Implementation Group Call for Consensus: Abstract Model Version 1.1

[Brett Peterson]

An Implementation Group call for consensus on the current state of the
NHIN Direct Abstract Model has been posted here:

http://nhindirect.org/message/view/NHIN+Direct+Abstract+Model/22692935

This will be discussed on the 4/13/2010 Implementation Group
conference call and results posted in the meeting minutes on the
Implementation Group wiki page:

http://nhindirect.org/Implementation+Group+%26+Our+Workgroups

As always, all community members are encouraged to voice their
opinions through relevant wiki discussion pages and on this forum.

Brett

[David Tao]

Coming in a bit late on this model, but I have a question about the
"awareness" of the HSP that is expected re the source and destination.
Originally, thinking about "point to point" and "push" I thought that
the idea was that a source would act as if they were sending directly
to the destination, just like I would write a letter to my mother. I
don't think of "sending a letter to the US Postal Service, who in turn
delivers it to my mother" although of course the USPS is in a role
similar to the "HSP" in the abstract model. Similarly, sending e-mails
I pick out someone's e-mail address from my address book and send it
to them, without thinking of the ISPs and who knows what else lies
between.

I notice that the User Stories that I've read don't mention the HSP,
so I assume that the goal is for the users (of the source and
destination) to be unaware of the HSP. Since the Abstract Model also
says that the source and HSP may be combined in a single EHR, thus the
HSP is not necessarily a "thing" (like an Health Info Exchange
organization) but just an "actor" similar to the IHE actors (document
consumer, content creator, etc.).

I don't feel well-enough established yet to directly edit the NHIN
Direct Wiki, but if you agree with what I just said, shouldn't the
definition of HSP be clarified? I suggest the following (my changes in
ALL CAPS).

HSP

* Health Service Provider. An actor that serves the onboarding and
backbone exchange needs of Source and Destination actors. An HSP will
typically be an HIO, but may be an IDN, federal provider, CAPABILITY
WITHIN AN EHR OR EHR MODULE, etc. In the current state of this
Abstract Model, an HSP should be thought of in the context of message
delivery/receipt and not in the context of governance
responsibilities.

Also, since the Internet itself is the essential "transport backbone"
and we're not reinventing that, I conclude that it's not really part
of the HSP, but that the HSP is assumed to sit on top of the
Internet.

Am I correct?

Finally, doesn't the term "onboarding" need to be defined?

Thanks,

David Tao

On Apr 12, 2:22 pm, Brett Peterson <Brett.Peter...@visionshareinc.com>
wrote:

[Brett Peterson]

Thanks for the comments.

Regarding the (now renamed) HISP definition: The area that you
highlight has been removed from the model in version 2.0 with the
intent that such context be specified in a non-normative form in the
Abstract Model Examples wiki page (still under construction).
Hopefully that page will help clarify without having to insert
assumptions about the type of software being used. The under-
construction examples page is here:

http://nhindirect.org/Abstract+Model+Examples

And yes... I see the Internet itself as an assumed transport backbone
on which transactions move.

Good point about the vague term "onboarding". I've added an item to
the 4/21/2010 Abstract Model workgroup agenda to discuss defining this
term more clearly or removing it. Any opinions on this topic are
welcome here or in the discussion page for the abstract model. Note
that the 4/21/2010 Abstract Model workgroup agenda is here:

http://nhindirect.org/Abstract+Model+WG+Meeting+Notes+042110

Brett