[Openid-specs-heart] New thread - What is in a clipboard resource set?
Adrian Gropper
agropper at healthurl.com
Fri Jul 22 02:28:23 UTC 2016
Thanks Eve for sending around the various links and for bringing the WEDI
document
<http://www.wedi.org/docs/default-document-library/virtual-clipboard-definition-and-design-document.pdf?sfvrsn=0>
to our attention. It represents a good point of reference for how some of
the industry will consider adoption of HEART. I will use the WEDI
terminology for this comment.
The WEDI document does not mention the reality that there is zero adoption
of federated patient identity services in healthcare today. I think HEART
needs to consider how to ease the path to broad adoption of OpenID Connect
while also staying focused on our charter. I'm not sure how to interpret
what WEDI calls "External Security Services" (figure on p13 - b).
Presumably it's an OpenID Connect IDP? Since patient identity federations
don't exist yet and it's not clear who would govern such a federation, we
need to be careful about our IDP assumptions and federations in general.
The rest of the diagram on p13 is much easier to understand if we treat the
Application Services (purple) as just another instance of an External
Application Partner (yellow). All four blocks have an Inter-application
path (isn't this just FHIR?) and a Security Communication Path (isn't this
just OAuth and UMA?).
The Mobile App (green) is pretty clear. No matter what, we can assume that
we need a mobile secure element under control of a mobile UI. Per the WEDI
model, the Mobile App needs to be recognized by the (optional) IDP and what
WEDI calls the Application Security Services (orange).
The Application Security Services (orange) is the only one containing an
Authorization Service so it presumably maps into HEART. Regardless of the
fact that WEDI put a gray rectangle around this and the Application
Services block, HEART does not benefit from adopting this combination.
Doing so is unnecessary and might even violate our charter.
We can conclude that WEDI would support an UMA AS as the manager of the
Security Communication Path that controls FHIR-standard resource access.
First and foremost, IMHO, HEART needs to consider how we can drive adoption
of a patient-centered authorization service without forcing all of the
External Application Partners, including the Application Services block, to
worry about also federating with an IDP. In other words, HEART must be
clear whether we are asking the FHIR endpoints (PHR, EHR, Payer,
whatever...) to trust the AS regardless of whether an IDP is trusted as
well.
This post about WEDI might be a separate thread from whatever we mean by
virtual clipboard but I put it here first because WEDI's is the only
definition of a virtual clipboard that we've seen.
Adrian
On Thu, Jul 21, 2016 at 5:15 PM, Nancy Lush <nlush at lgisoftware.com> wrote:
> The Argonaut Group has implemented (or will soon complete) the following
> resources in FHIR:
>
> · Patient (demographics)
>
> · Allergies
>
> · Problems & Health concerns (Condition/Problem)
>
> · Vital Signs
>
> · Labs
>
> · Smoking Status
>
> · Care Team (Some vendors have Care Plan of which Care Team is a
> subset.)
>
> · Medications
>
> · Immunizations
>
> · Goals
>
> · UDI (Device)
>
> · Procedures
>
> · Plan of Treatment
>
> · Assessment
>
> These pretty much match the Common Clinical Data Set. Why don’t we start
> with that list?
>
> Yes, Eve, it does include labs.
>
>
>
> The virtual clipboard reference is a bit askew since the virtual clipboard
> initial pilot seems much smaller in scope. Perhaps it is just me, but I am
> confused by that reference added in.
>
>
>
> -Nancy
>
>
>
> *From:* Openid-specs-heart [mailto:
> openid-specs-heart-bounces at lists.openid.net] *On Behalf Of *Debbie Bucci
> *Sent:* Thursday, July 21, 2016 12:07 PM
> *To:* Eve Maler <eve.maler at forgerock.com>
> *Cc:* openid-specs-heart at lists.openid.net
> *Subject:* Re: [Openid-specs-heart] New thread - What is in a clipboard
> resource set?
