Vertical modeling: analysis of competing risks data with a cure fraction.

Area of Science:

• Biostatistics
• Survival Analysis
• Epidemiology

Background:

• Analysis of survival data with competing risks is crucial in many fields.
• Existing models may not adequately account for populations with a proportion of individuals who will never experience certain events (cure fraction).

Purpose of the Study:

• To extend the vertical modeling approach for competing risks survival data to include a cure fraction.
• To develop a method that models the proportion of cure, the overall risk of failure, and the relative risk of specific causes of failure.

Main Methods:

• The proposed method integrates three components: cure proportion, overall failure risk, and cause-specific relative risk.
• Covariates can influence all three components.
• Regression parameters are estimated using the Expectation-Maximization (EM) algorithm.
• The approach is a natural extension of the semi-parametric mixture cure model to competing risks.

Main Results:

• A simulation study was conducted to evaluate the performance of the proposed estimators.
• The method provides a more nuanced analysis compared to existing mixture cure models for competing risks.
• The model was illustrated using a melanoma cancer dataset.

Conclusions:

• The developed vertical modeling approach effectively incorporates a cure fraction into the analysis of competing risks survival data.
• This method offers a flexible and statistically sound framework for understanding disease progression and outcomes in populations with potential immunity to certain events.
• The findings have implications for cancer research and other fields dealing with complex survival outcomes.