Thanks Astounding!   Your last post clarified the last couple points I was getting hung up on and I was able to make the few changes I needed to get the program to run. Thanks for also explaining the different styles or arrays and the advantages of using the style you suggested. ... View more

Astounding,   Thanks! I created a macro with the strings in it I want to search for and then tried to include it in the data step. Can you please tell me if this looks right to you:   %macro check_gastro_hemg(dx); (substr(&dx,1,4) in ('5310', '5312', '5314', '5316', '5320', '5322', '5324', '5326', '5330', '5332', '5334', '5336', '5340', '5342', '5344', '5346', '5693')) or (&dx in ('53082', '53783', '53784', '56985', '56986')) or (substr(&dx,1,3)='578') or (substr(&dx,1,3)='535' and substr(&dx,5,1)='1') %mend;     data want; set have; array diags {9} c_dx1-c_dx8 c_hcpc; found = 0; do _n_=1 to 9 until (found=1);   if %check_gastro_hemg(diags) then then found=1; end; run; ... View more

Astounding,   Thanks for pointing this out. "280" should match strings that start with "280" that may then also have additional numbers in the fourth and fifth position. For example, "280", "2801", 2802" and "28041" should all be recognized as matches. Conversely, "12802" should not be a match. ... View more

Reeza,   Thanks for the quick reply. I took a look at the example you directed me to on page 10 and see that as a good starting point. I have started to modify it to suite my needs but don't think I can get it all the way there. There are a few questions I have: 1) The example has a unique key for each item (1 key for each item). I have the same set of 7 keys for all 9 items. Does that make a difference in how the problem needs to be coded? 2) My 7 keys are actually strings. Thus when I search c_dx1, if row 3 has the value 2814, I want that to be recognized as a match (with string 281). In the below example, I don't think 2814 would be scored a 1 (and it needs to be).   Here is what I have so far:   Array Raw {9} c_dx1, c_dx2, c_dx3, c_dx4, c_dx5, c_dx6, c_dx7, c_dx8, c_hcpc; Array Key {7} $ _temporary_ ('28730', '28739', '2875','280', '281', '282', '283'); Array Score {9} ; Do i = 1 to 9; if raw{i} eq key{i} then score{i}=1; else score{i}=0; End;   Total = sum( of score1 – score9 ); if total ge 1 then dis2 =1; else if total dis2 =1;   ... View more

Hello,   I have a similar task that I am trying to perform that I was hoping you could also provide some insight on. I want to state up front that I am new to SAS (and this community) so I apologize if my question is not clear or if I am not asking it in the correct forum.   I have a data set with ~10 million records/rows. I want to check for the presence of any one of multiple strings across 9 of the variables in the data set. I then want to create a new variable (dis2) in my data set that will take the value of either 0 or 1. Dis2 should have a value of 1 if at least one of the 9 variables has any one of the strings and 0 if none of the 9 variables have any of the strings.   I created a macro statement to hold the strings I want to seach for: %let dc2 = %str('28730', '28739', '2875','280', '281', '282', '283');   I then want to check each of these 9 varibales to see if any one of the strings are in any one of the variables: c_dx1, c_dx2, c_dx3, c_dx4, c_dx5, c_dx6, c_dx7, c_dx8, c_hcpc   I was thinking that I could put the 9 variables in an array but I am not sure how to put everything together and create dis2. Any help is greatly appreciated. If you can provide example code of how to solve this problem that would be ideal. Something like:   %let dc2 = %str('28730', '28739', '2875','280', '281', '282', '283');   data example_w_dis2;      set example;      .......;      .......;   Thanks!   ... View more