Table 3

Linear regression of VA

Parameter	Estimate	SE	t Value	p Value	Adj. R²	F	DF	Model P
Cross-sectional model, regression of VA quadratic in CA
Age	1.1622	0.0886	13.122	<0.0001	0.4267	80.88	6.638	<0.0001
(Age)²	1.1482	0.0917	12.519	<0.0001
(Age)²: tobacco	0.5994	0.2300	2.606	0.0094
(Age)²: amphetamine	0.8288	0.3582	2.314	0.0210
Age: tobacco	0.2074	0.0954	2.173	0.0301
Parameter	Parameter values					Model parameters
Parameter	Value	SE	DF	t Value	p Value	AIC	BIC	LogLik
Longitudinal model, regression of VA quadratic in CA
Age	1.2181	0.0835	205	14.5874	<0.0001	395.0924	437.9048	−188.5462
(Age)²	1.2253	0.0822	205	14.9025	<0.0001
Age: amphetamine	0.2488	0.1074	205	2.3174	0.0215
Longitudinal model, regression of VA/CA, cubic in CA
(Age)²: BMI	−0.2086	0.0401	164	−5.1965	<0.0001	391.9747	536.0531	−164.9874
(Age)³: BMI	−0.1147	0.0347	164	−3.3023	0.0012
BMI	0.2725	0.0882	164	3.0915	0.0023
(Age)²: Days: BMI: Amphetamine	−2.1703	0.8548	164	−2.5391	0.0120
Days: Amphetamine	3.1828	1.2905	164	2.4664	0.0147
Age: tobacco	7.2217	2.9621	164	2.4380	0.0158
Age: BMI: tobacco	−2.2018	0.9296	164	−2.3686	0.0190
(Age)³: Days: BMI: Amphetamine	−0.7505	0.3333	164	−2.2519	0.0257
(Age)³: days: BMI: non-smokers	0.0130	0.0060	164	2.1683	0.0316

AIC, akaike information criterion; BIC, bayesian information criterion; BMI, body mass index; CA, chronological age; DF, degrees of freedom; LogLik, Log-Likelihood; VA, vascular age.