Table 4.

Predictors of Peripheral Arterial Disease

Variable	Univariate	Full Model^*		Subset of Full Model^†		Reduced Model^‡
Variable	OR (95% CI)	OR (95% CI)	P Value	OR (95% CI)	P Value	OR (95% CI)	P Value
Gender (women/men)	1.12 (0.66, 1.89)	0.70 (0.34, 1.44)	.33	0.89 (0.49, 1.59)	.69
Education (<High School/ ≥High School)	0.76 (0.45, 1.29)	0.49 (0.21, 1.13)	.09
Currently married (y/n)	0.78 (0.46, 1.32)	1.06 (0.51, 2.24)	.87
Income (<$5000/≥$5000)	1.23 (0.61, 2.48)	0.94 (0.39, 2.27)	.89
Current work status (n/y)	2.06 (1.00, 4.25)	2.52 (0.92, 6.92)	.07			2.67 (1.04, 6.86)	.04
Antiplatelet (y/n)	1.36 (0.79, 2.33)	1.00 (0.47, 2.15)	.99	1.00 (0.55, 1.83)	.99
Arthritis (y/n)	1.16 (0.69, 1.97)	1.13 (0.54, 2.36)	.74	1.09 (0.61, 1.95)	.77
High cholesterol (y/n)	0.97 (0.57, 1.64)	0.90 (0.44, 1.84)	.77	0.89 (0.50, 1.57)	.68
Diabetes (y/n)	2.34 (1.37, 3.98)	3.52 (1.67, 7.41)	.0009	2.64 (1.47, 4.73)	.001	3.42 (1.68, 6.95)	.0007
Hypertension (y/n)	2.32 (1.20, 4.51)	1.27 (0.26, 6.12)	.77	0.64 (0.19, 2.22)	.48
MI (y/n)	1.59 (0.76, 3.30)	1.35 (0.51, 3.55)	.55	1.36 (0.60, 3.05)	.46
Stroke (y/n)	1.98 (0.78, 4.83)	0.80 (0.25, 2.61)	.71	1.64 (0.62, 4.37)	.32
BP medication (y/n)	3.02 (1.52, 5.97)	2.36 (0.54, 10.42)	.26	3.30 (0.94, 11.51)	.06	2.69 (1.14, 6.37)	.02
Current smoker (y/n)	2.13 (1.17, 3.86)	2.63 (1.17, 5.89)	.02	2.83 (1.45, 5.55)	.002	2.37 (1.11, 5.06)	.03
Ever smoker (y/n)	1.77 (1.01, 3.09)
Race
black/white		1.59 (0.67, 3.79)	.30	1.57 (0.79, 3.15)	.20
Latino/white		1.85 (0.62, 5.55)	.27	1.01 (0.47, 2.20)	.98
Age (continuous)		1.03 (0.98, 1.08)	.25	1.03 (0.99, 1.07)	.12
BMI (continuous)		0.93 (0.88, 1.00)	.04			0.94 (0.88, 0.99)	.03

* Full model contains the following variables: gender, education, married, income, current work status, antiplatelet, arthritis, high cholesterol, diabetes, hypertension, myocardial infarction (MI), stroke, blood pressure (BP) medication, current smoker, race (using dummy variables with white as referent), age, and body mass index (BMI). There are missing data for BMI (49 observations), work status (56 observations), and income (24 observations)
† Subset of full model contains the following variables: antiplatelet, arthritis, high cholesterol, diabetes, hypertension, MI, stroke, BP medication, current smoker, race (using dummy variables with white as referent), and age.
‡ Reduced model is built using stepwise procedure with variables having P ≤ .20 entering the model and those with P ≤ .05 remaining in the model. Order of variables entered: diabetes, current smoking, age, current work, age removed, BP medication entered, BMI entered, education entered, and then education removed.