On estimating the relationship between longitudinal measurements and time-to-event data using a simple two-stage...

Area of Science:

• Biostatistics
• Longitudinal Data Analysis
• Survival Analysis

Background:

• Joint modeling and regression calibration are used for longitudinal and time-to-event data.
• Ye et al. (2008) proposed a two-stage regression calibration method.
• This method may introduce bias due to informative dropout and measurement error.

Purpose of the Study:

• To demonstrate substantial bias in Ye et al.'s (2008) regression calibration approach.
• To propose an alternative regression calibration method to alleviate bias.
• To validate the new approach through simulations.

Main Methods:

• Developed an alternative two-stage regression calibration approach.
• Applied the method to both discrete and continuous time-to-event data.
• Conducted simulations to compare bias with Ye et al.'s (2008) method.

Main Results:

• The proposed regression calibration approach significantly reduces bias compared to Ye et al. (2008).
• The method is applicable to both discrete and continuous time-to-event data.
• Simulations confirmed the substantial bias in the original method.

Conclusions:

• The revised regression calibration approach offers a less biased alternative for analyzing longitudinal and time-to-event data.
• The proposed method is implementable with standard statistical software.
• This approach avoids complex estimation techniques required by joint modeling.