Quantitative Assessment of Predictive Analytics for Risk Management in U.S. Healthcare Finance Systems

Authors

  • Rifat Chowdhury Executive MS in Data Science, University of the Cumberland, Williamsburg, KY, USA Author

DOI:

https://doi.org/10.63125/x4cta041

Keywords:

Predictive Analytics Capability, Risk Management Effectiveness, Healthcare Finance Systems, Actionability, Governance and Data Integration

Abstract

This study addresses the problem that U.S. healthcare finance workflows on cloud and enterprise platforms face denial, improper payment, and audit exposure risk, while predictive analytics is not always embedded into control decisions. The purpose was to test whether Predictive Analytics Capability (PAC) improves Risk Management Effectiveness (RME) and how Data Quality and Integration (DQI), Governance and Compliance Readiness (GCR), and Actionability (ACT) contribute. A quantitative cross sectional, case-based design surveyed an enterprise revenue cycle and payment integrity environment, producing N = 214 usable responses (89.2% completion). Measures were 1 to 5 Likert composites with strong reliability (Cronbach’s alpha: PAC .88, DQI .84, GCR .86, ACT .85, RME .90). Descriptives indicated high PAC (M = 3.74) and high RME (M = 3.81), alongside moderate data integration readiness (DQI M = 3.46). The analysis plan used descriptive statistics, Pearson correlations, and hierarchical multiple regression with role controls. All predictors were positively associated with RME (PAC r = .62; DQI r = .49; GCR r = .53; ACT r = .58; all p < .001). In regression, adding PAC increased explained variance by ΔR² = .33 (total R² = .39) with a large effect (β = .58, p < .001). The final model explained 51% of variance (R² = .51) and retained independent effects for PAC (β = .31, p < .001), ACT (β = .29, p < .001), GCR (β = .18, p = .006), and DQI (β = .12, p = .041). A risk value map showed the strongest analytics impact in denials and rework (M = 3.94) and fraud or waste or abuse screening (M = 3.86). Pipeline reliability was moderate (PRR M = 3.44) and separated outcomes (mean RME 3.28 at low PRR versus 4.09 at high PRR). Implications are that risk leaders should prioritize workflow actionability, governance traceability, and cross system linkage, and strengthen outcome tracking so predictive signals translate into sustained risk reduction in practice.

Downloads

Published

2025-04-29

How to Cite

Rifat Chowdhury. (2025). Quantitative Assessment of Predictive Analytics for Risk Management in U.S. Healthcare Finance Systems. ASRC Procedia: Global Perspectives in Science and Scholarship, 1(01), 1570–1602. https://doi.org/10.63125/x4cta041

Cited By: