The CDA Book is SHIPPING from Amazon! See here for Errata.

2013 Ask Me a Question Archive

      1. Hello Keith, for MU Stage 1, CDA entries with codes(level 3) was required for Allergies, Medication, Labs, Patient Problems.. What additional sections need to have entries(level 3) for MU stage 2 when generating c-CDA? It's not clear from the final rule. Also, other than the above 4 sections, what others need to be parsed for MU Stage 2. Thanks, -sri
        1. You'll still need to be able to "parse" problems, allergies, meds and labs to incorporate them into the EHR. There's other data that you might also want to use, but no "incorporate" clause that requires you to do much more than display it. With regard to what you need to put into your C-CDA, it's the common data set (defined in the rule) plus a few other things. For that, I suggest you have a look at the testing procedures.
      2. Hi Keith, Does the top validation tool at http://xreg2.nist.gov/cda-validation/validation.html --------------- Name: CDA R2 Description: HL7 CDA R2 (with no extensions) --------------- test the latest consolidated CDA (CCDA) corresponding to the Dec 2001 implementation guide? Thank you
        1. I think you mean the December 2011 guide (which was updated to 1.1 in 2012). You want this link:http://hit-testing.nist.gov:9100/ttt/
      3. I have a project coming up that requires an HL7 v3 specification to be written. I have a good grasp of v2 (certified) but have never worked with v3. Do you have any suggestions on where to start learning about v3? Or, how you would go about doing this? Just from the little research I've done so far I like v3. I have a history of programming in Java and appreciate the OO nature of v3. Thank you for any advice, Doug
        1. HL7 has a couple of books available. The understanding V3 guide is short, but is a good start. The CDA Book covers a lot of V3 in the first half.
      4. Hi Keith, I see, for example Results section of the CCDA, that you include the templateIds for both 'Coded Entries Required' and 'Coded Entries Optional'. Is it because the Required conforms to the Optional? So basically the Required option requires both templateIds?. Thank you in advance.
        1. I did it because it conforms to both, and declaring both means that systems that only deal with "Coded Entries Optional" could still deal with coded entries required. I don't believe "Coded Entries Optional" is required in "Coded Entries Required" in C-CDA.
      5. Hi, There is a way in CDA to reprsent the parent document of an addendum using the relatedDocument actrelationship (Source: Addendum to Target: parent). But how to represent all the addenda that belong to a CDA document ? Thanks
        1. In CDA, you would use the parentDocument element to represent the document that was addended by the addendum.
      6. Hello Keith, So this question is probably quite similar to the one that prompted you to give us Hashtag Soup. The question is related to some of the existing industry standards for defining a Clinical Data Model/ Repository to store clinical data. Some of the examples that I can think of are - CEDD, Query Health CEDD,Query Health CIM, mini Sentinel Common Data Model, Clinical Information Modeling Initiative, FHIM, QDM and others. It would be great if you can provide some insight on 1. Widely followed standards for a Clinical Data Model/ Repository 2. What are some of the advantages / disadvantages for each 3. When to use what OR typical use ? Much appreciated. Warm Regards, Shyam
        1. Your repository data model needs to enable views of the data that can support multiple clinical and quality data models. There's no "one model to rule them all" yet. CEDD, Query Health, and QDM have a lot of overlaps, as do CEDD and CIMI. When designing your repository, you have to design it for what you want to with the data you put in it, not for what format that data originated in. See The map is not the territory.
      7. Hello Keith - is there information available on how to represent preliminary vs. final laboratory results in a Consolidated CDA? If not, could you provide some guidance?
        1. Good question. The laboratory observation of Consolidated CDA was really intended to record "Completed" (e.g., final) observations, and usually has the statusCode set to "completed" in those cases. However, you could set statusCode to "active" to indicate that it hasn't been finalized. I'm not sure many systems would make that distinction though.
      8. Hi Keith, I'm new in this area so the question is: 1. what are the existing standards for encoding, structure and semantics of clinical documents for exchange? I just know the CDA (Clinical Document Architecture) from HL7 because there is much information. 2. If there are others witch is more used? Best regards, José
        1. CDA, obviously. OpenEHR/CEN13606 is another. HL7 Version 3 (of which CDA is an implementation of) includes several based on the HL7 RIM, for example the Care Record standard. The ASTM CCR is a third. Which is more used depends on where you are. CDA is used pretty widely internationally, US, Canada (just getting started), South America, Europe, Asia and Australia. OpenEHR/CEN13606 is used in Australia, parts of Europe and South America. CCR is pretty much US only, and not as widely used as CDA in the US.
      9. Hello Keith, The MU2 document states this "Section views. Extract and allow for individual display each additional section or sections (and the accompanying document header information) that were included in a transition of care/referral summary received and formatted in accordance with the standard adopted at § 170.205(a)(3)" So for additional sections( other than Allergies, Patient Problems, Meds which needs to be parsed and reconcilled ), what does "Extract" mean? Do you think its acceptable to store/show the entire C-CDA with easy navigation to additional sections and display narrative block( like a clickable table of content)?
        1. That would seem to work, but you might want to look at the NIST Test procedures for Meaningful Use Stage 2. They might provide greater clarity.
      10. Hi Keith, We are looking for CDA Consolidation data (you described in (http://motorcycleguy.blogspot.ru/2012/03/liberating-consolidated-cda-templates.html) Do you know why hl7 closed access to this data? Can you send us the copy?
        1. I believe there is a way to export, but haven't been able to get that to work, and haven't spent much time with it. I don't know why HL7 closed access to it, and no I cannot just send out a copy of it.
      11. Hi Keith, I'm wondering if you know the answer for us to interpreting the "90 continuous day reporting period" requirement for providers participating in MU for their first year. The CQMs refer to the measurement period as the calendar year and it seems many assume the reporting period would be a full calendar year as well. Example - Measure NQF 0018/CMS165 High Blood Pressure Denominator qualifying events must occur in the first 6 months and numerator events in the last six months. In this case, if the provider were reporting on a 90 day, not calendar year, a person could be in the denominator in the first 6 months and the numerator in the last 6 months, so reporting on a random 90 days, could have one in either a denominator or numerator but not both. We asked CMS for clarification, but so far, no word. If you have any insight, we would appreciate it. Thanks & Best Regards, Anne 
        1. Quite honestly, CMS is the best source to seek your answer from, as this is more about policy rather than technology.
      12. Hello Keith, Is there something on the C-CDA header that indicates if the document is a Transfer of Care or Clinical Summary? Not sure if the 8 different type for C-CDA ( other than CCD ) indicates if its TOC or Clinical Summary. Also is TOC at Chart level and Clinical Summary at a specific Encounter Level? Thanks, -sri
        1. No, there is nothing at the chart level that tells you functionally how the C-CDA is being used for Meaningful Use. From my own perspective, every Clinical Summary should contain the same content you would use for a Transfer of Care, that way, you have only one code path.
      13. Help me Obi Wan. You are my last hope. I have used your link to the ONC testing criteria, but I also have received a document from someone from NIST for EHR certification testing criteria. I have searched the NIST site (http://healthcare.nist.gov/use_testing/tools.html)extensively to try to determine if this is an "official" site that vendors should be using. I noticed a discrepancy between the race and ethnicity standards required for cancer registry testing through NIST. NIST says CDC standards are to be used while MU requires OMB standards for race and ethnicity. What are the official sites that should be used?
        1. Comparing OMB standards to CDC standards is like comparing a pebble to an X-Wing fighter when using the force. It is not the vocabulary terms that matter, but rather their arrangement. The CDC vocabulary is structured in the way that OMB requires. So, use the [CDC] terms Soroya. As far as who to trust, NIST is on the light side of the force. You can trust them.
