Question Posted 10/11/17:

I would like to confirm how the DX variable from the ADNIMERGE CSV file is computed. I am trying to replicate it using the DXCURREN, DXCHANGE and DIAGNOSIS variables in the Diagnostic Summary CSV file, but I do not get the same proportion of patients with dementia. Could you please provide me more information on how the DX variable is derived in ADNIMERGE CSV file?
Response posted 10/16/17 by Danielle:
The DX variable in the ADNIMERGE file is based on the DXCURREN, DXCHANGE, and DIAGNOSIS variables. If you are seeing differences in diagnosis for a given participant between ADNIMERGE and what you get when you look at the DXSUM_PDXCONV_ADNIALL.csv file, it would be helpful if you send specific examples.