time to first medication switch as time-dependent covariate in Cox model

[Merdock]

I need help with setting up my dataset to perform survival analysis with a time-dependent covariate. I read the Therneau et al. paper but I'm still unsure of how to set up the counting process format for my data.

Below is a test dataset similar to what I have.

Data: Liver transplant dataset where some patients are in group 1 (if on immunosuppressive drug A at baseline), others are in group 2 (if on drug B at baseline).

Goal: I want to know how/if the time of first switch from med A to med B impacts graft survival (graft failure is the event of interest). The expectation is that the switch to drug B wouldn't significantly impact graft survival but it's important to show this because drug B would be preferable as it has less side effects, better tolerance profile and is more cost effective than drug A.

GROUP=1, if on drug A; 2 if on drug B, and 3 if they switched from A to B;

CENSOR=1, if event (graft failure) happened, 0 otherwise;

FUTIME = total follow-up time (months) (=CENSDATE – transplant date), where CENSDATE is either date of event, or date of last follow-up/death)

SWITCHTIME=time of first switch from group 1 to group 2 (=date when they switch– transplant date); this will be 0 for those patients who start off on drug B at baseline, and it is FUTIME for those who start off on drug A and never switch (per investigators's request).

data have;
input ID$ TRANSDATE:mmddyy. CENSDATE:mmddyy. FUTIME CENSOR SWITCHTIME GROUP$;
format TRANSDATE mmddyy10. CENSDATE mmddyy10.;
datalines;
001	7/2/2002	7/11/2003	344	1	344	1
002	8/23/2001	7/30/2002	311	0	311	1
003	3/27/2002	10/1/2007	1984	1	1984	1
004	1/28/2000	3/26/2011	4045	1	0	2
005	8/6/2002	12/16/2005	1198	0	1198	1
006	1/23/2001	9/8/2005	1659	0	12	3
007	4/10/2003	3/10/2006	1035	0	1035	1
008	8/26/2002	3/26/2008	2009	0	2009	1
009	8/14/2014	9/15/2016	733	0	18	3
010	11/3/2000	11/10/2003	1072	0	1072	1
011	4/11/2001	9/12/2007	2315	0	2315	1
012	2/5/2002	1/2/2006	1397	0	1397	1
013	7/11/2000	1/2/2008	2701	0	0	2
014	5/7/2002	11/10/2008	2349	0	2349	1
015	9/23/2001	2/7/2005	1203	0	6	3
016	5/3/2000	3/26/2003	1027	0	1027	1
017	3/25/2003	12/20/2006	1336	1	1336	1
018	6/24/2004	12/7/2006	866	0	866	1
019	3/22/2006	2/17/2012	2128	0	13	3
020	5/27/2000	11/14/2006	2332	0	2332	1
021	8/23/2004	4/2/2008	1288	0	1288	1
022	4/10/2001	2/5/2013	4289	1	4289	1
023	9/13/2001	12/10/2020	6998	0	54	3
024	8/24/2012	2/3/2020	2689	0	0	2
025	5/9/2000	6/8/2000	0	0	0	1
026	4/12/2000	5/12/2000	0	0	0	2
027	6/9/2005	11/24/2008	1234	0	13	3
028	10/23/2008	4/24/2012	1249	0	1249	1
029	5/22/2003	11/17/2008	1976	0	38	3
030	12/13/2000	12/8/2005	1791	0	0	2
031	10/28/2003	8/9/2007	1351	0	0	2
032	3/7/2006	7/12/2007	462	0	462	1
033	3/5/2006	3/22/2013	2544	0	6	3
034	3/5/2000	3/13/2013	4726	1	5	3
035	5/7/2003	3/28/2008	1757	0	0	2
036	8/31/2005	2/5/2009	1224	0	1224	1
037	10/2/2002	3/18/2013	3790	0	94	3
038	7/2/2010	1/8/2019	3082	0	0	2
039	3/27/2002	2/9/2004	654	0	23	3
;
run;
proc print data=have; run;

[ballardw]

I think that you need to seriously examine your text describing your variables FUTIME and EVTIME as months. If Futime is months, as stated, then the value of 6998 would be in excess of 583 years. If you derived those values with subtraction of SAS date values then the unit would be days not months.

How do we know, from the values shown, when graft failure occurs? I don't see a statement that clearly states that.

You do not define group=3 at all in terms of the problem. Since this is the most common value that might be a severe short fall in any analysis.

---

@Merdock wrote:

I need help with setting up my dataset to perform survival analysis with a time-dependent covariate. I read the Therneau et al. paper but I'm still unsure of how to set up the counting process format for my data.

Below is a test dataset similar to what I have.

Data: Liver transplant dataset where some patients are in group 1 (if on immunosuppressive drug A at baseline), others are in group 2 (if on drug B at baseline).

Goal: I want to know how/if the time of first switch from med A to med B impacts graft survival (graft failure is the event of interest). The expectation is that the switch to drug B wouldn't significantly impact graft survival but it's important to show this because drug B would be preferable as it has less side effects, better tolerance profile and is more cost effective than drug A.

GROUP=1, if on drug A; 2 if on drug B, and 3 if they switched from A to B;

CENSOR=1, if event (graft failure) happened, 0 otherwise;

FUTIME = total follow-up time (months) (=CENSDT – transplant date), where CENSDT (not shown here, is either date of event, or date of last follow-up/death)

EVTIME= time of event (months); EVTIME=FUTIME when CENSOR=1, else EVTIME= N/A;

SWITCHDATE (not shown here) =date of first switch from drug A to drug B;

SWITCHTIME=time of first switch from group 1 to group 2 (=SWITCHDATE – transplant date); this will be 0 for those patients who start off on drug B at baseline, and it is FUTIME for those who start off on drug A and never switch (per investigators's request).

data have;
input ID$ CENSOR$ FUTIME EVTIME SWITCHTIME GROUP$;
datalines;
001	0	1229	.		333		3
002	0	1659	.		343		3
003	0	733	 	.		509		3
004	0	6998	.		1630	3
005	1	1005	1005	558		3
006	1	4726	4726	147		3
007	0	3790	.		2856	3
008	1	672		672		504		3
009	0	5224	.		2648	3
010	0	4143	.		149		3
011	0	4973	.		500		3
012	1	3626	3626	3624	3
013	0	4296	.		3998	3
014	0	977		.		0		2
015	0	2898	.		331		3
016	0	1382	.		1187	3
017	1	1164	1164	1164	1
018	0	1232	.		336		3
019	1	1599	1599	143		3
020	0	1795	.		111		3
021	0	3171	.		3171	1
022	1	483		483		333		3
023	1	824		824		351		3
024	0	662		.		662		3
025	1	597		597		0		2
026	0	1269	.		0		2
027	0	4120	.		4120	1
028	1	621		621		0		2
029	0	3452	.		3452	1
030	0	1842	.		0		2
;
run;

---

[Merdock]

@ballardw, thanks! I updated my dataset and post. The FUTIME and EVTIME are in months, as I have used the intck function with 'months', not subtraction to calculate them. Graft failure occurs whenever you see CENSOR=1.

I don't understand what you mean by your last statement: "You do not define group=3 at all in terms of the problem".

Please let me know if this is clearer or you need any additional clarifications.