Table 4.

Full model: two-part model effects between fasting insulin and CRP among adults without diabetes

Covariate	Logit Model		GLM Model		Combine two-part model Predictive margins
	OR	p-value	β	p-value	Predicted CRP mg/L (CI)
Age, years	1.02	.002	0.01	.015
20–39					3.45 (3.17, 3.73)
40–59					4.01 (3.70, 4.33)
60–85					4.40 (3.98, 4.82)
Sex
Male	Reference		Reference		3.67 (3.23, 4.11)
Female	1.62	.046	0.96	<.001	4.09 (3.81, 4.36)
Race
Non-Hispanic White	Reference		Reference		3.76 (3.45, 4.08)
Mexican/other Hispanic	2.12	.027	0.38	.241	4.30 (3.62, 4.99)
Non-Hispanic Black	0.83	.447	0.41	.094	4.54 (4.07, 5.01)
Other race/multi	0.81	.528	0.11	.620	3.32 (2.87, 3.76)
Poverty-income ratio	1.01	.835	− 0.14	.004
< 130 FPR					4.63 (4.20, 5.06)
130–349% FPR					4.09 (3.77, 4.41)
>=350% FPR					3.39 (3.09, 3.69)
Smoking
Non/former smoker	Reference				3.71 (3.39, 4.03)
Current smoker	1.40	.121	1.22	<.001	4.50 (3.88, 5.12)
Physical activity
No	Reference				4.87 (4.39, 5.34)
Yes	1.07	.796	− 0.52	.050	3.52 (3.21, 3.83)
Waist circumference, cm	1.13	<.001	0.07	<.001
Females <=88.9 cm (35 in)					2.53 (2.25, 2.82)
Females > 88.9 cm (35 in)					5.11 (4.75, 5.47)
Males <=101.6 cm (40 in)					2.62 (2.31, 2.93)
Males > 101.6 cm (40 in)					4.93 (4.28, 5.58)
Fasting Insulin, pmol/L	1.00	.307	0.01	.012
< 60					2.94 (2.74, 3.14)
60–119					4.42 (4.05, 4.79)
≥ 120					6.39 (5.28, 7.50)

Mexican/other Hispanic: Mexican or other Hispanic ethnicity; Other race/multi: other race or Multiracial; CRP: C-reactive protein; OR: odds ratio; CI: confidence interval; FPR: federal poverty ratio

Full Model: controlling for age, sex, race, socioeconomic status (poverty-income ratio), smoking status, physical activity, waist circumference and fasting insulin

Two-part regression model used; reference group is the reference group used for each categorical variable

Each level of categorical variable used to provide a predicted CRP based on the model

National estimates based on complex survey design