Related Experiment Video
Updated: Aug 7, 2026
05:37
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
Improved risk prediction via cross-domain calibration in a retrospective case-control study
Ge Zhao1, Yanyuan Ma2, Yaqi Cao3,4
1Department of Mathematics and Statistics, Portland State University, Portland, U.S.
The Canadian Journal of Statistics = Revue Canadienne De Statistique
|August 6, 2026
Summary
This study introduces a constrained maximum likelihood method to improve risk prediction by incorporating biomarker data from different populations. The approach ensures accurate predictions without fully modeling the biomarker, enhancing breast cancer risk assessment.
Area of Science:
- Biostatistics
- Epidemiology
- Medical Informatics
Background:
- Standard risk prediction models often lack biomarker integration, especially when data originates from different populations.
- Accurately modeling biomarkers from external populations presents significant statistical challenges.
- Improving predictive accuracy in disease risk assessment is crucial for effective public health interventions.
Purpose of the Study:
- To propose a novel constrained maximum likelihood approach for integrating external biomarker data into existing risk prediction models.
- To develop a method that enhances prediction accuracy without requiring a complete model of the biomarker.
- To address challenges posed by differing distributions of standard risk predictors between source and target populations.
Main Methods:
- A constrained maximum likelihood estimation framework was developed.
- The method incorporates biomarker data sampled from a different population.
- Constraints were implemented to ensure the averaged risk with the biomarker approximates that of the standard model.
- Large sample theory was derived for parameter estimation and predictive accuracy measures.
Main Results:
- The proposed method effectively incorporates biomarker data from external populations.
- The approach allows for differing distributions of standard risk predictors.
- Simulation studies confirmed the finite sample performance of the method.
- Application to a breast cancer study demonstrated the utility of including mammographic density as a biomarker.
Conclusions:
- The constrained maximum likelihood approach offers a robust strategy for improving risk prediction models using external biomarker data.
- This method is valuable when biomarker distributions differ across populations or when full biomarker modeling is infeasible.
- The approach has practical implications for enhancing disease risk prediction, as shown in the breast cancer case study.
Related Concept Videos
01:12Relative Risk
Relative risk (RR) is a statistical measure commonly used in epidemiology to compare the likelihood of a particular event occurring between two groups. This metric is important for evaluating the relationship between exposure to a specific risk factor and the probability of a particular outcome. It plays a crucial role in medical research, public health studies, and risk assessment. Relative risk quantifies how much more (or less) likely an event is to occur in an exposed group compared to an...
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Biopharmaceutical studies constitute a vital field aiming to enhance drug delivery methods and refine therapeutic approaches, drawing upon diverse interdisciplinary knowledge. In research methodologies, the choice between controlled and non-controlled studies significantly influences the study's reliability and accuracy.
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
01:29Bias in Epidemiological Studies
Biases can arise at various stages of research, from study design and data collection to analysis and interpretation. Recognizing and addressing these biases is essential to ensure the validity and reliability of epidemiological findings.Broadly speaking, biases in epidemiology fall into three main categories: selection bias, information bias, and confounding. A more detailed description of possible biases is: