topic Re: Help creating a time-varying dataset in SAS Health and Life Sciences

Help creating a time-varying dataset

b00z_84 — Thu, 26 Nov 2015 22:03:24 GMT

Hello,

I am new to SAS and am having trouble creating my dataset so it fits a time varying cox model and I am in desperate need for help.

My dataset looks as following;

Vx_State = a categorical variable, if 1 it stays in that state rest of the study

Vx_Date = Date when the variable changes from 0 to 1

t1 = starting time in days for that row

t2 = ending time in days for that row

ID Start_date End_date V1_State V1_Date V2_State V2_Date V3_State V3_Date V4_State V4_Date t1 t2

1 2010-01-15 2013-12-31 0 2012-05-19 1 . 0 2010-11-15 1 . 1 1446

2 2010-05-13 2012-10-07 1 . 0 . 1 . 0 2012-08-17 1 878

But I want to split the rows for each time a variable (Vx_State) changes by the patient ID ;

ID Start_date End_date V1_State V1_Date V2_State V2_Date V3_State V3_Date V4_State V4_Date t1 t2

1 2010-01-15 2010-11-14 0 2012-05-19 1 . 0 2010-11-15 1 . 1 303

1 2010-11-15 2012-05-18 0 2012-05-19 1 . 1 2010-11-15 1 . 304 853

1 2012-05-19 2013-12-31 0 2012-05-19 1 . 1 2010-11-15 1 . 854 1446

I hope that my problem fairly clear, at least, and any help would be appreciated!

Best Regards

Re: Help creating a time-varying dataset

Reeza — Thu, 26 Nov 2015 22:28:52 GMT

I'm afraid I don't understand and I can't quite get the format of your data, it's garbled in the window. Can you post it as a data step?

Re: Help creating a time-varying dataset

FreelanceReinh — Thu, 26 Nov 2015 22:56:13 GMT

I think I know what you mean. But your example target dataset does not quite match what I would have expected:

Two examples:

For ID 1 the start date is 2010-01-15. I take it that this is day 1 (because t1=1). Right?

Therefore, I thought 2010-11-14 should be day 304 (as calculated by %put %sysevalf('14NOV10'd-'15JAN10'd+1);). But you have t2=303 in the target dataset. Similarly, 2012-05-18 is 854 days after 2010-01-15, hence would seem to be day 855 rather than t2=853. ...

V3_State changes from 0 to 1 from the first to the second row of the target dataset. This is logical because of the V3_Date.

By the same logic I would have expected V1_State to switch from 0 to 1 from the second to the third row. But you left it at 0.

(Edit: Sorry, I'll be back tomorrow. It's getting late here in Central Europe.)

Re: Help creating a time-varying dataset

FreelanceReinh — Fri, 27 Nov 2015 16:20:14 GMT

I think it's possible to obtain the restructured dataset within a single data step. However, it depends on the handling of special cases, as to how complicated this data step will be:

Vx_Date=Vy_Date (for x not equal to y)
Vx_Date=Start_date
Vx_Date=End_date

That said, it is possibly not even necessary to create the time-varying Vx_State variables (which I assume would be used as covariates in your Cox regression model) the way you suggested. Are you aware of PROC PHREG's capabilities for handling time-dependent covariates? (I wasn't when I was "new to SAS.") You could have a look at Example 64.6 Model Using Time-Dependent Explanatory Variables in the online doc.

As you can see there, PROC PHREG allows for IF-THEN statements, which make the definition of time-dependent binary indicators like your Vx_State variables fairly easy.