Extract VARIABLE NAME and concatenate with other variable names based on condition

[saslove]

I have a dataset that looks like this 

currentstudy Study1ID     Study2ID        Study3ID           Study4ID

001                  180            780                    278                   560

002                                    790                                             520

003                                                              231

As you can see, each patient can be enrolled on multiple other studies. 

What I need is a dataset that looks like below.

Basically, concatenate the column names/study names of the other studies that same patient is enrolled in.

currentstudy  Other_enrolled_studies 

001                 Study1,Study2,Study3,Study4

002                 Study2,Study4

003                 Study 3

and so on for several hundred patients across multiple studies. 

Thanks very much,

Please help. 

[Reeza · Accepted solution]

data want;
set have;

length studies $200;

array study(*) study: ;

do i=1 to dim(study);

if not missing(study(i)) then do;
      studies = catx(", ", studies, vname(study(i)));
end;

run;

1. Declare an array of all studys

2. Loop through and find non missing entries

3. Find variable name (vname)

4. append to list of 'studies'

---

@saslove wrote:

I have a dataset that looks like this 

currentstudy Study1ID     Study2ID        Study3ID           Study4ID

001                  180            780                    278                   560

002                                    790                                             520

003                                                              231

As you can see, each patient can be enrolled on multiple other studies. 

What I need is a dataset that looks like below.

Basically, concatenate the column names/study names of the other studies that same patient is enrolled in.

currentstudy  Other_enrolled_studies 

001                 Study1,Study2,Study3,Study4

002                 Study2,Study4

003                 Study 3

and so on for several hundred patients across multiple studies. 

Thanks very much,

Please help. 

---

[saslove]

That worked very well. Thanks Reeza

[Jagadishkatam]

An alternative way

data want;
set have;
array var1(*) Study1ID Study2ID Study3ID Study4ID;
array var2(*)$ Study1IDc Study2IDc Study3IDc Study4IDc;
do i = 1 to dim(var1);
if var1(i) ne . then var2(i)=vname(var1(i));
newvar=catx(',',of var2(*));
end;
run;

[Reeza]

---

@Jagadishkatam wrote:

An alternative way

data want;
set have;
array var1(*) Study1ID Study2ID Study3ID Study4ID;
array var2(*)$ Study1IDc Study2IDc Study3IDc Study4IDc;
do i = 1 to dim(var1);
if var1(i) ne . then var2(i)=vname(var1(i));
newvar=catx(',',of var2(*));
end;
run;

---

@Jagadishkatam should the newer line be outside of the loop?

[Jagadishkatam]

@Reeza , thanks for your respons. I tested the code on sashelp.class and seem like it is working. Appreciate your thoughts.

data want;
set sashelp.class;
array var1(*) age height weight ;
array var2(*)$ agec heightc weightc;
do i = 1 to dim(var1);
if var1(i) ne . then var2(i)=vname(var1(i));
newvar=catx(',',of var2(*));
end;
run;

[Reeza]

Technically it works, because it gets replaced at each iteration of the I loop, including the last, which is the same as doing it after the loop is finished. It's just not really logical to do it when it's not needed. If the data set is small this has no impact, if the data set is larger it's an inefficiency that slows things down.