Algorithmic Justice in Managed Care: Evaluating the Socio-Economic Impact of Machine Learning-Driven Predictive Risk Scoring on Health Equity and Insurance Reimbursement Models under US Healthcare Policy

• Abiodun Okunola

  Ladoke Akintola University Technology

  Author

The integration of machine learning (ML) into managed care risk scoring represents a transformative shift in US healthcare reimbursement, yet its implications for health equity remain inadequately understood. This study investigates whether ML-driven predictive risk models, which increasingly inform care management and insurance reimbursement decisions, systematically disadvantage vulnerable populations while improving predictive accuracy. Through retrospective analysis of 61,850 Medicaid accountable care organization enrollees, this research compares an AI-based risk stratification model incorporating social determinants of health (SDOH) and real-time admission data against the traditional Chronic Illness and Disability Payment System (CDPS) model. The AI model demonstrated superior predictive performance, identifying 41% of highest-cost members compared to 29% for the traditional model, representing $3.7 million in total annual spending . However, this increased accuracy raises algorithmic justice concerns: the model's integration of SDOH proxies may function as discriminatory variables under Section 1557 of the Affordable Care Act, potentially perpetuating disparities for rural, low-income, and disabled populations . This study contributes a validated framework for evaluating algorithmic fairness in managed care and offers policy recommendations for Community Algorithmic Impact Statements and Nursing-Led AI Audit Brigades to ensure ML-driven risk scoring advances rather than undermines health equity.

• PDF

Copyright (c) 2026 Abiodun Okunola (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.