I still do not understand the question, but perhaps if I try to restate the information you have posted we can figure out what the question is?
You posted an EXCEL file, but when I browsed it there is just one value. So here it is in a more usable form.
data dv ;
term = 'Subject 00000000 baseline labs were collected on Day -1,'
|| ' 00SEP0000, instead of on Day -0 per protocol. The baseline'
|| ' labs should not have been drawn as the baseline visit was'
|| ' < 00 days from screening.'
;
run;
It this point it would have been helpful to post what you expected the answer to be. Say a new data step that creates the expected output dataset.
So looking at the code you posted I see three steps. (1) split TERM into words using space as the delimiter. (Note this would eliminate any extra spaces that might have been between the words.) (2) Add up how many characters the words would take. It seems to goal was to force the combined length to be less than 200 characters. (3) Put the words back together into longer strings.
So now I am getting the feeling that the question is
How can I split a long string into shorter strings of 200 characters or less?
But since the string that was in the Excel file was only 199 characters I am not even sure if I have the right question because there is nothing to do to shorten it to strings less than 200 characters long. So let's remake the problem data to have something longer than 200.
data dv ;
term = 'Subject 00000000 baseline labs were collected on Day -1,'
|| ' 00SEP0000, instead of on Day -0 per protocol. The baseline'
|| ' labs should not have been drawn as the baseline visit was'
|| ' < 00 days from screening.'
|| ' And more so that it is longer than 200 characters.'
;
len = lengthn(term);
put len=;
run;
Now let's split it into two string. The last sentence should be in the second string.
data want ;
set dv ;
length dterm $200 word $200 ;
group=1;
do i=1 by 1 until (word=' ');
word = scan(term,i,' ');
if length(catx(' ',dterm,word)) > 200 then do;
output;
group+1;
dterm=word;
end;
else dterm=catx(' ',dterm,word);
end;
if dterm ne ' ' or group=1 then output;
run;
Most likely the issue in your original program stems from always adding 1 to the length of the word. You should only add the one when you are appending it to an existing word since you then need to account for the space that you are adding. For example the first two words have length 6 and 8. Together they have length 15 since there needs to be a space between them, but that does not mean the first word has length 7 or that the two combined have length 16.
