394814009 |General practice (specialty) (qualifier value)| is a child of 394658006 |Clinical specialty (qualifier value)|, while 394802001 |General medicine (qualifier value)| is a child of the (higher) 394733009 |Medical specialty (qualifier value)|.
Is General medicine the term of Internal medicine? i.e. the specialty for Hospitalists? If that’s the case, what’s the difference between these two concepts and 419192003 |Internal medicine (qualifier value)|?
Vs General Practice for a General Practitioner (general outpatient care?
Dear Dr. Taleya Girvan, thank you for your enquiry. These are good questions. Much of the content in SNOMED CT is legacy content and very often there is overlap in terms with same meaning but different names in different countries, that in common parlance are used interchangeably.
I agree that these concepts need review as possibly redundant. However, in the first instance, we would have to agree on definitions for Medical specialty as it is distinguished from Clinical specialty.
Initial clarification must be sought from the Community of Practice. I expect someone will comment hereon.
If a valid change is necessary, a formal request must be routed through the National Release Centre (NRC) of the submitting country where you reside. Stay posted and let’s see if anyone else has an opinion and post it here. best wishes, Monica Harry
The clinical specialty hierarchy below <<394733009 Medical specialty (qualifier value) has become something of a mess!
Up until 2006 there were only 135 such codes, of which 131 had been absorbed into SNOMED from the UK Clinical Terms Version 3. This included all of:
394814009 |General practice (specialty) (qualifier value)|
= a physician working in family medicine / primary care / general practice, especially if in the UK model where such care is provided out of physical clinic facilities that are located in the community, with minimal or zero on-site diagnostic or imaging facilities, and that frequently also involve consults delivered in the patients own home.
394802001 General medicine (qualifier value)
= a hospital physician who is not primarily a specialist (e.g. a neurologist, endocrinologist, oncologist, cardiologist etc) but a physician generalist who admits into hospital care and then provides treatment to patients with conditions spanning all of the above (and more) - especially when as urgent cases - and who may or may not then refer to such specialists only those more complex cases that require it
394609007 General surgery (qualifier value)
= a hospital surgeon who is not primarily a specialist (e.g. neurosurgery, colorectal surgery, orthopaedics, plastic surgery) but a surgeon who admits into hospital care and then provides treatment to patients with conditions spanning all of the above (and more) - especially when as urgent cases - and who may or may not then refer on to such specialists those cases requiring specialist care, surgical skills or facilities.
Only 4 “clinical specialty” codes that were already present back in 2005 were “non UK” in their origins.
However, over the following 20 years, those original 135 have now been joined by 225 newer codes, added (I think) mostly in response to requests from other non-UK jurisdictions and mostly during five distinct periods of time:
33 in 2006
53 in 2012
34 in 2019/20
47 in 2022
22 in 2024
The end superimposed result now contains many unresolved “collisions of clinical culture” between the original UK-specific terminology for UK specific ways of organising and delivering care, and newer but equally parochial perspectives from other jurisdictions (but chiefly I suspect from the US).
One of the other more obvious limitations of the existing taxonomy is that it does not currently support simple taxonomically-driven recovery of all hospital-based specialties that would be considered “surgical” or all those considered “non-surgical” (and which in the UK are therefore commonly termed the “medical specialties”). It is however possible that there isn’t international agreement on which specialties are “surgical”!
The SNOMED Editorial advisory Group has struggled on the definition of surgical vs non-surgical procedures and the medical occupations that might perform them and have not come up with a suitable set of criteria. The discussion has gottem to the point of considering that this distinction between medical and surgical is so fuzzy as to be not very useful.
The does not seem to be a uniform set of medical or surgical specialties that is internationally accepted and most, if not all authoritative lists are national in scope. Much like the issues with the Occupation hierarchy in SNOMED, what consitutes a medical specialty is jurisdictionally dependent. The Member Forum had identified a few hierarchies in SNOMED that do not seem to meet international needs and are being considered for relegation back to national extensions, much like the recent inactivation of Race and Ethnicity content from the international release (on the recommendation of a Member Forum study group).
