Hello. I have the following dataset per subject ID where i need to calculate difference in hours based on each day. Logic: For day = 0, where i need to take the MAX DRUG_stop_DTM - Enrolled_DTM. Next days (EXCEPT THE LAST DAY for that subjectID), the difference in hours needs to be static set to 24 hours.  Last day: final_stop_DTM - first drug_start_dtm for that last day. If the "Final_stop_dtm" is missing then use max drug_stop_dtm for that day    Have: PatientID Enrolled_DTM days Drug_start_DTM Drug_stop_STM WEIGHT final_stop_DTM 1232 26MAR2020:13:30:00 0 26MAR20:13:30:00 26MAR20:23:59:59 74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 1 27MAR20:00:00:00 27MAR20:23:59:59 74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 1 27MAR20:03:46:00   74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 1 27MAR20:20:04:00   74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 2 28MAR20:00:00:00 28MAR20:13:00:00 74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 2 28MAR20:04:11:00   74.2 28MAR2020:21:56:00 1232 26MAR2020:13:30:00 2 28MAR20:04:46:00   74.2 28MAR2020:21:56:00 4456 02JAN2020:16:11:00 0 02JAN20:18:16:00 02JAN20:23:59:59 80.9 . 4456 02JAN2020:16:11:00 1 03JAN20:00:00:00 03JAN20:15:50:00 80.9 . 4456 02JAN2020:16:11:00 1 03JAN20:15:50:00 03JAN20:23:59:59 80.9 . 4456 02JAN2020:16:11:00 2 04JAN20:00:00:00 04JAN20:21:11:00 80.9 .   want; PatientID days WEIGHT hours_difference 1232 0 74.2 10.5 1232 1 74.2 24 1232 2 74.2 22 4456 0 80.9 7.82 4456 1 80.9 24 4456 2 80.9 21.18   What I have done so far is break the dataset by days. for day = 0 but then i wasn't sure how to do the "last" by group per patient   any help is appreciated.    ... View more

Hi. Overlay histogram works but i would like to have certain color appear on top of another.    proc univariate data=work.data; WHERE var1 between 0 and 100; class class2; var var1; histogram var1 / endpoints = 0 to 100 by 10 normal overlay VAXISLABEL = "Percentage of patients" ; run;   Note: class1 = 0 is blue color and class1 = 1 is red color.  I want the blue to appear on top of red. how do i do that? ... View more

I have the following code . I would like to do is add BOLD headers for LATEX output. I am not getting the same results in LateX   Group Body System Rate               ods tagsets.simplelatex file="file_destination\file_name.tex" (notop nobot) stylesheet="sas.sty" (url="sas"); options linesize=256;   proc report data=work.SAE  nowindows headline headskip ; column group body_system_soc rate_percent _rate ; define group / group "Treatment" width =14 style(header)={font_weight=bold}; define system _soc / display "Body Systems" width =14 style(header)={font_weight=bold}; define rate_percent/ display "Rate" width =14 style(header)={font_weight=bold}; run; ods tagsets.simplelatex close; ... View more

I have the following dataset with overlapping time constraints id time_start time_stop measure1 measure2 1 1/2/2020 0:00 1/2/2020 21:12 1 3.4 1 1/2/2020 10:20 1/2/2020 10:45 2 4.2             Since one time frame is a subset of longer time frame, i need to divide them to something like this:    id time_start time_stop measure1 measure2 1 1/2/2020 0:00 1/2/2020 10:20 1 3.4 1 1/2/2020 10:20 1/2/2020 10:45 2 4.2 1 1/2/2020 10:45 1/2/2020 21:12 1 3.4   I am not sure how to start this but this is my attempt   data work.two;       merge work.one work.one(keep=id time_start measure1 measure2   firstobs=2                                 rename=(id=_id time_start=_start_date));         format new_stop_date datetime20.;       if id=_id and (_start_date < time_stop and time_stop > time_start) then new_stop_date=_start_date;       else new_stop_date=time_stop;         rename time_start=new_start_date;       drop time_stop _:; run; ... View more

I have the following tables were subjects have overlapping date/time but i need to cut them by certain time to calculate different measures   have table. For example row 1 has a longer duration but row 2 has short. I need to assign row 2 start time as new stop time for row 1and create a row 3 that has row 1 as new stop dt but with row 2 stop dt time as new start time for row 3. It can have more than 2 observations.    patient start_time stop_time value1 value2 1 12/1/2019 0:00 12/1/19 22:12 1 4 1 12/1/2019 10:12 12/1/19 10:42 0.5 4   Want: patient start_time stop_time value1 value2 1 12/1/2019 0:00 12/1/2019 10:12 1 4 1 12/1/2019 10:12 12/1/19 10:42 0.5 4 1 12/1/19 10:42 12/1/19 22:12 1 4   I have no idea how to start to be honest,    I tried _n = n + 1 for start_time < stop_date but i have no idea ... View more

I have the following dataset   WHat I want to do is PER Patient PER DAY, if two stop date/times are the same, then I would like to make the second date/time as the new stop date time for the previously record BUT second start and stop date remain the same. If there are no same stop dates, then keep the original   patientid method start_date stop_date weight day 1 2 1/2/19 11:20 1/2/19 20:12 21 1 1 3 1/2/19 18:15 1/2/19 20:12 10 1 1 2 1/3/19 1:00 1/3/19 21:01 10 2 2 2 1/4/19 9:12 1/4/19 14:01 12 1 2 2 1/5/19 9:16 1/5/19 14:01 12 1   Something like this:   patientid method start_date stop_date weight day new_start_date new_stop_date 1 2 1/2/19 11:20 1/2/19 20:12 21 1 1/2/19 11:20 1/2/19 18:15 1 3 1/2/19 18:15 1/2/19 20:12 10 1 1/2/19 18:15 1/2/19 20:12 1 2 1/3/19 1:00 1/3/19 21:01 10 2 1/3/19 1:00 1/3/19 21:01 2 2 1/4/19 9:12 1/4/19 14:01 12 1 1/4/19 9:12 1/4/19 14:01 2 2 1/5/19 9:16 1/5/19 14:01 12 1 1/5/19 9:16 1/5/19 14:01   I have no idea how to start this! 😞 ... View more

I have the following table. What I would like to do is report the sum of all 1's overall by subjectID. Only 1's, no 2's, 0's, missing   have SubjectID value1 value2 value3 1 1 0   1 0 1 2 1 0 0 0 1 0 0 0 2 0 1 0 2 0 0 0 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 2 3 0 0 0 4 0 0 0 4 1 0 1 4 1 1 0           SubjectID total 1 2 2 2 3 0 4 4   i tried the following proc sql; select subjectid, count(subjectid) as total from a.have where value1 = 1 or value2 = 1 or value3= 1;quit; ... View more

I am looking for the following format age_group gender race Frequency Percent CumFrequency CumPercent 1 female white         1 male white         2 female white         2 male black         3 female asian         3 male black         ods output onewayfreqs= test_owf proc freq data=test ; table gender race/missing; by age_group; run; ods listing;   I get what I want but i was wondering if there is an easier way because when i do ods output , it always ends up outputting the results. i just want the output in different dataset rather than showing in results view   ... View more