Table 4: The suggested cut points, sensitivity, specificity, and accuracy from the different statistical analyses. OLR: ordinary linear regression; MIX: mixed model regression; ROC 1: cut point with the highest product of sensitivity and specificity; ROC 2: cut point with the highest accuracy.

Cut point Sensitivity Specificity Accuracy 3 METs OLR 720 0.923 0.489 0.826 MIX 612 0.936 0.422 0.821 ROC 1 1646 0.795 0.911 0.821 ROC 2 1310 0.878 0.733 0.846 6 METs OLR 5779 0.125 0.984 0.915 MIX 4980 0.375 0.957 0.910 ROC 1 3061 0.938 0.686 0.706 ROC 2 7220 0.063 1.000 0.925