Stream: CARIN IG for Blue Button®
Topic: X12->CPCDS->FHIR mapping
Ryan Howells (Sep 14 2019 at 18:32):
For organizations who need to map from their source system -> CPCDS -> FHIR. . . . . Please see the attached initial draft of an 837 to CPCDS to FHIR mapping. We would welcome your input! Our collective input will be sent to X12 so they can publish an 'official' mapped version that we will point to in our final implementation guide. 837-BB_CARIN_Mapping-Version-1.xlsx
Varvara (Oct 18 2019 at 20:35):
Hi, I am also working on mapping 837 to FHIR, I am very interested why are you mapping 837 to EOB and not to Claim - do you have adjudicated 837 claims , is that the reason?
Amol Vyas (Oct 19 2019 at 20:22):
@Varvara : The CARIN Blue Button IG is about allowing consumers to access their adjudicated claims (aka EOBs) data at their Health Plans, through an app of their choice. To construct the complete EOB data model (present in both CPCDS data model and EOB FHIR profiles) from X12 transactions (instead of Facets or any other Health Plan claims adjudication system of record), I believe you'll also need the 277CA and 835 in addition to the 837.
Ryan Howells (Oct 29 2019 at 13:46):
For organizations who need to map from their source system -> CPCDS -> FHIR. . . . . Please see the attached initial draft of an 837 to CPCDS to FHIR mapping. We would welcome your input! Our collective input will be sent to X12 so they can publish an 'official' mapped version that we will point to in our final implementation guide. 837-BB_CARIN_Mapping-Version-1.xlsx
This should in no way be viewed as official in any way and really should not be used. We continue to work with X12 to finalize an official mapping between X12 and FHIR. We're meeting with them in early November to finalize an MOU. Please reach out to us to learn more.
Josh Mandel (Oct 29 2019 at 14:35):
Thanks @Ryan Howells ! A couple of things that would make this content much easier to review:
-
Stable location with version control so people know where to find the latest and can track changes over time.
-
Next to the mapping file, a few examples of inputs and outputs to see how the mapping applies to real data.
Josh Mandel (Oct 29 2019 at 14:36):
Without this, it's very hard for me (as a non-expert in X12, to put it mildly) to understand what's happening.
Grahame Grieve (Oct 29 2019 at 19:11):
@Ryan Howells who is 'we' in the sentence
We're meeting with them in early November to finalize an MOU
Ryan Howells (Oct 29 2019 at 19:35):
Ryan Howells who is 'we' in the sentence
We're meeting with them in early November to finalize an MOU
@Grahame Grieve Dave Lee and I from Leavitt are meeting with them to finalize an MOU on behalf of the CARIN Alliance. They also are looking to sign something with #DaVinci if they haven't already. @Jocelyn Keegan may know more.
@Josh Mandel That's the plan once we have something to review. Right now, there is no mapping from X12 to FHIR for Blue Button but we do have folks who started to work on it very recently. When there is something to review, we plan on posting the information for the HL7 community to review on our CARIN Blue Button page on Confluence (https://confluence.hl7.org/pages/viewpage.action?pageId=55941223).
Grahame Grieve (Oct 29 2019 at 19:37):
@Wayne Kubick FYI
All the talk about joint work with X12 I've heard of has them publishing the mappings in secret. (I mean, behind their IP access rules)
Ryan Howells (Oct 30 2019 at 14:13):
Wayne Kubick FYI
All the talk about joint work with X12 I've heard of has them publishing the mappings in secret. (I mean, behind their IP access rules)
One of the reasons we want to sign the MOU is so CARIN and the HL7 community can have a more significant voice regarding how they should construct the mappings. Since the X12 IP is still proprietary, they will still be the only ones who can officially publish the X12 -> FHIR mapping but our hope is we can influence that process in some way. We will ultimately point to the 'official' X12 mapping in our IG.
Keep in mind we're not mapping all of X12 to FHIR to start. We're going to begin with CARIN Blue Button and the 270/271 eligibility transactions and possibly move on to others. #DaVinci may have additional transactions they are looking at. Open to your thoughts @Grahame Grieve @Wayne Kubick and others . . .
Wayne Kubick (Oct 30 2019 at 18:46):
Hi Ryan, we all certainly would prefer if X12 mappings could be made available. There were a couple of attempts (from Chuck Jaffe and me) to reach out but we weren't able to proceed. I certainly support your effort and appreciate any influence you can exert.
