topic Removing Observations Based on Varible Value in SAS Programming

Removing Observations Based on Varible Value

eabc0351 — Tue, 06 Feb 2018 09:06:14 GMT

Hello,

I have a dataset of hospital admissions with multiple observations per patient. I'm looking to delete patient records based on the value of a binary variable (RANK) in the dataset. The first record for each patient should have a value 1 for RANK. If there are records within the patient ID that have a 0 before a 1 for RANK, I want to delete them. Any records within each patient id with a 0 after 1 for RANK should be kept. Here is an example of the data:

data have:

RANK ID

1 1

1 2

0 2

1 3

0 4

1 4

0 4

1 5

0 6

1 6

1 7

0 8

1 8

0 8

data want:

RANK ID

1 1

1 2

0 2

1 3

1 4

0 4

1 5

1 6

1 7

1 8

0 8

I'm not sure how to go about this -- any coding suggestions are appreciated.

Thank you.

Re: Removing Observations Based on Varible Value

Kurt_Bremser — Tue, 06 Feb 2018 09:42:59 GMT

Sort by ID and date (or whatever you got for a time series).

Do data step with

date want;
set have;
by id;
retain flag;
if first.id then flag = 0;
if rank = 1 then flag = 1;
if flag;
drop flag;
run;

Re: Removing Observations Based on Varible Value

eabc0351 — Tue, 06 Feb 2018 14:52:32 GMT

yes that does it, thank you.