Early additions to the Medical specialty hierarchy were most likely done to support members that did not have an extension, but we see what that has resulted in. The notion of GP as a specialty is not recognized in some countries, such as the US and Canada where they are called Family practitioners or Primary care providers.
A review of the usefulness of this hierarchy by the Member Forum might be in order.
As far as use-cases go, in our system we use specialty concepts to encode the following EHR data
specialty associated with provider
specialty associated with a department
specialty associated with an encounter
Other things to consider:
Specialty is a key information element stored in almost all EHR systems (as an attribute of a provider, encounter, care site, and healthcare service)
Patients and providers use software build around specialty to help them search for providers, clinics, and appointments
The LOINC DO contains specialty (SMD) and the specialty concepts will likely be needed in any record artifact concept model used in the LOINC Extension
OMOP uses specialty concepts in the provider table and it’s likely many health systems have mapped their data locally to SNOMED-CT concepts, so inactivation would mean
Substantial rework for OMOP implementers
Gaps in available concepts for OMOP implementers (which in turn will require us to create custom concepts in OMOP - another case of OHDSI being forced to become a standards developer)
I believe this content is within the stated scope of SNOMED-CT and I believe this content can (and should) be represented in an internationally acceptable way. We just need to stop and think about a specialty semantically (what it is) rather than in terms of how countries organize them in their local classification of specialties.
There are a finite set of anatomic systems, functional systems, and disorders and a finite set of high level approaches to preventing, detecting, measuring, and treating human disease. Specialties are really just different ways of mixing and packaging knowledge of these discrete classes of things, right?
Aren’t we terminologists? Isn’t this the kind of problem we just LOVE solving?!
Also….
There are some important missing high level concepts in the hierarchy right now that would need to be added if SI decided to work on the hierarchy:
Health specialty (should be highest level concept subsuming Medical specialty)
Specialty is also uses in Canada in our EHR landscape to define healthcare provider specialty in concert with the healthcare provider/practitioner role (223366009 |Healthcare professional (occupation)|) and healthcare services provided (224891009 |Healthcare services (qualifier value)|). We also for the most part leverage specialty concepts agnostic of categorization of whether it is surgical or non-surgical . We do often run into the quandary of what is a specialty /subspecialty versus more of a specialized “area of focus” or better defined as a “healthcare service” . The feedback from providers is usually that as their care provision is defined in the real world within the rubric of these practicing specialties, it should be represented in the terminology as such.
We saw a similar issue with “medical” v “clinical” pairs (e.g., “medical oncology” v “clinical oncology”).
The distinction (in the US at least) is apparently based on the class of treatments used (medical = chemotherapy and immunotherapy, clinical = chemotherapy, immunotherapy, and radiation therapy).
As you say, different countries may use ‘clinical’ and ‘medical’ to distinguish between two closely related specialties, but do it the opposite way!
So, probably a better idea to name concepts using FSNs that describe the specialty in terms of the values of the specific classes of things (disorders, anatomic or functional systems, interventions, age groups, etc.) that define the specialty.
Hello, Given this discussion thread, I would like to confirm if SI accepting requests for lateral promotion for clinical/medical specialty? Our stakeholders would like to request some content for promotion from the Korean extension from this hierarchy. Thanks!
Lateral promotion implies movement from one extension to another extension. Promotion to the international release is more of a vertical promotion as extensions are dependent upon it. We do not have any plans at this point to do much with the medical specialty hierarchy, unless there is a call from the members to do so.
We would consider promoting such concepts if they meet the standard requirements for submission through the CRS. However, given the unresolved issues regarding their meaning, use of these concepts should be regarded as caveat emptor.
@jcase . It is essentially vertical, correct. We would normally submit a request to the relevant NRC for promotion to core, they then submit a CRS request to SI for (if request is accepted). We will discuss and decide whether to proceed. Thanks
