AI-DRIVEN HEALTHCARE MANAGEMENT IN PUBLIC SECTOR HOSPITALS: AN EMPIRICAL PERFORMANCE AND SOCIO-TECHNICAL EVALUATION
DOI:
https://doi.org/10.66021/pakmcr1102Keywords:
Artificial Intelligence; Hospital Administration; Public Sector; Delivery of Health Care; Machine Learning; Decision Support Systems, Clinical; Medical Records Systems, Electronic; PakistanAbstract
Background: Public sector hospitals in developing nations face unprecedented patient overcrowding, operational bottlenecks, and severe resource constraints. While Artificial Intelligence (AI) offers data-driven optimization, empirical research evaluating its integration within resource-limited public health frameworks remains scarce.
Objective: Grounded in Socio-Technical Systems (STS) Theory, this study evaluates the empirical impact of AI-based systems on operational efficiency, triage accuracy, and predictive clinical risk modeling within public sector hospitals.
Method: A mixed-methods research design was deployed across five public sector hospitals in Peshawar, Pakistan, utilizing a curated dataset of 5,000 anonymized electronic health records (EHRs). Supervised machine learning pipelines—specifically Logistic Regression and Random Forest algorithms—were developed, trained, and cross-validated using a 70:30 data split. Social subsystem dynamics were captured qualitatively to analyze human-technology interactions.
Results: Quantitative evaluation demonstrated that AI integration significantly enhanced hospital performance indicators. Triage accuracy increased by 28%, and mean patient waiting times decreased by 22%. The Random Forest model achieved an 87% predictive accuracy (F1-score = 0.86) in identifying high-risk clinical deterioration cases. Furthermore, hospitals recorded a 19% improvement in bed utilization and a 15% reduction in diagnostic errors. Qualitative observations revealed that system efficacy depends directly on institutional technical readiness and user workflow adaptation.
Conclusion: AI-driven healthcare management systems significantly improve operational throughput and clinical precision in low-resource settings. To secure sustainable digital transformations, health policies must balance technical deployment with structured training paradigms for the healthcare workforce.
