I am trying to use the LEDD information for other analysis in PPMI. I found the LEDD information from the file, LEDD_Concomitant_Medication.csv. I am struggling with the EVENT_ID, which has useless info (ED and LOG). Also, I tried to match the REC_ID in other data files to get useful EVENT_ID but did not work. I guess that I have to the date information. I wonder if anyone has tips for matching the data using the dates.
head( data.frame.LEDD )
REC_ID PATNO EVENT_ID PAG_NAME LEDTRT TOTDDA LEDDSTRMG LEDDOSSTR LEDDOSE LEDDOSFRQ STARTDT STOPDT LEDD ORIG_ENTRY LAST_UPDATE
1 IA248838 3001 ED LEDDLOG Carbidopa/Levodopa CR NA 100 25/100 mg PO 2 1 02/2023 01/2024 150.0000000000000000 03/2023 2024-09-25 00:00:00.0
2 IA98224 3001 ED LEDDLOG Amantadine NA 100 100 mg PO 1 1 03/2022 07/2022 100.0000000000000000 09/2022 2023-11-14 00:00:00.0
You could try using the most recent datacut, which does have LEDD for each corresponding visit for each PATNO, which was probably determined from the dates as you suggested. Someone has already gone through the trouble for you - the file is available among the LONI PPMI files, and is usually titled something like, “PPMI_Curated_Data_Cut_Public_20240129”.
My other suggestion would be to use the start date and stop date for each medication to determine if these medications are being taken at each visit after the initial baseline visit.
Wow, really great to see this @marcojrusso, as I went through this trouble and was ready to answer @jaeyoon.chung to talk to me so I could provide him the code I’ve created to do this. But that’s great to know and thank you!
The same question was sent to the MJFF data resources team which I responded to as below. However, I also agree that it’s worth taking a look at the curated data cut as well…
"The LEDD values are not synchronised with the study visits. The participant’s medication can change at any time and this means the LEDD value may increase of decrease in between study visits, potentially several times. To match the LEDD value with other measurements taken at the study visits, some options are: (a) take the LEDD value at the time of the study visit; (b) calculate the mean LEDD value in the period between the previous and current study visits; or (c) depending on the nature of your research you might consider another value such as the maximum LEDD value since the previous study visit.
LEDD is discussed in some detail in the PPMI data user guide, section 7.1 and you may find script 7 useful in order to determine the LEDD value for a participant at a point in time. To determine the LEDD value for a participant at the time of a study visit, first determine the date of the study visit from the INFODT field on whichever measurement table you are using, then find the latest STARTDT (for the participant) in the output from script 7 that is less than or equal to the INFODT. For example, if, for an individual participant you are looking at MDS-UPDRS study visit V06 with an INFODT of 05/01/2014, then you need to identify the maximum STARTDT on the derived LEDD table (i.e. the output of script 7 filtered by the PATNO of interest) that does not exceed 05/01/2014, and look up the corresponding LEDD value."
