Assessing individual and population variability in degenerative joint disease prevalence using generalized linear...

Area of Science:

• Bioarchaeology
• Paleopathology
• Statistical Modeling

Background:

• Degenerative joint disease (DJD) is a common condition with implications for understanding past populations.
• Traditional statistical methods for analyzing DJD prevalence in bioarchaeological contexts have limitations in data granularity.

Purpose of the Study:

• To introduce and evaluate generalized linear mixed models (GLMM) as an advanced statistical tool for bioarchaeological research.
• To compare the efficacy of GLMMs against traditional methods in assessing DJD prevalence.

Main Methods:

• Analysis of DJD prevalence in a Spanish population (15th-18th century) focusing on appendicular joints and spine.
• Application of contingency tables, logistic regression, and logistic GLMM for data analysis.
• Utilizing GLMMs to analyze data at the individual bone level.

Main Results:

• GLMM results generally align with traditional statistical methods.
• GLMMs identified DJD associations not apparent with aggregated data analysis.
• Analysis at the individual bone level provided deeper insights into DJD prevalence.

Conclusions:

• GLMMs provide a more nuanced and robust analysis of DJD prevalence in bioarchaeological datasets.
• The ability of GLMMs to handle hierarchical correlations and varied data distributions enhances comparability across studies.
• GLMMs offer unbiased prevalence estimates, improving the accuracy of bioarchaeological interpretations.