BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
mayasak
Quartz | Level 8

Hi,

 

I have this SAS code that I need to use so I can have the snomed codes to cover all the types of result texts that I have in excel sheet.

 

if test_loinc="94558-4" then do; /*Note: this is an antigen test*/
result_snomed=10828004;
where (test_result="Positive"
or test_result="POSITIVE"
or test_result="positive");
result_snomed=260373001;
where (test_result="Detected"
or test_result="DETECTED"
or test_result="detected");
result_snomed=260385009;
where (test_result="Negative"
or test_result="NEGATIVE"
or test_result="negative");
result_snomed=260415000;
where (test_result="Not Detected"
or test_result="NOT DETECTED"
or test_result="not detected"
or test_result="Not detected");
end;
else if test_loinc="94759-8" then do; /*Note: this is a PCR test*/
result_snomed=10828004;
where (test_result="Positive"
or test_result="POSITIVE"
or test_result="positive");
result_snomed=260373001;
where (test_result="Detected"
or test_result="DETECTED"
or test_result="detected");
result_snomed=125154007;
where (test_result="Unsatisfactory"
or test_result="UNSATISFACTORY"
or test_result="unsatisfactory");
result_snomed=260385009;
where (test_result="Negative"
or test_result="NEGATIVE"
or test_result="negative");
result_snomed=260415000;
where (test_result="Not Detected"
or test_result="NOT DETECTED"
or test_result="not detected"
or test_result="Not detected");
result_snomed=280416009;
where (test_result="Indeterminate"
or test_result="INDETERMINATE"
or test_result="indeterminate");
result_snomed=419984006;
where (test_result="Inconclusive"
or test_result="INCONCLUSIVE"
or test_result="inconclusive");
end;

 

  • First thing I got this error (after each if statement):

if test_loinc="94558-4" then do; /*Note: this is an antigen test*/
220 result_snomed=10828004;
221 where (test_result="Positive"
222 or test_result="POSITIVE"
223 or test_result="positive");
ERROR: WHERE clause operator requires compatible variables.

  • Also is there another simpler way to do this rather than having this long code?

Thank you.

 

1 ACCEPTED SOLUTION

Accepted Solutions
gamotte
Rhodochrosite | Level 12
I don't understand what you are trying to do.

You do exactly the same thing for the different values of test_loinc so why distinguishing several cases ?

Also, for a given value of test_loinc all selects after the first one are redundants since
they test the same condition (although expressed differently).

=> All your code could be replaced by (under the assumption that there are no other values for test_loinc) :
select (upcase(test_result));
when ("DETECTED") result_snomed = 260373001;
when ("NEGATIVE") result_snomed = 260385009;
when ("POSITIVE") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;

View solution in original post

10 REPLIES 10
PaigeMiller
Diamond | Level 26

You are trying to compare a numeric variable named TEST_RESULT to a character string. You can't do that. You have to compare numeric to numeric, or character to character.

 

it is possible that numeric TEST_RESULT is formatted to appear as character; that would still require you to compare it to a numeric value. Use PROC CONTENTS to find out how TEST_RESULT is formatted.

--
Paige Miller
gamotte
Rhodochrosite | Level 12

Hello,

 

The where statement in a data step has a global effect and drops the observations that don't

match the condition before executing the data step intructions.

Thus, it does not make sense to condition these where statements as they will all be taken into account

at the program execution.

 

Before resolving this type mismatch problem, you might want to reconsider the logic of

your program.

 

Can you explain on a simplified example what you are trying to achieve ?

 

Kurt_Bremser
Super User

You have two serious mistakes here:

  1. trying to have a conditional WHERE statement. This is not possible, see the documentation.
  2. having multiple WHERE statements. Only the last WHERE statement will be compiled by the data step compiler.

guess that you should use the conditions in the IF statements; for more detailed help, supply some example data in usable form (DATA step with DATALINES, NOT Excel files or screenshots!), and what you intend to have as a result from that.

mayasak
Quartz | Level 8
data;
   input name $ test_loinc $ test_result $;
   datalines;
John 94759-8 Positive
Michelle 94558-4 Not detected
Grace 94759-8 DETECTED ;


So the issue is that we're getting an excel sheet from a hospital that have different variables (in additition to the test_result and test_loinc) and we need to import into to our system. They use only 2 loincs (tests) (used in the example) and the result comes in different formats such as DETECTED, Detected, Positive, positive... etc. What I needed to do is to code all different result formats into the standard snomed (results) codes such as 10828004 for all postive results because our system accepts only codes for loincs and snomeds. BTW I'm not sure if the test_result is a character or numerical but I'm assuming it's a character in the example! Thank you

     

 

PaigeMiller
Diamond | Level 26

Can you show us a portion of the data set from the imported spreadsheet by following these instructions , and not via any other method. Do not attach the Excel file, most people will not download it.

--
Paige Miller
Kurt_Bremser
Super User

Start with a "brute force" approach like this:

select (test_loinc);
  when ("94759-8") do;
    select (upcase(test_result));
      when ("POSITIVE") result_snomed = 260373001;
      when ("DETECTED") result_snomed = 260385009;
      otherwise result_snomed = 10828004;
    end;
  end;
end;

expand this code snippet for the other tests and results.

mayasak
Quartz | Level 8

Thank you,

I expanded this to:

select (test_loinc);
when ("94759-8") do;
select (upcase(test_result));
when ("DETECTED") result_snomed = 260373001;
when ("NEGATIVE") result_snomed = 260385009;
when ("POSITIVE") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
select (lowcase(test_result));
when ("detected") result_snomed = 260373001;
when ("negative") result_snomed = 260385009;
when ("positive") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
select (propcase(test_result));
when ("Detected") result_snomed = 260373001;
when ("Negative") result_snomed = 260385009;
when ("Positive") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
end;
when ("94558-4") do;
select (upcase(test_result));
when ("DETECTED") result_snomed = 260373001;
when ("NEGATIVE") result_snomed = 260385009;
when ("POSITIVE") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
select (lowcase(test_result));
when ("detected") result_snomed = 260373001;
when ("negative") result_snomed = 260385009;
when ("positive") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
select (propcase(test_result));
when ("Detected") result_snomed = 260373001;
when ("Negative") result_snomed = 260385009;
when ("Positive") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
end;
end;

 

Does it look fine? Especially for the use of end;

Thank you,

Razina

 

gamotte
Rhodochrosite | Level 12
I don't understand what you are trying to do.

You do exactly the same thing for the different values of test_loinc so why distinguishing several cases ?

Also, for a given value of test_loinc all selects after the first one are redundants since
they test the same condition (although expressed differently).

=> All your code could be replaced by (under the assumption that there are no other values for test_loinc) :
select (upcase(test_result));
when ("DETECTED") result_snomed = 260373001;
when ("NEGATIVE") result_snomed = 260385009;
when ("POSITIVE") result_snomed = 10828004;
otherwise result_snomed = 260415000;
end;
mayasak
Quartz | Level 8

There are 2 values for "test_loinc" and more values for "test_result" such as results with all lower case and results with mixedletters (first letter only upper case).

Thank you,

Razina

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 10 replies
  • 1532 views
  • 2 likes
  • 4 in conversation