      14. Hi Keith Question regarding report status in OBR-25 for Meditech Magic. I have a customer who wants to recieve "Electronically Signed" reports. To my knowledge Meditech posts F-Final, S-Signed, D-Draft and P-Pending. Do you know if there is a specific module in Meditech that displays the Electronically Signed status? Thanks!
        1. For this, you are going to need to talk to Meditech, as I'm neither familiar with their products, nor in a position to speak about them.
      15. The test scripts for Stage 2 2014 for the clinical information reconciliation (which focus on importing data from a CDA document) reference specifically that they want us to display the "last modification date" which they further define as "for medication list, last modification date should be interpreted as the last date the medication was document, ordered, prescribed, refilled, dispensed, or edited." We are really struggling with what value at the element level should be used for this in the CDA specification. The start time is not necessarily the last time it was "modified or verified".... do you have any suggestions here?
        1. Indeed. This would be captured in the participation time of the author associated with the medication entry.
        2. There is similar language in the clinical information reconciliation test script for problems and medication allergies as well. Are you suggesting that an element be utilized for each individual entry (i.e. each medication, each problems, and each medication allergy)and the corresponding time value of the element be used as the "last modified or verified" date?
        3. The above should read Are you suggesting that an author element be utilized. Apparently you can't use html tags.
        4. The "time of last update" of an entry can be determined by the time it was authored, so, you would use an author element for that, yes. You shouldn't need an author element for each entry, since the author element for the document represents the author of each entry unless you need to override that for a specific entry. I don't know exactly how NIST will be testing for this. The XPath from an entry to the time of participation of its first author is: ./ancestor-or-self::cda:author[1]/cda:time
        5. The tests have different "last modify dates" for each entry to be consolidated (each medication, allergy and/or problem) How can we handle this date at the record level in the CCDS instead of the author for the entire document. Thank you.
      16. Hello Keith, For MU stage 2, do we need to automatically match the patient on the C-CDA to a patient in the Org, OR is manual picking of patient suffice? Thanks, -sri
        1. See the rule, which states: (A) Correct patient. Demonstrate that the transition of care/referral summary received is or can be properly matched to the correct patient.
      17. Hi Keith, Can you provide any guidance if you are able to represent comments associated with a LAB result on the CCDA? I'm not able to see any guidance in the spec or companion guide. Thanks, Jon
        1. Certainly you can represent comments associated with a lab result. These would be represented using the Comment Activity [act: templateId 2.16.840.1.113883.10.20.22.4.64(open)] attached to the result observation using an entryRelationship with typeCode='SUBJ' and inversionInd='true'. The general pattern can be seen here, just be sure to use the new C-CDA template ID.
        2. Thank you very much for your response.
      18. Hi Keith, I had the following questions Question 1 -------------- We will be routing the emails sent to a patients DIRECT address to the Patient Portal Inbox. Alternately the 'Making the Clinical Summary' clause available to a patient on Patient Portal to view also exists. We are thinking if a patient anyway has access to his Summary on Patient Portal whats the point in routing the same clinical summary via DIRECT and making it sit in the Portal Inbox just to include it in the numerator calculation for 50% Clinical Summary? Are we missing something here ? Question 2 ----------- How would we handle an outgoing fax in context of Transitions of Care and Secure Messaging ? Thank you, Chinmay
        1. Under Meaningful Use, the certified EHR has to enable the patient to View, Download or transmit their summary to a third party. I don't think you can pick and choose which of those capabilities to implement, since they are all covered under the same criterion. Doing any of those three enables that patient to be counted in the numerator. With regard to an outgoing fax, I don't believe that counts under transmit because fax isn't one of the standards allowed for transmission under the certification criteria.
      19. Question 1: A certified EHR must have the capability to enable patients to View, Download, and Transmit their data. These are separate activities, each of which can count towards the numerator calculation (if a patient does any one of those actions, it counts in the numerator, but never counts more than once). Question 2: Transitions of care requires structured data as well. I don't think "FAX" counts.
        1. Thank you that helps. I had another question. Assuming that the EHR is certified on DIRECT, is it compulsory to always send the Clinical Summary out to the patients DIRECT email address ? If a patient wants to just not use his direct address and wants the summaries routed to his personal email address. Should that be allowed ? If so, will the transport protocol now switch from DIRECT protocol to just pure Secure Messaging ?
        2. The mechanism to transmit to other providers (for the transfer of care) is DIRECT, but you can also use DIRECT to transmit to the patient. My personal e-mail address IS a DIRECT address, because I have a secure certificate for it, which is all that you really need. I've even successfully sent myself a DIRECT e-mail using the Open Source tools and my little go-cart implementation. Under the HIPAA Ombibus rule, if the patient wants their data to be e-mailed, even though that may be unsecure, they have a right to have that done. That isn't discussed in Meaningful Use, but if you are implementing DIRECT, there's no reason why you couldn't implement e-mail as an alternative.
      20. Hello Keith, I have a question specifically for Encounter Diagnosis...Per MU Stage2 Encounter Diagnosis is ICD, but Encounter Diagnosis in C-CDA uses Problem Observation which is SNOMED. So if we capture Encounter Diagnosis as ICD( for billing purposes ), do you think it would be ok to put a translations in C-CDA for ICD, OR does the system have to capture both ICD and SNOMED? Thanks, -sri
        1. If you check my crosswalk, you'll see that you are allowed to use ICD-10-CM or SNOMED-CT. And yes, you could put in a translation for ICD-10-CM if you wanted to. It isn't prohibited.
      21. Keith: a couple of questions related to allergies. First - MU requirements and testing are centered on medication allergies and the test cases provide specific drugs by RxNorm# allowing one to feed Drug Interaction engines. However, many patients report ambiguous allergies like "mycins" or "cillins". Are you aware of any 'good work' being done by groups to be able to deal with these ill-defined or certainly inadequately specified allergies? A second and related question is whether you know of good consensus sources of lists of other allergens (foods, environmental, other) that need to be 'actionable' as they rise to the level of medical importance? Thanks for the tremendous work that you do.
        1. The C-CDA also allows classification of allergens using NDF-RT and UNII. While I know of nothing presently that specifies how one would communicate this source of information well, one could use an appropriate NDF-RT code for Penicillins to specify an allergy to the class of Penicillin drugs. The reality of this is that I don't believe that there are many systems out their using NDF-RT well though, and such communication might be lost. I don't know of a good, consensus developed sources for allergens either. SNOMED-CT has some coverage. ICD-9 is very limited (see the V-14 codes) in its coverage. It's been a long standing gap in standards.
      22. I need to transform a CCR to a CCD. Is a mapping between a CCR and CCD available? Ids there software that already does this?
        1. I know that several folks have developed CCR to CCD transforms, but know of nothing in the public domain.
      23. Hi Keith, Can you point me to good detail on MU2 requirement for Direct Message Exchange. We are a RHIO worrking with our hospitals and planning to implement a Direct / HISP solution to help them reach MU2
        1. See the MU requirements for a Certified EHR to Transmit Receive using Direct in the Meaningful Use regulation for the requirements. You could also look at the Direct specifications and Reference Implementation
      24. Hi Keith, We have started implementing SVS IHE Profile. We found opensource client side application (on OpenHealthTools). But we cant find any server side implementations. Have you seen any? Thank you! Dmitry.
        1. I built a simple implementation that was based on the Trifolia database. It's pretty straightforward. While you might not be able to get the Trifolia database anymore, you can figure out the necessary data schema from my implementation. I only handled the RESTful implementation, but the SOAP implementation isn't that much more difficult. The real challenge is getting at the data. Keith
      25. Hi Keith, Do we have a vocabulary requirement for Medication Allergy Reactions?
        1. According to C-CDA, the value element should be coded using SNOMED-CT.
        2. Hi Keith, I checked the 'Reaction Observation' element from the Implementation Guide. It says the following for 'Code' attribute "The value set for this code element has not been specified. Implementers are allowed to use any code system, such as SNOMED CT, a locally determined code, or a nullFlavor (CONF:16852)." Please let me know if this is what you are referring to. Thanks for all the help !
