Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Multiple comparison of inverse probability weighted survival curves

Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Posted 07-23-2019 02:49 PM
(954 views)

Hello,

I am using SAS 9.4 TS1M5 and have not been able to obtain a multiple comparison of inverse probability weighted Kaplan Meier failure curves for my three exposure groups using the adjust=sidak option in proc lifetest. The multiple comparison test ran successfully when I did not include the inverse weight variable, see code1:

```
code1:
proc lifetest data= bb notable plots=survival( nocensor failure test );
where 66 <= ageatdiag < 75;
strata pdc2 / test=logrank adjust=sidak ;
time lenfol*death365day(0);
run;
```

However, when I added the weight statement to the code, see code2 below, the output did not include the multiple comparison output :

```
proc lifetest data= bb notable plots=survival( nocensor failure test );
where 66 <= ageatdiag < 75;
strata pdc2 / test=logrank adjust=sidak ;
time lenfol*death365day(0);
weight ps_wt;
run;
```

Is there something else I need to do in order to generate the multiple comparison test results?

Thanks.

2 REPLIES 2

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

Any interesting notes in the log?

Without data we can't see what is generated, not generated or test options.

Instructions here: https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-dat... will show how to turn an existing SAS data set into data step code that can be pasted into a forum code box using the {i} icon or attached as text to show exactly what you have and that we can test code against.

- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content

I am not allowed to post any of our data, but perhaps these additional details will be helpful: The exposure variable, pdc2 is a categorical variable with 3 levels; lenfol is the time to event in days, and death365day is a Y/N flag for the event. My question is centered on the difference between the output I receive after I add the weight statement to the code below, which runs without error and produces the standard output for proc lifetest, as well as the Adjustment for Multiple Comparisons for the Logrank Test table.

```
%dummytbl(sheet_name=WtdKMyr_Longterm66743);
proc lifetest data= bb_ps_w_pdc2 notable plots=survival( nocensor failure test atrisk(atrisktickonly outside)=(0,30,90,180,365)) maxtime=365;
title 'IPW Failure curve for ages 66-74 by PDC Group- weight removed';
where 66 <= ageatdiag < 75;
strata pdc2 / test=logrank adjust=sidak ;
time lenfol*death365day(0);
run;
```

The Multiple Comparison table gave me the strata comparison, chi-square test results and the p-Values (raw and Sidak). However, when I added "weight" statement (in order to generate inverse probability weighted curves), as shown in the code below:

```
proc lifetest data= bb_pswt_pdc_adj2 notable plots=survival(cl nocensor failure test atrisk(atrisktickonly outside)=(0,30,90,180,365)) maxtime=365;
title 'IPW Failure curve by age2 strata pdc with adjusted weight';
by age2;
strata pdc2 / test=logrank adjust=sidak ;
time lenfol*death365day(0);
weight ps_wt_adj;
run;
```

the Multiple comparison table was not generated. The only output I got were the Rank Statistics, Covariance Matrix, and Test of Equality over Strata tables, along with the plot. I did not receive any error and the log output was the same for both sets of code and was the following: NOTE: The LOGLOG transform is used to compute the confidence limits for the survivor function. To suppress using this transform, specify CONFTYPE=LINEAR in the PROC LIFETEST statement.

NOTE: The PROCEDURE LIFETEST printed pages 13-18.

NOTE: PROCEDURE LIFETEST used (Total process time):

Thanks.

Barbara

**Available on demand!**

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.