Conversational Agent for Cancer

(ECA-PRO Aim 3 Trial)

Overseen By Jasmine Malave, BS

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: Tufts Medical Center

Disqualifiers: Incarceration, Leaving Boston, ECA screener

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

Compensation: Varies

Number of Visits: Unknown

Duration: 6 weeks

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial tests a virtual assistant called an Embodied Conversational Agent (ECA), which serves as a digital companion to help people with cancer manage their health. The researchers aim to determine if this virtual helper can improve personal health outcomes by engaging users through conversation and gestures. Participants may use the ECA on their smartphones, while another group will complete standard online surveys. This trial suits individuals with gastrointestinal or head and neck cancer who are undergoing chemotherapy or radiation and can read and speak English. As an unphased trial, it offers a unique opportunity to explore innovative digital health tools that could enhance cancer care management.

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 the Embodied Conversational Agent is safe for patients with cancer?

Research shows that Embodied Conversational Agents (ECAs) are generally well-received by users. These agents are computer characters that mimic real conversations using voice and hand movements. Studies have found that ECAs are effective in healthcare, especially for people with chronic illnesses like diabetes and cancer.

While specific safety data for ECAs in cancer care is limited, research suggests they can help patients better understand health information and manage treatment side effects. Patients often find them useful with few negative effects.

Overall, evidence suggests that ECAs are safe for use, especially since they have been tested in various healthcare settings. However, staying updated on new research findings is important. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the Embodied Conversational Agent (ECA) for cancer care because, unlike traditional methods that often rely on face-to-face consultations or static online resources, this approach leverages cutting-edge technology to provide a dynamic, interactive experience. The ECA is a smartphone-based system that offers real-time patient interaction, clinician tools for swift configuration of new measurement systems, and data visualizations for both patients and clinicians. This innovative approach could enhance patient engagement and provide timely, personalized feedback, potentially leading to improved management of symptoms like depression, anxiety, and fatigue commonly associated with cancer.

What evidence suggests that the Embodied Conversational Agent is effective for enhancing patient-reported outcomes in cancer patients?

Research has shown that virtual characters, known as Embodied Conversational Agents (ECAs), can improve outcomes for cancer patients. In this trial, some participants will interact with an ECA system, which includes a smartphone-based patient interface and tools for clinicians. Studies have found that these virtual helpers can enhance quality of life, reduce pain and anxiety, and keep patients more engaged in their care over time. ECAs are particularly beneficial for individuals who find health information difficult to understand. They use voice, gestures, and empathy to create a conversational experience. Early results suggest that ECAs are valuable for gathering patient feedback, which is crucial for tailoring care to individual needs. ^{1 5 6 7 8}

Who Is on the Research Team?

Michael Paasche-Orlow, MD, MPH

Principal Investigator

Tufts Medical Center

Are You a Good Fit for This Trial?

This trial is for English or Spanish-speaking adults with cancer who can consent independently, have good hearing and vision (with correction), and are patients of oncologists at Boston Medical Center. There are no specific exclusion criteria.

Inclusion Criteria

Must have adequate corrected vision to use the ECA system (based on a 1 minute ECA functional screener)

Provider subjects must be practicing oncologists at BMC

I can hear well enough to use special medical equipment.

Exclusion Criteria

Not able to use the ECA screener

Subject is incarcerated

Subject plans to leave the Boston area in the next 6 weeks

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Treatment

Participants are randomized to provide data for 6 weeks using ECA-PRO or standard web administration (REDCap)

6 weeks

Weekly data collection

Follow-up

Participants are monitored for safety and effectiveness after treatment

4 weeks

What Are the Treatments Tested in This Trial?

Interventions

• Embodied Conversational Agent

Trial Overview The study tests an Embodied Conversational Agent (ECA) tool designed to improve patient-reported outcomes in cancer care by simulating face-to-face conversation. It also uses REDCap surveys to collect data on its effectiveness.

How Is the Trial Designed?

Group I: Embodied Conversational Agent Experimental Treatment 1 Intervention

The ECA systems compromises 1) a smartphone based ECA patient interface; 2) clinician authoring tool, to enable new measurement systems to be rapidly configured 3) clinician and patient data visualizations; and 4) a central server with relation database, administrative user interfaces and ability to send asynchronous notifications to users' smartphones.

Group II: RedCap Survey Active Control 1 Intervention

An internet-based measure will use PROMIS Profile measures which will include depression, anxiety, fatigue, pain intensity and interference, sleep disturbance, physical function and satisfaction with social role

Who Is Running the Clinical Trial?

Tufts Medical Center

Lead Sponsor

Trials

264

Recruited

264,000+

Boston Medical Center

Collaborator

Trials

410

Recruited

890,000+

Published Research Related to This Trial

In a study involving 142 women with breast cancer, the artificial conversational agent (chatbot) Vik provided answers that were as satisfying as those given by a group of physicians, with a success rate of 69% compared to 64% for physicians, demonstrating noninferiority.

This research suggests that chatbots like Vik can effectively inform patients about their treatment, potentially reducing the need for in-person consultations for minor health concerns, allowing doctors to focus on patients who require more direct care.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A Chatbot Versus Physicians to Provide Information for Patients With Breast Cancer: Blind, Randomized Controlled Noninferiority Trial. Bibault, JE., Chaix, B., Guillemassé, A., et al.[2020]

A study involving 14 clinicians revealed that motivations for reporting serious oncology-associated adverse drug reactions included scientific curiosity and public health concerns, but feedback from pharmaceutical manufacturers was predominantly negative.

