If your source data was a text file and you copy something like the first 5 lines and paste them into a text box (critical!) we may be able to show you how to read that file directly to create that data structure.
If your source data was in a different format tell us what type file.
Consider this example, which only has 3 sets of values consisting of two "words" and 5 numeric values:
data example;
infile datalines dlm=',' dsd;
informat worda wordb $10. value1-value5 best.;
input id @;
do i=1 to 3;
input worda wordb value1-value5 @;
output;
end;
input;
drop i;
datalines;
1,abc,pdq,1,0,0,1,1,zz,top,0,0,0,0,1,john,doe,1,1,1,1,0
1,aaa,ggq,1,0,0,1,1,z3,bgt,0,0,0,0,1,mary,pit,1,1,1,1,0
;
The infile statement would point to a text file but here it shows some options if the original data were comma separated (CSV). The DLM option tells SAS to expect a comma between values. The DSD option means that if there are two successive delimiters to treat that variable as missing.
The Informat describes how to read each variable, worda and wordb as character, and then 5 numeric variables as numbers (not actually needed as a default but to make basis for when more complicated things come up). The ID could be numeric or character.
The first input reads just the ID value. The @ symbol at the end of an input means "hold the reading pointer on the current record where it is", which in this case would be right after the ID.
Since we KNOW we have 3 sets of variables we can use a loop to specify that number of times to read, still using the @ to hold the pointer. The OUTPUT statement says to write the current data to the output data set.
After the Loop we need an explicit Input to advance to the next row of data.
The I value isn't needed, unless you want a line or group indicator for some reason. If so pick a name that means more than the "i" and don't use the Drop statement.
If not all 3 groups are present you will get some blank lines currently. That could be fixed by using "If not missing(worda) then output;" instead of the simple output.
Another option might be to EXPORT your existing data to CSV and write a program similar to above to read it.
You did mention other variables. Where they appear is going to make a difference but as long as the pattern is regular it is quite doable, just somewhat cumbersome code.
For example if you had indication1 otherindication1 indication2 otherindication2 etc as the column order then the Input statement would just list those.
The headaches come if the "other" are mixed in after the the drug type and name group of values.
