第一次做COX回归分析的时候,把包括年龄在内的一些影响因素都设置成了协变量,结果如下:
Omnibus Tests of Model Coefficients(c)
Overall (score) Change From Previous Step Change From Previous Block
Step -2 Log Likelihood Chi-square df Sig. Chi-square df Sig. Chi-square df Sig.
1a 24.146 15.141 1 .000 15.553 1 .000 15.553 1 .000
2b 19.808 16.711 2 .000 4.338 1 .037 19.891 2 .000
a. Variable(s) Entered at Step Number 1: age
b. Variable(s) Entered at Step Number 2: PTH
c. Beginning Block Number 1. Method = Forward Stepwise (Conditional LR)
然后又把年龄剔除了,结果如下:
Omnibus Tests of Model Coefficients(d)
Overall (score) Change From Previous Step Change From Previous Block
Step -2 Log Likelihood Chi-square df Sig. Chi-square df Sig. Chi-square df Sig.
1a 35.902 5.282 1 .022 3.797 1 .051 3.797 1 .051
2b 30.880 8.894 2 .012 5.023 1 .025 8.820 2 .012
3c 24.969 12.523 3 .006 5.911 1 .015 14.731 3 .002
a. Variable(s) Entered at Step Number 1: hsCRP
b. Variable(s) Entered at Step Number 2: AASI
c. Variable(s) Entered at Step Number 3: ARV
d. Beginning Block Number 1. Method = Forward Stepwise (Conditional LR)
这种结果提示什么?有意义的变量完全不一样,这是说明,hsCRP、AASI、ARV对结果产生影响不独立于年龄存在么?那为什么第二次分析的变量中又没有PTH了?
谢谢各位前辈!