Difference between revisions of "Minutes - Medication WG 2021-02-02"

From Health Level 7 Belgium Wiki
 
(5 intermediate revisions by the same user not shown)
Line 22: Line 22:
  
 
===== '''Agenda''' =====
 
===== '''Agenda''' =====
* compose logical datamodel
+
* magistral preparations
* magistral preparations by Jan
+
* first draft of logical datamodel medication record
  
 
===== '''Minutes''' =====
 
===== '''Minutes''' =====
* magistral preparations: Jan has not been able to work on it due to some internal deadlines
+
* '''Magistral preparations''': Jan has not been able to work on it due to some internal deadlines. Tom Henkens prepared some things on the magistral preparations. [https://drive.google.com/file/d/1P5kWeMCNl1ogeXeLaUN01c4B3blg13_d/view?usp=sharing See the summary here]. Recip-e can prescribe 3 things:
* Tom Henkens prepared some things on the magistral preparations. Recip-e can prescribe 3 things:
 
 
::* pre-packed product
 
::* pre-packed product
 
::* INN
 
::* INN
Line 33: Line 32:
 
::::* structure by defining  one or more compounds
 
::::* structure by defining  one or more compounds
 
::::::* substances
 
::::::* substances
::::::* sformulary reference
+
::::::* formulary reference
 
::::::* pre-packed product (as compound)   
 
::::::* pre-packed product (as compound)   
::::* structure: only formulary reference
+
::::* structure: defined by formulary reference only
 
::* How to translate this in a model ?
 
::* How to translate this in a model ?
::* magistral preparation: how to indicate: as a flag or derive it from one of the used codes, f.e. naming system = BE magistral preparation
+
::* How indicate that it is a magistral preparation: as a flag or derive it from one of the used codes, f.e. naming system = BE magistral preparation
 
::* 1 CNK can point to multiple GTIN, 1 GTIN can never link to multiple CNK's
 
::* 1 CNK can point to multiple GTIN, 1 GTIN can never link to multiple CNK's
::* discussion on quantium satis : there seems no need to add this in our FHIR profile
+
::* discussion on quantum satis : there seems no need to add this in our FHIR profile
 
::* do we have to encode this in such a detail ?
 
::* do we have to encode this in such a detail ?
  
* HL7 FHIR: offset is in minutes, so it could be resolved by changing the start date
+
* HL7 FHIR (international): offset is in minutes, so it could be resolved by changing the start date
::* can we put an example on this ? So that we can point out the need for an extension for it
+
::* can we put together an example on this ? So that we can point out the need for an extension for it
 
::* Belgium can also make his own extension for it
 
::* Belgium can also make his own extension for it
  
* Draft model for the medication record (we are not going to use the medication scheme as this is a reserved word in Belgium)
+
* '''Draft model for the medication record''' (we are not going to use the medication scheme as this is a reserved word in Belgium)
 
::* from architectural update
 
::* from architectural update
::* José is using the use cases for the medication record - for which there were some challenges
+
::* José is using the use cases for the medication record - for which there were some challenges...
::::* very simple summaries vs complex summaries
+
::::* Context: you can have very simple summaries vs complex summaries
 
::::* medication scheme can be updated by GP, pharmacists, nurse, ... and updates are pushed to vaults
 
::::* medication scheme can be updated by GP, pharmacists, nurse, ... and updates are pushed to vaults
::::* what if dispenses are sent to the vault ? is vault handling the update ? at the vault update is put next to the current scheme and GP or pharmacist have to determine what is the right one
+
::::* what if dispenses are sent to the vault ? is the vault handling the update ? at the vault, the update is put next to the current scheme and GP or pharmacist have to determine which is the right one
::::* a compiled summary step migth be needed = the need for a process of automated reconciliation will have to be confirmed and if so, defined  
+
::::* a compiled summary step might be needed = the need for a process of automated reconciliation will have to be confirmed and if so, defined  
 
::::* would/could it be better to work from medication dispenses ?
 
