Social capital factors in healthcare AI implementation: a qualitative systematic review

Fragmented healthcare systems often hinder the seamless adoption of artificial intelligence (AI). Traditional frameworks frequently view AI adoption as a discrete technical acquisition, overlooking the socially negotiated processes and team-level dynamics crucial for integration. Social capital theory provides a vital lens to understand how relational networks and trust translate digital capabilities into sustainable clinical execution. To synthesize existing qualitative evidence on how social capital shapes the sustainable integration of AI in healthcare, primarily from healthcare providers’ and implementation actors’ perspectives, while examining their accounts of wider stakeholder relationships. A qualitative systematic literature review with thematic synthesis was conducted following the PRISMA reporting guidelines. Five databases (PubMed, Web of Science, Embase, Cochrane, and CINAHL) were searched for literature published between January 2020 and April 2025. Methodological quality was appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for qualitative studies and the Mixed Methods Appraisal Tool (MMAT). From 14,703 retrieved records, 39 studies were included, comprising 32 qualitative studies and 7 mixed-methods studies. Thematic synthesis yielded four overarching categories: (1) physical structure, (2) accountability structure, (3) AI in social networks, and (4) regulation and law. Physical structure and resource availability were interpreted as contextual antecedents rather than direct expressions of social capital. Findings reveal that social capital critically moderates AI implementation across dimensions. Structurally, professional networks linking medical, technical, and administrative actors, together with hierarchical accountability arrangements, established the institutional channels through which AI was introduced and sustained. Relationally, interpersonal trust between clinicians and AI tools, alongside patient-provider bonds reshaped by algorithmic mediation, determined the acceptability and ethical legitimacy of adoption. Cognitively, shared norms, collective attitudes, and value alignment around AI use shaped practitioners’ interpretive frameworks and willingness to integrate AI into clinical practice. The findings suggest that AI integration in healthcare is shaped by sociotechnical conditions in which infrastructure, accountability relationships, professional networks, trust, and shared norms interact. PROSPERO CRD420250653624.