cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
sri21592
‎02-14-2020
  • 6 Posts
  • 0 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: Finding the continous and discontinous months enrollment over the ...

by Fluorite | Level 6 in SAS Programming
‎12-31-2019 02:11 PM
‎12-31-2019 02:11 PM
Hi @ed_sas_member,   Thanks for the quick response and it worked as magic. I just need to find the way to display the discontinued no.of times.   Thanks again! ... View more

Re: combining overlap date ranges based on the effective run dates and...

by Fluorite | Level 6 in SAS Programming
‎12-31-2019 01:17 PM
‎12-31-2019 01:17 PM
Hi @ed_sas_member ,   I am not actually losing any information since that is a duplicate entry. They might have entered multiple entries for the different drug codes. They initially might have a drug code 'A' program but in between they might have shifted to new drug_code 'B' for few months and they might come back again to Drug code 'A' program. Effective run date tells us which is the latest entry in the system since those are updated after 6 months to 1 year of actual program enrollment dates.   e.g. for patient 699 -> I will lose the drug intake of drug BB between  nov 2015 and dec 2015, because same person cannot have more than more than one drug at same span. Nov and Dec 2015 are captured by the first record with latest entry date which means that the first record is true for nov and dec 15.   you might have question on how can two drug codes can be at same effective date. (Current system captures the  date but not the timestamp for the entries, it might change in future but we have very less cases of same effective dates compared to millions of overlap date ranges). ... View more

Re: combining overlap date ranges based on the effective run dates and...

by Fluorite | Level 6 in SAS Programming
‎12-31-2019 10:07 AM
‎12-31-2019 10:07 AM
Hi @ed_sas_member,   Thanks for the quick response and trying to help me. Greatly appreciated. Here is the actual data set for combined scenarios 2.1,2.2 and 2.3 together.   Data set for scenarios - 2.1,2.2 and 2.3:   ID agent_id drug_code beg_date end_date effective_run_date 11 450 BC 11/1/2016 12/31/2016 11/2/2016 11 450 BD 11/1/2016 12/31/2016 12/2/2016 1 580 A1 1/1/2014 5/31/2018 8/29/2014 1 580 586 11/1/2017 11/30/2017 5/7/2018 1 580 567 12/1/2017 10/31/2018 5/7/2018 3 757 O1 8/1/2017 1/31/2019 6/26/2017 3 757 P1 8/1/2018 8/31/2018 8/3/2018 6 999 P1 9/1/2015 11/30/2016 10/20/2015 6 999 E1 10/1/2015 11/30/2015 10/20/2015 8 686 AL 1/1/2016 8/31/2016 1/4/2016 8 686 BL 6/1/2016 6/30/2016 1/23/2017 8 686 779 8/1/2016 8/31/2016 1/26/2017 8 686 775 8/1/2016 1/31/2017 1/12/2017 8 686 779 10/1/2016 1/31/2017 1/12/2017 9 99 BA 11/1/2015 12/31/2015 11/30/2015 9 99 CA 12/1/2015 12/31/2015 11/30/2015 10 699 F1 3/1/2015 12/31/2015 11/29/2016 10 699 BB 11/1/2015 3/31/2016 11/7/2016     Expected results: (I included comments and scenario type) ID agent_id drug_code beg_date end_date effective_run_date comments Scenario 11 450 BD 11/1/2016 12/31/2016 12/2/2016 I don’t want the first record since both records date ranges are same but effective date should be latest Scenario 2.1 1 580 A1 1/1/2014 10/31/2017 8/29/2014 (Changed the end date since the effective date is not latest compared to the other dates and I am not missing any unique months by changing the date). Scenario 2.2 1 580 586 11/1/2017 11/30/2017 5/7/2018 1 580 567 12/1/2017 10/31/2018 5/7/2018 3 757 O1 8/1/2017 7/31/2018 6/26/2017 I split the first record into two records since there is other reecord with one month with latest effective date. I am not losing any unique months again. Scenario 2.2 3 757 O1 9/1/2018 1/31/2019 6/26/2017 3 757 P1 8/1/2018 8/31/2018 8/3/2018 6 999 P1 9/1/2015 9/30/2015 10/20/2015 Here, when effective dates are same, I want to keep the overlap date ranges and split the non overlap months into three different records. Scenario 2.3 6 999 P1 10/1/2015 11/30/2015 10/20/2015 6 999 P1 12/1/2015 11/30/2016 10/20/2015 6 999 E1 10/1/2015 11/30/2015 10/20/2015 8 686 AL 1/1/2016 5/31/2016 1/4/2016 you can this case is three scenarios combines, first looking if they have same date ranges and removing the non latest effective date, changing the end dates based on the effective run dates keeping all the month date ranges and finally scenario 2.3 splitting the date ranges when they have same effective date ranges. combined scenarios 2.1,2.2 and 2.3 8 686 AL 7/1/2016 7/31/2016 1/4/2016 8 686 BL 6/1/2016 6/30/2016 1/23/2017 8 686 779 8/1/2016 8/31/2016 1/26/2017 8 686 775 9/1/2016 9/30/2017 1/12/2017 8 686 775 10/1/2016 1/31/2017 1/12/2017 8 686 779 10/1/2016 1/31/2017 1/12/2017 9 99 BA 11/1/2015 11/30/2015 11/30/2015 I am just splitting the first record into two date ranges since there is overlap and splitting non overlap date ranges since both have same effective run dates. Scenario 2.3 9 99 BA 12/1/2015 12/31/2015 11/30/2015 9 99 CA 12/1/2015 12/31/2015 11/30/2015 10 699 F1 3/1/2015 12/31/2015 11/29/2016 Here, second record date range is overlapped but since the effective run date is not latest, I changed it after the end date of the first record. Scenario 2.2 10 699 BB 1/1/2016 3/31/2016 11/7/2016 ... View more

