topic Re: Survival analysis w/ data structure of each row being separate measurement over time... in Statistical Procedures

Survival analysis w/ data structure of each row being separate measurement over time...

mhtoto — Mon, 23 May 2016 04:48:11 GMT

Basically I need a way to find the first entry in the Censoring_Event column (based on MINIMUM time), and then delete any other subsequent entries.

Patient Censoring_Event Time

1 1

1 3

1 4

1 0 7

(this pt does not pose a problem as the first 3 entries will just be ignored when using proc lifetest)

2 1

2 6

2 1 8

2 9

2 0 14

(even if the pt had the event a 0 shows up at their last f/u date)

3 1

3 2

3 5

3 1 11

3 1 19

(some pts experience the censoring even more than once)

Any help would be appreciated.

Re: Survival analysis w/ data structure of each row being separate measurement over time...

PGStats — Mon, 23 May 2016 05:48:27 GMT

One simple way to do this:

data have;
input Patient Censoring_Event Time;
datalines;
1 . 1
1 . 3
1 . 4
1 0 7
2 . 1
2 . 6
2 1 8
2 . 9
2 0 14
3 . 1
3 . 2
3 . 5
3 1 11
3 1 19
;

data want;
do until(last.patient);
    set have; by Patient;
    if not drop then output;
    if not missing(censoring_event) then drop = 1;
    end;
drop drop;
run;

proc print noobs; run;

Re: Survival analysis w/ data structure of each row being separate measurement over time...

mhtoto — Mon, 23 May 2016 15:56:35 GMT

Thanks PG for your timely response. Exactly what I needed. Worked perfectly.