Analysis of Familial Aggregation Using Recurrence Risk for Complex Survey Data

Area of Science:

• Biostatistics
• Epidemiology
• Statistical Genetics

Background:

• Familial aggregation of diseases is crucial for understanding genetic etiology.
• Recurrence risk is a key measure of family aggregation.
• Household surveys with network sampling are valuable for disease prevalence and recurrence risk estimation.

Purpose of the Study:

• To extend the quadratic exponential model (QEM) for analyzing complex survey data.
• To enable simultaneous estimation and testing of parameters and recurrence risk for multiple family relationships.
• To adjust for confounding by individual-level covariates using propensity score weighting.

Main Methods:

• Extension of composite-likelihood estimation for QEM to complex sample designs.
• Simultaneous estimation and testing for multiple family relationships and covariates.
• Application of propensity score weighting to adjust for confounding.
• Simulation studies to evaluate parameter estimation, variance estimation, and hypothesis testing.

Main Results:

• The extended QEM provides robust estimation for recurrence risk in complex survey data.
• The methods effectively handle complex sample designs and adjust for confounding.
• Simulation results demonstrate good performance of parameter estimation and hypothesis testing.

Conclusions:

• The enhanced QEM is a powerful tool for studying familial disease aggregation and genetic etiology.
• The methodology is applicable to complex health survey data, improving the accuracy of recurrence risk estimation.
• The study provides a framework for more nuanced analysis of disease patterns within families.