Page tree
Skip to end of metadata
Go to start of metadata

 13:00 AEDT

Meeting 42

Attendees

Jared Davison

Andrew McIntyre

David McKillop

Isobel Frean

kyle macdonald

Michael Legg

Rueben Daniels

Dalisay Giffard

Andy Bond

Angus Millar

Danielle Tavares-Rixon


Apologies

Eric Browne

Anthony Cruice

Discussion

Follow up from Isobel's email:

  1. Apart from being complicated and slow, what exactly are you suggesting needs to be fixed re the QA model for pathology?
    1. Andrew: 
      1. QA for healthcare messaging. Pointless pushing pathologist for correct messages if they don't work at the receiving endpoints.
      2. Michael : requirement in GP standard. not just about putting up on the screen pathology result taken from atomic data.
        1. receiving systems should have similar requirements on them to comply as senders do
      3. Andrew: receiving systems QA is priority since the are dependent on receivers compliance capability
      4. Angus: senders currently have more capability than receivers. if receivers have capability then senders are more likely to send them.
      5. Andy: dependent on maturity of order sending 
      6. Isobel: what position of working group to solve compliance?
        1. Michael: using existing accreditation systems. e.g. GP should be done under the existing GP accreditation system
          1. already done in pathology through SPIA.
          2. Public health have capacity to legislate rules
        2. Andrew: there is no shortage of ability to regulate the clinical side of medicine. 
        3. Michael: can't lock software into standards e.g. TGA natural experiment in USA, blood banking was controlled by FDA. Triple G locked done accredited system. In Australia then was safer because it was locked down.
          1. Danger to lock down of software too. 
          2. has standard list of pathology messages that can be shown on the screen to be sure system can be viewed correctly
        4. Testing software can handle compliant messages. 
        5. Andrew: Currently no testing required to release clinical related software
        6. Michael: currently the stick is only on the pathologist side, and is not on the side of the receiver. if a pathology org can demonstrate that it 
          1. units, time, consistent order. actually been able to read the message
        7. Andy: funding has been a problem for testing. no one has been able to make a unsustainable business. Formal NATA accredited software testing was cost prohibitive.
        8. Andrew: costs of doing the job relatively low compared with the NEHTA/ADHA burocracy
          1. start with self testing, using a test suite set of requirements and set of messages
          2. software have to declare they have passed those tests
          3. 2nd stage is to verify software vendor is reporting their results truthfully
          4. no test suite 
          5. low hanging fruit is to develop a test suite and require software vendors to self declare
        9. Kyle:
          1. only part of the story
          2. most will try to cut as many corners as possible because they aren't required
        10. Andrew:
          1. required self declaration. 
          2. senders such pathologists could contribute messages that 
        11. Andy; if it was more cost effective or better long term value. rather than a carrot or stick approach. the testing govt for govt service access is hugely costly. by using providing libraries.
        12. Andrew: thinks there needs to be a requirement. similarly to doctors having to pass an exam. same should apply to software vendors. having a library of test messages should be a huge boon to software developers to help them test suite.
        13. HL7 could put up a strawman testing 
        14. Angus: asks why not use FHIR? where can you make quality improvement impact.
        15. Michael: This is about risk and safety can be done now. and are independent of the data format
        16. Andrew: need culture of software quality of what is in use. patients lives are dependent on data being transition

Resolution

OO Meeting is supportive of Copies to being profiled into FHIR profile for Diagnostic request.

https://github.com/hl7au/au-fhir-base/issues/670 The concept of Copy to Doctors is essential for safety and clinical. We want to be able to move between FHIR and HL7v2 losslessly. 

Orders and Observation request to review FHIR profile work for orders and observations. Request presentation from profile developers ADHA.

Next Meeting

  • Organise impromptu meeting when Isobel has paper ready.

Next meeting to be determined.

  • No labels