        3. For code, use ASSERTION from HL7 ActCode (2.16.840.1.113883.5.4). For value, use SNOMED-CT.
      26. Can you recommend a resource that helps identify what data should be put in an element. The CCDA standard is very good at letting a user know the datatype, number of occurences, whether a element is required, etc. However, it is not always clear what data should be in an element, especially date/time elements. One example is the effectiveTime element of the Smoking Observation template. What kind of date is it looking for? Is the the date the patient started smoking or the date the observation was recorded or something else? Based on one of the examples, I am inferring that it the observation date but it would be good to not infer but to know. Note that this is one example of many. Thanks for all of the helpful information on your blog.
        1. Most of the data elements in the CCDA came out of the HITSP C83 specification. Every data element in the C83 specification went back to a data element definition in C154, which has at least one sentence describing the data that goes in there. You might check out those two documents.
      27. Keith We are now working through the results section for the CCD and need an interpretation for lab results (like digoxin) that aren't really a battery or a cluster. These tests have only one observation but we cannot null because we have a code and id. What classcode should be used for these types of lab tests that have single observations? Is there a crosswalk that links the LOINC code with the corresponding classcode that should be used? Thanks!
        1. I think you are asking about the lab results organizer, which uses BATTERY and CLUSTER, but are dealing with a case where you have just one observation. In that case, just pick one of the two codes. It's the code inside the observation that is truly the important one.
      28. Keith, thank you so much for the helpful information! By the way I think you may have a typo on the replacement template id for the Problem Concern Act. According to my notes the replacement template is "2.16.840.1.113883.10.20.22.4.3". The id you have on your guide is below. Thanks!
      29. Hi Keith, Is there a CCDA specification document anywhere on teh internetz? If so, could you please point me to it? Thanks
        1. HL7 Members can access it here: http://www.hl7.org/documentcenter/public/standards/dstu/CDAR2_IG_IHE_CONSOL_DSTU_R1dot1_2012JUL.zip You can purchase a copy for $50 here: https://www.hl7.org/store/index.cfm?ref=nav
      30. Keith, I've got to get this off my chest. I just started reading the HL7 Implementation Guide: CDA Release 2 – Continuity of Care Document (CCD). I ran into this statement in the Introduction: "Constraints are expressed in a technology-neutral formalism. section 7.1.4 Sample CCD Validating Style Sheet provides a non-normative example of how one might implement the normative conformance statements, by expressing them as assertions within a Schematron schema." Totally discouraged, I stopped. I do understand the need for formality and preciseness; but this is a perfect example of why so many people shudder at the name HL7; and why it's so incredibly difficult to actually implement HL7-based interfaces. No two human beings can possibly read a statement like that and come away with the same understanding. And of course, that's just one statement in the entire document. And that document isn't even the CDA standard. Anyway, I have to convert a CCR to a CCD. I don't know where to start, except that starting with the "HL7 Implementation Guide: CDA Release 2 – Continuity of Care Document (CCD)" seems the wrong place. So I thought that maybe I should start by getting a good grasp of the fundamentals. That is where I hope your "CDA Book, 2011" will help. My hope is that I will find the book less impenetrable and more human-friendly. Perhaps by the end, I'll actually be prepared to read the "HL7 Implementation Guide: CDA Release 2 – Continuity of Care Document (CCD)". Sorry to vent. Hope your book helps!
        1. I hope my book helps too. You might look at some of the sample CCD documents that NIST provides with the Stage 1 validation tools. That should be a big help.
        2. Keith, The "Stage 1 Validation tools" link you gave is invalid. One other thing: What is NIST and what is it's relation to the HL7 CDA and CCD standards?
        3. That link just worked for me. You might want to try it again. NIST is one of the Federal agencies here in the US responsible for developing test procedures for Meaningful Use. As CCD and CCDA are specified standards under that regulation, they have developed tests for conformance. NIST does quite a bit of work in the area of conformance testing.
      31. Keith, Have you heard of any interest in transmitting communication preference via ADT?
        1. I've not heard of it, but given that providers want to capture communication preferences during check-in, it seems likely that they'd want to exchange those from a PM system to an EHR. What I would recommend would be to transmit communication preference in an OBX segment in the ADT message. You could use SNOMED-CT "Preferred Mode of Communication" code (428330008) in the OBX-3.
      32. Hi Keith, I had the following question : We are defining personal information in HL7 V3. Our particular question is with the ADDRESS information. To define the type of the address HL7 V3 define to use: ..... But what happen if we require to define new differents types from "H", "HP", "WP" or "HV" ... ? or a new set of clasification with our own categories, ? ... What whe should do ? ... 1. Include new categories in the current set or 2. Define a new list of categories using code ... How we can do that ? Thanks. Fernando Delete
        1. The simple answer is that you cannot. These code systems and value sets are fixed according to the HL7 Data Types specification. If you could explain what kind of extra information you want to impart, I may be able to help further.
      33. Keith or others, This has to do with the ClinicalDocument/documentationOf/serviceEvent element in a CCD. On page 14 of the "HL7 Implementation Guide: CDA Release 2 - Continuity of Care Document (CCD), April 01, 2007" CONF-3 states: "The value for “ClinicalDocument / documentationOf / serviceEvent / @classCode” SHALL be “PCPR” “Care provision” 2.16.840.1.113883.5.6 ActClass STATIC." I know how to assign "PCPR" to the classCode attribute of the serviceEvent element: . But what do I do about "Care provision" and 2.16.840.1.113883.5.6? NOTE that I found an example on page 75 of the "HL7 Implementation Guide for CDA R2: IHE Health Story Consolidation, DSTU R1.1". A portion is below. It looks like "Care provision" and 2.16.840.1.113883.5.6 could be included as attributes of the CODE element. BUT....I can't find anything that documents the CODE element as a child of the . It isn't even documented in the document I got the example from. Can you provide a list of the valid attributes of the element, and also the valid child elements of ? Can you also suggest how I should encode “Care provision” and 2.16.840.1.113883.5.6? Thanks! Tom Sadlowski
        1. You need not be concerned with where care provision or the ccde system OID go in this case, since classCode is a special attribute of classes in the RIM. The code system OID (2.16.840.1.113883.5.6) for classCode is fixed and need not be specified. Also, classCode display names (such as "Care Provision") are not transmitted for these special RIM attributes.
      34. Hi Keith, I observed a deviation in the the Minimum Data requirements mentioned by the ONC in the final edition test data that got released and the Transport Testing Tool validation. I was able to successfully validate the Clinical Summary without including sections such as Diagnostic Tests Pending, Future Scheduled Tests, Future Appointments, Recommended Patient Decision Aids which are required as minimum data as mentioned in the Final Edition Test Data for clinical summary . Is there anything I'm missing here ? -Chinmay
      35. Hi Keith I was able to follow your guide to change a C32 to a CCDA. I was able to run it through the Latana Validator that checks for Dec 2011 compliance and I do not have any errors. One thing you need to add on your guide is the template changes for Vital Signs unless I just did not see it. Also, you need to add the the classcode change for a manufactured product to include the required "MANU". Finally, do you know a site that has an official CCDA Validator for July 2012 CCDA Specs? Thanks!
        1. Try this site. It is the site hosting the tool used for Meaningful Use testing.
      36. Keith, In the 2nd paragraph of your July 30, 2012 article "Moving from HITSP C32 to CCD 1.1", you reference the "HL7 Consolidation Guide". I went to the HL7 website, searched it, and could not find anything titled "HL7 Consolidation Guide". Can you give the specific reference name that can be found on th HL7 website? COMMENT/REQUEST: I have found that the names of itemss to buy that are listed on the HL7 store web page really represent "containers" that have many files in them; and that typically, the names of the files are completely different from the "container" names. This just adds to the confusion of trying to find documents on the website "by name", since you can't even be sure of the name! Can you discuss this with the folks at HL7? I mean, how can a "Standards Committee" cause such "name" confusion?
