TABLE 2.

Multiple linear regression analyses of the associations between serum cotinine (≥10 ng/mL) and lipid-related indices.

	Independent variable cotinine levels (≥10 ng/mL)a n = 32
Dependent variables	β	SE	P	R 2	f 2
CHOLb	0.001	0.002	0.40	0.12	−
HDL-cb [log]	−0.001	0	0.04*	0.39	0.65**
LDL-cb	0.002	0.001	0.29	0.17	−
TGb [log]	0	0.001	0.48	0.27	−
CRI-Ib [log]	0.001	0	0.03*	0.38	0.62**
ACb [log]	0.001	0.001	0.03*	0.39	0.64**

*P is considered significant at <0.05. Bolded values mean statistically significant results.

^aLevel for cotinine (≥10 ng/mL) were considered reliable indicators for cigarette smoke.

^bDependent variables in separate multiple linear regression equations with covariates including age, diabetes medications, gender, statins, and waist circumference.

AC, atherogenic coefficient; β, regression coefficient; CHOL, total cholesterol; CRI-I, Castelli risk index-I; f², Cohen’s effect size; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; R², coefficient of determination; SE, standard error, TG, Triglycerides.

**f² = R²/1-R² calculated for significant regression models.