Insulin metabolism markers are predictors of subclinical atherosclerosis among overweight and obese children and adolescents

Table 2 Associations of insulin metabolism markers with carotid intima media thickness according to sex

	Multiple linear regression		Multiple ordinal regression^*
	Standardized β	P-value	OR (95% CI)	P-value
Fasting insulin (μU/ml)
Boys
Unadjusted	0.299	< 0.001	1.49 (1.10–2.01)	0.009
Model 1^†	0.351	< 0.001	1.87 (1.24–2.81)	0.003
Model 2^‡	0.344	< 0.001	1.84 (1.22–2.79)	0.004
Girls
Unadjusted	0.010	0.911	1.21 (0.93–1.58)	0.154
Model 1^†	−0.090	0.344	0.93 (0.67–1.29)	0.672
Model 2^‡	−0.116	0.242	0.89 (0.63–1.25)	0.506
HOMA-IR
Boys
Unadjusted	0.308	< 0.001	1.51 (1.12–2.04)	0.006
Model 1^†	0.350	< 0.001	1.81 (1.22–2.70)	0.003
Model 2^‡	0.343	< 0.001	1.80 (1.20–2.68)	0.004
Girls
Unadjusted	−0.006	0.951	1.18 (0.91–1.55)	0.215
Model 1^†	−0.096	0.314	0.91 (0.66–1.27)	0.601
Model 2^‡	−0.122	0.217	0.87 (0.61–1.23)	0.425
QUICKI
Boys
Unadjusted	−0.288	< 0.001	0.82 (0.63–1.06)	0.126
Model 1^†	−0.305	0.001	0.63 (0.42–0.93)	0.022
Model 2^‡	−0.298	0.001	0.64 (0.43–0.95)	0.027
Girls
Unadjusted	−0.034	0.704	0.70 (0.49–1.00)	0.051
Model 1^†	0.051	0.581	0.96 (0.60–1.54)	0.859
Model 2^‡	0.077	0.430	1.00 (0.61–1.64)	0.994

HOMA-IR homeostatic model assessment of insulin resistance, QUICKI quantitative insulin sensitivity check index
^*Multiple ordinal regression was used with carotid intima media thickness as response with three ordered categories
^†Adjusted for body fat mass percent, waist circumference, and physical activity
^‡Additionally adjusted for Tanner stages

ISSN: 1471-2431