Work

From Models to Management: A Framework for Predictive Analytics in Health Systems

Year: 2025

Type: article

Abstract: Health systems increasingly deploy predictive analytics to improve patient outcomes and operational performance, yet many projects stall at the interface between model output and managerial action. This review looks at real-world deployments connecting clinical prediction with market and operational levers, staffing, bed flow, outreach, and scheduling, and distills an integration framework spanning data architecture, model selection, real-time pipelines, governance, and evaluation. Evidence emphasizes measurable effects on process and, in selected contexts, outcomes when models are embedded in event-driven workflows and governed with clear decision rights, calibration monitoring, and explainability support. Because much of the empirical literature originates in the United States, generalizability is assessed using compact international implementations (United Kingdom stroke AI, Singapore’s C3 command center, and India’s TB computer-aided CXR triage). The review argues that impact depends less on algorithmic novelty than on socio-technical integration: reliable data plumbing, execution discipline, and incentives aligned to net clinical and operational utility

Source: Journal of Medical Science Biology and Chemistry

DOI: 10.69739/jmsbc.v2i2.1149

Authors: Chizube Obinna Chikezie, Chibuzo Okechukwu Onah, Chukwudi Anthony Okolue, Mary-Jane Ezinne Ugbor, Emmanuel Fagbenle, Olukunle Akanbi

Institutions: University of Ulster, Georgia State University, Vanderbilt University, University of Port Harcourt, University of New Hampshire, National Louis University

Language: English

Topic: Artificial Intelligence in Healthcare

Subfield: Health Information Management

Field: Health Professions

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: hybrid

OpenAlex ID: W4416245142

FWCI:

Cites:

Cited by:

Related to: