Skip to main content

Advertisement

Table 2 Predicted class memberships of the clinically abnormal biomarkers cut-off values for the 4 latent classes model. Estimated class population shares for the four different LCA classes

From: Metabolic profiles to predict long-term cancer and mortality: the use of latent class analysis

LCA-derived Classes Class 1 Class 2 Class 3 Class 4
% on the population 63% 22% 9% 6%
Biological interpretation Normal Lipids Liver Iron/ Inflammation
ApoB/ApoA-I ≥ 1.00 b 0.1320 0.6840 0.4519 0.2480
Log (Triglycerides/HDL) ≥ 0.50 b 0.0126 0.5436 0.3852 0.1421
Glucose ≥6.11 mmol/L 0.0342 0.2401 0.2174 0.0919
Fructosamine ≥2.60 mmol/L 0.0039 0.0967 0.0555 0.0280
ALT ≥50 IU/L 0.0051 0.0107 1.0000 0.0291
GGT Elevated36–72 IU/L 0.0848 0.2532 0.3521 0.1732
GGT Highly elevated ≥72 IU/L 0.0240 0.0843 0.4098 0.0619
AST ≥ 45 IU/L 0.0052 0.0045 0.3168 0.0180
CRP > 10 mg/L 0.0282 0.0715 0.0771 0.2740
Albumin < 35 g/L 0.0007 0.0022 0.0024 0.0114
Leukocytes ≥109 cells/L 0.0265 0.0786 0.0438 0.1344
Iron low μmol/L 0.0001 0.0040 0.0281 0.5527
Iron high μmol /L 0.0404 0.0155 0.0712 0.0000
TIBC low mg/dL 0.2201 0.2807 0.2622 1.0000
TIBC high mg/dL 0.2438 0.1707 0.2984 0.0000
Creatinine low μmol /L 0.0022 0.0037 0.0041 0.0051
Creatinine high μmol /L 0.0822 0.1765 0.1166 0.1116
Phosphate low mmol/L 0.0078 0.0041 0.0063 0.0098
Phosphate high mmol/L 0.0425 0.0611 0.0544 0.1110
Calcium low mmol/L 0.0124 0.0092 0.0099 0.0458
Calcium high mmol/L 0.0121 0.0253 0.0299 0.0135
  1. aHigh probabilities of the biomarkers to belong to a class are highlighted.
  2. bRatios are dimensionless.
  3. Higher proportions of the abnormal biomarkers belonging to a class is indicated in the table in boldface