一般cox,结局变量都是不利的事件,比如死亡时间,疾病复发时间,等等。如此作出的HR>1;
我做cox时,结局变量都是有益的事件(出院的时间),结果,cox出来的HR都<1,
这个讨论和解释起来本无可厚非,但不符合业界习惯!这样的表达也不被认可...>-< b (解释为该风险,比如吸烟,对出院的贡献为负)
请问各位老师,我该怎样修改代码,能把HR扭过来??使吸烟等风险>1 ??
谢谢老师!我的代码如下(因为是根据help自学的,(T-T)有错误一定要指出啊,谢谢老师!!)
. stset ex, failure( ex ) //ex即结局变量:出院时间
failure event: ex != 0 & ex < .
obs. time interval: (0, ex]
exit on or before: failure
------------------------------------------------------------------------------
1714 total observations
0 exclusions
------------------------------------------------------------------------------
1714 observations remaining, representing
1714 failures in single-record/single-failure data
2845.103 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 0
last observed exit t = 30
. stcox diameter3 blood2 age2 as
No. of subjects = 1114 Number of obs = 1114
No. of failures = 1714
Time at risk = 2845.1029
LR chi2(11) = 293.13
Log likelihood = -1153.947 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
diameter3 | .9261015 .0203681 -3.49 0.000 .8870289 .9668951
blood2 | .9861023 .0053189 -2.59 0.009 .9757323 .9965825
age2 | .9485863 .0184377 -2.72 0.007 .9131287 .9854207
as | .5867303 .0992673 -3.15 0.002 .4211395 .8174309
------------------------------------------------------------------------------