        1. In the HL7 Store, it is called: HL7 Implementation Guide for CDA® Release 2: IHE Health Story Consolidation, Release 1.1 - US Realm
        2. Thank you, Keith.
      37. Keith, Is there an HL7 "documentation tree" that relates the HL7 documents to each other? It would be good to know a) if a particular document supercedes another, or b) if a partucular document was developed in response to or because of some other document.
        1. Some of the data you are looking for is in the product briefs accessible from the HL7 product matrix.
      38. Hi Keith, Posting again.... :) I observed a deviation in the the Minimum Data requirements mentioned by the ONC in the final edition test data that got released and the Transport Testing Tool validation. I was able to successfully validate the Clinical Summary without including sections such as Diagnostic Tests Pending, Future Scheduled Tests, Future Appointments, Recommended Patient Decision Aids which are required as minimum data as mentioned in the Final Edition Test Data for clinical summary . Is there anything I'm missing here ? -Chinmay
      39. The typical content of the telecom element is: [telecom value="tel:(816)276-6909"/]. And in a CCD, the telecom element is a required child element of patientRole. But what if you don't have the patient's phone number? How do you write/code the telecom element then? Thanks! 
        1. <telecom value='tel:' nullFlavor='UNK'/> is one way.
      40. Hi Keith - Reading your post today about inline examples, there are a few sections defined for the 2014 Criteria I could use for C-CDA. Here's my question... Difference between Discharge Instructions vs. Clinical Instructions vs. Patient Decision Aids? I've looked at HL7s Implementation Guide for C-CDA, S&I Framework Companion Guide, and various spreadsheets, which I'm uncertain at this point where I got them from, to better understand this topic. To the provider, it would seem Discharge Instructions and Clinical Instructions are one in the same, however, to ONC this is really dependant on the setting; Hospital vs Clinic. Reading the ONC test data, they are very similar - providing the patient with a set of instructions. From reading the guides above, it appears Discharge Instructions use Template ID 2.16.840.1.113883.10.20.22.2.41. However, for Clinical Instructions, it appears to be using Template ID 2.16.840.1.113883.10.20.22.2.45. To make this more confusing, you throw in Patient Decision Aids, which uses the same section Template ID as Clinical Instructions, but to differentiate between the two you can utilize the sub-section template ID 2.16.840.1.113883.10.20.22.4.20 and then define a SNOMED-CT code to define the instruction type - patient , immunization, or cancer instruction. Then I would assume there is SNOMED-CT codes for other "instruction" types. Am I even on the correct path? When I think I have it, I read something new and get more confused. Thanks Patrick (aka Dazed and Confused)
        1. I am also facing same question, will you please guide us
        2. I think you do have it.
      41. Hi Keith, Thanks for being so generous with sharing your wisdom on this blog and Q&A section. I am building a database to integrate medical data from multiple sources. My current approach is to store the data in both a NoSQL DB (as JSON) and a MySQL database, and I'd like to design the data structure following the C-CDA model. Since the JSON/RDBMS data structures don't map directly to XML (short of storing the XML as a blob), do you have any suggestions for resources/guidelines to follow for this design? Does this approach make sense? Thanks! Jeremy
        1. I'm afraid you aren't going to like my answer.
      42. We have a lot of "legacy" clinical documents from many institutions in our domain. As we integrate these into enterprise repositories and compile further into various CDA formats, they often don't comply with the current data standards. Once particular pain point is dates without time components. This is especially common for outpatient/ambulatory records which tended to capture DATE of service only. I have searched for an acceptable expression of "null" for the time components in the standard date/time formats. Not surprisingly, I haven't found one. When displaying such documents in our current applications, we have established a convention for handling them as if they had a timestamp of 0000. This at least provides predictable behavior for chronlogical views, but isn't accurate, since 0000 represents a specific time. Simply rejecting the documents as non-compliant is technially an option, of course - but certainly doesn't meet patient care or operational needs. Another internal suggestion we heard was to use a bogus value, like 7777. (I'm relatively sure that this would be rejected by any reasonable process.) Any suggestion for how to express or handle this "data reality" going forward with the new standards? Thanks for any input you may have.
        1. CDA uses the HL7 TS data type, which uses the ISO 8601 format for date and time (W3C Schema uses a variant of that same ISO standard): The format is: CCYYMMDDhhmmss.sss[+|-]ZZZZ Where CC is the century, YY is year within century, MM is month, DD is day, hh is hours, mm is minutes, and ss is seconds (and .sss is fractional seconds and can be to any precision you like), and [+|-]ZZZZ is the offset from GMT in hours and minutes. The timestamp for my response to you right now would be formatted like: 20130228180939.04-0500 You can record that in any precision you want in CDA, so if your date/time is only precise to the day, you can use CCYYMMDD (20130228). And if your precision is less than to the hour, don't include the time zone. In transitioning old documents to the CDA format for interchange, that level of precision is certainly sufficient. It's only when you transition to something like the Consolidated CDA (for Meaningful Use Stage 2) where more precision is recommended, but even there it is not required. The only time you would use null would be when there is no date/time information at all. In that case, you would represent that as in the following examples in CDA: <time nullFlavor='UNK'/> <effectiveTime nullFlavor='UNK'/> <birthTime nullFlavor='UNK'/> I hope that helps.
      43. Keith, Is there an "official" schema for the CCD; and if so, where can I find it? If there isn't a schema for the CCD, how can I create one from the CDA, since the CCD is a "constrained" CDA? Comment: If there isn't an official CCD schema, why doesn't HL7 create one and help everyone?
        2. Keith, If there is no schema for a CCD, then can you point me to a document that details the specific constraints that a CCD places on the the "base CDA Schema"? I have the document "HL7 Implementation Guide: CDA Release 2 - Continuity of Care Document (CCD), April, 2007". But its focus is on converting a CCR to a CCD; and not on the constraints a CCD places on the CDA schema.
      44. Keith, Where can I find the valid entries for a codeset referenced in the CDA? For example, the Participation Function code set 2.16.840.1.113883.5.88 is used to specify the exact function an actor had in a service in all necessary detail. One valid value is PCP. But what are the other values? Where can I find them?
        1. You can find these values in the vocabulary section that comes with the HL7 CDA Standard.
      45. Where can I get the C-CDA implementation guide? It is near impossible to adopt and implement these standards without easy access to documentation and tools. It would be nice if there is a single location where you can get all the information. Here are some challenges 1) Where is the C-CDA implementation guide 2) I found a companion guide the implementation guide on the blue button site. I guess the implementation guide is so complex that a companion guide is required 3) Are there validation tools: How do I validate a CCDA? 4) Where are there development tools to generate/consume CCDAs 5) How do I get a hold of the terminology sets should be used to code data. I am sure it's all "out there" if you know where to find it - some of it is public and some not. If the goal of MU2 is to encourage interoperability it would greatly help if there is a single source for all the information to make it happen.
        1. You can find the C-CDA implementation guide here. The companion guide addresses details not found in the C-CDA guide because they appeared in regulation after that guide was published. You can find validation tools here. The same tools used behind the scenes for for testing and validation (MDHT) can also be used to generate/consume C-CDA documents. As for terminology, see the C-CDA consolidation guide for references. You can find some of it documented here, other parts of it are in the referenced standards (e.g., HL7 CDA), and terminology for Clinical Quality Measures can be found here. All of this information is publicly available, and as of April 1st, it will also be free (which is not the same as public). As to your last issue (finding it all), that's one of the reasons I started this blog.
        2. Appreciate the response. I followed your first link for the C-CDA IG but found several documents there. Are you referring to CDA R2 IHE Health Story Consolidation? That seems to be the one suggesting "Consolidated" The other references are very helpful, particularly, MDHT and VSAC. Thanks!
