A Predictive Business Intelligence Framework for Evaluating the Financial Viability of Accountable Care Organizations (ACOs) within U.S. Public Health Systems

Abstract

The transition from fee-for-service to value-based payment models has positioned Accountable Care Organizations

(ACOs) as central pillars of U.S. public health system reform. However, many ACOs struggle with financial

sustainability due to unpredictable shared savings, high initial infrastructure costs, and complex patient risk

stratification. This study addresses the gap in real-time, predictive tools for assessing ACO financial viability. The

purpose is to design and validate a Predictive Business Intelligence (BI) framework that integrates machine learning

forecasting with operational financial metrics. Using a quantitative, design-based research methodology, we simulate

five years of Medicare Shared Savings Program data across 12 public health system ACOs. Key findings indicate that

a hybrid model combining random forest regression for cost prediction and Monte Carlo simulation for risk-adjusted

revenue forecasting achieves 89.4% accuracy in predicting financial viability 18 months in advance. The conclusion

underscores that predictive BI frameworks can reduce financial uncertainty, improve resource allocation, and enhance

patient outcomes, offering a replicable model for public health administrators.