AE Programming query

[tarheel13]

no worries. I gave you new code. Please see if the new screenshot matches your desired output. I did not understand your requirement initially. Apologies.

[abhinayingole]

Perfect.... thank you very much @tarheel13 .

please don't mind if I bother you again incase I stuck up 🙂

Thank you again.

[tarheel13]

haha it's never a bother! I learn so much on this forum!

[abhinayingole]

Hi @tarheel13  @Kurt_Bremser 

   My supervisor has another bizarre request with RANK value in this data 😞

This is how we develop RANK value up till now. but my client need highlighted value for obs. 7 and 8 as 5 and not 1

like this...

Reason: He says...while deriving  RANK value for obs. 7, he wanted to check latest of LLT = urea value, which is obs. no. 4 and AENDT of obs. no. 4 is not equal to ASTDT of obs. no. 7, so we need to keep adding RANK +1.

currently while deriving  RANK value for obs. 7, we are considering obs. no. 3.

can we modify our existing code for this requirement.

sorry to trouble you with this again.

I am also attaching the datalines for this.

data test;
 
input @1 usubjid $3.
      @5 ae $8.
      @14 llt $8.
      @23 astdt date9.
      @33 aendt date9.
;

format astdt aendt date9.;

datalines;
410	amino    alan     19JUL2022	08AUG2022
410	amino    aspar    01AUG2022	08AUG2022
410	Toxicity urea     26JUL2022	18AUG2022
410	Toxicity urea     02AUG2022	15AUG2022
410	Toxicity prot     08AUG2022	21SEP2022
410	Toxicity creat    11AUG2022	22AUG2022
410	Toxicity urea     18AUG2022	18AUG2022
410	Toxicity urea     18AUG2022	02SEP2022
410	Toxicity minuria  29AUG2022	21SEP2022
410	Toxicity urea dec 15SEP2022	21SEP2022
720	Toxicity estim    27DEC2022	17JAN2023
720	Toxicity urea     03JAN2023	17JAN2023
720	Toxicity creat    10JAN2023	24JAN2023
720	Toxicity estim    17JAN2023	28FEB2023
720	Toxicity urea     31JAN2023	14FEB2023
720	Toxicity creat    31JAN2023	14FEB2023
720	Toxicity urea     21FEB2023	28FEB2023
720	Toxicity estim    28FEB2023	21MAR2023 
;
 
run;

[tarheel13]

sorry I don't know how to do that. I would suggest maybe making a new thread with the new requirement. since a solution was already accepted on this thread, there is no chance for another SAS user to have their solution accepted. I think @Kurt_Bremser probably knows hash object better than me anyway.

[Tom]

Can you explain WHAT is the medical significance of this calculation?  What are you trying to "count" or "rank"?  Are you trying to collapse the conflicting date ranges into a single contiguous range?  Are you trying to collapse multiple different reported terms into a single record based on the coded Lowest Level Term from the MEDDRA vocabulary? 

Do you really need all of the records back out? Or would it be better to just collapse them?

How did you collect this data?  Why do you have overlapping AE dates for the same AE term?  Are you sure that all events with overlapping dates can/should be collapsed?  What if they got stung by two different bees?

Normally when you collect adverse events from patients (or their doctors) you get a start date and either an end date or an indication that the AE is ongoing.  Do you have the ONGOING flag for the individual records?

If you collected the data some other way (examining records or billing data perhaps?) then it is more likely you will multiple reports of the same event with conflicting or date ranges that need to be consolidated.