Skip to main content
. 2016 Apr 11;34(16):1898–1904. doi: 10.1200/JCO.2015.63.3511

Table 2.

Unadjusted Analysis by Time-Varying Model and Conventional Cox Proportional Hazards Regression Analysis

Metformin Use Deaths/Total Median Survival (95% CI)* Time-Varying Model Conventional Cox Model
HR (95% CI) P HR (95% CI)* P
Overall .2819 .0763
 Never 564/614 8.9 (8.0 to 9.6) Ref Ref
 Ever 311/366 9.9 (8.2 to 10.7) 0.93 (0.81 to 1.07) 0.88 (0.77 to 1.01)
Resectable .2249 .0908
 Never 153/187 19.4 (16.7 to 23.5) Ref Ref
 Ever 70/97 22.8 (18.6 to 30.9) 0.84 (0.63 to 1.12) 0.78 (0.59 to 1.04)
Locally advanced .0093 .0058
 Never 211/222 8.1 (6.9 to 9.5) Ref Ref
 Ever 111/132 10.2 (8.9 to 12.9) 0.74 (0.58 to 0.93) 0.72 (0.58 to 0.91)
Metastatic .4370 .7096
 Never 199/204 5.3 (4.6 to 6.1) Ref Ref
 Ever 130/137 4.9 (3.8 to 5.9) 1.09 (0.88 to 1.36) 1.04 (0.84 to 1.30)
Metformin naïve .8889 .5374
 Never 564/614 8.9 (8.0 to 9.6) Ref Ref
 Ever 72/85 10.2 (8.4 to 12.9) 1.02 (0.80 to 1.30) 0.93 (0.72 to 1.18)

Abbreviations: HR, hazard ratio; Ref, reference.

*

Kaplan-Meier method.

Cox proportional hazards regression model.

Log-rank test.