Skip to main content
. 2012 Oct 15;7(10):e47339. doi: 10.1371/journal.pone.0047339

Table 4. Univariate and multivariate analyses of outcomes.

Outcomes
Quality of Life, Burn Specific HealthScale – Brief (BSHS-B) Depressive Symptoms, Beck Depression Inventory (BDI) Post-Traumatic Stress Symptoms, Impact of Event Scale-Revised (IES-R) Employment, Time off work
Variable Coefficient (95% CI) p-value Coefficient (95% CI) p-value Coefficient (95% CI) p-value Odds Ratio (95% CI) p-value
Univariate analysis
Burn Injury −4.50 (−6.45, −2.55) <0.001 5.63 (1.31, 9.95) 0.01 6.61 (−1.90, 15.12) 0.13 7.67 (2.73, 21.56) <0.001
Constant 33.66 (32.26, 35.07) <0.001 8.52 (5.56, 11.48) <0.001 25.24 (19.40, 31.08) <0.001
Multivariate analysis
Burn Injury −1.17 (−4.79, 2.44) 0.51 −0.28 (−5.53, 4.96) 0.91 1.58 (−10.78, 13.94) 0.79 4.04 (1.11, 14.63) 0.03
Gender −1.02 (−4.43, 2.40) 0.54 2.43 (−2.41, 7.27) 0.31 −4.97 (−16.46, 6.51) 0.38 0.83 (0.25, 2.77) 0.76
Age 0.01 (−0.23, 0.24) 0.94 −0.04 (−0.38, 0.30) 0.80 0.22 (−0.57, 1.01) 0.57 1.08 (0.98, 1.20) 0.12
Race −8.29 (−18.45, 1.87) 0.10 −9.65 (−24.58, 5.27) 0.19 −2.63 (−39.17, 33.91) 0.88 0.44 (0.02, 12.71) 0.63
Marriage 2.45 (−1.17, 6.06) 0.18 3.59 (−1.60, 8.79) 0.17 −1.52 (−14.30, 11.25) 0.81 0.61 (0.18, 1.99) 0.41
Pre-fire employment −0.68 (−5.13, 3.78) 0.76 −0.46 (−6.88, 5.96) 0.88 4.21 (−10.83, 19.25) 0.57 0.86 (0.19, 3.82) 0.84
Number of children −0.19 (−1.72, 1.33) 0.80 −1.53 (−3.63, 0.57) 0.14 3.91 (−1.00, 8.82) 0.11 0.79 (0.46, 1.34) 0.38
Time off work −2.34 (−5.98, 1.29) 0.20 −1.51 (−6.91, 3.88) 0.57 −1.85 (−14.64, 10.95) 0.77
Post-fire employment −0.85 (−4.08, 2.39) 0.59 −0.81 (−5.48, 3.86) 0.72 2.14 (−8.88, 13.16) 0.69
BDI −0.39 (−0.64, −0.14) 0.004 1.74 (1.05, 2.43) <0.001 1.02 (0.92, 1.14) 0.66
IES-R 0.03 (−0.09, 0.16) 0.58 0.31 (0.19, 0.44) <0.001 0.96 (0.92, 1.01) 0.14
BSHS-B −0.80 (−1.32, −0.29) 0.004 0.40 (−1.06, 1.85) 0.58 0.87 (0.73, 1.03) 0.11
Constant 44.78 (32.01, 57.55) <0.001 39.89 (11.37, 68.42) 0.008 −15.60 (−93.92, 62.72) 0.68
a

In the univariate analysis an empty cell occurs because the logistic regression analysis for the binary outcome employment does not produce a constant. In the multivariate analysis empty cells occur where the outcome and variable are the same.