Skip to main content
. 2018 Oct 3;6:e5740. doi: 10.7717/peerj.5740

Table 4. Cox proportional-hazards regression analysis of factors affecting LOS in patients with burns.

Variable Variable assignment B SE Wald P-value HR (95%CI)
Age Continuous variables −0.008 0.002 10.461 0.001 0.993 (0.988–0.997)
Gender 1 = Male; 0 = Female 0.210 0.094 4.976 0.026 1.234 (1.026–1.484)
Degree of burn Dummy variable 66.404 <0.001
 Superficial partial thickness 1.00
 Deep partial thickness 1.140 0.146 60.851 <0.001 3.128 (2.349–4.166)
 Full thickness 0.583 0.134 19.037 <0.001 1.791 (1.379–2.327)
TBSA Dummy variable 81.342 <0.001
 <10% 1.00
 10–29% 1.381 0.227 37.078 <0.001 3.978 (2.551–6.205)
 ≥30% 0.581 0.231 6.292 0.012 1.787 (1.135–2.813)
Etiology 1 = Heat; 0 = non-heat* 0.404 0.118 11.770 0.001 1.497 (1.189–1.886)
Skin grafting 1 = Yes; 0 = No −1.083 0.147 54.553 <0.001 0.339 (0.254–0.451)
Inhalation burn 1 = Yes; 0 = No 0.073 0.204 0.127 0.721 1.076 (0.721–1.605)
Complications 1 = Yes; 0 = No −0.199 0.162 1.518 0.218 0.819 (0.597–1.125)
Medical treatment ≤24 h 1 = Yes; 0 = No −0.032 0.112 0.083 0.773 0.968 (0.777–1.207)

Notes:

Gender, TBSA, degree of burn and etiology were risk factors for LOS, while age and skin grafting were protective factors of LOS.

B, regression coefficients; SE, standard error; HR, hazard ratio; CI, confidence interval.

*

Heat etiology including hot liquid, fire, hot solids, and hot gases; all other factors were non-heat etiology.