mlhim-owners team mailing list archive

[Tim Cook <timothywayne.cook@xxxxxxxxx>]

On Fri, Apr 29, 2011 at 8:38 PM, Tim Cook <timothywayne.cook@xxxxxxxxx>
wrote:

      1) Invalid: The original definition is "The value as represented
in the instance is not a member of the set of permitted data values in
the constrained value domain of a variable". This whole thing about "the
constrained value domain of a variable" is far-fetched. My suggestion is
using just "variable" instead. Everybody in the healthcare data world
will understand that. So, my suggestion for the entire definition of the
Invalid type of EV would be: "The value as represented in the instance
is not a member of the set of permitted data values in the variable".

Implemented.

     2) Not Applicable: The definition is "No proper value is applicable
in this context e.g., last menstrual period for a male". It is OK, but
the researchers in the gender field might dispute the example. I would
use an example with zero probability of misinterpretation, such as "the
number of cigarrettes smoked per day by a non-smoker subject".

Implemented.


     3) Derived: The definition is "An actual value may exist, but it
must be derived from the provided information; usually an expression is
provided directly". I didn't understand the definition very well. Does
that mean that the information that was received was already received in
a derived way or does that mean that the information was received in a
original format, and then derived by the receiver? I need to understand
that to make the second question, about what "usually an expression is
provided directly" really means.


21090 doesn't really go into any more depth on this except to say that it is one of the special nullFlavors. Remember that these are actually the HL7 NullFlavor table entries.  So, I have the same questions you have.  Maybe someone with access to HL7 documents can shed more light on it.  But then, they wouldn't be allowed to tell us would they.  :) 
Actually, I can't see how a Derived value should be a NullFlavor or ExceptionalValue.  But as an informative flag maybe it is useful?


     4) Other: The definition is "The actual value is not a member of the permitted data values in the constrained value domain of a variable. (e.g., concept not provided by code system)". Same as (4). My suggestion is: "The actual value is not a member of the permitted data values in the variable. (e.g., when the value of the variable is not provided by the coding system)"


Implemented.


 

     5) Unencoded: The definition is "No attempt has been made to encode
the information correctly but the raw source information is represented,
usually in original text". What does "usually in original text" means?
Does that mean that the raw information is represented in free text? If
yes, then my suggestion for the definition is "No attempt has been made
to encode the information correctly but the raw source information is
represented, usually in free text".


Implemented. 
 
 

    create a new cathegory called "Asked but Refused", wich definition
might be "Information was sought but refused to be provided (e.g.,
patient was asked but refused to answer)".


Implemented as ASKR. 


    Trace: So, I would allow negative traces, defining it as "The
content is greater or less than zero but too small to be quantified".


Implemented.


** Changed in: mlhim-specs
       Status: In Progress => Fix Committed

-- 
You received this bug notification because you are a member of Multi-
level Healthcare Information Modelling Owners, which is subscribed to
MLHIM Specifications.
https://bugs.launchpad.net/bugs/729437

Title:
  Some suggestions for the EV package

Status in MLHIM Constraint Definition Designer:
  Triaged
Status in MLHIM Specifications:
  Fix Committed

Bug description:
  I've just translated the *.po file for the EV package, but there are
  some suggestions I'd like to give for this package.

  OK, I understand that we need to keep compatible with ISO 21090, but I
  had trouble understanding some definitions, I noticed that other
  people could have trouble understanding some definitions, and there
  are "missing cathegories" of EVs. Let's go:

  1) Invalid: The original definition is "The value as represented in the instance is not a member of the set of permitted data values in the constrained value domain of a variable". This whole thing about "the constrained value domain of a variable" is far-fetched. My suggestion is using just "variable" instead. Everybody in the healthcare data world will understand that. So, my suggestion for the entire definition of the Invalid type of EV would be: "The value as represented in the instance is not a member of the set of permitted data values in the variable".
  2) Not Applicable: The definition is "No proper value is applicable in this context e.g., last menstrual period for a male". It is OK, but the researchers in the gender field might dispute the example. I would use an example with zero probability of misinterpretation, such as "the number of cigarrettes smoked per day by a non-smoker subject".
  3) Derived: The definition is "An actual value may exist, but it must be derived from the provided information; usually an expression is provided directly". I didn't understand the definition very well. Does that mean that the information that was received was already received in a derived way or does that mean that the information was received in a original format, and then derived by the receiver? I need to understand that to make the second question, about what "usually an expression is provided directly" really means.
  4) Other: The definition is "The actual value is not a member of the permitted data values in the constrained value domain of a variable. (e.g., concept not provided by code system)". Same as (4). My suggestion is: "The actual value is not a member of the permitted data values in the variable. (e.g., when the value of the variable is not provided by the coding system)"
  5) Unencoded: The definition is "No attempt has been made to encode the information correctly but the raw source information is represented, usually in original text". What does "usually in original text" means? Does that mean that the raw information is represented in free text? If yes, then my suggestion for the definition is "No attempt has been made to encode the information correctly but the raw source information is represented, usually in free text". If that's not what that means, then I didn't understand at all.
  6) Asked but Unknown: The definition is "Information was sought but not found (e.g., patient was asked but did not know)". That's OK, but epidemiologists know that there's a difference between "patient was asked but did not know" and "patient was asked but refused to answer". So, my suggestion is to keep "Asked but Unknown" the way it is and create a new cathegory called "Asked but Refused", wich definition might be "Information was sought but refused to be provided (e.g., patient was asked but refused to answer)".
  Trace: The definition is "The content is greater than zero but too small to be quantified". The vast majority of traces are really greater than zero, but you can have negative traces, such as negative temperatures and some concepts in health economics and demography. So, I would allow negative traces, defining it as "The content is greater or less than zero but too small to be quantified".