Table 3: Survey logistic regression models for the association between outpatient visits and OTC consumption.*

Effect		Reference	Model 1	Model 2	Model 3
OTC consumption		No OTC	0.62 (0.55,0.70)	0.61 (0.54,0.69)	0.45 (0.39,0.53)
Gender		Female		1.41 (1.24,1.60)	1.05 (0.89,1.25)
Age				1.01 (1.00,1.02)	1.00 (0.99,1.01)
Education level	Primary school	No school		1.19 (1.02,1.38)	1.13 (0.98,1.29)
	Middle school			1.16 (1.00,1.35)	1.16 (0.98,1.36)
	College and above			1.27 (0.79,2.05)	1.35 (0.81,2.25)
Marital status	Others	Married		1.08 (0.92,1.27)	1.03 (0.87,1.23)
Place of residence	Urban	Rural		1.05 (0.89,1.24)	1.13 (0.93,1.38)
Drinking	More than once/month	Never			0.66 (0.54,0.81)
	less than once/month				0.98 (0.74,1.30)
Smoking	Smoking	Never			0.86 (0.74,1.01)
	Quitted				1.14 (0.94,1.39)
Income	1500-12500	< 1500			0.94 (0.81,1.09)
	12500-35000				1.03 (0.88,1.22)
	> 35000				1.16 (0.95,1.41)
Insurance	Urban employee medical insurance	No insurance			1.32 (0.93,1.86)
	Urban resident medical insurance				0.97 (0.73,1.29)
	New cooperative medical insurance				1.14 (0.85,1.53)
Self-evaluated health	Fair	Good and above			1.59 (1.37,1.86)
	Poor and worse				2.66 (2.17,3.25)
N of chronic diseases					1.33 (1.25,1.42)
CESD					1.03 (1.02,1.04)

^*Model 1 was a univariate model; Model 2 adjusted for gender, age, education level, marital status, and place of residence (urban/rural); Model 3 also included household incomes, life style (smoking, drinking), self-reported health status, number of chronic diseases and depression.