On 22/12/2009 14:19, Greg Caulton wrote:
David Rowen brought up an issue with the modeling of Diagnosis as being
a specialized Problem. I concur and the comments in CKM are shown
below.
The question is how do these issues get resolved - or will they not get
resolved - once modeled can an archetype never change due to
architecture e.g. the AQL requires the diagnosis to be in a specific
place in a specific archetype so there will be resistance to change?
it can change, in strictly controlled ways. What cannot change:
- existing structure & paths,
- meanings of existing at-codes
What can change:
- new structure can be added
- anything can be done to the meta-data elements, i.e. description,
translations etc etc
These conditions guarantee that queries based on the previous revision
of an archetype will not be invalidated by a more recent issue.
- thomas beale
----------------------------------------------------------------------------------------------------------------------------------------------------------------
david.rowed [Show user profile]
(07-Jul-2009 08:46) [david.rowed]
Issue/Problem vs Diagnosis
I don't accept Diagnosis is a specialisation of Problem in the "is a"
sense of specialisation.
The two are quite appropriately used in the same way by clinicians
under many circumstances but their intrinsic meanings are diferent.
When used like this, 'Problem' is the intended concept and Diagnosis
has a relationship (which needs to thought out ) to the problem. A
pathologist would not consider Diagnosis as a specialisation of Problem.
I made some of the following comments on the Issue archetype:
Problem is a specialisation or property of Issue. We need to sort out
Issue, Problem, and Diagnosis. They mean different things according to
strict definition, but their use overlaps. Problem is a type of issue.
Diagnosis can be an aspect (attribute or relationship) of a problem.
This causes untold debate in clinical IT groups and could be put to
rest if we could just come up with clear definitions of the concepts
and info models which capture these usages and separate out the
different components of each concept.
A Diagosis may clearly exist but not be a problem (depending on owner,
context, circumstance, values and perception). It may "be a" problem
for a while then, although itself persisting, cease to "be a" problem.
Specialisation, then is not the right relationship to express this, but
what is? Is it that a diagosis takes on the role of a problem and that
this role is allowed to commence, terminate, qualify etc?
Problems and Issues exist because of perceptions and values while
diagnoses exist becuse of matches against criteria, terminology and
definitions. Like the recognition/Creation of Issue and Problem, this
matching that declares a Diagosis is evaluative and so Dx, like Problem
correctly belongs in the Evaluation group of archetypes ---as we
currently have it.
gregory.caulton [Show user profile]
(01-Jun-2009 18:55) [gregory.caulton]
Re: Issue/Problem vs Diagnosis
I agree, well spotted. Two different concepts which may have a
relationship but should be modelled separately. Nursing problems and
nursing diagnoses are distinct too.
heather.leslie [Show user profile]
(12-Jun-2009 05:06) [heather.leslie]
Re: Issue/Problem vs Diagnosis
Good discussion - brings out some good points.
I think that there is enormous value in having a 'softer' problem and a
more 'rigorous' differentiation to create diagnosis. I wonder if the
discussion point here might be more about the naming of 'problem' and
'diagnosis' and 'issue' and their semantic loading.
In practice (and therefore paralleled in modelling) I think that there
is a very close relationship between 'problem' and 'diagnosis' -
largely the same data elements are required by both, with additional
elements required to support the more rigorous nature of a formal
diagnosis. This is why they have been modelled here as a parent and
child specialisation.
To me, 'issues' are softer again - highlighting things that are of
significance to the patient but not formalised in clinical terms, and
as author of the CLUSTER.issue archetype, the intent was to allow for
_expression_ of a soft concern of the patient as part of a history taking
process, not an evaluative process. In my parlance, identification of
an evaluative 'problem' is likely to be the assessment of a clinician
that follows the patient's _expression_ of a perceived issue within a
clinical consultation.
In my mind a Current Condition or Current problem list (again, whatever
your favorite phrase) will then comprise a combination of 'softer'
problems' and more rigorously defined diagnoses.
sundaresan.jaganathan [Show user profile]
(07-Jul-2009 08:07) [sundaresan.jaganathan]
Re: Re: Issue/Problem vs Diagnosis
Suggested definition/description for diagnosis.
Diagnosis is the name or label assigned to the conclusion/clinical
judgement reached through the process of clinical reasoning which is
obtained from the presentation, assessment &findings and
investigations & results, on which further management is based.
Is codeable and may appear in ICD10, ICF, NANDA, SNOMED-CT Read, ICNP
etc
Problems, issues, concerns etc are the Factors which may instigate the
purpose of an encounter with a health professional.
These would be the Reason for Encounter (different from Purpose of
Encounter) which may include symptoms and signs, a known diagnosis,
impairment, health needs, social problems etc.
Purpose of encounter would be : for consultation, treatment (eg
chemotherapy sessions), investigation, admission, check up, advice,
education, antenatal classes etc
david.rowed [Show user profile]
(07-Jul-2009 08:46) [david.rowed]
Re: Re: Re: Issue/Problem vs Diagnosis
Yes, but as a definition it needs the qualification that the name or
label must be drawn from a predetermined set of agreed pathological
states, ie it must appear in one of : {some authoratitive clinical
work, ICD10, SNOMED-CT, ......, etc.}. I can't arbitrarily
assign/invent a new name or label on the fly for the clinical
abnormalities in the case of Diagnosis but I can for a Problem.
As an aside, we commonly see things like "Abdominal Pain" under the
heading of Diagnosis on hospital discharge summaries. This is an abuse;
really there is often no Diagnosis, just a Problem which probably = the
Reason For Encounter.
david.rowed [Show user profile]
(07-Jul-2009 08:57) [david.rowed]
Re: Re: Re: Re: Issue/Problem vs Diagnosis
Problems, Issues and Concerns are references to situations which have
the possibility of evolving into, or indicating the existence of,
something impacting on well-being.
A Problem specialises / constrains an Issue to one where the impact
would be unfavourable. A Concern is probably the same as an Issue ie
more general.
I don't have a good word for an Issue that where the impact would be
favourable --ie the opposite of a Problem ----perhaps something like
an Advantage?
sundaresan.jaganathan [Show user profile]
(07-Jul-2009 09:20) [sundaresan.jaganathan]
Re: Re: Re: Re: Re: Issue/Problem vs Diagnosis
If I am not mistaken, there is a recommendation that where there is no
formal diagnosis made, the presenting problem, signs, symptoms are to
be recorded as the diagnosis as they appear in the authoratative pieces
of work.
Yes, I agree it is strictly an abuse. It should be probably be recorded
as diagnosis: None
Medicine is constantly evolving, new dignoses do appear. For a start,
the authoratative pieces of work like ICD10, SNOMED-CT do fulfil a
majority of the requirements and I would assume they undergo reviews
and change control.
We have to accept that almost nothing is 100 % achievable
Can you give me an example where you have to invent or create a new
diagnosis?
Problems, issues, concerns and needs are used for want of a better term.
Regards
Jag
--
Gregory Caulton
Principal at PatientOS Inc.
personal email: caultonpos@gmail.com
http://www.patientos.com
corporate: (888)-NBR-1EMR || fax 857.241.3022
_______________________________________________
openEHR-clinical mailing list
openEHR-clinical@openehr.org
http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-clinical
|
