Skip to main content
. 1999 Oct;14(10):599–605. doi: 10.1046/j.1525-1497.1999.09258.x

Table 3.

Factors Associated with Site of Usual Health Care for Homeless Cohort *

Factor Emergency Department (n= 112) RR (95% CI) No Source of Usual Care (n= 38) RR (95% CI) Shelter-Based Clinic (n= 29) RR (95% CI) Nonspecifiable (n= 8) RR (95% CI)
Uninsured 3.10 (1.8, 5.5) 3.95 (1.8, 8.5) 3.21 (1.3, 7.7) 1.85 (0.3, 10.41)
Unmarried  2.94 (1.0, 8.3)  1.20 (0.4, 4.0)  3.12 (0.4, 24.9) 0.93 (0.1, 8.6)
Homelessness >2 years  2.37 (1.2, 4.5)  1.42 (0.5, 4.0)  2.95 (1.2, 7.4) 1.58 (0.2, 14.9)
No health care past 6 mo  2.12 (1.2, 3.8)  7.48 (1.7, 32.9)  1.57 (0.6, 3.9) 4.55 (0.5, 39.5)
Nonveteran status  1.92 (1.1, 3.4)  2.49 (1.0, 6.0)  0.97 (0.4, 2.2) 0.90 (0.1, 6.3)
Race (non–African American)  1.00 (0.5, 1.9)  1.59 (0.6, 3.9)  3.27 (1.4, 7.6) 3.97 (0.9, 17.7)
Gender (male)  0.94 (0.4, 2.0)  0.29 (0.1, 1.3)  0.49 (0.1, 2.4) 5.62 (0.9, 34.0)
*

Items in boldface were significant in the polychotomous logistic regression. The comparison group was traditional ambulatory care sites. RR indicates relative risk; CI, confidence interval.

The overall pseudo-R2for the model was .10.