@juju_p  You may try adding options=nofmterr; in the data step when you read your data file to avoid that format error. libname mylib "c:\the path to your data file"; data mydata; set mylib.datafilename; options=nofmterr; run; Then import your format file in LIBRARY.  libname library "c:\path to your format file"; Then you can access the value labels in your PROC steps. proc freq data=mylib.mydata; tables var1; run;     ... View more

Try applying PROC SORT before doing your PROC REPORT.  PROC SORT is great for eliminating duplicate rows.  Here is generalized code based of the report you provided, fill in actual table and column names.  Be careful and name a new output table with OUT= so your original table is not over written.   proc sort data=source_table out=new_table nodupkey dupout=table_with_duplicates; by lesiontype timepoint totalsum dateofexam lesioneval overalleval; run;     ... View more

There are multiple ways to read Excel files into SAS e.g. Comparing SAS LIBNAME Engines for PC Files Data  ... View more

Looking at the updated description for the output, I would keep the duplicates all in one data set: data vials; set have; array viallab {10}; do whichtest=1 to 10; if viallab{whichtest} ne . then do; vial = viallab{whichtest}; output; end; end; keep vial subjid visitt visitd whichtest; run; proc sort data=vials; by vial subjid; run; data want; set vials; by vial; if first.vial=0 or last.vial=0; run; /* optionally */ proc print data=want; by vial; id vial; var subjid visitt visitd whichtest; run; It's untested code, but should do the trick. ... View more