Table 3.

Regression analysis predicting beliefs about e-cigarette harms and benefits by e-cigarette use status, 2016

	Never users (n=1326)			Tried but not in the past 30 days (n=422)			Used at least once in the past 30 days (n=308)
Beliefs about harms	B	95% CI	p	B	95% CI	p	B	95% CI	p
Experimental Condition
Positive (referent)
Negative	0.219*	[0.033,0.405]	0.021	0.028	[−0.288,0.343]	0.863	0.264	[−0.156,0.685]	0.217
Conflicting	0.048	[−0.138,0.233]	0.613	0.023	[−0.303,0.350]	0.889	0.145	[−0.264,0.553]	0.486
Control	0.166*	[0.007,0.325]	0.041	0.042	[−0.233,0.317]	0.766	0.141	[−0.232,0.514]	0.458
Beliefs about benefits	B	95% CI	p	B	95% CI	p	B	95% CI	p
Experimental Condition
Positive (referent)
Negative	−0.324**	[−0.523, −0.125]	0.001	0.008	[−0.306,0.322]	0.958	−0.268	[−0.619,0.084]	0.135
Conflicting	−0.221*	[−0.422, −0.021]	0.03	−0.031	[−0.354,0.291]	0.849	−0.165	[−0.508,0.178]	0.345
Control	−0.210*	[−0.382, −0.038]	0.017	−0.083	[−0.354,0.189]	0.549	−0.265	[−0.579,0.048]	0.097

Notes. Above models controlled for ethnicity.

^*p<0.05,

^**p<0.01.