Abstract: Pilot Randomized Controlled Trial of a Spanish-Language AI-Powered Chatbot to Support Maternal Mental Health Among Latina Mothers: Rosie the Chatbot En Español (Society for Social Work and Research 31st Annual Conference)

Schedule:

Saturday, January 23, 2027

Golden Gate 4 (Hilton San Francisco Union Square)

* noted as presenting author

Alexis Hunter, MA, PhD Student, University of Maryland at College Park

Francia Ximena Marin Gutierrez, MSW, Research Coordinator, University of Maryland at College Park, College Park, MD

Thu Nguyen, PhD, Associate Professor, University of Maryland at College Park

Xin He, PhD, Associate Professor, University of Maryland at College Park, College Park, MD

Jordan Boyd-Graber, PhD, Professor, University of Maryland at College Park, College Park, MD

Heran Mane, Data Scientist, University of Maryland at College Park, MD

Xiaohe Yue, Research Analyst, University of Maryland at College Park, College Park, MD

Neha Srikanth, Doctoral Student, University of Maryland at College Park, College Park, MD

Pankaj Dipankar, PhD, Postdoctoral Fellow, National Institutes of Health (NIH), Bethesda, MD

Daniel Smolyak, PhD, Postdoctoral Fellow, National Institute of Nursing Research, Bethesda, MD

Quynh Nguyen, PhD, na, University of Maryland at Baltimore, Baltimore, MD

Elizabeth Norell, PhD, MSW, LCSW-C, Associate Professor, University of Maryland at College Park, College Park, MD

Background and Purpose: Latina pregnant women and mothers of infants in the United States experience elevated risk for perinatal anxiety, stress, and depression, along with persistent structural and systemic barriers to culturally and linguistically responsive care. These inequities are compounded by limited access to mental health services, immigration-related stressors, and under-resourced support systems that contribute to unmet need during a critical developmental period for both mothers and infants. Artificial intelligence (AI) powered digital health tools offer a potentially scalable approach to expand access to support; however, evidence on their acceptability and effectiveness for Latina mothers remains limited. Directly addressing the conference theme, “Social Work Research at the Service of Democracy: Advancing Rigorous Inquiry for the Common Good,” this study examines how culturally and linguistically tailored digital interventions can expand equitable access to maternal mental health support among Latina mothers. This study evaluates the feasibility, acceptability, and preliminary efficacy of Rosie, a Spanish-language AI-powered chatbot designed with and for Latina mothers, to inform equitable approaches to maternal mental health support and healthcare utilization outcomes.

Methods: We conducted a mixed-methods pilot randomized controlled trial with 30 Spanish-speaking Latina participants who were pregnant or parenting an infant younger than six months. Participants were recruited through community-based organizations and healthcare settings and randomized 1:1 to the Rosie intervention or a comparison Book Club condition that provided monthly children’s board books. Surveys administered at baseline and three-month follow-up assessed depression (PHQ-9), anxiety (GAD-7), perceived stress (PSS), and healthcare utilization. Quantitative analyses examined within-group and between-group changes using appropriate statistical tests for small samples, and qualitative feedback was collected through open-ended responses to assess acceptability, usability, and user experience.

Results: Baseline demographic characteristics and mental health scores were comparable between groups. Over three months, participants in the Rosie group demonstrated statistically significant within-group reductions in anxiety (mean change −2.00, p = 0.02) and perceived stress (mean change −12.20, p = 0.003); no statistically significant changes were observed in these outcomes among the comparison group. Depression scores declined modestly in both groups, with no significant between-group differences. No significant differences were observed in maternal emergency room utilization or adherence to well-baby visits. Engagement with Rosie was high, with consistent use over the study period. The qualitative feedback indicated that participants found the chatbot useful, accessible, culturally relevant, and supportive.

Conclusion: Findings suggest that a culturally and linguistically tailored AI-powered chatbot is feasible, acceptable, and shows preliminary promise for reducing anxiety and stress among Latina mothers during the perinatal period. For social work, these results highlight the potential of AI-enabled tools to expand access to culturally responsive, low-barrier mental health support for underserved populations facing structural barriers to traditional services. Integrating these tools into maternal and child health and community-based service settings may strengthen outreach, engagement, and continuity of care while complementing existing supports. Adequately powered trials are needed to determine effectiveness and guide implementation within social work practice and service systems to advance health equity and improve maternal mental health and healthcare utilization outcomes.