# Production of a risk mitigation matrix and four-grade threshold scale using Cox Regression to predict osteoarthritis development

## TL;DR

This study uses statistical models to identify biomarker thresholds for predicting osteoarthritis progression and offers a risk classification system for early intervention.

## Contribution

A novel risk mitigation matrix and four-grade threshold scale using Cox Regression for OA risk stratification.

## Key findings

- IL-6 ≥ 6.95 pg/mL, TNF-α ≥ 40.51 pg/mL, and MPO ≥ 5.45 pg/mL indicate early-stage OA.
- Hazard thresholds from the model correlate with advanced disease risk.
- Combining Cox Regression with DFA improved risk stratification accuracy.

## Abstract

Prognostic markers have become essential in predicting disease progression, particularly in osteoarthritis (OA), where early detection can improve clinical outcomes. This study applied survival analysis, specifically the Cox proportional hazards model, to evaluate the role of biomarkers (IL-6, TNF-α, and MPO) in OA progression. A risk mitigation matrix was developed to stratify individuals based on Xbeta values obtained through Cox Regression, allowing for an evidence-based approach to risk classification. A four-grade threshold scale was constructed to refine diagnostic precision by categorising risk into four levels. The study found that IL-6 concentrations ≥ 6.95 pg/mL, TNF-α ≥ 40.51 pg/mL, and MPO ≥ 5.45 pg/mL were indicative of early-stage OA, while hazard thresholds marked advanced disease risk. The risk mitigation matrix aligned with these findings, demonstrating its applicability in clinical decision-making. Integrating Cox Regression with Discriminant Function Analysis (DFA) improved validation, ensuring robust risk stratification. These findings contributed to advancing OA diagnostics by providing quantitative thresholds for early intervention.