Related Experiment Videos
Correlation analysis of twin data with repeated measures based on generalized estimating equations
J S Grove1, L P Zhao, F Quiaoit
1Biostatistics Program, School of Public Health, University of Hawaii, Honolulu 96826.
Genetic Epidemiology
|January 1, 1993
Summary
Repeated measures in twin studies can reveal changes in lab practices. Analysis of male twin triglyceride levels showed shifts in mean and variance across exams.
Area of Science:
- Twin studies
- Biostatistics
- Epidemiology
Background:
- Twin correlation analysis relies on assumptions that can be tested with repeated measures.
- Longitudinal data collection in twin studies can be affected by external factors.
Purpose of the Study:
- To assess the utility of repeated measures in testing assumptions of twin correlation analysis.
- To investigate changes in serum triglyceride levels within individuals over time.
Main Methods:
- Analysis of log serum triglyceride levels in male twins from the National Heart, Lung, and Blood Institute (NHLBI) cohort.
- Application of generalized estimating equations (GEE) to account for repeated measures.
Main Results:
- Significant shifts in both the mean and variance of log serum triglyceride levels were observed across repeated examinations.
- These shifts suggest potential changes in laboratory procedures or environmental factors influencing triglyceride measurements over time.
Conclusions:
- Repeated measures provide a valuable tool for identifying and accounting for systematic changes in longitudinal biometric data.
- Findings highlight the importance of considering laboratory practice variations in twin studies and other longitudinal research.