I would like to apologise as I am unable to post the data sample here for you all, as the dataset I am working is in restricted environment (a virtual computer) which does not allow me to extract sample data. But I will make a effort to describe it.
I have a dataset where participants eGFR observations are longitudinal in nature. That is 1 person might have many eGFR observation. However, I am only interested in the values three years prior and one year after the treatment date in order for them to define cases. I have arrived at a point where I have kept all the observations by each person who have eGFR values recorded three year before date_record and one year after date_record.
In order to define cases, I am interested if the observation only has egfr<=60 three year prior and one year later from txt_date, if there is observation with egfr>60 for a participant within three year prior and one year after txt_date, then I wish to exclude participants from my data. At the moment I am having problem to code this part. I am creating a sample data below
again I am very sorry to post it without data, but please od understand my limitations.
Id | txt_date | date_ckd | egfr | case |
1 | 2007 | 2004 | 63 | exclude |
1 | 2007 | 2005 | 46 | exclude |
1 | 2007 | 2006 | 55 | exclude |
1 | 2007 | 2008 | 67 | exclude |
2 | 2008 | 2005 | 66 | yes_case |
2 | 2008 | 2006 | 75 | yes_case |
2 | 2008 | 2006 | 55 | yes_case |
2 | 2008 | 2009 | 55 | yes_case |
3 | 2009 | 2006 | 42 | exclude |
3 | 2009 | 2006 | 43 | exclude |
3 | 2009 | 2007 | 71 | exclude |
3 | 2009 | 2008 | 77 | exclude |
3 | 2009 | 2010 | 30 | exclude |
0 Response to Help to define cases in longitudinal data
Post a Comment