Using a Multilingual AI Care Agent to Reduce Disparities in Colorectal Cancer Screening for Higher Fecal Immunochemical Test Adoption Among Spanish-Speaking Patients: Retrospective Analysis
- Meenesh Bhimani 1, R Hal Baker 2, Markel Sanz Ausin 1, Gerald Meixiong 1, Rae Lasko 1, Mariska Raglow-Defranco 1, Alex Miller 1, Subhabrata Mukherjee 1, Saad Godil 1, Anderson Cook 1, Jonathan D Agnew 3, Ashish Atreja 4
- 1Hippocratic AI, Palo Alto, CA, United States.
- 2WellSpan Health, York, PA, United States.
- 3School of Population and Public Health, University of British Columbia, Vancouver, BC, Canada.
- 4University of California Davis Health, Davis, CA, United States.
- 0Hippocratic AI, Palo Alto, CA, United States.
Related Experiment Videos
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Multilingual artificial intelligence (AI) significantly boosted colorectal cancer (CRC) screening engagement for Spanish-speaking patients. This AI outreach challenges assumptions about technology disadvantaging non-English speakers, showing higher opt-in rates for CRC screening tests.
Area Of Science
- Health Informatics
- Public Health
- Artificial Intelligence in Healthcare
Background
- Colorectal cancer (CRC) screening rates are lower in Hispanic and Latino populations.
- Artificial intelligence (AI) in healthcare may disadvantage non-English speakers due to bias.
- Addressing disparities in CRC screening access is crucial.
Purpose Of The Study
- To evaluate a multilingual AI care agent's effectiveness in engaging Spanish-speaking patients for CRC screening.
- To compare engagement rates between Spanish-speaking and English-speaking patients.
- To assess AI's role in overcoming language barriers in preventive care.
Main Methods
- Retrospective analysis of an AI outreach initiative at WellSpan Health.
- Included 1878 patients (517 Spanish-speaking, 1361 English-speaking) eligible for CRC screening.
- A multilingual AI agent conducted personalized calls to promote fecal immunochemical test (FIT) kit requests.
Main Results
- Spanish-speaking patients had significantly higher FIT test opt-in rates (18.2% vs 7.1%).
- Higher connect rates (69.6% vs 53.0%) and longer call durations (6.05 vs 4.03 minutes) were observed in Spanish-speaking patients.
- Spanish language preference independently predicted higher FIT test opt-in (aOR 2.012).
Conclusions
- AI-powered outreach significantly improved engagement among Spanish-speaking patients for CRC screening.
- Language-concordant AI interactions can help reduce disparities in preventive care access.
- Findings challenge the notion that technology inherently disadvantages non-English-speaking populations.
Related Experiment Videos
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.

