Skip to main content
. 2017 Jan 20;12(1):e0170714. doi: 10.1371/journal.pone.0170714

Table 2. Models of County-level Correlates of Higher HIV Rates, Southern Counties with Historically Black Colleges and Universities (HBCUs), American Community Survey, United States, 2014 (n = 69 counties).

Bivariable Models Multivariable Model
Outcome: Log (HIV Diagnosis Rates) Outcome: Log (HIV Diagnosis Rates)
Predictor Estimate (95% CI) p-valuea Estimate (95% CI) p-value
Percent Black 0.02 (0.01, 0.03) <0.0001 0.02 (0.01, 0.02) <0.0001
Percent White -0.02 (-0.03, -0.02) <0.0001
Percent Hispanic 0.01 (-0.01, 0.02) 0.43
Median Household Income -0.000 (—) 0.8178
Unemployment Rate 0.06 (0.01, 0.12) 0.024
Percent Owner-occupied Housing Units -0.06 (-0.07, -0.05) <0.0001 -0.02 (-0.04, -0.01) 0.004
Education 0.01 (-0.06, 0.07) 0.8202
Percent Uninsured 0.004(-0.04, 0.05) 0.8705
Percent Female Head of Household 0.07 (0.05, 0.10) <0.0001 0.01 (0.003, 0.05) 0.026
Percent below Poverty Level 0.03 (0.004, 0.05) 0.0251
Percent Male -0.10 (-0.23, 0.030) 0.133
Urban 0.74 (0.33, 1.14) 0.0005 0.55 (0.31, 0.79) <0.0001
Primary Care Provider Rates 0.01 (0.00, 0.02) 0.0727
Percent Vacant Housing 0.02 (-0.02, 0.05) 0.31

aCandidate variables used in the Multivariable model have bolded p-values. CI = confidence interval.