The research highlights the need for improved pharmacovigilance mechanisms, such as big data and registries, to enhance the reporting and documentation of adverse drug events, addressing barriers like complacency and indifference among clinicians.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Caveat Medicus: Clinician experiences in publishing reports of serious oncology-associated adverse drug reactions. Bennett, CL., Schooley, B., Taylor, MA., et al.[2020]

The pilot version of the Rosa chatbot, developed over two years with input from patients and healthcare professionals, aims to assist breast and ovarian cancer patients in understanding genetic BRCA testing, highlighting the importance of user-centered design in health technology.

Despite initial challenges with AI matching user questions to answers, the final version of the chatbot significantly expanded its database and received positive feedback from users regarding its layout and trustworthiness, indicating potential for real-world application in healthcare settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Ask Rosa - The making of a digital genetic conversation tool, a chatbot, about hereditary breast and ovarian cancer. Siglen, E., Vetti, HH., Lunde, ABF., et al.[2022]

A study involving 63 patients (20 using a chatbot and 43 receiving usual care) found that the chatbot significantly reduced emergency department visits and unscheduled hospitalizations for patients undergoing chemotherapy for gynecologic cancers.

Patients using the chatbot had an adjusted incidence rate ratio of 0.27 for ED visits and 0.31 for unscheduled hospitalizations, indicating a strong potential for chatbots to improve patient management and reduce healthcare burdens during cancer treatment.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Using a chatbot to reduce emergency department visits and unscheduled hospitalizations among patients with gynecologic malignancies during chemotherapy: A retrospective cohort study. Huang, MY., Weng, CS., Kuo, HL., et al.[2023]

A multidisciplinary expert group identified key signals that a new AI system should emit to enhance the management of adverse drug events (ADEs) in patients taking oral antineoplastic agents, including educational reminders for patients and alerts for necessary tests and examinations.

In a 6-month observational study at a university hospital, a total of 3,641 ADEs were reported, highlighting the importance of patient education and clinical monitoring in improving patient safety and quality of life.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Analysis of adverse drug events as a way to improve cancer patient care. Vicente-Oliveros, N., Gramage-Caro, T., Corral de la Fuente, E., et al.[2022]

A study evaluating responses to 200 medication-use questions found that a medication database provided more clinically correct, complete, safe, and useful answers than an AI chatbot, with 97% of questions answered by the database compared to 80% by the chatbot.

While the chatbot's clinical correctness was comparable to the database at 85%, it lagged behind in completeness (65%), safety (71%), and usefulness (68%), indicating a need for further improvements in chatbot technology for reliable clinical use.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Evaluation of inpatient medication guidance from an artificial intelligence chatbot. Beavers, J., Schell, RF., VanCleave, H., et al.[2023]

In a study assessing the quality of responses from AI chatbots ChatGPT and Bing regarding cancer-related questions, ChatGPT scored an average of 3.9 while Bing scored 3.2, indicating that ChatGPT provided higher quality information overall (P < 0.001).

Despite the potential of AI chatbots in healthcare, the study revealed that 3% of ChatGPT's responses and 15% of Bing's responses had serious shortcomings, highlighting the need for ongoing improvements to ensure accurate and supportive information for cancer patients.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Physician Assessment of ChatGPT and Bing Answers to American Cancer Society's Questions to Ask About Your Cancer. Janopaul-Naylor, JR., Koo, A., Qian, DC., et al.[2023]

Citations

1. clinicaltrials.gov clinicaltrials.gov/study/NCT05948618

Improving PRO for Patients with Cancer Using ECAs and ... Over the past decade, the investigators have developed and tested a tool with the potential to enhance PROs at the individual level - embodied conversational ...

2. jmir.org jmir.org/2025/1/e76968

The Effectiveness and Feasibility of Conversational Agents ... Specifically, the objectives were to (1) assess their effectiveness in improving patient outcomes, including quality of life, pain, anxiety, ...

3. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC10806449/

Embodied Conversational Agents for Chronic Diseases This scoping review aims to identify the current practice in the development and evaluation of ECAs for chronic diseases.

4. researchgate.net researchgate.net/publication/394437783_The_Effectiveness_and_Feasibility_of_Conversational_Agents_in_Supporting_Care_for_Patients_With_Cancer_Systematic_Review_and_Meta-Analysis

The Effectiveness and Feasibility of Conversational Agents ... Objective: This study aimed to evaluate the feasibility and effectiveness of CAs in supporting care for patients with cancer and to summarize ...

5. withpower.com withpower.com/trial/phase-malignancies-3-2022-4f24c

Conversational Agent for Cancer (ECA-PRO Aim 3 Trial) Trial Overview The study tests an Embodied Conversational Agent (ECA) tool designed to improve patient-reported outcomes in cancer care by simulating face-to- ...

6. clinicaltrials.gov clinicaltrials.gov/study/NCT05948618

Improving PRO for Patients with Cancer Using ECAs and ... Over the past decade, the investigators have developed and tested a tool with the potential to enhance PROs at the individual level - embodied conversational ...

7. jmir.org jmir.org/2024/1/e47134/PDF

Embodied Conversational Agents for Chronic Diseases In health care, ECAs have been designed to assist with various tasks such as providing diabetes self-management education [10], promoting cancer.

8. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC12374140/

Systematic Review and Meta-Analysis - PMC - PubMed Central Narrative synthesis suggested that CAs have the potential to improve patients' acquisition of health information and help manage treatment-related side effects.