FHIR Chat · Responsible Consultant in NHS Encouner · uk

Stream: uk

Topic: Responsible Consultant in NHS Encouner


view this post on Zulip Alan Pinder (Feb 10 2022 at 13:18):

Hi folks, can we get some advice on where to put the two key clinicians the vendor hospital system supports for Encounter? We have two HCPs - the responsible Consultant (who may or may not even be in the hospital at the time), and the Attending Consultant (one delivering day-to-day care on the ward). What type of Participants should these be - are these ADM and ATND respectively?

view this post on Zulip Richard Kavanagh (Feb 10 2022 at 16:41):

And this is what UK core should have focussed on - not just renaming ressources :disappointed:

view this post on Zulip Kevin Mayfield (Feb 10 2022 at 17:05):

The best I can offer at the moment would be to follow https://build.fhir.org/ig/HL7/v2-to-fhir/ConceptMap-segment-pv1-to-encounter.html

and yes that would align with ADM and ATND.

view this post on Zulip Kevin Mayfield (Feb 10 2022 at 17:08):

The prerelease seems to match that https://simplifier.net/guide/UKCoreImplementationGuide0.3.0-STU1/Home/ProfilesandExtensions/ProfileUKCore-Encounter#participant-030-duplicate-2

view this post on Zulip Rik Smithies (Feb 10 2022 at 18:22):

None of the codes seem ideal.
ADM is: The practitioner who is responsible for admitting a patient to a patient encounter.
The concept we appear to want is: The practitioner who has overall responsibility for the patient encounter (and probably every encounter on that shift).
It doesn't even seem to be patient specific.
ADM could be thought of as the person directly responsible for admitting, and not the person who carries the can. ADM could be the receptionist or triage person.
I don't think it can be said that UK Core aligns with ADM and ATND. It just re-iterates the list and is silent on which codes to use for these concepts.
I also don't think that V2 map helps us. It just equates the V2 concept of admitter to the FHIR concept to admitter. I don't think this concept is really "admitter" at all.
I think we should seek clarity on this from PA workgroup.

view this post on Zulip Vassil Peytchev (Feb 10 2022 at 21:07):

This might be a case of changing healthcare practices in the US, where "admitting provider," or a provider with "admitting privileges" seems to have been close to "responsible consultant". From here:

[...] the concept of admitting privileges is that your doctor would be able to admit you with some paperwork and a quick phone call.

They would then be able to come to the hospital — often once in the morning and once in at night on “rounds” — to coordinate your care. They would order tests, prescribe medications, and schedule procedures.

At least that’s the concept.

view this post on Zulip Alan Pinder (Feb 11 2022 at 12:10):

so ... is it good to advise ADM = Responsible HCP for a UK system for now? Or should I wait from clarity from the PA workgroup?

Rik Smithies said:

None of the codes seem ideal.
ADM is: The practitioner who is responsible for admitting a patient to a patient encounter.
The concept we appear to want is: The practitioner who has overall responsibility for the patient encounter (and probably every encounter on that shift).
It doesn't even seem to be patient specific.
ADM could be thought of as the person directly responsible for admitting, and not the person who carries the can. ADM could be the receptionist or triage person.
I don't think it can be said that UK Core aligns with ADM and ATND. It just re-iterates the list and is silent on which codes to use for these concepts.
I also don't think that V2 map helps us. It just equates the V2 concept of admitter to the FHIR concept to admitter. I don't think this concept is really "admitter" at all.
I think we should seek clarity on this from PA workgroup.

view this post on Zulip Rik Smithies (Feb 11 2022 at 12:39):

Does it seem correct to you Alan? It doesn't seem correct to me, from the definition.

It is up to us to see if the standard concept fits our use case (first yours, which you are the expert for, and then UK wide, but it is probably the same thing) and actively seek guidance if not. Most PA people won't see this on the UK stream, so I will cross post something.

view this post on Zulip Rik Smithies (Feb 11 2022 at 13:07):

see Implementers https://chat.fhir.org/#narrow/stream/179166-implementers/topic/Supervising.20clinician.20for.20.20an.20encounter

view this post on Zulip Kevin Mayfield (Feb 11 2022 at 13:40):

I think we are mixing up use cases here.

I am aware of two.

Health admin and clinical

CareConnect focused on the clinical/primary care only and I know caused confusion in acute settings who are predominantly mapping from HL7v2.

I've asked about how these should be handled in the past. My thoughts is to keep the two use cases separate (maybe two profiles?), health admin encounter/episodes are top level constructs and hold details like admitting doctor.

The clinical encounters don't hold this data and are either partOf the health admin encounter or encounters in an episodeOfCare.
Going from the NHS (England) v2 definition, ADM would be

Screenshot-2022-02-11-at-13.39.57.png

view this post on Zulip Alan Pinder (Feb 11 2022 at 14:56):

Thanks @Kevin Mayfield , @Rik Smithies . I am just trying to accurately process what the existing UK-based 'PAS' system is doing - which is sharing encounter data designed to fit the NHS / HL7v2 description of an Admitting Doctor (Responsible HCP). So in that (Health Admin / HL7v2) context, ADM in the FHIR encounter doesn't seem too wrong to me (though in a plain-english context, its plain wrong!). I guess my use case is quite narrow at the moment - make Hospital IP / OP / ED data that would normally be HL7v2 appear appropriately in FHIR resources. I'm watching now for activity on the PA group version of this message!

view this post on Zulip Kevin Mayfield (Feb 11 2022 at 15:09):

I've been going for something like this.

No resource references just using GMC/GMP codes.

        {
            "type": [
                {
                    "coding": [
                        {
                            "system": "http://terminology.hl7.org/CodeSystem/v3-ParticipationType",
                            "code": "ADM",
                            "display": "admitter"
                        }
                    ]
                }
            ],
            "individual": {
                "identifier": [
                    {
                        "system": "https://fhir.nhs.uk/Id/gmc-number",
                        "value": "C12345678"
                    }
                ],
                "display": "Dr Sandra Goose"
            }
        }
    ]

view this post on Zulip Rik Smithies (Feb 11 2022 at 15:48):

If the question is how to put the ADM code (from "legacy" V2 data) into FHIR, then I expect the answer is to continue to use ADM.

If the question is should this have been ADM in the first place, or should it be that going forwards, then I think that is different.

If the definition is "Admitting Doctor" then it may go one way. If the definition is "the responsible Consultant (who may or may not even be in the hospital at the time)" then that seems a different case to me. I would not call that second person the Admitting Doctor, but I don't know all the terminology.

I am not sure what the difference between the admin and the clinical case is.

view this post on Zulip Kevin Mayfield (Feb 11 2022 at 16:20):

One is around admission and discharges in v2.4. You will start a stay with a ADT_A01 and finish with an ADT_A03. This is often the case for inpatient, outpatient and emergency. We have a lot of data NHS collection around these events.

During that stay you will have many patient care focused encounters.

So for the former you would probably populate Encounter.hospitalisation, etc. For ward round encounter I don't believe you would - just refer to the admission encounter.

view this post on Zulip Kevin Mayfield (Feb 11 2022 at 16:21):

I don't think we can class this as legacy data. A lot of it is required and part of the NHS (England) Data Dictionary

view this post on Zulip Kevin Mayfield (Feb 11 2022 at 16:22):

We do need feedback from PAS/regional/ICS suppliers on how they plan to handle this?


Last updated: Apr 12 2022 at 19:14 UTC