>
>
>
>
>
>
>
> On Thu, Jul 21, 2016 at 11:39 AM, Eve Maler <eve.maler at forgerock.com>
> wrote:
>
> Just to level-set, there are three links in the current use case to
> sources that describe potentially interesting data (though not directly how
> they map to FHIR, as far as I can see):
>
> - Use case
> <https://bitbucket.org/openid/heart/wiki/Alice_Shares_with_Physicians_and_Others_UMA_FHIR>
>
>
> - PGHD
> <https://www.healthit.gov/policy-researchers-implementers/patient-generated-health-data>
> - Common Clinical Data Set
> <https://www.healthit.gov/sites/default/files/commonclinicaldataset_ml_11-4-15.pdf>
> - Virtual clipboard
> <https://www.wedi.org/home/virtual-clipboard-to-improve-patient-data-collection>
> (this link
> <http://www.wedi.org/docs/default-document-library/virtual-clipboard-definition-and-design-document.pdf?sfvrsn=0>
> is actually more helpful -- I've changed it in the use case)
>
> And based on one of our telecon conversations, we provided this list right
> in the use case, where the first item might possibly encompass some of the
> others by definition:
>
> 1. Virtual clipboard
> 2. Basic patient profile /Patient)
> 3. Medication history (/MedicationDispense, /MedicationAdministration)
> 4. Immunizations (/Immunization.vaccineCode , /Immunization.status)
> 5. Allergies (/AllergyIntolerance)
> 6. Problem list /Condition.code, /Condition.status)
> 7. Labs (/DiagnosticReport.code - LOINC
> /DiagnosticReport.codedDiagnostics - SNOMED)
> 8. Basic insurance info? EOB?
> 9. (what else? Keep it simple)
>
> Here was my guess about directions, given that a use-case-based definition
> of this resource set is meant to encompass baseline information before a
> first visit.
>
>
>
> Agree - basic info to start the conversation - vs. a clinical referral for
> specific illness - that may be a follow-on request.
>
>
>
>
>
> First, we want to keep firmly in our minds that Alice's motivation to
> share this information at all is as a convenience vs. sharing it in a more
> onerous, brittle fashion (e.g., having to keep updating her meds list
> later, spending time printing forms, arriving early to the first
> appointment...). So we don't "win" if we don't hit that mark.
>
>
>
> If we focus on convenience first - then other benefits may follow.
>
>
>
> Second, labs and EOBs look like the "odd items out" on that list for a
> first visit. Am I right? And what's missing? And what's the "FHIR
> concordance" for all this? Can one actually be done?
>
>
>
> Agree no labs and EOB but in US basic insurance (payer as
> noted below) info will be needed. Is emergency contact part of
> demographics (need to look). How do capture both OTC meds, herbs ect?
> (some of that may be covered as well) Some information ex: - why are you
> making the appointment and other things of that nature would be out of
> scope - or perhaps an extension?
>
>
>
>
>
>
>
> Third, there would be some kind of a "matrix of requesting-side
> appropriateness" for what is to be shared:
>
> - *Payment info:* E.g., there's a particular type of insurance
> infrastructure in the US, but in Canada, it would look different. So the
> "insurance info" reference in #8 would need to vary or be missing per
> jurisdiction.
>
>
> -
> - *Provider services:* There *may* be a fundamental mismatch between
> the nature of health information available and what the provider can even
> do. When it comes to, say, chiropractic, is there an appropriateness or
> privacy concern about providing meds/allergy information? It's a complex
> technical question to imagine leaving this info out per provider type (akin
> to the segmentation challenges discussed on the recent thread titled
> "Patient Consent", so perhaps we decide to put it out of scope in favor of
> time constraints on usage, notifications of what's missing, usage
> constraints, etc.), but I'm listing this for discussion completeness.
>
> Finally, I'm assuming that for interoperability and attractiveness of
> standardization, we'd want to say that (barring the above concerns if
> solved) if a data item is *available* in the record, and Alice chooses to
> share it as part of the "virtual clipboard" resource set context, then it
> *must* be included. In technical terms, this would equate to the resource
> server supporting an API call that extracts out a bunch of potentially
> disparate data fields and provides them, because it supports the use case
> (i.e., provides a lot of value). We could separately think about whether
> extensibility is okay or not in the context of HEART.
>
>
>
> Agree
>
> _______________________________________________
> Openid-specs-heart mailing list
> Openid-specs-heart at lists.openid.net
> http://lists.openid.net/mailman/listinfo/openid-specs-heart
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-specs-heart/attachments/20160721/db269165/attachment.html>
More information about the Openid-specs-heart
mailing list