Skip to main content
. 2003 Nov;18(11):921–928. doi: 10.1046/j.1525-1497.2003.20909.x

Table 4.

Multiple Variable Logistic Regressions Predicting Unmet Need Among 974 Homeless Women: Separate Regressions With and Without Perceived Barriers

Without Barriers With Barriers
Odds Ratio 95% Confidence Interval Odds Ratio 95% Confidence lnterval
Predisposing Variables
Education (0–12 reference)≥ 13 years 2.04 1.38 to 2.99* 2.06 1.42 to 3.00*
Mental health/substance abuse
 Lifetime history of alcohol abuse 1.22 0.67 to 2.22 1.31 0.74 to 2.31
 Lifetime history of drug abuse 1.65 1.03 to 2.63 1.63 1.04 to 2.56
Homeless history
 Homeless ≥ 1 year 0.41 0.24 to 0.70* 0.39 0.22 to 0.67*
Usual place of stay (shelter, reference)
 Limited housing 2.31 1.12 to 4.75 2.08 0.91 to 4.76
 Street 1.25 0.40 to 3.88 1.40 0.44 to 4.52
 Exit housing 0.96 0.59 to 1.57 0.98 0.60 to 1.60
Enabling variables
 Regular source of care 0.35 0.21 to 0.58* 0.44 0.28 to 0.70*
 Any insurance 0.72 0.44 to 1.16 0.86 0.52 to 1.42
 Any children with her 1.93 1.21 to 3.10§ 1.71 1.06 to 2.74
 Log of income 1.07 0.99 to 1.16 1.09 1.01 to 1.19
Need variables
 Three or more serious symptoms 2.23 1.45 to 3.42* 1.94 1.31 to 2.88*
 Fair/poor health 2.75 1.77 to 4.29* 2.86 1.79 to 4.56*
Perceived barriers
 Not knowing where to go for care 2.27 1.40 to 3.69*
 Long office waiting times 1.89 1.27 to 2.83§
 Too sick to seek care 2.03 1.14 to 3.62

.10

.05

§

.01

*

.001.