topic Re: Unique outcomes across visits in SAS Programming

Unique outcomes across visits

rfarmenta — Fri, 20 Mar 2015 05:43:00 GMT

I have data from 5 study visits that assessed health behaviors. The same questions were asked at each visit and I would like to determine how many unique participants reported one behavior. I have the data sets appended in long format, so I can run a proc freq and get that 110 participants reported the behavior across all visits but I want to know how many of those are unique participants versus the same participants reporting the behavior at each visit. Ideally I would like to know how many participants reported the behaviors at all visits, how many at any visit, and how many reported 1, 2, 3, and 4 visits. I previously asked a similar question but was not able to figure this out. Any suggestions would be greatly appreciated.

Re: Unique outcomes across visits

ballardw — Fri, 20 Mar 2015 17:09:31 GMT

If by long you mean you have data as:

ID VisitDate Behavior or similar:

Then

Data temp;

set have;

behaviorcount = (behavior='behavior value of interest');

run;

proc summary data=temp nway;

class ID;

var behaviorcount;

output out=want sum=;

run;

will have the number of visits with the specified behavior across all visits for each ID. ID could be several variables that uniquely identify the patient.

Re: Unique outcomes across visits

Reeza — Fri, 20 Mar 2015 18:05:32 GMT

You can also use either count distinct in a proc sql or use a double proc freq - first time include the participant ID, the second you don't.

For more help please post sample data and expected output that corresponds to your sample data.

Re: Unique outcomes across visits

rfarmenta — Fri, 20 Mar 2015 21:02:36 GMT

I am not sure I understand the code from ballardw and I tried count distinct but that doesn't seem to give me what I want either. I can't attach an excel right now for some reason but below is a short example of what I would want to do. I just made up the example data. I would want to know how many people did the behavior at 1 visit, at 2 visits, at 3 visits, etc. so from the data the results would be this, 1=yes, 2= no in the code for behavior:

behaviorcount

0=1 (did not report)

1=1

2=3

3=1

4=0

5=1

pid	behavior	visit
1	0	1
1	1	2
1	1	3
1	0	4
1	0	5
2	0	1
2	0	2
2	0	3
2	0	4
3	1	1
3	1	2
3	1	3
4	1	1
4	0	2
4	0	3
4	0	4
4	0	5
5	1	1
5	1	2
5	1	3
5	1	4
5	1	5
6	0	1
6	1	2
6	1	3
6	0	4
6	0	5
7	1	1
7	1	2

Re: Unique outcomes across visits

ballardw — Fri, 20 Mar 2015 21:34:49 GMT

With that coding:

proc means data=have sum;

class pid;

var behavior;

run;

If 1 represents have the behavior then the sum would be how many visits exhibited the behavior for each PID value.

Re: Unique outcomes across visits

Patrick — Fri, 20 Mar 2015 23:48:09 GMT

A SQL approach:

proc sql;

/* create table want as*/

select pid, behavior, count(behavior) as count

from have

/* where behavior='1'*/

group by pid,behavior

;

quit;