Stream: implementers
Topic: Patient matching
David Hay (Mar 23 2016 at 06:13):
I was asked by a colleague if there was any output from this HIMSS Patching Matching event...
Grahame Grieve (Mar 23 2016 at 06:14):
http://www.slideshare.net/GrahameGrieve/patient-matching-in-fhir is the only online thing that I know
David Hay (Mar 23 2016 at 06:22):
ta - needed this link for some reason http://www.slideshare.net/GrahameGrieve/patient-matching-in-fhir?qid=bc7bedb4-e33c-474d-b0fd-81919e57e820&v=&b=&from_search=1
Jeremy Morris (Mar 23 2016 at 07:43):
Hi @Grahame Grieve - I am the said colleague, and had found your slides as part of my research, thanks for that :)
I guess what I was looking for was a "what did we learn....?" out of the HIMSS innovation experiment with regard to how FHIR could improve patient matching. I'd really appreciate if you could elaborate on this if possible?
Related question: Is the goal of the https://www.hl7.org/fhir/valueset-patient-mpi-match.html extension to give us the option to record how an MPI/EMPI assessed patient demographics at a point in time? What type of value does this introduce?
Thanks in advance :)
J
Grahame Grieve (Mar 23 2016 at 07:51):
see either http://hl7-fhir.github.io/patient.html#match, or http://hl7-fhir.github.io/patient-operations.html#5.1.16.1 for how this stuff works in practice
Grahame Grieve (Mar 23 2016 at 07:51):
it's about an MPI match interchange. @Jason Walonoski might want to ask his Mitre colleagues to talk about other forms of MPI matching - I'm not on top of that
Grahame Grieve (Mar 23 2016 at 07:55):
as for wider outcomes: I would say they were two-fold: if you're not going to have a national patient identifier, you're going to need really good matching infrastructure. And the other: no one is sufficiently interested, even with the CHIME challenge
Grahame Grieve (Mar 23 2016 at 07:56):
in fact the chime challenge might be setting things back by creating a competitive urge instead of a collaborative one
Andy Gregorowicz (Mar 23 2016 at 13:14):
MITRE is doing work for ONC on patient matching, but the perspective is flipped a bit. Our focus is on how patient matching/MPI systems would interact with clinical data stores using FHIR to determine patient matches. We are building a test harness, so that multiple patient matching systems can be benchmarked against the same data set, residing in a FHIR server. The interface for this test harness is available now: http://mitre.github.io/test-harness-interface/. We're working on the software implementation, which should be ready for testing soon.
Peter Bernhardt (Mar 23 2016 at 16:26):
One could argue that CommonWell has already solved this in the USA. Then again, I'm biased.
Adam Culbertson (Mar 23 2016 at 19:06):
Well in terms of output we confirmed that patient matching is a really hard problem. We did a follow-up event pre-HIMSS 16. Work is focused on development of test data sets and also trying to integrate matching solutions within the FHIR specifications. Works is still in progress.
Adam Culbertson (Mar 23 2016 at 19:19):
@Peter Bernhardt would love to learn more about how CW has solved this? Message me would enjoy discussing further.
Grahame Grieve (Mar 23 2016 at 19:24):
Ewan is from the CW team that worked on this, right? he was in Cleveland...
Peter Bernhardt (Mar 23 2016 at 19:53):
@Grahame Grieve you're probably thinking of Evan Schnell, a colleague of mine at RelayHealth. He was in Cleveland but not involved in CW.
Peter Bernhardt (Mar 23 2016 at 19:56):
We have a non-CW use case involving matching patients for our lab order services.
Grahame Grieve (Mar 23 2016 at 20:00):
I was thinking of Evan
Last updated: Apr 12 2022 at 19:14 UTC
