HI all, not sure where I have gone wrong with this. I have an actual code, but created this dummy code, and the same issue occurs.  Note that I'm asking cells to be red if the test date is missing (.) AND test result is not blank. However, I'm getting red cells in patients 1, 2 even though they have a test date. Based on this code, I should only have a red cell at test result for patient 5. What am I missing?   data test; input patient_id $ test_date :mmddyy10. test_result $15.; format test_date mmddyy10.; datalines; 1 06/01/2023 Negative 2 07/13/2023 Positive 3 09/25/2023 . 4 11/21/2023 . 5 . Positive ; run; proc report data=test; compute test_result; if test_date = . and test_result ne '' then call define(_col_,'style','style={background=red}'); endcomp; run; ... View more

I have a long code, but made a mock one for our purposes here, which creates the same error. I have 5 out of 10 patients that were evaluated, yet I get 60% instead of 50% evaluated when using "else 0". But when I use "else .", I get 50%, which is correct. I'm wondering why? Problem solved after 2 days, but perhaps someone can help me understand why using else 0 vs. else .  would make a difference?   Mock data:   data test; input patient_id evaluated $; datalines; 1 Yes 2 Yes 3 Yes 4 Yes 5 Yes 6 No 7 No 8 No 9 No 10 No ; run;   First code that gives me 60% with "else 0" in the second subquery:   proc sql; create table test1 as select "Evaluation Complete" as Variable, (select (Count(distinct patient_id)) from test where evaluated = "Yes") as Count, (select (Count(distinct case when evaluated="Yes" then patient_id else 0 end)/(Count(distinct patient_id)*1.0)) from test) as Percent format percent8.1 from test(obs=1); quit;   Second code where we replace "else 0" with "else ." and gives us 50% for patients evaluated (correct):   proc sql; create table test2 as select "Evaluation Complete" as Variable, (select (Count(distinct patient_id)) from test where evaluated = "Yes") as Count, (select (Count(distinct case when evaluated="Yes" then patient_id else . end)/(Count(distinct patient_id)*1.0)) from test) as Percent format percent8.1 from test(obs=1); quit;   ... View more

Getting  error in the log: "Expecting ;.  The symbol is not recognized and will be ignored". If I comment out compute discharge date section in the compute block, the error does not show. I am attempting to traffic light so if there is a hospital admission date, then we should also have a discharge date. Similarly, if we have a discharge date, we should have an admission date.  I don't see where I'm missing a semi-colon. I have seen SAS throw errors that are not applicable before though.   Error when running this: proc report data=Cohortb1_2022Q3Q4; columns record_number admission_date discharge_date; define admission_date / Display 'Admission Date To Diagnostic Facility'; define discharge_date / Display 'Discharge Date From Diagnostic Facility'; define record_number / 'Record Number'; compute admission_date; if admission_date = . and discharge_date ne . then call define(_col_,'style','style={background=red})'); endcomp; compute discharge_date; if admission_date ne . and discharge_date = . then call define(_col_,'style','style={background=red})'); endcomp; run;   No error when commenting out compute discharge date section: proc report data=Cohortb1_2022Q3Q4; columns record_number admission_date discharge_date; define admission_date / Display 'Admission Date To Diagnostic Facility'; define discharge_date / Display 'Discharge Date From Diagnostic Facility'; define record_number / 'Record Number'; compute admission_date; if admission_date = . and discharge_date ne . then call define(_col_,'style','style={background=red})'); endcomp; /*compute discharge_date;*/ /* if admission_date ne . and discharge_date = .*/ /* then call define(_col_,'style','style={background=red})');*/ /*endcomp;*/ run;   ... View more

I'm running the following to get the number of males and females /*Number of males and females*/ proc sql; create table number_birth_sex as select birth_gender 'Birth Gender', count(birth_gender) as Count from tb_yearly_report1 group by birth_gender; quit; and I get this  birth_gender Count Female 11 Male 27   What I want to do is add another variable called "Percent" that calculates percent of Females (11/38) and Males (27/38) from the total. So ideally like this:   birth_gender Count Percent Female 11 28.9% Male 27 71.1%   I can do it with a data step but figured there's a way I can do it within this same section of code.   ... View more

I'm not new to SAS, but new to using arrays. I have a SAS table set up like this except for the final_race variable, which is what i want:   patient_id race_white race_black race_asian race_am_indian race_pac_islander race_other race_unk final_race 1 1 0 0 0 0 0 0 white 2 1 1 0 0 0 0 0 multiracial 3 0 0 0 0 0 0 0 unknown 4 0 0 1 0 0 0 0 asian   A 1 means a person is of that race. Some people could have 1s in multiple columns because they're multiracial. Some have all 0s meaning we don't know their race.  I need to sort through all the combos.. if a person has a 1 for only one race, then I need final_race to be that race. Else, if there are multiple 1s, then final_race "Multiracial". If all 0s, then "Unknown".  I tried to build something off https://communities.sas.com/t5/SAS-Procedures/Race-looping/td-p/182117 but I couldn't figure it out with the differences I need. Thank you ... View more