Using the general linear mixed model to analyse unbalanced repeated measures and longitudinal data

Area of Science:

• Statistics
• Biostatistics
• Data Analysis

Background:

• General linear mixed models (GLMMs) are powerful tools for handling complex data structures.
• Unbalanced repeated measures and longitudinal data present analytical challenges.
• Advances in statistical methods and software have made GLMMs more accessible.

Purpose of the Study:

• To introduce the theory behind general linear mixed models.
• To provide guidelines for model fitting and checking.
• To illustrate GLMM application through case studies.

Main Methods:

• The tutorial focuses on the theory and practical application of general linear mixed models.
• It emphasizes model specification and goodness-of-fit assessment.
• Case studies involve a clinical trial and an epidemiological survey.

Main Results:

• General linear mixed models accommodate correlated observations, offering flexibility beyond traditional regression.
• The tutorial demonstrates effective strategies for fitting and validating mixed models.
• Case studies highlight the model's utility in real-world scenarios with dropouts and longitudinal follow-up.

Conclusions:

• General linear mixed models are a valuable and flexible statistical approach for complex data.
• Proper model specification and checking are crucial for reliable analysis.
• The tutorial provides practical insights for statisticians using mixed models in research.