        3. Yes, and make sure you get version 1.1.
      46. Hi Keith - Thanks for the amazing work you put into this blog - it's really great to have a "live" person who is actually responsive and engaged in this area to bounce issues off of. I have a question relating to the separation of Patient from Person in a C-CDA (or any CDA for that matter). I'm trying to indicate a Patient/RecordTarget's education level, living arrangement, and disability designations. It would seem like the reasonable place to put this in is on the attributes of Person, but I can't figure out how to associate the Person attributes with the Patient. From my research, it appears that the preferred method would be to indicate these data elements as observations in, for example, the Functional Status or Social History sections, but the attributes off of Person are just too alluring for me to "let it go." I'm almost positive I'm missing something, but I can't figure out what it is. Any ideas? Appreciate the help!!
        1. I'm afraid you will have to let it go. What you are missing is that you could make those kinds of observations about any subject, including family members of the patient, or their healthcare provider (and education level even makes some level of sense in that context). The way that V3 works, you have acts (including observations) and the patient is simply one of many possible subjects that could be associated with them.
        2. I had a feeling you would say that, but the confirmation is certainly helpful. As with all standards-based work, I don't have to like it, but as long as I know, I can make it work. Thanks again!
        3. Hi Keith unable to find xml schema files(xsd files) for implementing C-CDA for MU2 170.314(b)(7). Can you please help where can I find it Thanks Kumar
        4. Kumar, There is a schema for the CDA standard that you can find here (if you aren't an HL7 member, wait until April 1). C-CDA is a restriction on it that cannot be represented using W3C XML Schema (see whyhere). Y
      47. Is there a template and templateID for encompassingEncounter?
      48. what is the recommended id type that need to populate patientrole id field (/cda:ClinicalDocument/cda:recordTarget/cda:patientRole/cda:id)? Should it be MRN (local id at the site) or MPI (master patient index at the RHIO)? please let me know if may question is not clear?
        1. just to clarify question further, the situation is where RHIO is importing a CCD sent by a site.
        2. The CCD you are generating could be used by a number of different people. The ID that has the most value to the stakeholders using the document is the one you should probably be using. That could be the MPI identifier in the RHIO, or it could be the MRN, you are a better judge of that than I.
      49. Hello, I have a question on the level of detail required for the Audit Reports in Meaningful Use 2. Is it necessary to record patients returned on reports that are generated? For example “show me all patients taking Lipitor”. This can be a large list and I am concerned about the number of entries that would need to be logged. Thanks, Shelly 
      50. Hi Keith - the Consolidated CDA guide from HL7 doesn't seem to be free of charge as was expected. Could you confirm please? I wasn't sure if I was missing something. Thanks, Bindu
        1. Never mind Keith - the document is now available from HL7 free of cost. Thx.
      51. Hi Keith, Do you have any thoughts as to a strategy for representing SNOMED post-coordinated expressions for the value/code in for instance a Problem Observation template? It appears that the values are constrained to a list of pre-coordinated SNOMED codes. I'm running into a number of scenarios where information would be lost if I only used the pre-coordinated codes. Thanks for your help. Your blog is invaluable! Josh
        1. Be careful here. You have both <code> and <value>. Code has an enumerated value set associated with it. <value> does not, and could use post-coordinated information. I'd need some details as to what you were trying to do.
      52. Hello Keith, the Mu Stage2 requirement states this "Electronically and simultaneously display (i.e., in a single view) the data from at least two list sources in a manner that allows a user to view the data and their attributes, which must include, at a minimum, the source and last modification date". What is the source and last modification date from the CCDA? - is it the author? custodian? componentOf/encompassingEncounter/location/healthcareFacility? performer? Authenticator? or do we need to look at the author of the individual sections? Thanks in advance. -sri
        1. If you are comparing two documents, the date associated with the "authorship" of the act is what you should consider to be the "modification" date for the act. The source could be considered to be many things, including the document being compared, the information system which created it, the provider who authored it, or the organization who created it. This would be a question for ONC, and not for me. Once you know which of those it is, you could find your answer in /ClinicalDocument/id, /ClinicalDocument/author//assignedAuthoringDevice, /ClinicalDocument/author/assignedAuthor, /ClinicalDocument/author/representedOrganization, et cetera.
      53. Hello Keith, Great Blog. I would be interested in hearing your opinion about which version of a summary document (CCD, C32v2.5, C-CDA, other) would you create if you were starting on the project now given: 1) You were a Public Health Agency starting down the data exchange path for the first time, and very few Public Health Agencies have done this. 2) There is no MU money for you but you want to try to be MU compliant 3) You want to be compliant with as many providers as possible so that your data can be consumed 4) Your core data that you capture about patients has a lot of unstructured data and/or values that are not mapped into a Vocabulary (The Omaha system classification for example) Thank you.
        1. Peter, I'd be starting from the C-CDA given that's where everyone else is starting from for 2014. Omaha has been mapped to LOINC, I believe, at least for questions.
      54. Hello Keith, Per the Mu Stage2 requirements, the provider must have the ability to select/deselect the items they export on to a CCDA. In your opinion what level of granularity would they have to select/deselect? Would the ability to select/deselect allergies/Problem/meds/labs/Encounter suffice OR would they have to have the ability to select/deselect items within an encounter - like Vitals, care plan, Functional status etc.? Thanks!! -sri
        1. See (iii) Minimum data from which to select. in the rule, which gives you the list of items that a user has to be able to select from.
      55. We're working on a report/results distribution system that uses the CDA. There seems to be no concept in the CDA header for an "ordering provider", beyond the loose connection of the "primary recipient". Is there any way to put the ordering provider into the CDA header?
        1. See how the Diagnostic Imaging Report handles this in the Consolidated CDA Guide. Essentially it is the referring physician associated with the service event or encounter.
      56. Hello Keith, We are working on an Ob/Gyn product and I am working on the interop design for its different modules. I have been struggling to find the relevant HITSP templates for OB History, Gyne History, Fertility History, Sexual History etc. but I dont seem to be able to find any pointers/references to them so far. Thus i would like to see if you can be of any help in this regard. Also since there can be many other sections for which predefined templates can not be found, is there a way for us to customize them for ourselves or alternatively how to we add unstructured/properitary format information for such sections in our CCD document? Kind Regards, Javed
      57. Hi Keith, I'm trying to build a local copy of the NIST CDA Validator (muCrValidation20130206) package. Do you know where I can find the appropriate jar files to resolve the following references: gov.nist.hitsp.validation.* gov.nist.hl7.v3.xml.util.* everything I seem to find is too old :(
        1. I sent an email to Andrew, hopefully he will help.
      58. Hello Keith, In many samples of CCDs in the Advanced Directives observation, there is a value element that gives the actual value of the code just previous. In the below case, the code is "Resuscitation" and the value is "Do not resuscitate". However, I cannot find the definition for this value element anywhere in any spec I can find. Do you happen to know the valueset for this value element? I had to remove the left and right brackets from the xml below because Blogger was not allowing them to be published. code code="304251008" codeSystem="2.16.840.1.113883.6.96" displayName="Resuscitation"/ statusCode code="completed"/ value xsi:type="CD" code="304253006" codeSystem="2.16.840.1.113883.6.96" displayName="Do not resuscitate" originalText reference value="#AD1"/ /originalText /value Thanks, Matt Wennerlund
        1. Matt, The CCD value sets for Observation/value are carried over from CCR, and don't necessarily make sense in the context of the observation/code valueset provided. The example you are looking at uses the Resuscitation code as recommended by CCD, but then uses values that are appropriate in SNOMED CT, rather than the CCD Value Set for observation/value. Keith P.S. To insert < or > you need to type &lt; or &gt; (and to insert an & you need to use &amp;
        2. So I need to look to CCR to find out the value set for Observation/value for CCD? Or is there just no place in CCD for the value element, and the example is bogus? Thanks, Matt
      59. One question I am asked often is how can we send free text 'clinical notes' through CCD (particularly HITSP C32)? Could you please guide how 'notes' are included in CCDs?