Ryan Howells (Oct 30 2019 at 19:20):
Hi Ryan, we all certainly would prefer if X12 mappings could be made available. There were a couple of attempts (from Chuck Jaffe and me) to reach out but we weren't able to proceed. I certainly support your effort and appreciate any influence you can exert.
Sounds good @Wayne Kubick . We are meeting with them next week where we hope to make some additional progress and will keep you posted.
Shane O'Neill (Nov 13 2019 at 02:41):
@Wayne Kubick @Ryan Howells Hi everyone. I am also deeply curious on the relationship between CPCDS to FHIR mappings. And I just have a few questions.
- Is CPCDS -> FHIR purely semantic, as in it just specifies what FHIR R4 fields to consume (EOB, etc)?
- Or perhaps is it a FHIR Profile which adds fields and other features?
Thanks!
Josh Mandel (Nov 13 2019 at 03:04):
Can you clarify this question? I'm not sure I understand what the distinction is.
Shane O'Neill (Nov 13 2019 at 03:33):
Can you clarify this question? I'm not sure I understand what the distinction is.
Sometimes FHIR Profiles add attributes. For example, the FHIR Profile known as US Core adds a new field birthSex to Patient. Is there anything concerning CPCDS that does things like this to R4 resources? Does a CPCDS ExplainationOfBilling add any new fields, or define new extensions? Or is it just a semantic distinction.
I appreciate your response and clarification!
Josh Mandel (Nov 13 2019 at 15:09):
Even if it does not add extensions, a profile like this defines specific pieces of supporting information that must be present -- and the modeling of supporting Info is very similar to the modeling of extensions. So I don't have a sharp distinction like this in my mind.
Josh Mandel (Nov 13 2019 at 15:10):
In other words http://build.fhir.org/explanationofbenefit-definitions.html#ExplanationOfBenefit.supportingInfo acts a lot like an extension
Josh Mandel (Nov 13 2019 at 15:12):
And profile constraints like https://build.fhir.org/ig/HL7/carin-bb/StructureDefinition-CARIN-BB-ExplanationOfBenefit-Inpatient-Facility-definitions.html#ExplanationOfBenefit.supportingInfo:billingnetworkcontractingstatus.id act a lot like extension definitions.
Josh Mandel (Nov 13 2019 at 15:16):
But I'm still struggling with the fundamentals of your question -- CARIN is essentially creating a guide that explains how to convey a specific set of financial data in FHIR; a guide like this necessarily constrains FHIR resource definitions by locking down vocabularies and so forth. I don't know what you mean by "purely semantic" -- but if you mean something like "just another set of labels", then no, I don't think this captures what's going on. I guess the struggle for me is that "CPCDS" is not a pre-existing or independently defined set of content; so CARIN is not so much providing "a mapping of CPCDS into FHIR" as "defining FHIR profiles for a consumer financial use case."
Lloyd McKenzie (Nov 13 2019 at 16:11):
As much as possible, an implementer of the FHIR spec should require no knowledge of the other specs being mapped to and from.
Ryan Howells (Nov 13 2019 at 20:27):
But I'm still struggling with the fundamentals of your question -- CARIN is essentially creating a guide that explains how to convey a specific set of financial data in FHIR; a guide like this necessarily constrains FHIR resource definitions by locking down vocabularies and so forth. I don't know what you mean by "purely semantic" -- but if you mean something like "just another set of labels", then no, I don't think this captures what's going on. I guess the struggle for me is that "CPCDS" is not a pre-existing or independently defined set of content; so CARIN is not so much providing "a mapping of CPCDS into FHIR" as "defining FHIR profiles for a consumer financial use case."
I agree with @Josh Mandel . That is our intent.
Christopher Marchand (Apr 25 2020 at 17:15):
@Ryan Howells Do you happen to have an updated mapping document that you would share? Would love if we could leverage this more broadly as well and put on the CPCDS IG itself!
Ryan Howells (Apr 26 2020 at 14:03):
Well, @Christopher Marchand I would love to say yes but the answer is no. X12 is a proprietary data set and we are respecting that by not publishing any guides online or in the IGs. We have reached out to X12 multiple times over the last few months to determine if they would be willing to work with us to publish a joint mapping but it hasn't yielded any meaningful progress despite their strong verbal desire to work with us. Your point is well taken. There is a strong interest in publishing something especially on the state Medicaid side. @Wayne Kubick : Any ideas here? Do others have thoughts?
Wayne Kubick (Apr 27 2020 at 15:13):
I haven't heard any more from X12 either, and they are likely backed up since they had a recent ransomware attack. I'll ask around.
Last updated: Apr 12 2022 at 19:14 UTC
