Using adnimerge and dxsum to identify changes to patient status

[Pete Philipson]

Hi,

Firstly, a huge thanks for putting together the R package. It is incredibly helpful and has made a seemingly daunting task much easier.

I am looking at patients who convert from MCI to dementia from the ADNI1 cohort and, in particular, need to know the time of the conversion. I have noticed that for some patient RIDs that there is a discrepancy between the EXAMDATE when the DX variable changes within adnimerge when compared o the EXAMDATE within dxsum for the same change (now to DXCHANGE). As a concrete example, consider RID 285: from adnimerge the conversion occurs at 2013-05-01 but in dxsum this is considerably later, namely at 2014-09-22. My thinking is that the dxsum data is more accurate in this instance but would appreciate it if someone could confirm this?

As a further query, I noticed (using the same example RID) that the exam dates differ after the first few visits between adnimerge and dxsum for some patients. Is the change in dx status monitored separately from the raft of repeated measures that are taken?

Many thanks, Pete.