NAVIGATING THE DIGITAL DIVIDE: IMPLEMENTATION AND MANAGEMENT CHALLENGES OF AI-BASED MEDICAL FACILITIES IN MODERN HEALTHCARE: A MULTI-SITE CASE STUDY OF PAKISTAN

Abstract

Background: Artificial intelligence (AI) has a transformative potential in healthcare delivery in low- and middle-income countries (LMICs); nonetheless, the process of operationalisation of AI-based medical systems in the limited socioeconomic settings is understudied. Pakistan, with 230-million people, disjointed health services, and immature digital governance platforms, is a crucial but understudied laboratory on these dynamics. Methods: The research was based on a sequential mixed-methods design, which included (i) a cross-sectional survey of 94 healthcare facilities in four provinces; (ii) semi-structured interviews with 48 key informants (clinicians, IT administrators, policymakers, and AI vendors); and (iii) a systematic review of 62 peer-reviewed articles (20162026). The structural equation modelling (SEM) was used to analyse quantitative data to develop a validated Pakistan AI Healthcare Adoption Model (PAHAM). Thematic analysis was performed on qualitative data, using the Consolidated Framework for Implementation Research (CFIR). Findings: The three prevailing barriers were regulatory vacuity (common in 91.2% of facilities), workforce AI illiteracy (82.7%), and infrastructural deficits (78.3%). The success rate of implementation of AI in private facilities was 71.4% vs. 38.9% in tertiary settings in the public. SEM found that organisational readiness ( = 0.63, p < 0.001) and policy enablement ( = 0.57, p < 0.001) were the best predictors of successful deployment. Significance: The paper provides the first nationally scoped framework, and empirically validated, the Five-Pillar AI Healthcare Governance Model (FPAHGM), to contextualise AI adoption barriers in Pakistan. Results can be applied to the global debate on AI health equity, providing policy levers to be implemented in the wider LMIC setting. The research recommends timely legislative measures, a national AI health plan, and long-term investment in the IS by the public and the private sector to prevent an increasing digital health divide.