::::* would/could it be better to work from medication dispenses ?
::::* medication record could have the medication summary + details (source data f.e. prescriptions, dispenses, vaccination records, ...)
+
::::* medication record could have the medication summary + details/facts (source data f.e. prescriptions, dispenses, vaccination records, ...)
 
::::::* not all the details would be relevant to the summary
 
::::::* not all the details would be relevant to the summary
 
::::::* all dispenses, even when not prescribed, should be part of the medication record
 
::::::* all dispenses, even when not prescribed, should be part of the medication record
 
+
::::* Will it be feasible to code all these "facts" ? Only medication related.
 +
::::* could it be a solution for a "solid pot" = personal wallet with all information
 +
::::* goal is not to implement this right now, but to foresee the possibilities, to make our architecture ready for the future
 +
::::* see [http://build.fhir.org/ig/hl7-be/hl7-be-fhir-medication/branches/master/StructureDefinition-be-model-medication-record.html draft proposal here]
  
 
'''Next meeting Feb 16'''
 
'''Next meeting Feb 16'''
* examples of magistral preparations of ingredients/products with codes to distinguish the different types
+
* examples of magistral preparations of ingredients/products with codes to distinguish the different types (incl offset)
* first draft of logical model for the medication scheme
+
* feedback on first draft of logical model for the medication record

Latest revision as of 11:35, 2 February 2021

Attendees
  • Anne Nerenhausen
  • Annemieke Vergauwe
  • Bruno Casneuf
  • Elhassan Baazizi
  • Hanne Vuegen
  • Jan Lenie
  • Jean-Louis Maggetto
  • Jens Penny
  • José Costa Teixeira
  • Karlien Erauw
  • Marc Buckens
  • Richard Franken
  • Tom De Backer
  • Tom Henkens
  • Will Van Norel
Excused
  • Jeroen De Wilde
  • Robin Bosman
  • Nils Devos
Agenda
  • magistral preparations
  • first draft of logical datamodel medication record
Minutes
  • Magistral preparations: Jan has not been able to work on it due to some internal deadlines. Tom Henkens prepared some things on the magistral preparations. See the summary here. Recip-e can prescribe 3 things:
  • pre-packed product
  • INN
  • Compounded drugs/magistral preparation: text as well as structure is permitted
  • structure by defining one or more compounds
  • substances
  • formulary reference
  • pre-packed product (as compound)
  • structure: defined by formulary reference only
  • How to translate this in a model ?
  • How indicate that it is a magistral preparation: as a flag or derive it from one of the used codes, f.e. naming system = BE magistral preparation
  • 1 CNK can point to multiple GTIN, 1 GTIN can never link to multiple CNK's
  • discussion on quantum satis : there seems no need to add this in our FHIR profile
  • do we have to encode this in such a detail ?
  • HL7 FHIR (international): offset is in minutes, so it could be resolved by changing the start date
  • can we put together an example on this ? So that we can point out the need for an extension for it
  • Belgium can also make his own extension for it
  • Draft model for the medication record (we are not going to use the medication scheme as this is a reserved word in Belgium)
  • from architectural update
  • José is using the use cases for the medication record - for which there were some challenges...
  • Context: you can have very simple summaries vs complex summaries
  • medication scheme can be updated by GP, pharmacists, nurse, ... and updates are pushed to vaults
  • what if dispenses are sent to the vault ? is the vault handling the update ? at the vault, the update is put next to the current scheme and GP or pharmacist have to determine which is the right one
  • a compiled summary step might be needed = the need for a process of automated reconciliation will have to be confirmed and if so, defined
  • would/could it be better to work from medication dispenses ?
  • medication record could have the medication summary + details/facts (source data f.e. prescriptions, dispenses, vaccination records, ...)
  • not all the details would be relevant to the summary
  • all dispenses, even when not prescribed, should be part of the medication record
  • Will it be feasible to code all these "facts" ? Only medication related.
  • could it be a solution for a "solid pot" = personal wallet with all information
  • goal is not to implement this right now, but to foresee the possibilities, to make our architecture ready for the future
  • see draft proposal here

Next meeting Feb 16

  • examples of magistral preparations of ingredients/products with codes to distinguish the different types (incl offset)
  • feedback on first draft of logical model for the medication record