        1. That takes me back to why I started this blog. You can put free text narrative in any section of a CCD. But if you want to use that CCD to qualify for Meaningful Use, then you need to have structured content.
      60. Could you please give the definition of the Problem - §170.314(a)(5)? Please give some examples of the Problems. How do you think what for and how Physician will use a Problem List (actual using of problem list)?
      61. Hi, What do you think about FHIR? http://hl7.org/implement/standards/fhir/
      62. can HITSP C32 contain dual coding for a problem, i.e. a problem coming with both corresponding ICD9 and ICD10 code?
      63. Hello- We are attempting to certify 2014 using MS HealthVault as our portal. Though they claim to be WCAG compliant, we have had a heck of a time getting any documentation from HV with regards to WCAG. Any ideas?
      64. Hi Keith, I have a question regarding 170.314 (a)(14) - Patient List Creation They are looking for a filter and sort based on "date/time", however they do not specify which date(s)/time(s) specifically they would like for this to be sorted/filtered on. Is there somewhere I can find information based on which date(s)/time(s) they are looking for here?
      65. Hi Keith, Am an EHR vendor. I would like to implement DIRECT in my native code. Below is my understanding. 1. I need to get direct address and certificate from HISP established in my state. 2. I need to publish the obtained direct address and certificate in DNS or LDAP servers 3. I need to get other physician direct address along with their public key certificate to start transferring PHI. Please do confirm my understanding or do let us know the steps to be followed that would be of great help to EHR vendor community. Thanks, Riya
        1. I believe HISP will provide you with digital certificate. It will also publish them in DNS or LDAP ( no need to publish in both). I am not sure if the HISP is supposed to be from your state. I assume you have customers (doctors) who use your product in many states. May be some one with experience can clarify this issue. I am trying to find ONC certified HISP my self. Do you know where I can get a list of ONC certified HISPs? If we use HISP not certified by ONC, we will not be certified for MU2.
      66. Hi Keith, I am looking at the HL7 CCD, HITSP C32 specs for summarization of episode note and just realised that not just in HITSP C32 / C83, but also in HL7 CCD, none of the sections are specified as required. I finally found that IHE PCC Medical Summary (1.3.6.1.4.1.19376.1.5.3.1.1.2) mandates three sections as required. However, it is not required to conform to IHE PCC Medical Summary template at all per HITSP C32. So with all this I am trying to figure out if any sections are indeed mandatory in HITSP C32. BTW, I am more interested in HITSP C32 as that is the one that is mandated by Meanigful Use Stage 1. Is my understanding correct? Thanks, Sridhar
        1. Sridhar, The answer is, it depends on what you are doing. If you are implementing the C32 in a program in the US, likely you are trying to meet the Meaningful Use requirements. That DOES have requirements on which sections must be present. The list of what you need for that can be found elsewhere on this Blog and in the NIST MU Stage 1 testing procedures and tools. However, why would you be implementing to MU Stage 1 now? I'd be working with C-CDA and Stage 2 requirements at this point. Anything prior to that has just a few more months of life in it.
      67. Hi Keith, I was poking around the "Open Source Standards Implementations" page and noticed the Everest framework isn't on the list. I was wondering if it could be added as some implementers have found it a useful starting point in their v3 projects. It supports the formatting/parsing/transmission of HL7v3 messages in Java as well as .NET and has an implementation of the v3 data types (there is a larger feature list available at the site) ... http://everest.marc-hi.ca/ Thanks -Justin
        1. Just added it. I don't know a lot about it.
      68. Hi Keith, For the following Transition of Care Display certification criteria: EHR technology must be able to electronically display in human readable format the data included in transition of care/referral summaries received and formatted according to any of the following standards (and applicable implementation specifications) specified in: § 170.205(a)(1), § 170.205(a)(2), and § 170.205(a)(3). Is it sufficient to display (in a styled and meaningful way) the human readable sections in a Level 2 CCD? That is the html content within the nodes? I've been searching but haven't found a clear indication of whether we need to parse and display the machine readable data (included in level 3 CCD) for display or just the human readable (which from what I understand is a Level 2 CCD). Thanks!
      69. Hi Keith, I'm seeing the errors posted below when validating (http://hit-testing.nist.gov/cda-validation/muCr.html) a sample CCD (based on CDA), but it doesn't look like the addr and telecom elements are actually required. I'm looking at an Excel spreadsheet that shows the requirements, and for each of the following segments, telecom and addr are optional ([0..*]). All patientRole, assignedAuthor, assignedEntity[not(parent::dataEnterer)] and associatedEntity elements SHALL have an addr and telecom element. All guardianOrganization, providerOrganization, wholeOrganization, representedOrganization, representedCustodianOrganization, receivedOrganization, scopingOrganization and serviceProviderOrganization elements SHALL have name, addr and telecom elements. Thanks, Erik
      70. I don't know where the spreadsheet came from. The constraints for the Cancer Report come from here. And that requires the content that you've specified above by referencing constraints from the General Header that was in the CDA4CDT H&P Guide, which says those words verbatim. So, put them in.
      71. Hi Keith, Do you know if PIX Add (v3) can carry information other than demographics like if patient admitted or discharged?
        1. The short answer is no. The long answer is also no. The reason is that the PIX Add message starts with the Patient Role, and includes No acts (an encounter is an act) in which you can put admission or discharge information.
      72. Hi Keith , First of all awesome blog with lot of information . I am very new to HL7 and CDA . I have a requirement where I need to import XML CDA into SQL Server 2008 database . And later generate XML file from Database . I did read this is not a good approach but there are some reasons we are doing it . Do you know any existing procedure or tool that I can use ? Do you know anywhere I can find database schema for this (Even mysql sample is fine)? Thanks in advance and I will appreciate your input . Thanks
      73. Hi Keith I was wondering if you had any thoughts how a vendor for a Radiology Information System program would implement clinical decision support for MU Stage 2. It seems like the MU CDS rules are geared more towards providers, in that it helps them decide on a course of action depending on a list of variables. However, when a patient arrives in a radiology department the decisions have already been made by the provider. There's not much else anyone in the radiology department can do. To make things even more difficult, the ONC's testing criteria "Test Procedure for §170.314(a)(8) Clinical decision support" states over and over that each CDS intervention be automatic and based on the patient's problem, medication, medication allergy, demographics, labs, and vitals. Like I said, all of those variables have already been sorted out by the time a patient has arrived in the radiology department, so do you have any thoughts on how a radiologist can incorporate CDS?
      74. Sorry for the novice question but I have looked at so many examples I have confused myself. I just need some clarification on defining a CCD. Is a CCD specified by the below syntax? typeId root="2.16.840.1.113883.1.3" extension="POCD_HD000040" where 2.16.840.1.113883.1.3 = RMIM and POCD_HD000040 = CCD Also, I'm seeing a lot of examples with additional templateIds at the ClinicalDocument level that say, templateId root="2.16.840.1.113883.10.20.22.1.1" templateId root="2.16.840.1.113883.10.20.22.1.2" But I can't find those OIDs in the HL7 OID registry. An explanation would be greatly appreciated.
      75. Hi Keith, I have a question on recording patient ethnicity Category codes. CDA is providing 2 codes: Code Code System Print Name 2135-2 Race and Ethnicity - CDC Hispanic or Latino 2186-5 Race and Ethnicity - CDC Not Hispanic or Latino Choose not to disclose is required from MU perspective: Should this be linked to a code and if so which (CDC) code? Best regards Arjo
      76. Hi Keith, I am using your book to give recommendations for mapping HL7 version 2 messages to CDA Release 2. The book have pretty good coverage and I am moving alone, but I can't find a good example of Allergy(AL1 segment) mapping to HL7. ADT^A19 - Patient Query Response is the hl7 that we are expecting back. Please let me know if you have any ideas about mappings. Thanks for all your help! Roman.