Finding the continous and discontinous months enrollment over the 13 m...

by Fluorite | Level 6 in SAS Programming
‎12-30-2019 03:43 PM
‎12-30-2019 03:43 PM
Hello SAS community,   I have 13 months span data but I need to know if the person is pre-enrolled in first month before the start of 12months span. For example: I have data from 201712 to 201812, 201712 is to flag if the person is enrolled before or the new enrollment.   I want to have new variables which are defined as: Total_Elig_Enrolled_Months - Total no.of months person is enrolled over the 12 months span excluding the pre-enrolled month. Total_missing_months - total no.of span months(default it is 12) subtracting with total_elig_Enrolled_Months  Enroll_Continous_months - total no.of months enrolled consecutively over 12 months span excluding the first month. No.of_times_discontinued - data discontinued for the consecutive months, if person has enrolled from 201801 to 201810 and discontinued for a month and then enrolled back again from 201812 then the no of discontinued times is '1', where as if the person is not enrolled  from 201801 but enrolled in 201805 to 201810 and then from 201812  then the no.of times discontinued should be '2'. Enrolled_all_12_months - If the person is enrolled in all 12 months of span. New_enrollment - If person is not enrolled in first month which is 201712 then the person is considered as new enrollment, if present in 201712 then person is not newly enrolled into the program.   I appreciate the help. I am new to SAS and in learning phase. Thanks again for the help.   Source data: Person_id region code Enroll_month 101 10001 201712 101 10001 201801 101 10001 201802 101 10001 201803 101 10001 201804 101 10001 201812 102 10003 201712 102 10003 201805 102 10003 201806 102 10003 201807 102 10003 201808 102 10003 201809 102 10003 201810 103 10009 201812   Expected Data: Person_id region_code Total_Elig_enrolled_months Total_Missing Months Enroll_Continous_Months No.of_times_discontinued Enrolled_all_12_months New_Enrollment 101 10001 5 7 4 1 No No 102 10003 5 7 5 2 No No 103 10009 5 11 1 2 No Yes ... View more

Re: combining overlap date ranges based on the effective run dates and...

