Hostility Modifies the Association between TV Viewing and Cardiometabolic Risk

Table 5

Cross-lagged associations from TV viewing to five cardiometabolic risk variables by levels of hostility.

Low hostility	WST 10	HOMA 10^a	TRI 10^a	HDL 10	SBP 10
TV viewing 5
B	0.009	0.046*	−0.079**	0.075**	0.119***
P value	0.63	0.03	0.005	0.002	<0.001
Low hostility	WST 15	HOMA 15^a	TRI 15^a	HDL 15	SBP 15
TV viewing 10
B	0.025	0.055**	−0.051	0.000	0.037
P value	0.14	0.009	0.08	0.99	0.10
Low hostility	WST 20	HOMA 20^a	TRI 20^a	HDL 20	SBP 20
TV viewing 15
B	0.005	0.024	−0.008	−0.056**	0.053*
P value	0.75	0.25	0.78	0.006	0.02
High hostility	WST 10	HOMA 10^a	TRI 10^a	HDL 10	SBP 10
TV viewing 5
B	0.035	0.064**	−0.02	0.065**	0.177***
P value	0.17	0.001	0.48	0.001	<0.001
High hostility	WST 15	HOMA 15^a	TRI 15^a	HDL 15	SBP 15
TV viewing 10
B	−0.032	0.002	−0.064*	−0.086***	−0.006
P value	0.09	0.95	0.03	<0.001	0.85
High hostility	WST 20	HOMA 20^a	TRI 20^a	HDL 20	SBP 20
TV viewing 15
B	−0.006	0.04	0.02	−0.034	0.160***
P value	0.75	0.68	0.51	0.07	<0.001

WST: waist circumference; HOMA: HOMA insulin resistance; TRI: triglycerides; HDL: HDL-cholesterol; SBP: systolic blood pressure. All cross-lagged panel models were adjusted for the same covariates as the clustered cardiometabolic risk model. Regression weights are standardized.
aVariables were log-transformed (natural log).
^∗𝑃<0.05 ; ^∗∗𝑃<0.01; ^∗∗∗𝑃<0.001.