2017-07-14 Learning Health Systems Call
Jump to navigation Jump to search
|Facilitator||Russell Leftwich||Note taker(s)||Emma Jones|
|John Roberts||Tennessee Department of Health|
|Asim Muhammad||Philips Research Europe|
|Laura Heermann-Langford||Intermountain Healthcare|
|Chris Melo||Phillips Healthcare|
|Benjamin Kummer||Columbia University|
- Chair: Russell Leftwich
- Scribe: Emma Jones
Rob hausam started valueset review of queried SNOMED codes.
- Ran exact string match against any valid description in SNOMED including synonyms. Gets the concept IDs and the exact term match
- Almost all are in occupation
- The two spots for the person hierarchy and not the occupation will be filled by the non-paid (caregiver and patient)
- The semantics seem to deal with the entity and the role of the entity
- This really need to be a role of the person doing it - not a paid occupation.
- SNOMED does not have any hierarchy for non-paid person which is role. It's always good to think about this
- How do we as a group need to look at the occupation hierarchy and find the really good terms - which do we need or not need? What would we do with what we identify?
- What about qualifier value - per Rob, used to provide structure.
- If we find qualifier matches lets see how we can use it
- When the specific can't be used, use the qualifier
- May want to use one or the other
- Attending physician may be a qualifier depending on the institution.
- CDA function code does not have multiples. Could use extensions for multiple. Lisa working with SDWG as an initial step for this. Will bring in others as needed to get an extension added for multiple
- Example - Function is "specialist" then 'Neonatologist' as the role code
- For now try to get as many of these listed as possible and then try to tweak out the problem ones.
- Emma will take the top and Lisa will take the bottom on the list and search for snomed codes - add neonatologist to the list
- CDA - concept as a starter set will make this value set dynamic - so when we have to, will have the ability to update it
- FHIR - valueset types: example; preferred, extensible, required. We should make it extensible
- With SNOMED, could have extensible so that it's all descendants of occupation. Not a problem with US realm but with when international may not be able to use SNOMED.
- US Core can extend the bindings
- Stephen was not awareness of this piece of work by Lisa and Rob - Should the end result spreadsheet be put together with the definition spreadsheet so we end up with one spreadsheet? Will need codes for care focused (care coordination, episodic, encounter, condition focus). Stephen will pull the LOINC codes for these.
- Need the right concepts lined up with the care team focused concepts.
- Legal and professional list - should point to them by OID in VSAC.
Next steps for LHS
- Update from eLTSS - developing a white paper so this work can be referenced. Will be developed over the next 6 mos
- Last week discussion - concept of a group of individuals fulfilling a care team role. What we call them will vary. VA calls it another care team.
- Things unique to the model.
- They need to have an electronic end point that represents the group rather than an individual and consent to the group. How the group fulfills their decisions, Need to be a lead.
- Public health on the care team - will have the Public Health role filled by that care team;
- Research team will follow the same pattern.
- Future work to represent use cases with these scenarios
- Mirror image of a group fulfilling as a back up - call group, planned rotation, other backup concepts, etc
- eLTSS data set does have a back-up plan or back up concepts
- Questions about handing off the consent or decision making to the back up team/person.
- Next week - Russ will be on an airplane. Will contact John Roberts to see if he will be available. Will sent notice out if need to cancel.