Table 4. Multiple logistic regression analyses of the associations between polychotomous vegetarian status and selected metabolic parameters as outcome variables.

	Outcome	variable
Covariate	MNCEP MetS (n  =  706)	IDF MetS (n  =  706)
Ovo-lacto-vegetarian (1–11 years vs. Non-VEG)	0.55 (0.32–0.92)(0.026)	0.46 (0.26–0.79)(0.006)
Ovo-lacto-vegetarian (≥ 11 years vs. Non-VEG)	0.58 (0.33–0.10)(0.054)	0.43 (0.23–0.76)(0.005)
Pure vegetarian (yes vs. Non-VEG)	0.21 (0.01–1.09)(0.138)	0.22 (0.01–1.15)(0.152)
Age (years)	1.08 (1.05–1.12)(< 0.001)	1.08 (1.04–1.12)(< 0.001)
Menopause (yes vs. no)	0.96 (0.50–1.88)(0.902)	0.84 (0.43–1.67)(0.615)
Activity (× 103 Kcal/day)	1.56 (1.28–1.92)(< 0.001)	1.84 (1.49–2.31)(< 0.001)
Nagelkerke R 2	0.139	0.170
HL GOF test p value	0.112	0.212
AUC	0.730	0.757

The statistics listed in each cell are the estimated adjusted odds ratio, (95% confidence interval of adjusted odds ratio), and (p value) respectively.

MNCEP MetS  =  modified national cholesterol education program criteria of the metabolic syndrome, IDF MetS  =  International Diabetes Federation criteria of the metabolic syndrome; non-VEG  =  non-vegetarian.

Vegetarian duration < 1 year was defined as “non-vegetarian” (Non-VEG) as the reference group. The n of non-vegetarians  =  315, n of ovo-lacto-vegetarians of 1–11 years  =  215, n of ovo-lacto-vegetarians ≥ 11 years  =  154, and n of pure vegetarian  =  22 respectively.

The vegetarian status includes 4 mutually exclusive categories, non-vegetarian, ovo-lacto-vegetarian for 1–11 years, ovo-lacto-vegetarian for ≥11 years, and pure vegetarian, which was defined by three dummy variables with non-vegetarian as the reference category.

HL GOF test: Hosmer-Lemeshow goodness-of-fit test, where p value > 0.05 indicated a good fit of logistic regression model to data.

AUC: Area under receiver operating characteristic curve. AUC of 0.7 and above is considered acceptable discrimination.