Table 4.

Percentage differences in cortisol decline associated with MetS components

	BP	HDL	Triglyceride	Glucose	WC	% Difference
% Differencea,b	0.42	0.23	−0.47	−0.06	0.31
	0.41	0.34	−0.67	−0.08	0.32
Two componentsc	X			X		−0.35
	X				X	0.35
	X	X				1.30
	X		X			−0.70
				X	X	1.38
		X		X		4.04*
			X	X		0.52
		X			X	−0.73
			X		X	0.19
		X	X			−0.71
Three components		X	X	X		−8.93d*
	X	X	X			0.47
		X	X		X	−0.88
	X	X		X		−
	X			X	X	−2.81
	X		X	X		−0.70
			X	X	X	1.31
	X		X		X	0.28
	X	X			X	−0.19
		X		X	X	1.72
Four components	X	X		X	X	3.99*
	X	X	X	X		2.57
	X		X	X	X	−0.10
	X	X	X		X	−0.05
		X	X	X	X	2.78
Five components	X	X	X	X	X	2.55+

BP, Blood pressure.

^aTop row indicates percentage difference associated with the presence of each individual component adjusted for age, race/ethnicity, gender, SES, physical activity, smoking, wake time, study day, and study site. Each component is included in a separate model.

^bSecond row indicates percentage difference associated with the presence of each individual component in model adjusted for all other components shown plus covariates listed above.

^cRemaining rows show results for specific pairs or clusters of components. The components in the pair or cluster are indicated by “X” in the corresponding columns. The rightmost column indicates the percentage differences in the hourly rate of cortisol decline associated with the presence of the pair or the cluster. For example, in the first row under “Two components,” the hourly cortisol decline was 0.35% lower among persons with both high blood pressure and high glucose compared to those who did not have high levels of both components. In the first row under “Four components” persons who simultaneously had high BP, low HDL, high glucose, and high WC had a 4% flatter decline than those who did not have all four components. Each model is adjusted for main effects of all five MetS components and covariates listed above.

^dCluster contained only one participant.

⁺P < 0.10.

^*P < 0.05.