Stream: implementers
Topic: Organizing my profiles
Dharani Ranasinghe (Nov 16 2021 at 01:55):
I'm creating observation types profiles these days and need a guidance to structure those. We have clinical parameters like, Body weight, birth weight, mother weight, father weight. They all have different snomed ct codes. All these are except the body weight are one time parameters. (patient has only one registration ideally) Birth weight is measured in grams. Most of the time body weight is measured in Kilo grams. But there are incidents where gram is also used. mother weight, father weight are always given in Kgs. Only Body weight has an additional attribute called, State of dress ( whether the patient is naked, half dressed, with shoes, etc) When creating profiles to above parameters what is the best approach ?
- Create individual profiles for each parameter ?
- Create one profile for all of these and add the code using a value set ?
- Create a WeightCore profile without specifying any code. Create 2 derived profiles using the core profile eg: WeightLIte profile & WeightRelative. WeightLite has a value set as the code and support both Birth weight and Body weight. WeightRelative has a value set including the codes of father weight and mother weight.
Lloyd McKenzie (Nov 16 2021 at 02:29):
You might also look at ObservationDefinition
Dharani Ranasinghe (Jan 04 2022 at 12:47):
I have a use case where the height, weight and head circumference of biological father and mother are used as observations of child's growth chart. There is no link to parents medical records to get those values. Staff just enters those values in Child's medical record as measurements of Parents. Now I'm creating profiles for weight of biological mother, Weight of biological father and etc. When creating observation profiles can we use the standard vital sign profile of FHIR to derive my profiles ? I have this concern because these are not vital signs of the patient, but his parents.
René Spronk (Jan 04 2022 at 13:12):
If you have Observation.focus as a reference to the father/mother, the remainder of the structure would be the same as the profiles you mention.
Dharani Ranasinghe (Jan 06 2022 at 12:44):
René Spronk said:
If you have Observation.focus as a reference to the father/mother, the remainder of the structure would be the same as the profiles you mention.
In the production ,values of above given parameters are saved in child's medical record using specific keywords. We need to retrieve those values and send through FHIR APIs using SNOMED codes. eg:
Weight of the biological father
SCTID: 72681000052105
Weight of the biological mother
SCTID: 72691000052107
René Spronk (Jan 06 2022 at 13:37):
.. which means that regardless of whether or not focus is populated, one will know what the observation is. No template problems whatsoever..
Dharani Ranasinghe (Jan 07 2022 at 06:04):
René Spronk said:
.. which means that regardless of whether or not focus is populated, one will know what the observation is. No template problems whatsoever..
I'm not sure if I understood your comment completely . But my question is can we derive this type of observation profiles( which shows measurements of parents) , from the FHIR standard 'Vital sign' profile ?
René Spronk (Jan 07 2022 at 08:20):
IMHO: yes, no problem
Dharani Ranasinghe (Jan 12 2022 at 03:45):
René Spronk said:
IMHO: yes, no problem
``
René Spronk : can we use following codes in those derived profiles (Observation.code)?
Weight of the biological father
SCTID: 72681000052105
Weight of the biological mother
SCTID: 72691000052107
René Spronk (Jan 12 2022 at 08:15):
In the http://build.fhir.org/vitalsigns.html profile, Observation.code has an extensible binding. As such codes for concepts not currently present bin the listed value set can be used. Currently LOINC is the preferred coding system, so to follow the spirit of the vitals sign profile you could try and use the LOINC codes for 'Weight of the biological father/mother' if LOINC equivalents exist. If not, you can extend the value set with SNOMED codes.
Last updated: Apr 12 2022 at 19:14 UTC
