iCARE

iCARE estimates individual absolute risk of developing a disease over a specified time horizon by combining relative-risk models, age-specific incidence rates, and target-population risk-factor distributions.

Key Features:

• Implementation: Implemented as an R package for constructing and evaluating absolute risk models.
• Relative-Risk Model Input: Accepts a model for relative risk specifying effect estimates for risk factors.
• Age-Specific Incidence Rates: Uses age-specific disease incidence rates as an input to absolute risk calculations.
• Risk-Factor Distribution: Incorporates the distribution of risk factors in the target population as an input.
• Missing-Data Handling: Manages missing individual risk-factor data via a coherent modeling approach with model averaging so all estimates derive from a unified model.
• SNP Integration: Integrates single nucleotide polymorphisms (SNPs) using published odds ratios and allele frequencies.
• Validation Metrics: Evaluates calibration, discrimination, and risk-stratification using independent validation datasets.
• Absolute Risk Estimation: Computes individual absolute risk over specified time intervals.

Scientific Applications:

• Individualized Absolute Risk Prediction: Estimates an individual's probability of developing a disease within a specified time frame by combining genetic and non-genetic risk factors.
• Breast Cancer Risk Modeling: Demonstrated application to breast cancer for building, validating, and applying comprehensive absolute risk models.

Methodology:

Uses three explicit inputs (relative-risk model, age-specific incidence rates, and target-population risk-factor distribution); manages missing data via a coherent modeling approach with model averaging; integrates SNPs via published odds ratios and allele frequencies; and assesses model calibration, discrimination, and risk-stratification on independent validation datasets.