-8 weeks

What Are the Treatments Tested in This Trial?

Interventions

• Nutritional Language Model

Trial Overview The study compares personalized nutritional messages created by AI language models (ChatGPT, Claude, Gemini, Co-Pilot) with those written by human nutrition experts. It assesses readability, relevance to themes and medicine, perceived effectiveness, and practical use in clinical settings.

How Is the Trial Designed?

Group I: Dietician Experimental Treatment 1 Intervention

Who Is Running the Clinical Trial?

University of Minnesota

Lead Sponsor

Trials

1,459

Recruited

1,623,000+

Citations

1. sciencedirect.com sciencedirect.com/science/article/pii/S2472630325000536

NLP for computational insights into nutritional impacts on ... This research introduces the Nutritional Impact on CRC Prediction Framework (NICRP-Framework), which combines Natural Language Processing (NLP) techniques with ...

2. clinicaltrials.gov clinicaltrials.gov/study/NCT06661590

Study Details | NCT06661590 | Nutritional Language Model Study Overview. Brief Summary. Colorectal cancer survivors often face unique nutritional challenges and require support in their recovery and long0term health.

3. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC11995809/

Evaluation of Large Language Models in Tailoring ... This study aimed to evaluate the overall performance of LLMs in generating tailored educational content for cancer survivors and their caregivers.

4. withpower.com withpower.com/trial/ai-generated-nutritional-messaging-for-colorectal-cancer-survivors-f9190

AI-Generated Nutritional Messaging for Colorectal Cancer ... This trial is for colorectal cancer survivors needing nutritional guidance. Participants should be open to receiving dietary advice from both AI language models ...

5. tandfonline.com tandfonline.com/doi/full/10.1080/10408398.2024.2360068

Still too little evidence: the role of diet in colorectal cancer ... This review provides a narrative summary of existing research on dietary factors affecting CRC-specific mortality, RFS, and disease-free survival (DFS).

6. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC11990115/

Bridging Gaps in Cancer Care: Utilizing Large Language ... Large language models (LLMs) like ChatGPT and Gemini demonstrate significant potential in expanding access to personalized dietary ...

7. nature.com nature.com/articles/s41467-024-55219-5

Diet-wide analyses for risk of colorectal cancer We report here on a systematic analysis of 97 dietary factors and subsequent colorectal cancer risk using a diet-wide association study

8. bmcpublichealth.biomedcentral.com bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-024-21123-2

Using large language models for extracting stressful life ... Using large language models for extracting stressful life events to assess their impact on preventive colon cancer screening adherence | BMC ...

9. academic.oup.com academic.oup.com/jnci/advance-article/doi/10.1093/jnci/djaf098/8124402?searchresult=1

Postdiagnosis dietary and lifestyle factors and mortality ... To optimize CRC survival, CRC patients may be recommended to adopt a diet rich in whole grains, coffee, milk, and dietary calcium; to avoid excessive alcohol, ...