Tech ‘front and centre’ for new Health NZ board

eHealth News
1

NEWS  - eHealthNews editor Rebecca McBeth

This is a companion discussion topic for the original entry at https://www.hinz.org.nz/news/728107/
2

I’m concerned that the focus seems to be on HR, Payroll, finance…
Those systems do not need to be where we strive for excellence.

The patient’s digital health record (all of it!) needs to be excellence is delivered. It is imperative that clinical staff (of all flavours) have secure access to a patent’s record regardless of setting and location.

1 Like
3

You are 100% correct, Chris.

But really we need excellence in both clinical and supporting systems, rather than one or the other.

1 Like
4

Aside from cybersecurity, the priorities for the coming years seem to be workforce, radiology and hospital desktops (little doubt that tech is front and centre of those :grinning_face: ). No sign of any investment in interoperability - certainly not SDHR which is another one-way extract to an unfiltered data lake which will be about as clean as the water in Wellington Harbour.

1 Like
5

Is it better to have access to a slightly dirty single health record than to have no access to a single health record?
I suspect it is - provided that labs, rads, meds is fairly accurate. I understand that since I left, at least Meds and Labs are available nationally and, in the SI PACS as well.

I remember Nigel Millar telling me jhow much he could determine about a patient by just looking at the patient’s meds history.

6

SDHR is a shared record (a copy of a subset of GP data) not a 'single" one and the first release will not include medications which seems strange to me given that they are held in a national data repository (the MDR which I helped to build when I was at Health NZ). I also don’t believe that pathology results are included in this release either. If you look at patient summary records in general, and the International Patient Summary in particular, the key data categories are current conditions/problems, allergies/intolerances, medications, and immunisations.

Poor quality data doesn’t help anyone or anything, including large language models, which will confidently hallucinate when using clinical data that isn’t structured and coded according to recognised standards.