Association Between Risk Factors and Colorectal Cancer Incidence: A Predictive Analytics Framework.

Colorectal cancer (CRC) is one of the most common cancer types and its increased incidence is related to a number of modifiable risk factors. There is a need for data-driven approaches to support public health policy making for CRC prevention. This paper presents a predictive analytics framework supporting country-level analysis of CRC incidence and risk factors through policy scenarios by linking population-level risk factor exposure to CRC incidence projections across multiple time horizons. The framework integrates Linear Mixed Effects Models, Generalized Additive Models and XGBoost to capture population-level trends and non-linear associations between risk factors and incidence rates. It is intended to serve as the analytical core for scenario-based policy analysis for CRC decision makers. All analyses rely on statistical associations and are designed for exploratory and comparative purposes rather than causal inference or direct policy prescription.