Patient diagnosis

[minh...@bcm.edu]

Hi all,

Which files of the following three should I use to get the most reliable and updated diagnosis for Alzheimer's disease: ADNIMERGE.csv, DXSUM_PDXCONV_ADNIALL.csv, and dxsum.sas? Also, if a person has dementia in ADNIMERGE.csv, does that one also have Alzheimer's disease?

For example, I have encountered mismatched diagnosis among these files:

- Diagnosis and exam dates in DXSUM_PDXCONV_ADNIALL.csv do not match with those in ADNIMERGE.csv (e.g. RID 230 at visit m36 has (NL to Dementia) in ADNIMERGE.csv but (5=Conv:MCI to AD) in the other; RID 984 has some mismatched examdates for different viscode).

- dxsum.sas indicates which dementia cases are due to either Alzheimer's disease or other etiology.  So there are patients like RID 552, at viscode m60, has MCI to AD in  DXSUM_PDXCONV_ADNIALL.csv; MCI to Dementia due to other etiology in dxsum.sas; and MCI to Dementia in ADNIMERGE.csv. 

Thank you,

Minh

[Danielle J Harvey]

Hi Minh,

In general, all of those files are using the same information. ADNIMERGE and the dxsum.sas file are both drawing from the DXSUM_PDXCONV_ADNIALL.csv file (dxsum.sas should just be reading in that file and formatting/labeling variables according to the data dictionary). 

When I checked RID 230 at m36, I see NL to dementia in all three files that you noted. 

DXSUM_PDXCONV_ADNIALL.csv has a variable for etiology (see, for example, DXDDUE and see the Data Dictionary file for coding information).

ADNIMERGE only contains the global categories of diagnosis (NL, MCI, dementia), so does not include the information about the etiology - it is best to go to the original data file, 

DXSUM_PDXCONV_ADNIALL.csv

Danielle

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of minh...@bcm.edu <minh...@bcm.edu>
Sent: Wednesday, March 22, 2017 1:40 PM
To: Alzheimer's Disease Neuroimaging Initiative (ADNI) Data
Subject: [adni-data] Patient diagnosis

 

--
You received this message because you are subscribed to the Google Groups "Alzheimer's Disease Neuroimaging Initiative (ADNI) Data" group.
To unsubscribe from this group and stop receiving emails from it, send an email to adni-data+...@googlegroups.com.
To post to this group, send email to adni...@googlegroups.com.
Visit this group at https://groups.google.com/group/adni-data.
For more options, visit https://groups.google.com/d/optout.

[minh...@bcm.edu]

Hi Danielle,

Thanks a lot for your response!

I figured that since the record  for RID 230 at m36 was from ADNI1, I should have looked at DXCONV and DXREV, instead of DXCHANGE (which states 5 = MCI to dementia) in DXSUM_PDXCONV_ADNIALL.csv.

I have another question about the etiology. For RID 984, dementia symptoms were due to Alzheimer's disease at m84 and m96 but later at m108 were due to trauma. Should I trust the later diagnosis (at m108) more than earlier diagnoses or RID 984's dementia was due to both Alzheimer's disease and dementia? 

Thank you,

Minh

[Danielle J Harvey]

Hi Minh,

 

In the DXSUM_PDXCONV_ADNIALL file that I just downloaded, DXCHANGE is missing for RID 230 at m36, so I only see information for DXCURREN, DXCONV, and DXCONTYP.

 

As for the reliability of the etiology of the dementia and how to handle this when it changes, I don’t really have a good answer. I’d recommend submitting a question through the Ask the Experts tool and direct it to the Clinical Core (http://adni.loni.usc.edu/support/experts-knowledge-base/ask-experts/). The Clinical Core folks might better be able to address this question (though I would say that any assessment of etiology is a guess until autopsy!)

 

Danielle

[minh...@bcm.edu]

Thank you very much, Danielle!

[Dominik Blum]

Hi Danielle,

I also observed a diagnosis mismatch between adnimerge and DXSUM_PDXCONV_ADNIALL for several subjects, e.g.

RID 4220:

ADNIMERGE, DXbl = 3 (examdate 23 Sep 2011) but
DXSUM_PDXCONV_ADNIALL, DXCHANGE = 2 at all visits incl. screening (examdate 2 Sep 2011) and baseline (examdate 12 Oct 2011).

How is this possible?

Thanks in advance

Dominik

[Danielle J Harvey]

Hi Dominik,

I just downloaded both the ADNIMERGE file and the Diagnostic Summary file. For RID 4220, ADNIMERGE has DX_BL="EMCI" and the Diagnostic Summary file has DXCHANGE=2 (which is MCI). In the Diagnostic Summary file, diagnosis only has the 3 global diagnosis categories (Normal/Cognitively Normal, MCI, or AD/Dementia) and whether there was a change in diagnosis from one category to another. EMCI is still MCI and SMC is still Normal. We get the EMCI or SMC from the ARM table, which is how they were identified at the screening visit (and were filling a particular recruitment arm).  So, at least for RID 4220, I am not seeing a diagnosis mismatch.

Danielle

---

From: adni...@googlegroups.com <adni...@googlegroups.com> on behalf of Dominik Blum <dominik...@gmail.com>
Sent: Monday, February 11, 2019 1:00 AM

To: Alzheimer's Disease Neuroimaging Initiative (ADNI) Data

Subject: Re: [adni-data] Patient diagnosis