TABLE 5.

IL6 rs10242595 in relation to fat parameters in Caucasian elderly men from the combined MrOS cohorts

Table 5. Linear regression analysis of the associations between rs10242595 SNP and fat parameters in Caucasian elderly men from the combined MrOS cohorts. Data are mean with confidence intervals (CI). P-values are calculated on log-transformed response (tissue fat mass measured by DXA) using linear regression under an additive model. Model I is adjusted for cohort and site, and Model II is adjusted for cohort, site, age, current physical activity, current smoking status, height, and total body lean mass.

Variables	MODEL I	MODEL II	P
	P	Effect size* (CI)
Total fat (kg)	0.005	-0.046 (-0.075 – -0.016)	0.002
Total fat (%)	0.006	-0.049 (-0.081 – -0.016)	0.003
Weight (kg)	0.02	-0.029 (-0.046 – -0.011)	0.002
BMI (kg/m2)	0.8	-0.033 (-0.053 – -0.015)	0.001
Lean body mass (kg)a	0.5	0.031 (0.005 – 0.056)	0.02
Trunk fat (kg)	0.01	-0.041 (-0.070 – -0.011)	0.007
Arm fat (kg)	0.0002	-0.065 (-0.095 – -0.034)	0.00003
Leg fat (kg)	0.006	-0.041 (-0.071 – -0.010)	0.009

^aLean tissue mass not included as a covariate.

^*Effect size (regression coefficient) and SEM values are expressed as change in SD units (CI) per additional A (minor) allele and presented for the P₂ model.