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: 
JJ27
a week ago
  • 6 Posts
  • 0 Likes Given
  • 0 Solutions
  • 0 Likes Received

count continuous enrollment in sas

by Fluorite | Level 6 in Statistical Procedures
a week ago
a week ago
I need to count number of patients with continuous enrollment after indexdt by month in claims data in SAS. Below is my data structure. startdt is the starting date of insurance coverage and enddt is the end date of insurance coverage. If there a gap between two start dates, the patient is categorized no continuous enrollment (continuous_enroll=0), otherwise has continuous enrollment (continuous_enroll=1).   data input; input ID $1. startdt:mmddyy10. enddt:mmddyy10. indexdt :mmddyy10. continuous_enroll; format startdt date9. enddt date9. indexdt date9. ; datalines; 1 1/1/2020 1/31/2020 3/12/2020 . 1 2/1/2020 2/28/2020 3/12/2020 1 1 3/1/2020 3/31/2020 3/12/2020 1 1 5/1/2020 5/31/2020 3/12/2020 0 2 8/1/2019 8/31/2019 3/16/2020 . 2 3/1/2020 3/31/2020 3/16/2020 0 3 4/1/2020 4/30/2020 4/13/2020 . 3 5/1/2020 5/31/2020 4/13/2020 1 3 8/1/2020 8/31/2020 4/13/2020 0 3 9/1/2020 9/31/2020 4/13/2020 1 ; run;   I need to count the number of unique patients with continuous enrollment 1 month after index date. the number of unique patients with continuous enrollment 2 month after index date. the number of unique patients with continuous enrollment 3 month after index date. .... All the way up to 12 months.   Let me know if my question is not clear. Thank you for your help! ... View more

calculate p value for ROC AUC by subgroups

by Fluorite | Level 6 in Statistical Procedures
‎12-15-2023 01:45 PM
‎12-15-2023 01:45 PM
0 I'm trying to see if there is a statistically significant difference in ROC AUC by subgroups in SAS. I can't use roccontrast because I only have one model. Here is my codes: proc logistic data=data plots(only)=roc; by Race; model htn (event='1') = bmi ; ods output roccurve=ROCdata; run; Race has 4 categories: hispanic, black, white and other I tried roccontrast, didn't work. I would really appreciate any guidance on this! ... View more

Re: exclude patients by counts and time

by Fluorite | Level 6 in SAS Programming
‎08-27-2021 01:27 PM
‎08-27-2021 01:27 PM
THANK YOU! ... View more

Re: exclude patients by counts and time

by Fluorite | Level 6 in SAS Programming
‎08-26-2021 09:40 AM
‎08-26-2021 09:40 AM
Hey @maguiremq    Thank you so much for doing this! i'm new to the community, will definitely make the reproducible form myself next time. And my date is in year-month-day format, you are correct!   So as I replied to @ChrisNZ , Patient ID 1 only had one record. I'm comparing the start_dt with fst_dt to see if the difference <12 months, if yes, then exclude, otherwise keep. start_dt is the diagnosis date for prioir_dx. fst_dt is the diagnosis date for dx. In this case, 2016-8-1 is within the 12-month window of 2017-1-1, so patient ID 1 should be dropped.    Patient ID 1 only has inpatient diagnosis (inpatient=1) so the exclusion criteria is at least one prior diagnosis within 12 months. But for patient ID 2, he/she had outpatient diagnosis (inpatient=0) only, the exclusion criteria is at least two prior diagnosis within 12 months. So it's keeped.    Thank you! Looking forward to your response!       ... View more

Re: exclude patients by counts and time

by Fluorite | Level 6 in SAS Programming
‎08-26-2021 09:31 AM
‎08-26-2021 09:31 AM
Hey ChrisNZ,    Patient ID 1 only had one record. I'm comparing the start_dt with fst_dt to see if the difference <12 months, if yes, then exclude, otherwise keep. start_dt is the diagnosis date for prioir_dx. fst_dt is the diagnosis date for dx. In this case, 2016-8-1 is within the 12-month window of 2017-1-1, so patient ID 1 should be dropped.   Hope it makes sense!   Thank you   ... View more

exclude patients by counts and time

by Fluorite | Level 6 in SAS Programming
‎08-25-2021 04:15 PM
‎08-25-2021 04:15 PM
Hi everyone!   I encountered a problem of excluding patients from claims data. My data is something like this: ID dx fst_dt prior_dx inpatient start_dt 1 1 2017-1-1 1 1 2016-8-1 2 1 2018-1-1 1 0 2016-1-1 2 1 2019-1-1 1 0 2017-1-1 3 1 2017-2-1 0 1 2015-2-1 3 1 2018-3-1 1 0 2018-1-1 3 1 2019-1-2 1 0 2018-12-1   My exclusion criteria is performing a 12 month lookback from fst_dt, if patients have at least one inpatient diagnosis (inpatient=1 and prior_dx=1) or patients have at least two outpatient diagnosis (inpatient=0 and priod_dx=1) then they will be excluded. Start_dt is the date of diagnosis of prior_dx. For example, patient ID 1 should be excluded, because this patient had an inpatient diagnosis within 12 months. Patient ID 2 should be included because the two outpatient diagnosis (inpatient=0) were before the 12 months interval. Similarly, patient 3 should be excluded.   So my output should be something like this: ID dx fst_dt prior_dx inpatient start_dt 2 1 2018-1-1 1 0 2016-1-1 2 1 2019-1-1 1 0 2017-1-1   I have tried multiple ways, but didn't work. Specifically, I need help with at least two outpatient diagnosis.  Any help is greatly appreciated!!!!   JJ ... View more
Contact Me
Online Status
Offline
Date Last Visited
a week ago
Quick links