Thank you all so much for the help.  I was able to get it working with this bit of code from Reeza found on a previous thread: Data Test_ER_2; set avoid1; by memid admitdt; retain continuous_group 1; prev_date=lag(admitdt); if first.memid then do;   prev_date=.;   continuous_group=1; end; format prev_date date9.; diff=intck('day',prev_date,admitdt); if intck('day', prev_date, admitdt)>1 then continuous_group + 1; run; Assigned a continuous_group number for each span for each memberID.  Was then able to aggregate using member ID and continuous_group.  I am going to try it with the code provided above as well.  Thank you everyone again for the help ... View more

If you google search "sas hash sugi", you should find easily what you are looking for. I haven't tested this code. If your lookup is actually a lookup with only 1 value to join, then try this. DATA D1; SET QIP.DHCS_AER; AF1=1; KEEP DIAG1 AF1; RUN; * etc. - until you#ve defined your 14 data sets, maybe use call execute or so to create code dynamically. no sorting required; Data Hash (Drop=rc:);   Length Group $15.;   Array rc{*} rc1-rc14;   Set ED2;   If _N_ =1 Then Do;     Declare Hash H1 (Dataset:'D1');     H1.Definekey('Diag1');     H1.Definedata(All:'y');     H1.Definedone();     Declare Hash H2 (Dataset:'D2');     H2.Definekey('Diag2');     H2.Definedata(All:'y');     H2.Definedone();     * etc.;   End;   rc1=H1.Find();   rc2=H2.Find();   * etc.;   If Sum(of rc{*}) gt 0 Then Group='UNAVOIDABLE';   Else Group='AVOIDABLE'; Run; ... View more