A flexible parametric model for combining current status and age at first diagnosis data.

Area of Science:

• Epidemiology
• Biostatistics
• Chronic Disease Research

Background:

• Cross-sectional studies often collect current disease status and recalled age at diagnosis.
• Analyzing these combined data types presents statistical challenges for understanding disease incidence.
• Existing methods may require restrictive assumptions or fail to handle missing data effectively.

Purpose of the Study:

• To present a flexible parametric modeling approach for analyzing combined current status and age at first diagnosis data.
• To develop methods that accommodate informatively missing data and allow inferences on disease incidence.
• To apply the proposed methods to a real-world dataset, such as uterine fibroids.

Main Methods:

• Utilizing a flexible parametric approach assuming nondecreasing log odds functions of time and covariates.
• Employing piecewise linear models to capture temporal changes in baseline disease odds.
• Incorporating methods for handling informatively missing current status data and landmark event analyses.

Main Results:

• The proposed formulation allows for straightforward maximum likelihood estimation.
• It avoids restrictive parametric or Markov assumptions, offering greater flexibility.
• The methods were successfully applied to analyze uterine fibroids data.

Conclusions:

• The developed flexible parametric approach provides a robust framework for analyzing complex chronic disease data.
• This method enhances the ability to study disease incidence and risk factors using combined data types.
• The application to uterine fibroids demonstrates the practical utility of the approach in epidemiological research.