Skip to main content
. 2014 Sep;24(9):956–962. doi: 10.1016/j.numecd.2014.03.009

Table 2.

Cross-sectional linear regression analysis of the associations of sedentary time with inflammatory variables in the full baseline sample.

Sedentary time
Interaction terms
Men (n = 184)
Women (n = 101)
B (95% CI) P value
β (95% CI) P value β (95% CI) P value
Log sICAM-1 (ng/ml)a
Model 1 0.01 (−0.04, 0.02) 0.642 −0.01 (−0.05, 0.04) 0.751
Model 2 −0.01 (−0.04, 0.02) 0.336 −0.02 (−0.07, 0.03) 0.409 0.03 (−0.01, 0.08) 0.171



Log IL-6 (pg/ml)a
Model 1 0.08 (0.00, 0.15) 0.049 0.12 (0.00, 0.24) 0.043
Model 2 0.07 (−0.01, 0.14) 0.097 0.08 (−0.05, 0.20) 0.228 −0.04 (−0.17, 0.08) 0.483



Log CRP (mg/L)a
Model 1 0.05 (−0.09, 0.19) 0.485 0.16 (−0.05, 0.36) 0.129
Model 2 0.02 (−0.11, 0.16) 0.768 0.06 (−0.14, 0.27) 0.537 −0.25 (−0.47, −0.03) 0.025



Log adiponectin (μg/ml)a
Model 1 0.00 (−0.07, 0.06) 0.937 −0.08 (−0.17, 0.02) 0.110
Model 2 0.00 (−0.06, 0.07) 0.979 −0.05 (−0.15, 0.05) 0.322 0.02 (−0.08, 0.11) 0.724

Regression results are presented as unstandardized coefficients (β) (95% CI) All models are adjusted for age, trial arm, smoking, deprivation score, lipid, glucose or blood pressure lowering medication, accelerometer wear time and MVPA. Model 2 is additionally adjusted for waist circumference. Interaction terms were performed on model 2.

a

Indicates exponentiated regression coefficients interpreted as proportionate change in outcome for 1 unit change in sed time.