TABLE 3.

Multivariable Logistic Regression of Predictive Model Types

Characteristic	Clinical models		Administrative models
	OR	P	OR	P
Experienced discrimination
No	Ref		Ref
Yes	0.85	.27	0.48	.001**
Transparent privacy policy
No	Ref		Ref
Yes	2.35	< .001***	1.68	< .001***
Sex
Male	Ref		Ref
Female	0.74	.01*	0.77	.07
Age in years
18-29	Ref		Ref
30-44	0.75	.25	0.92	.8
45-59	0.51	.01*	0.58	.09
≥ 60	0.67	.12	0.73	.32
Race/ethnicity
White	Ref		Ref
Black	0.88	.47	0.76	.25
Hispanic	1.16	.36	1.27	.22
Asian	1.01	.98	2.05	.07
Other	0.68	.23	1.27	.51
Education
Less than high school	Ref		Ref
High school	1.55	.29	1.48	.41
Some college	1.78	.15	1.15	.76
Bachelor’s degree or more	2.26	.05	1.43	.45
Annual household income
< $50,000	Ref		Ref
≥ $50,000	1.15	.26	0.93	.65
Health insurance coverage
No	Ref		Ref
Yes	1.54	.1	0.96	.88
Self-reported health
Poor to fair	Ref		Ref
Good	0.90	.48	0.87	.46
Very good to excellent	0.84	.29	1.01	.94
Last health care visit
Longer than 1 year	Ref		Ref
Within past year	0.95	.76	1.38	.14
Cancer diagnosis
No	Ref		Ref
Yes	1.25	.2	1.42	.09
Family history of cancer
No	Ref		Ref
Yes	1.09	.45	1.01	.96

Ref, reference.

^*P < .05;

^**P < .01;

^***P < .001.