Background Colorectal cancer (CRC) screening rates remain disproportionately low among Hispanic and Latino populations. While artificial intelligence (AI) has shown promise in healthcare delivery, its impact on health equity remains unclear.

Objective To evaluate the effectiveness of a bilingual generative AI voice agent outreach program in engaging Spanish-speaking patients for CRC screening compared to English-speaking patients.

Methods We conducted a retrospective analysis of AI-powered outreach calls for CRC screening at a large integrated health system serving central Pennsylvania and northern Maryland in September 2024. The study included 1,878 patients (517 Spanish-speaking, 1,361 English-speaking) eligible for colorectal cancer screening. The AI care agent conducted personalized phone calls in the patient’s preferred language to discuss screening and facilitate fecal immunochemical test (FIT) kit requests. The primary outcome was FIT test opt-in rate. Secondary outcomes included call connect rates and duration.

Results Spanish-speaking patients demonstrated significantly higher engagement across all measures compared to English-speaking patients: FIT test opt-in rates (18.2% vs. 7.1%, p<0.001), connect rates (88.8% vs. 53.3%, p<0.001), and call duration (6.05 vs. 4.03 minutes, p<0.001). In multivariate analysis, Spanish language preference remained an independent predictor of FIT test opt-in (adjusted OR 2.012, 95% CI 1.340-3.019, p<0.001) after controlling for demographic and other factors (gender, age, state of residence, and call duration).

Conclusions Contrary to concerns about technology exacerbating disparities, AI-powered outreach achieved significantly higher engagement among Spanish-speaking patients. These findings suggest that language-concordant AI interactions may help address healthcare disparities and improve preventive care engagement in traditionally underserved populations.

MB, MSA, GM, RL, MRD, AM, SM, SG and AC are employees of Hippocratic AI, which provided funding for this study. RHB is an employee of WellSpan, which provided data for this study. JDA is an Adjunct Professor at the University of British Columbia and received compensation for work performed on this project. AA is an employee of UC Davis Health and received compensation for work performed on this project. All authors have reviewed and approved the manuscript and materials included in this submission.

This research was supported by Hippocratic AI, Inc.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

Data supporting the results can be accessed by contacting the corresponding author.