topic Re: Remove duplicates for two field unique identifier in SAS Procedures

Remove duplicates for two field unique identifier

shellp55 — Sat, 26 Oct 2013 00:59:40 GMT

I am still plugging away on the readmits program and I figured it out but I want to remove multiple readmits for the same patient's episode. If I use the following code, it works but it removes all entries for a chart number when really I only want to remove when the same chartno with the same admit date is duplicated. So if the same chart number had a different admit date I don't want it removed.

data single dup;

set temp;

by chartno;

if first.chartno and last.chartno then output single;

else output dup;

run;

So is it possible to do have a first observation with two fields like first.chartno admdate and last.chartno admdate? Thanks.

Re: Remove duplicates for two field unique identifier

Jagadishkatam — Sat, 26 Oct 2013 01:05:49 GMT

Please try

data single dup;

set temp;

by chartno admit_date;

if first.admit_date and last.admit_date then output single;

else output dup;

run;

Thanks,

Jagadish

Re: Remove duplicates for two field unique identifier

shellp55 — Sat, 26 Oct 2013 02:44:43 GMT

Thanks so much Jagadish but that didn't work. I had a case with 3 readmits noted for one chart number, 2 were the same admit date and the third wasn't but the code wiped out all but one.

Any other suggestions? Thanks so much.

Re: Remove duplicates for two field unique identifier

Jagadishkatam — Sat, 26 Oct 2013 02:46:38 GMT

I hope, this will help you

data single dup;

set temp;

by chartno admit_date;

if first.admit_date then output single;

else output dup;

run;

Thanks,

jagadish

Re: Remove duplicates for two field unique identifier

shellp55 — Sun, 27 Oct 2013 20:23:02 GMT

Thanks so much, that seems to have worked!