Data Usability Workgroup

Data Provenance and Traceability of Changes

Viewing 2 reply threads
  • Author
    • #33470 Reply
      Hera Ashraf

      There are many things that can happen between a clinician documenting a piece of clinical data in one system, and you seeing that data in your own system. “Provenance” refers to the origin of a piece of data and what has happened to it as it has been transmitted between systems, which may include the name of the clinician who originated a piece of data, their organization, or modifications that have been made to the data. 

      Reply to this post with your answers to these questions:
      1. Is it important for you to know all users who have touched/reconciled the information, only the originator or only the most recent? 
      a) Does this requirement change for different types of data — e.g. labs vs. Problems/diagnoses
      b) What do you consider important provenance information: the clinician’s name, credentials, specialty, the name of the hospital or clinic?
      2. Which situations are the most important for receiving an updated piece of clinical data?

    • #33686 Reply
      Andrea Pitkus, PhD, MLS(ASCP)CM

      Regarding laboratory data, there are specific CLIA requirements by law which overlap with the questions. CLIA is more specific than HL7/ONC data provenance requirements too.

      Each laboratory accrediting body (i.e. The College of American Pathologists, Joint Commission) may have additional requirements such as interface checks ensuring what the performing lab sent to the first downstream entity (i.e. EHR, HIE, another lab, public health) doesn’t have any truncation or data issues, decimal points are in the right place, etc. There required data elements such as the performing laboratory address, etc. When data are updated there are also requirements to retain the original and updated information, communicate the change, etc.

      See CLIA §493.1291 Standard: Test report.

    • #34592 Reply
      Riki Merrick

      In additon to Andrea’s points:
      What data is important depends on the use case and what the data will be used for, but at minimum you would need to know who changed the data and when and in some cases why. WHat elements to us to identify the who (could be an id like the NPI, but if it’s not something unique we probably would need name (last and first) and associated organization at a minimum.) I would look at the provenance resource and check with the owning WG at HL7 for input.
      obviously any time a change in data affect clincial treatment or impacts public health that change needs to be communicated.

Viewing 2 reply threads
Reply To: Data Provenance and Traceability of Changes
Your information:

Skip to toolbar