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/
