An overview of methods to address distinct research questions on environmental mixtures: an application to persistent organic pollutants and leukocyte telomere length

Table 1 Summary results of health models for the unsupervised methods: (K-means clustering, hierarchical clustering, PCA, and EFA)

Variable	β	95% CI	P-value
K-means clustering
Cluster 1 (high exposure)	0.08	0.03, 0.13	0.001
Cluster 2 (medium exposure)	0.05	0.02, 0.09	0.005
Cluster 3 (low exposure)	Reference		—
Hierarchical clustering
Cluster 1 (high exposure)	0.05	0.01, 0.10	0.03
Cluster 2 (medium exposure)	0.03	-0.00, 0.07	0.06
Cluster 3 (low exposure)	Reference		—
Principal Component Analysis
PC1	-0.01	-0.02, -0.01	<0.001
PC2	0.001	-0.01, 0.01	0.87
PC3	0.002	-0.01, 0.02	0.76
Exploratory Factor Analysis
FA1	-0.003	-0.03, 0.03	0.86
FA2	0.03	0.00, 0.05	0.02
FA3	0.03	0.01, 0.05	0.01
FA4	-0.02	-0.04, 0.00	0.06

ISSN: 1476-069X