Proportional hazards model for competing risks data with missing cause of failure

Area of Science:

• Biostatistics
• Survival Analysis
• Medical Data Analysis

Background:

• Competing risks data present challenges in causal inference.
• Missing cause of failure is a common issue in medical studies.
• Semiparametric proportional hazards models are widely used for survival data.

Purpose of the Study:

• To develop robust statistical methods for analyzing competing risks data with missing failure causes.
• To estimate regression parameters in cause-specific hazard models under missing data.
• To evaluate the performance of proposed estimators through simulations.

Main Methods:

• Utilizing semiparametric proportional hazards models for cause-specific hazards.
• Proposing inverse probability weighted (IPW) and augmented inverse probability weighted (AIPW) estimating equations.
• Establishing theoretical properties for statistical inference.
• Conducting simulation studies to assess estimator performance.

Main Results:

• The augmented inverse probability weighted estimator demonstrates double robustness.
• The proposed methods are suitable for practical application in survival analysis.
• Comparison with multiple imputation methods shows the efficacy of the proposed estimators.
• The methods are illustrated using bone marrow transplant data.

Conclusions:

• The developed IPW and AIPW methods provide reliable tools for analyzing competing risks data with missing failure information.
• The AIPW estimator offers robustness, making it a valuable approach for biostatistical analysis.
• The findings support the practical utility of these methods in real-world medical research, such as in bone marrow transplant studies.