Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: stir and comparison of mortality rate between 2 groups


From   Hello Hello <pvindustry@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: stir and comparison of mortality rate between 2 groups
Date   Thu, 28 May 2009 19:28:30 -0700 (PDT)

Please bear me if the question is too simple. don’t have much experience doing survival analysis and can not find the answer in the achive 
  
Can I use stir command to compare the mortality rate of 2 groups of patients? Patients had different followup time. 
stir gave me the p-value for testing the incidence rate difference and incidence rate rate ratio. It seems if the incidence rate difference not equal zero, then the different. in my results, the p-value<0.05 (stir) but the CI of the 2 mortality rates (stptime) are overlapping with each other... 
  
 
  
If not, what is the appropriate command to use? 
  
as I know, one test is: 
sts test var, logrank
 Is it the right way to test?
 
 
.. stptime, per(1000) by(x)
     x   |  person-time   failures        rate   [95% Conf. Interval]
-----------+-----------------------------------------------------------
         0 |       5003.8          7   1.3989368   .6669206    2.934418
         1 |       1361.4          7   5.1417658   2.451254     10..7854
-----------+-----------------------------------------------------------
     total |       6365.2         14   2.1994596   1.302635     3.71372
  
  
.. stir x 
 
note:  Exposed <-> x==1 and Unexposed <-> x==0
  
                 |   Exposed   Unexposed  |      Total
-----------------+------------------------+------------
         Failure |         7           7  |         14
            Time |    1361.4      5003.8  |     6365.2
-----------------+------------------------+------------
                 |                        |
  Incidence Rate |  .0051418    .0013989  |   .0021995
                 |                        |
                 |      Point estimate    |    [95% Conf. Interval]
                 |------------------------+------------------------
 Inc. rate diff. |         .0037428       |   -.0002046    .0076903 
 Inc. rate ratio |         3.675481       |    1.100107    12.27986 (exact)
 Attr. frac. ex. |         .7279268       |    .0909976    .9185658 (exact)
 Attr. frac. pop |         .3639634       |
                 +-------------------------------------------------
                     (midp)   Pr(k>=7) =                     0.0103 (exact)
                     (midp) 2*Pr(k>=7) =                     0.0205 (exact) 
  
.. sts test x, logrank
Log-rank test for equality of survivor functions 
      |   Events         Events
x     |  observed       expected
------+-------------------------
0     |         7          10.63
1     |         7           3.37
------+-------------------------
Total |        14          14.00 
            chi2(1) =       5.15
            Pr>chi2 =     0.0232 
  
.. stcox x 
Cox regression -- no ties 
No. of subjects =          409                     Number of obs   =       409
No. of failures =           14
Time at risk    =  6365.199998
                                                   LR chi2(1)      =      4.39
Log likelihood  =   -79.204237                     Prob > chi2     =    0.0362 
------------------------------------------------------------------------------
          _t | Haz. Ratio   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        x |   3.155904   1.687085     2.15   0.032     1.106848    8.998284
------------------------------------------------------------------------------

  
Thanks. 
  
Sara 


      


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index