Hello,   I am trying to determine the best way to handle programming for a clinical data set I have. Below is a simplified view of the current structure of the data I have. What I want is a transposed version of it. The reason I haven't been able to figure out how to transpose it is because a client can have multiple assessments at the 'PRN' timepoint. For similar data, I have been able to output separate datasets based on timepoints and then merge them back together (which worked well because there was one assessment per timepoint - no 'PRN'). I also am not sure how to best handle var1-4 (these are 4 variables that are for each assessment) - which has also made this difficult. There should be one row per unique ClientID. In addition to this, I would like to (eventually) be able to bring the dataset in SAS and update a master existing SAS dataset (based on the desired structure) when a new ClientID is detected OR there is a new survey for an existing client. Any insight would be greatly appreciated.    Current Structure: ClientID timepoint surveydate totalscore var1 var2 var3 var4 1 5th-7th Pregnancy Visit 6/1/2022 1 Yes no yes yes 1 12 Weeks Postpartum 6/2/2022 2 Yes no yes yes 1 Child 16 Months 6/3/2022 3 Yes no yes yes 1 prn 6/4/2022 7 Yes no yes yes 1 prn 6/5/2022 8 Yes no yes yes 2 5th-7th Pregnancy Visit 6/6/2022 8 Yes no yes yes 2 Child 16 Months 6/7/2022 9 Yes no yes yes   Desired Structure: Clientid timepoint1 surveydate1 totalscore var1 var2 var3 var4 timepoint2 surveydate2 totalscore2 var1 var2 var3 var4 timepoint3 surveydate3 totalscore3 var1 var2 var3 var4 timepoint4 surveydate3 totalscore4 var1 var2 var3 var4 timepoint5 surveydate5 totalscore5 var1 var2 var3 var4 1 5th-7th Pregnancy Visit 6/1/2022 1 Yes no yes yes 12 Weeks Postpartum 6/2/2022 2 Yes no yes yes Child 16 Months 6/3/2022 3 Yes no yes yes prn 6/4/2022 7 Yes no yes yes prn 6/5/2022 8 Yes no yes yes 2 5th-7th Pregnancy Visit 6/6/2022 8 Yes no yes yes Child 16 Months 6/7/2022 9 Yes no yes yes                                           ... View more

Any suggestions for which procedure to use for relative risk in which the sample size is small and two-by-two table has 0 values? My table, top to bottom/left to right: 10/0/1/3. I've used the relrisk option with proc freq. Wasn't sure if it may be best to use the logistic model and if the firth option would best handle this. Suggestions are welcomed! ... View more