A characterization of missingness at random in a generalized shared-parameter joint modeling framework for...

Area of Science:

• Biostatistics
• Statistical Modeling
• Health Data Analysis

Background:

• Joint modeling of longitudinal and time-to-event data is crucial for comprehensive analysis.
• Handling missing data in such models presents significant challenges.
• Existing methods may not fully capture the complexities of data with both types of outcomes.

Purpose of the Study:

• To develop an extended shared random effects joint model.
• To establish a robust characterization of missing at random (MAR) within this new framework.
• To illustrate the utility of the proposed model using a real-world health dataset.

Main Methods:

• Conceptual correspondence between missing data and joint modeling was established.
• An extended shared random effects joint model was formulated.
• The MAR definition was adapted and validated within the joint modeling context.

Main Results:

• The proposed extended joint model provides a unified approach to longitudinal and time-to-event data with missingness.
• A novel characterization of MAR, consistent with traditional missing data theory, was derived.
• The framework demonstrated advantages over conventional joint models in a liver cirrhosis study.

Conclusions:

• The extended joint model offers a more comprehensive statistical framework for analyzing complex health data.
• This approach enhances the understanding and handling of missing data in joint modeling scenarios.
• The findings have implications for improving statistical analyses in clinical and epidemiological research.