We've been running throuh the "normal" and possible ranges of values for
lab tests and found that the archetype I mention in the subject does not
state but a "bigger than 0" restriction for Haemoglobin, RCC, etc.
I decided to take a look because ranges provided to me indicated
diferent "possible" ranges for male and female patients and was
wondering how would archetypes model this issue.
This is a very similar question to another one I did some time ago
regarding normal, abnormal, dangerous, etc. ranges. But in this case my
question refers to the "possible" values the test may return.
In order to differentiate male from female ranges assertions would need
to be introduced?
Or do you believe this is too a matter of decission support?
Archetypes are clinical concept models.
They communicate and express domain information models about clinical concepts.
Archetypes are ways to express clinical knowledge on what has to be recorded documenting the provision of healthcare.
Archetypes (at this moment) are not designed to deal with clinical knowledge about the interpretation of documented clinical facts. Decision support.
Fields in archetypes that are specified are containers to be used for documentation.
One type of the family of archetypes is the Observation.
One of the possible observations is the Lab-test.
For each lab-test the following things will have to be (can be) recorded about the test itself and its results:
name of the test
value/outcome/result
units of measurement
normal values
interpretation
comment
→ relevance<-
Each healthcare provider that is using this archetype will be able to record the indicated information about any lab-test.
The name of the test plus units of measurement are linked to each other.
Result is the number that is the result of a measurement that is recorded and can vary.
Normal values are dependent on local circumstances. Each lab has its own normal values for all of the tests they perform.
The interpretation of the test result and its normal values are dependent on local arrangements, clinical speciality, and patient related contexts.
The comment is additional information that has to be recorded about the test.
Thinking about it I foresee the need for a flag indicating that this lab test result is considered relevant or irrelevant by the Observer. The reason for this is: Suppose all fields are filled but the comment field states that the blood was not collected properly. Then all data that is recorded is interpreted as less relevant/reliable but can not (must not) be disregarded fully. Perhaps all Archetypes of the Observation Type need this extra field I call “Relevance”.
To answer your e-mail.
I think that the normal value and the interpretation are NOT part of the archetype definition. Within the archetype specification one can not deal with all situations that influence the normal values dependent on local contexts that vary from point in time to point in time, from place to place and from context to context. Normal values have to be provided by the lab. The interpretation has to be provided by the Observer (e.g. the lab or the physician). Or in the case of Clinical decision Support the module that handles this type of clinical knowledge.
Gerard
– –
Gerard Freriks, arts
Huigsloterdijk 378
2158 LR Buitenkaag
The Netherlands
Rodrigo
The archetype has absolute possible ranges in medicine - when we know these for sure we can add them, when we do not we can put some very large or small number to make sure it is OK and at least prevent crazy values.
The reference model deals with reference_ranges - and has a normal range as one of these. If there are different ranges for different genders or ages these can be recorded as part of the result. There is only ONE normal range - which may not be filled if the state information is not known in the lab.
This is an important point - archetypes are not about normal ranges. Cheers, Sam