Advancing Human-Centered AI in Clinical Decision Support: Sociocognitive Human-in-the-Loop Study in HIV Care

Clinical Snapshot

Key Findings

• P Value: Not applicable — qualitative study; descriptive statistics used for survey data only

• Effect Size: Not applicable — qualitative study; no effect sizes reported

• Primary Outcome: Three primary thematic findings: (1) Clinicians use familiar clinical indicators (e.g., CD4 count, viral load) as cognitive anchors when interpreting AI predictions; (2) Social determinants of health (SDOH) are central to clinician risk assessment and clinical decision-making, yet are perceived as underrepresented in AI model outputs; (3) Clinician trust in AI is conditional and develops incrementally, with explainability and actionability identified as critical prerequisites for translating AI predictions into clinical interventions

• Nnt Or Sensitivity: Not applicable — qualitative design study; no diagnostic or therapeutic outcome metrics reported

• Confidence Interval: Not applicable — qualitative study

Clinical Application

The design principles identified (cognitive anchoring, SDOH integration, explainability, actionability, conditional trust scaffolding) are feasible to implement in iterative CDSS development cycles. However, operationalising SDOH data within EHR-linked ML models remains a significant technical and data governance challenge. Adoption feasibility will depend on workflow integration, clinician training, and institutional AI governance frameworks. In Australia, HIV care is delivered through a network of sexual health clinics, infectious disease outpatient services, and general practice under the RACGP and ASHM (Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine) guidelines. The PBS subsidises antiretroviral therapy including integrase inhibitors and long-acting injectable formulations. The Australian Digital Health Agency's National Digital Health Strategy 2023–2028 explicitly supports AI-enabled clinical decision support, making the design principles from this study directly relevant to Australian CDSS development. The emphasis on SDOH integration is particularly pertinent given the disproportionate HIV burden among Aboriginal and Torres Strait Islander communities, where social, cultural, and structural determinants are paramount. TGA oversight of AI-based Software as a Medical Device (SaMD) would apply to any deployed AI-CDSS in the Australian context. RACGP's position on AI in general practice further underscores the need for explainable, clinician-centred AI tools consistent with this study's findings. HIV care clinicians in multidisciplinary settings, including infectious disease physicians, nurse practitioners, clinical pharmacists, social workers, and case managers working with people living with HIV (PLHIV) in outpatient and ambulatory care contexts

Abstract

BACKGROUND: AI-powered clinical decision support systems (CDSS) have shown promise in improving prediction, monitoring, and treatment optimization across clinical domains, including HIV care. However, translating AI outputs derived from electronic health records into clinically meaningful, trustworthy, and actionable decision support remains challenging, underscoring the need for more human-centered and socioecologically grounded CDSS design. OBJECTIVE: This study aimed to explore how we can effectively translate the outputs of machine learning models based on HIV electronic health records into a real AI-powered CDSS for HIV care. Using the human-in-the-loop method, we engaged a set of stakeholders, including HIV physicians, nurse practitioners, infectious disease pharmacists, social workers, and case managers. Stakeholders interacted with an AI-powered CDSS prototype to identify barriers and challenges to adoption, as well as to inform a more holistic and context-aware AI-powered CDSS design. METHODS: We conducted a field study at Prisma Health in South Carolina that included pre- and postsurveys, interactive usability testing sessions, think-alouds, and in-depth interviews with 16 clinicians providing HIV care between March and September 2025. We analyzed survey responses using descriptive statistics, and then transcribed and analyzed think-aloud and interview data using an etic and emic approach. RESULTS: Clinicians identified multiple challenges and design considerations for AI-powered HIV CDSS, demonstrating that clinician-AI interaction is inherently sociotechnical and embedded across multiple socioecological levels. While clinicians relied on familiar clinical indicators as cognitive anchors for interpreting AI predictions, they emphasized that social determinants of health were central to their own risk assessment and clinical decision-making. Additionally, clinicians' trust in AI is conditional and develops over time, with explainability and actionability emerging as critical factors for translating predictions into meaningful clinical interventions. CONCLUSIONS: Findings highlight the need to move beyond technically accurate predictions toward AI-powered CDSS designs that align with clinicians' cognitive practices and socioecological realities of HIV care. By extending a sociocognitive framework through empirical grounding in HIV clinical practice, this study offers design insights for developing AI-powered CDSS that are trustworthy, context-aware, and capable of supporting actionable decision-making in HIV care settings and beyond.

References

1. 1.Wu, D., Vera, V., Vuruma, S. K. R., Aust, L., Yaddanapalli, B. S., Zhang, J., Markanti, R., Zhang, J., Li, X., Weissman, S., & Olatosi, B. (2026). Advancing human-centered AI in clinical decision support: Sociocognitive human-in-the-loop study in HIV care. Journal of Medical Internet Research. PubMed ID: 42536982. [DOI requires verification against published record]