Chatbot to Maximize Hereditary Cancer Genetic Risk Assessment

NCT05562778 | Recruiting

• 1 other identifier: 21-11024123
• Study Type: interventional
• Target: 150
• Locations: 1 country
• Sites: 4

Brief Summary

In this study, the investigators aim to compare a mobile health platform, known as a 'chatbot,' that leverages artificial intelligence and natural language processing to scale communication, to 'usual care' that patients would receive. This comparison will enable the investigators to determine if the chatbot system can improve rates of recommendation for genetic testing among patients at elevated risk of harboring a familial cancer syndrome in an all-Medicaid gynecology clinic. Furthermore, the investigators aim to evaluate facilitators of inequity in regard to patient access to and utilization of genetic testing services.

Conditions

Gynecologic Cancer; Hereditary Cancer Syndrome

Outcome Measures

Primary Outcomes (1)

• Proportion recommended genetic testing

  The proportion of high-risk patients that are recommended genetic testing for familial cancer syndromes in the chatbot vs. usual care arms.

  2 years

Secondary Outcomes (9)

• Proportion completed recommended genetic testing

  2 years

• Facilitator of inequity in the utilization of genetic services

  2 years

• Facilitator of inequity in the utilization of genetic services

  2 years

• Facilitator of inequity in the utilization of genetic services

  2 years

• Facilitator of inequity in the utilization of genetic services

  2 years

Study Arms (2)

Chatbot

EXPERIMENTAL

Subjects will receive a text message initiating a chatbot conversation that relies on natural language processing to gather personal and family cancer. Subjects are identified by the chatbot as meeting National Comprehensive Cancer Network (NCCN) high-risk criteria. Next, subjects undergo pre-test genetic counseling via the chatbot and then clinicians are notified (via the chatbot portal) that the subject meets high-risk criteria. For subjects meeting high-risk criteria (based on the chatbot evaluation), the clinician will complete genetic counseling and recommend genetic testing during the visit. For subjects interested in genetic testing, the clinician will order genetic testing.

Other: Chatbot

Usual Care

NO INTERVENTION

Personal and family cancer history will be collected by the clinician during the subject's visit. Clinicians will evaluate the patient's personal/family history according to National Comprehensive Cancer Network (NCCN) high-risk criteria. For subjects recognized by the clinician as meeting NCCN criteria, the clinician will complete genetic counseling and recommend genetic testing. For subjects interested in genetic testing, the clinician will order genetic testing.

Interventions

Chatbot OTHER

A chatbot is a software program designed to simulate human conversation, typically via text. Chatbots utilize natural language processing to gather patient data, anticipate questions, and predict responses. In this study, the Chatbot will simulate a text-like conversation with patients via a smartphone and use this platform to deal with the time-consuming nature of family history collection. The chatbots can then triage the collected family history against medical guidelines to determine which patients warrant genetic testing. For those patients meeting criteria for genetic testing, the chatbot can offer pre-test education and assist physicians with ordering genetic testing for interested patients meeting high-risk criteria.

Chatbot

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• years of age or older.
• Scheduled for a New Patient appointment in the gynecology clinic. Speaks and reads in English.
• Access to a telephone with texting capacity.
• Has not had prior genetic testing for hereditary cancer syndromes.

You may not qualify if:

• Under 18 years of age
• Has had previous genetic testing for hereditary cancer syndromes
• Does not read/speak in English
• Does not have access to a phone with texting capabilities

Sponsors & Collaborators

• Weill Medical College of Cornell University: lead

Study Sites (4)

NYP Brooklyn Methodist Hospital

Brooklyn, New York, 11215, United States

RECRUITING

NYP Lower Manhattan Hospital

New York, New York, 10038, United States

RECRUITING

Weill Cornell Medicine

New York, New York, 10065, United States

RECRUITING

NYP Medical Group Queens

Queens, New York, 11375, United States

RECRUITING

Related Publications (1)

• Bull LE, Webster EM, McDougale A, Howard D, Ahsan MD, Levi S, Grant B, Chandler I, Christos P, Sharaf RN, Frey MK. Protocol for Health Risk Information Technology-Assisted Genetic Evaluation (HeRITAGE): a randomised controlled trial of digital genetic cancer risk assessment in a diverse underserved gynaecology clinic. BMJ Open. 2024 Sep 5;14(9):e082658. doi: 10.1136/bmjopen-2023-082658.

  PMID: 39237276 DERIVED

MeSH Terms

Conditions

Neoplastic Syndromes, Hereditary

Condition Hierarchy (Ancestors)

Neoplasms; Genetic Diseases, Inborn; Congenital, Hereditary, and Neonatal Diseases and Abnormalities

Study Officials

• Melissa K Frey, MD

  Weill Medical College of Cornell University

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Melissa K Frey, MD, MS

CONTACT

212-746-3049; mkf2002@med.cornell.edu

Isabelle Chandler

CONTACT

212-746-3049; irc4005@med.cornell.edu

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: NONE
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: 2-arm randomized controlled trial of a chatbot for personal/family history intake vs. usual care (intake via the clinician interview). Enrolled patients will be randomized in a 1:1 ratio.

Study Record Dates

• First Submitted: September 27, 2022
• First Posted: October 3, 2022
• Study Start: January 15, 2023
• Primary Completion (Estimated): December 1, 2026
• Study Completion (Estimated): December 1, 2026
• Last Updated: July 28, 2025

Record last verified: 2025-07

Data Sharing

• IPD Sharing: Will not share

Locations

US (4)