Table 3.

Multivariate logistic regression models of tea consumption and diabetic retinopathy.

	Number of		Model 1a			Model 2b			Model 3c
	DR	Non-DR	B	OR (95% CI)	p	B	OR (95% CI)	p	B	OR (95% CI)	p
Duration of tea consumption (y)	—	—	-0.02	0.98 (0.96–1.00)	0.059	-0.03	0.97 (0.95–1.00)	0.041	-0.03	0.97 (0.95–1.00)	0.046
Duration of tea consumption group
Non-tea consumption	39	292	0	1	Ref	0	1	Ref	0	1	Ref
1–19 years	9	41	0.55	1.74 (0.77–3.91)	0.183	0.44	1.55 (0.66–3.65)	0.318	0.57	1.76 (0.65–4.77)	0.263
≥20 years	6	133	-0.95	0.39 (0.15–1.00)	0.051	-0.99	0.37 (0.14–0.97)	0.042	-1.22	0.29 (0.09–0.97)	0.044
Frequency of tea consumption
Non-tea consumption	39	292	0	1	Ref	0	1	Ref	0	1	Ref
1–5 times/week	3	16	-0.39	0.68 (0.32–1.45)	0.319	-0.54	0.58 (0.27–1.28)	0.179	-0.7	0.50 (0.19–1.29)	0.15
>5 times/week	12	158	0.37	1.44 (0.40–5.19)	0.576	0.4	1.48 (0.41–5.41)	0.549	0.86	2.37 (0.57–9.84)	0.236
Tea type
Non-tea consumption	39	292	0	1	Ref	0	1	Ref	0	1	Ref
Green tea	13	155	-0.26	0.77 (0.37–1.61)	0.487	-0.39	0.68 (0.32–1.45)	0.315	-0.31	0.74 (0.30–1.79)	0.5
Non-green tea	2	19	-0.15	0.86 (0.19–3.88)	0.847	-0.19	0.83 (0.18–3.76)	0.806	-0.79	0.46 (0.05–4.01)	0.479

Note. DR: diabetic retinopathy; OR: odds ratio; 95% CI: 95% confident interval. ^aModel 1 was adjusted for age and gender. ^bModel 2 was adjusted for age, gender, and covariates with p < 0.2 in univariate analysis in Table 3, including individual monthly income, FBG, and SBP. ^cModel 3 was adjusted for age, gender, and covariates with p < 0.5 in univariate analysis in Table 3, including occupation, educational level, individual monthly income, smoking, alcohol consumption, duration of diabetes, BMI, FBG, TG, HDL-C, and SBP. Bold type indicates statistical significance (p < 0.05).