      77. Hi Keith, can we please use your Clinical Quality Measure diagram, of course with acknowledgement for a presentation @AMIA... Best regards, Fauzia Khan/ Alere
      78. Hi Keith, Would you have any information you can share with regards to exchanging Microbiology Results in a CCDA document? How does CCDA support linking the sensitivity tests performed for the associated organism? In HL7 we have used OBX-4 to tie together results that should be grouped together. Best Regards,Cecilia
      79. Hi Keith, We are preparing for a conversion of multiple EMR systems to a single system. We decided to use CDA as a middle ground. We thought if we get required data into a CCD then we should be able to go from the CCD to the new EMR system. The current challenge is translating legacy data to the standard. Today's challenge, medication frequency for values like Q4-6H, PRN, TIDCC, Q3HPRN, and the like. Do you have any advice on coding medication frequency from legacy data? It has been difficult to find examples of the cases listed above. Any assistance would be appreciated. Thanks.
        1. See these examples from IHE. Q4-6H is dubious at best though, and against current prescribing recommendations.
      80. Keith, thanks for all your work in maintaining this site. It is a great resource. I am a HIT architect for a Medicaid enterprise with a programming and networking background. I've been working with CDA for a couple years now and have implemented the HITSP C32 and new CCDA summary template. My question is about the CMS e-CQM submission standards. A while back, I reviewed the CMS requirements for e-CQM submission. As I remember, I came away from that thinking I wasn't hype about implementing this because there wasn't a way to track patients longitudinally. I don't think I saw a patient identifier, let alone one based on an MRN at the EHR implementation. Without this, e-CQMs aren't an attractive option for the money it would cost to implement. Our attestation portal that gathers manually-inputted nums/denoms would be fine. Am I in error on this? My CMS contacts are assuring me that an e-CQM implementation would supply patient identifiers that I could use across years of attestation for longitudinal studies. I have read some of the threads on your site and there are references to different levels of QRDA. What level of QRDA supplies patient IDs that can be tracked longitudinally and is this level included in the ONC EHR cert requirements for 2014? Got a hardcopy of your CDA book, great reference. Thanks for any input you can supply, jason.a.phipps@gmail.com
      81. Hi Keith, The organization that employees me is trying to attain MU2 compliance and the particular piece that I am currently working on is the CQMs. We currently trap all encounters in our EMR using CPT-4 codes and descriptions exclusivley. We are looking into ways to accommodate the requirement to trap Face-to-Face Interactions recorded in Snomed CT for all CQMs and work that into our current CPT-4 only encounter table. We are looking into CPT-4 to Snomed CT mapping but currently are just in the discovery phase of this. Could you give us some insight as to how other developers have handled this problem? Any advice would be much appreciated. Mike
      82. We are using CCDA to communicate between the EHR and our CQM system. Several of the QDM elements are not supported by CCDA. E.g. Source of payment. For some elements we are surprised to find that there is no CCDA support. E.g. an Indication why a procedure was not done. You addressed this issue in your blog (http://motorcycleguy.blogspot.com/2013/05/ours-is-to-reason-why-what-others-do-or.html). Our proposed solution for the ‘reason not done’ is to use the Indication element. This kind of fits. We would use ‘finding’ for the problem type code and the appropriate SNOMED code to indicate the reason the procedure was not performed. We do have a concern that it will become confusing if you want to indicate both the reason why you would have performed the procedure (positive reason) and the reason why it nevertheless was not performed. Who should we reach out to to discuss the validity of our solution? Who can we talk to work towards adding this into a future version of the CCDA? Thanks Jonathan
        1. Jonathon, you would want to ask the HL7 Structured Documents workgroup about your solution. In general, the proper way to indicate WHY a procedure (or any other act) is not done is to use negationInd='true' on the act to indicate it wasn't done, and then provide a coded observation inside it using the Reason act relationship: <act ... negationInd='true'>  ...  <entryRelationship ... typeCode='RSON'>   <observation ...>...</observation>  </entryRelationship> </act>
      83. Thanks to Keith for the great resources! Question for this blog's community as well as for Keith. To help with capacity planning, I'm wondering what the typical number of documents found in an XDS repository via XDS RegistryStoredQuery for a given patient might be, and also what the normal range of returned payload sizes of XDS RetrieveDocumentSet requests might be. Any thoughts on this?
        1. We're told a single provider spends 10-15 minutes per patient (less time face to face). That covers 32 encounters a day in an 8 hour day. I know radiologists can generate 50-60 reports a day. A two page report would generate about 20kb of XML in UTF-8. I know from previous experience that the average number of in-patient and out-patient documents per patient maintained at a hospital is somewhere between 1 and 2 per year, and closer to 1 than 2. There's enough data there to give you an idea. A single medium sized hospital can generate 5 million documents a year, a single provider might generate 10,000 documents a year. Your Mileage may vary.
      84. Hello Keith, Are you aware of any MU standards/certifications related to using biometrics as authentication for ePrescribing? Voice or fingerprint? Thanks!
        1. I'm aware that there are in fact no standards identified in Meaningful Use for these purposes.
      85. Hi Keith, I recently finished work on a C-CDA compliant implementation (C#) to generate C-CDA CCDs. However, now I need to retrofit (or rewrite) the implementation to pass the NIST HITSP/C32 validator. In other words, I need to (temporarily at least) generate HITSP/C32 compliant CCDs in the least painful way from a C-CDA implementation. I understand there are structural differences beyond just Template ID differences. Any advice on the least painful way to retrofit this? Or by chance has anybody written a transform that you know of that will take a C-CDA CCD and go backwards to a HITSP/C32 CCD compliant document? Thanks, John King
        1. Nobody I know of has done that. You might want to look here at this series of posts to figure out how to work in reverse. You might want to read them backwards, but don't blame me if it summons something you don't like.
        2. It turned out not to be too terribly bad. Got it done in a day or two. Your posts were very helpful, thanks. Now I really appreciate Consolidation! John K.
        3. Glad to hear that! Care to share your experiences or the code? Possibly do a guest post? If so, let me know.
      86. Hello! Can You help me how to set up Direct certificates discovery through LDAP. Have a lot of manuals, but really don't know what of them is more helpfull. May be You know usefull links or documentation? Best regards, Ivan Knyazev IT Department, Developer. 3047 Avenue U Brooklyn, NY 11229
      87. What is an Encounter Diagnosis supposed to be [as opposed to or in relation to a Problem]? If it is supposed to be a billing code then would you know why is Snomed-CT required for it? I am looking for a description that I can explain to doctors and then build my EMR to that. Thanks!
      88. Keith, I am looking for the best way to differenciate a CCDA document from a CCD document. In the header section of the document are the templateID/@root values. Are these values unique to the type of document it is i.e. CCDA is type ( 2.16.840.1.113883.10.20.22.1.1 or 2.16.840.1.113883.10.20.22.1.2 ) ?
      89. Hi Keith. I am having trouble understanding some of the validation errors we are receiving. Below are examples. Any suggestions on where to start with these would be helpful. Thanks! The 'validateII' invariant is violated on 'org.openhealthtools.mdht.uml.hl7.datatypes.impl.IIImpl@a977cc2 The 'validateURL' invariant is violated on 'org.openhealthtools.mdht.uml.hl7.datatypes.impl.TELImpl@47988c63
      90. I have just started looking at the reporting numerator requirements for MU stage 2 and wanted to know for Transition of Care when a provider refers a patient to another provider electronically how can we confirm that the document sent is a Transition of Care document and not another type of document which means it shouldn't be counted in the numerator? Do we have to rely on the metadata of the message content?
