Descriptions:
| Term | description type | Language/acceptability | Langauge/acceptability | Case significance |
|---|---|---|---|---|
| Leukoencephaloathy due to [disease] (disorder) | FSN | us:P | gb:P | ci |
| Leukoencephaloathy due to [disease] | SYN | us:P | gb:P | ci |
| White matter disorder due to [disease] | Syn | us:P | gb:P | ci |
Concept model:
| Attribute Cardinality | Attribute | Value | role group number | Role group cardinality |
| 1..1 | 0 | N/A | ||
| 0..1 | 0 | |||
| 1..1 | 1 | 1..1 | ||
| 0..1 | 1 | |||
| 0..1 | 1 | |||
| 0..1 | 1 |
Definition status:
900000000000073002 |Defined (core metadata concept)|
Applies To:
Template Language
362977000 |Sequela (disorder)| + 87486003 |Aphasia (finding)| :
[[~1..1]] { [[~1..1]] 42752001 |Due to (attribute)| = [[+id ( <64572001 |Disease (disorder)|) @dueTo ]] }, [[~1..1]] { [[~1..1]] 255234002 |After (attribute)| = [[+id ( <64572001 |Disease (disorder)|) @dueTo ]] } Rules for description generation:
- Remove semantic tag from [disorder] concept
8 Comments
Bruce Goldberg
Hi Peter G. Williams. Please modify template language.
Thank you.
Jim Case
This one looks OK as long as we are satisfied that we do not want the causal disorder to be co-occurrent
Bruce Goldberg
These decisions are hard to make.
Jim Case
Bruce Goldberg , any suggested guidelines?
Bruce Goldberg
Jim Case An example of when it is easy is 48440001 |Arthritis due to gout (disorder)|. This as the FSN implies is modeled with due to gout. Every reference I have come across however states that gout is a kind of arthritis, so the model I had proposed does not include a due to relationship but rather incorporates definitions for both arthritis and gout. If Leukoencephalopathy is part of a genetic disorder or is a common manifestation of a multisystem disorder then model as co-occurrence.
That is the best I can come up with. I think it is a good idea for these combined disorders to be investigated one by one and templates devised based on decisions that we both agree upon or upon consensus reached during ECE/EAG calls. What do you think?
bruce
Jim Case
Bruce Goldberg, I agree that these would need to be addressed on a one-by-one basis to determine the nature of the relationships between the two disorders. I think you are getting to the crux of the issue with your mention of multiple disorders caused by the same underlying cause being co-occurrent, while disorders that are directly in the causal chain with another disorder would be modeled with DUE TO. We have a continuing issue as to when to model something as "co-occurrent and due to" vs. only due to. I was responding to inquiries from OHDSI related to renal disease caused by another disorder as to why some terms did not have the due to disorder also as a IS A parent. The best argument I can come up with is that the focus of the concept is not the causal disease, but the resulting disorder, i.e. there is a difference between the causal disorder itself and the disease the causal disorder causes.
Bruce Goldberg
Jim Case, I thought we had decided to no longer use the co-occurrent and due to pattern ? Your argument explains why you would not use the causal disease as a parent but it sounds like the question from OHDSI is when would you use the co-occurrent and due to pattern. Unless we are reconsidering using co-occurrent and due to we should make it clear that we abandoned this model in favor of just using co-occurrent and due to separately.
Jim Case
Bruce Goldberg , yes, we have abandoned the co-occurrent and due to pattern because we could not come up with specific guidelines to use it consistently. That was what I tried to explain to the OHDSI folks. We reasoned that if you wanted to see what disorders were caused by another disease you would just us ECL to search for diseases DUE TO disease. I hoped I had answered their question satisfactorily.