Table 2.

Multiple linear regression models with cognitive domains as dependent variables.

Effect	Estimate	SE	95% CI		P value
			LL	UL
Model 1: Global functiona
Intercept	0.225	0.733	−1.229	1.679	0.760
Glucose variabilityb	0.093	0.259	−0.420	0.606	0.720
Age	−0.015	0.004	−0.024	−0.007	0.000
Sex	0.115	0.114	−0.112	0.342	0.319
Education	0.022	0.020	−0.018	0.061	0.285
Group	–0.267	0.119	−0.503	−0.031	0.027
Model 2: Working memorya
Intercept	0.205	0.880	−1.541	1.951	0.816
Glucose variabilityb	0.147	0.311	−0.469	0.764	0.636
Age	–0.016	0.005	−0.026	−0.006	0.002
Sex	0.104	0.138	−0.169	0.376	0.452
Education	0.017	0.024	−0.031	0.065	0.478
Group	−0.316	0.143	−0.600	−0.032	0.029
Model 3: Executive functiona
Intercept	0.455	0.925	−1.380	2.290	0.624
Glucose variabilityb	−0.160	0.327	−0.808	0.488	0.626
Age	−0.009	0.005	−0.020	0.001	0.072
Sex	0.168	0.144	−0.119	0.454	0.248
Education	0.029	0.025	−0.022	0.079	0.260
Group	−0.216	0.150	−0.514	0.082	0.154

CI confidence interval, LL lower limit, UL upper limit, SE standard error.

N = 110 (Insulin-sensitive = 57, Insulin-resistant = 53). Sex (0 = male, 1 = female), Group (0 = insulin-sensitive, 1 = insulin-resistant).

^aScores were z-standardized for better interpretation.

^bGlucose variability was measured by the coefficient of variation (CV) which was log-transformed to correct right-skewed data.