      91. Hi Keith - I have a question regarding the classification of procedures for inclusion in the CCDA. The CCDA Implementation Guide describes three procedure categories: act, observation, and procedure. Is there guidance or a mapping for determining the category for procedures in SNOMED or CPT? I know that either can be used for MU2 purposes, but I'm unsure if the categories should be arbitrarily assigned or if the assignments are pre-determined in some kind of mapping.
      92. HL7, are there any plans to add a segment (ADT) to indicate whether a patient wants to opt-in or opt-out of the HIE?
      93. Hello, If an EHR or patient portal receives lab test results via an ORU from a CLIA certifed application(the lab), does the EHR or patient portal then fall under the CLIA regulations as they are taking direct lab feeds?
      94. hello, We are moving to MU2. But currently we are using ICD9 code for diagnosis. But for setting encounter diagnosis in CCDA as part of transfer of care, what data to be placed? from where we get the SNOMEDCTcode for ICD9.
      95. This is a followup to your June 30, 2013 Post - And Odd-Essay Through Time I'm hassling with the definition of "starts after end" as it applies to CMS 153/NQF0033 Chlamydia Screening which has an exclusion with the "starts after end." In the Cypress test data, the end date of the source and the start date of the target are shared and evaluate to true for "starts after end." This seems to challenge common sense but I can't seem to find anything that discusses the boundary condition for these particular operators (though I have such discussion for "during" and "concurrent." Before we implement this, I would like some documentation to provide to my team. Any documentation you can provide would be appreciated.
      96. Hi Keith! Thanks so much for all the information you share on your site! It has been a welcome resource for me. My question: I would like to modularly certify for a number of criteria - the list includes the medication list, problem list, medication allergy list, smoking status, clinical information reconciliation, Transition of care, and demographics. I see that I need to also certify for 170.314(g)(1) (Automated Numerator), but am confused and cannot find a very clear definition of when to certify for 170.314(g)(2) (Automated Measure calc). On Drummond's 2014 Edition EHR Modular Certification Registration - I checked off the items I would like to certify and it did not 'pre-select' 170.314(g)(2). However, when I look at the Test procedure for both criteria - I cannot decipher what is mandatory. And obviously - I do not want to screw up! :) Do you know which decoder ring I should wear in order to interpret the Test procedures to get the answer I am looking for? Thanks~! Gretchen Ellis
        1. This is another question for which I'd be digging in the CMS FAQs or the NIST test procedures, rather than the standards for. I'm afraid I don't know the answer.
      97. Hi Keith, I appreciate your time in maintaining this blog. It answered so many of my questions. In HL7 CDA R2, what OID should we use for ClinicalDocument/id/@root? Can we use Organization/Facility OID for ClinicalDocument/id/@root or ClinicalDocument/id/@extension? What will be the best place to put the Organization OID? Can we use custodian id tag? Thanks.
      98. Hi Keith, I am in the process of mapping the data in our existing MEDITECH EHR to the values sets required to be used in the C-CDA. I have come across elements in our system that can not be mapped to the standard value set entries and I don't think we should change the system to conform either. I feel we should still include the values in the C-CDA document though. 1. Should we create custom entries or map to closest value? 2. If we should create custom entries, should we create codes similar to other entries within the value sets? (ex. Race has a 5 digit numeric code, Relationship has a alpha shorthand code.) Ex. PersonalRelationship: We have "Guardian" and "Ward" entries in our system and the value set does not have an equivalent value at this time. I have seen some recommendations that new values be added, but how should we handle with the current version of the value sets. Notes: Value Set Name Personal Relationship Role Type Value Set Value Set OID 2.16.840.1.113883.1.11.19563 Thanks, Mat
      99. Hi Keith, Is it correct that MU stage 2 does not require the use of HL7 Version 3 messaging standard for electronic submission of data? Also, can we still comply with MU stage 2 without fully adopting HL7 v3 RIM and v3 messaging standards? Thank you! Linda
      100. Hi keith, for CCDA MU stage 2 am displaying Race and Ethnicity Along with Patient Information . when Validating Getting Validation Message.can you give me link to validate CCDA.
      101. Hi Keith, Do infant security or wandering management systems contain PHI and fall under HIPAA security? These systems clearly identify an individual patient by name, location (room, ward, building) and display or send (SMS, Wifi) the individual's info to users for security alerts and alarms. but the system does not have any info related to the patient's health condition. Is the mere fact that the security system is used within a hospital (CE) classify the system as having health information. Thanks, Steve.
      102. I have a CCDA that has instuctions for the medications defined as directions in narrative text and not under instructions. While parsing this CCDA how should I able to read the instructions? Can the narrative text use its own way (directions, intructions or any other word) to deifine how the medications are being administered (like "2 tabs a day") etc. What is the right way of this implementation?
      103. Keith, I am writing a program to transport a PDF report to an hospital information system. I am CDA newbie and I am devouring your posts. My current plan is to include the PDF (either by reference or value) as nonXMLbody in a CDA document. Thanks for your post on this; it was very informative. I have a few questions: What is the best practice for transmitting the CDA document itself? Via HL7 message? If sending via HL7 message is the optimal method, then what type of message? If not HL7 message, what other options exist? Publish the CDA document on a web server? Any other alternatives? Do you know of any examples that illustrate PDF encapsulation in CDA? Also, if I have the PDF encapsulated in DICOM then is there a way to include that DICOM object in CDA? Thx, -- Jay P.S. Does your book address any of the topics/questions in this email?
      104. Hi keith, I am trying to generate QRDA1 using MDHT model and have stuck up with the setting of the sdtc:valueset attribute to a code or value tag. For Example: value code="4855003" codeSystem="2.16.840.1.113883.6.96" xsi:type="CD" sdtc:valueSet="2.16.840.1.113883.3.526.3.327" Tag for code,codesystem are present in MDHT but there is no tag for valueSet in library. I have used the latest MDHT library "org.openhealthtools.mdht.uml.cda.runtime.R20121220-1425" and "org.openhealthtools.mdht.uml.cda.consol.runtime.R20121220-1425" but cannot find appropriate tag for valueset. Can anyone suggest the right way to solve the issue????
      105. Hi Keith, First of all, a great initiation from your side helping development in Health standards development. I also got the chance to work on healthcare projects and while doing some research came across this link. Well on to my queries now. I got following questions where you might help 1) Is there any special or standard in which we show CCDA document type information. 2) Can I get few sample documents for CCDA document types. I already got one for each type given on HL7 website. 3) Is there any parser available where I can parse CCDA document types and extract information from it. Thanks in advance.
      106. Hello Keith, We are trying to submit CCDA to NWHIN Connect endpoint with all security (SAML, certificates etc) Currently this end point expects XDR only and not directly CCDA. Is there any tool available to convert CCDA to XDR and submit to Connect? How about Open Health Tools Bridge? Any idea about compatibility? Mahesh
      107. Keith, We have an HIE client that is wishing to include the 42 CFR Part 2 Disclosure in a narrative note of sorts in the CDA Header (C32 or CCDA (CCD). Consent is given but the disclosure language needs to be inserted. If in the header, do you have any thoughts on the best way to do that? Or would the annotation comment be a better place to try and put the disclosure? Thanks, Doug
      108. Keith, In the implementation guide, there are statements like: "SHALL contain exactly one [1..1] value (CONF:14263)" Any idea on where one can find what (CONF: 14263) means? Or for that matter, even if it's pertinent. Thanks - Steve
      109. Keith, I've got code for MU-2 that extracts Medication, Problems, and Allergy content from a C-CDA. Would support for the IHE content profiles likely be as simple as looking for different templateIds? Thanks. Elliott
        1. Yes and see http://motorcycleguy.blogspot.com/p/moving-on-from-c32-to-ccda.html
Share | | |
Subscribe to: Posts (Atom)