by Fluorite | Level 6 in SAS Programming
‎12-30-2019 01:38 PM
‎12-30-2019 01:38 PM
code for Scenario 1: Code for scenario 1: (which I achieved exactly as I wanted) data WORK.new_end_dates_v1 (drop=id2 drug_code2 rename=(beg2=beg_dos end2=end_dos edbc2=edbc_dos)); set WORK.sort; retain beg2 end2 edbc2; agent_id2=lag1(agent_id); id2=lag1(id); drug_code2=lag1(drug_code); if agent_id2=agent_id and id2=id and drug_code2=drug_code and beg_date le(end2+1) then do; beg2=min(beg_date,beg2); end2=max(end_date,end2); edbc2=max(edbc_run_date,edbc2); end; else do; seg+1; beg2=beg_date; end2=end_date; edbc2=edbc_run_date; end; format beg2 end2 edbc_run_date mmddyy10.; run; data WORK.new_end_dates_v2(drop=agent_id2 seg beg_dos end_dos edbc_dos); retain beg_date end_date edbc_run_date; set WORK.new_end_dates_v1; by agent_id id seg; format beg_date end_date mmddyy10.; if first.seg then do; end; if last.seg then do; beg_date = beg_dos; end_date = end_dos; edbc_run_date = edbc_dos; output; end; run; ... View more

combining overlap date ranges based on the effective run dates and kee...

by Fluorite | Level 6 in SAS Programming
‎12-30-2019 01:37 PM
‎12-30-2019 01:37 PM
Hi Data set: I have a data set with variables as ID agent_id drug_code beg_date end_date effective_run_date I want to combine the date ranges if rows have same id, agent_id,drug_code without any gaps between beg_date and end_dates of next rows with same data variables with bringing the max effective run date. Scenario 1: ID agent_id drug_code beg_date end_date effective_run_date 1 A1 100 1/1/2018 12/31/2018 12/1/2018 1 A1 100 1/10/2018 3/31/2019 3/1/2019 1 A2 100 1/1/2018 5/31/2018 4/1/2018 1 A2 100 2/1/2018 3/31/2018 3/1/2018 1 A2 200 4/1/2018 12/31/2018 12/1/2018 1 A2 200 1/2/2019 5/31/2019 5/1/2019 Expected data for scenario 1: ID agent_id drug_code beg_date end_date effective_run_date 1 A1 100 1/1/2018 3/31/2019 3/1/2019 1 A2 100 1/1/2018 5/31/2018 4/1/2018 1 A2 200 4/1/2018 12/31/2018 12/1/2018 1 A2 200 1/2/2019 5/31/2019 5/1/2019 which I achieved successfully for scenario 1 Scenario 2: From here the scenarios are tricky since we have same ID,Agent_ID can have enrolled for different drug_codes and we want to keep all the unique records based on the latest effective run dates: Scenarios 2.1: ID agent_id drug_code beg_date end_date effective_run_date 1 A3 100 1/1/2018 3/31/2019 3/1/2019 1 A3 200 1/1/2018 3/31/2019 3/31/2019 In the above case we have two different drug_codes but with same beg_date and end_date but different effective date. In this case I want to remove the first record since the effective date is not the latest. Expected data for scenario 2.1: ID agent_id drug_code beg_date end_date effective_run_date 1 A3 200 1/1/2018 3/31/2019 3/31/2019 Scenarios 2.2: ID agent_id drug_code beg_date end_date effective_run_date 1 A4 301 1/1/2018 3/31/2019 3/1/2019 1 A4 302 2/1/2018 3/31/2019 3/31/2019 In the above case we have same id and different drug_code and also the beg_dates are not same, in this scenario I do not want to delete first record since it has one month which is unique. I want to change the end_date of the first record (next record beg_date -1 ) when the effective_run_date is not latest. Expected data for scenario 2.2: ID agent_id drug_code beg_date end_date effective_run_date 1 A4 301 1/1/2018 1/31/2018 3/1/2019 1 A4 302 2/1/2018 3/31/2019 3/31/2019 Scenarios 2.3: ID agent_id drug_code beg_date end_date effective_run_date 1 A5 450 2/1/2018 5/31/2019 2/1/2019 1 A5 459 11/1/2018 5/31/2019 2/1/2019 In the above case, as we have two drug_codes with different date ranges but same effective_run_date then I want to have my date range of first record to split based on the overlap date range. Expected data for scenario 2.3: ID agent_id drug_code beg_date end_date effective_run_date 1 A5 459 2/1/2018 10/31/2018 3/31/2019 1 A5 459 11/1/2018 3/31/2019 3/31/2019 1 A5 450 11/1/2018 3/31/2019 3/31/2019 Actual data have more than two drug_codes and millions of records in the source data set. I couldn't achieve the scenario 2.1,2.2,2.3.  Any help will be greatly appreciable. ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎02-14-2020 12:43 PM
Quick links