Please edit this (is a wiki) or add more as replies so that we nail this down clearly!
The Problem
Digital health information resources (e.g. forms) are not being effectively shared in Aotearoa NZ.
Background
In NZ (and the world) there is a need to configure digital health systems in order to empower workflows. This configuration often requires the generation of forms or similar, each of which is individual workflow specific. This requires significant investment of IT and clinical resource.
The data required and actual physical workflows tend to be very similar in different places. However, there is little sharing of the working forms due to:
Multiple products across the country in each space
Different instances of the same product siloād from one another
Vendors provide only rudimentary forms
Absence of a useful sharing mechanism of existing forms
Absence of an accessible national collaboration platform for this sort of things
Proposed Solution(s)
GitHub space for the sharing of information resources:
I think there are a few more aspects to the problem statement that we should be considering before jumping to the answer (I hate how I sound like an IT person in saying this):
What is the artefact(s) we are producing? A standard? A prototype? A FHIR conformant json/xml document? A user interface design?
We have this problem in lots of spaces - eg. standards, code, FHIR IGs etc do we need to think about those as well?
When we say formās what do we mean specifically - assessments? Discharge summaries, PROMs, calculators?
What are the integration requirements for the forms (patient identification, pre-population, location it will be saved to, availability, authorship rights)?
What connection does this have to clinical pathways and networks, should we do the care pathway first (how do you manage versioning)?
Perhaps the place to start is with something like the smoking cessation protocols. Relatively simple, low risk, affects primary care and hospitals and is currently very inconsistent but could have a big impact?
To add a little more - I think one of the big problems we have is agreeing the content on the forms - which comes from clinical networks agreeing, which comes from clinical practice being standardisedā¦ Given the lack of traction nationally for the hospital pathways toolset I will double down on the smoking idea - 1. because in this conversation we have two anaesthetists (who care about this somewhat). 2. I also know that Al Rumball-Smith cares about this too which brings Chch/Wellington and Cortex into the conversationā¦
I agree Jon but I wouldnāt want to hold up progress.
The three concerns from my view are:
the ability to get clinical collaboration going.
Share first, then national templates, then alignment. Smoking cessation protocols is a good example. It starts with business / clinical process and data, thinking broadly, reusing patterns (then later D&D helps with interoperability, then UI/UX design and implement). Standards, whole 9 yardsā¦
The national directory and collab forum is the starting point.
This is the ādoing it properlyā ideal.
Efficiency gains from lifting and shifting forms and data models that work from one setting to another.
Itās a transitional approach to the point above (acknowledged not the same as doing it properly) but given we are so slammed for resourcing itās still a good thing.
Forms and content can be tweaked on copy/paste regionally if needed. Still requires development but you get to reuse the learning.
Cost savings and efficiency gains from copying working code.
Ditto to the point above ā¦ perhaps less flexibility if you wanted to maintain some notion of supportability but you get something now.
maybe a conversation is best. today has been a challenging dayā¦
I think if we make this too big, then weāll become overwhelmed at the size of the task and lose traction.
In my opinion, step 1 would be to create a place where people can contribute their solutions and where everyone can go to first when they need a new solution - to see if itās already been done and then borrow with thanks.
I agree with David - share first, then align. Going for national alignment from the first implementation adds a massive task, and we donāt all have the exact same systems. By sharing, we will get a good step towards alignment since weāll be working from the same base.
In response to your questions Jon:
My mindset / approach comes from the app with which I am most familiar - in this case, weād be producing an XML file that when imported into Orionās Clinical Portal becomes a form template. We arenāt procluded from adding other additional artifacts though - eg) images etc can also be stored in Git repos.
Standards would be handled by the NZHTS. Iām not sure what is meant by FHIR IGs (is that implementation guides?). But yes, they would be something that could be added to a national Git service as separate repositories.
A mix of different forms for different use cases - can be all of the above; however, PROMS would be captured potentially differently. Calculators could be built into the forms; and any calculations implemented would benefit by being available in a shared repository as it allows more eyes to review and validate that the calculation is correct.
Each form / solution added to the repository would need to have an implementation guide / assumptions of data availability. The idea would be that if I submit a form I have developed, then I include a blurb and instruction set detailing the requirements so that it can be implemented elsewhere around the country. There should be a key contact attached to each form as a person to refer back to.
This would provide us with a tool to share the clinical pathways by sharing the forms and configuration needed.
Iād argue that Smoking Cessation may not be the best place - around the country, the Districts using Patientrack have Smoking Cessation being recorded in there. It highlights that the ideal state would be for a central patient repo to define the required data fields, and for local implementations to then be able to feed the required data to the central patient repository. Over time, we can then reconcile everywhere into the same form & system (if needed) - however, until we have the same apps implemented everywhere, thatās going to be very tough to do.
I understand the desire to share the xml documents. If you request access there is already a national Te Whatu Ora Github account so you can request access and we can have a private repo there: Te Whatu Ora (github.com)
I think the tricky thing will be everyone knowing that what is shared in there is not already the ānationally agreedā version.
My thought was that we should start on something reasonably simple and ideally something where primary care also is interested (coz they do it differently) hence smoking. Ironically this is about to become important as we will want to use this information for the planned national lung cancer screening so there is incentive to get this done. We also already have a fhir repository available that can be connected to with json defined parameters: tewhatuora/fhir-fe-poc-pages: Github pages hosting of the fhir-fe-poc project developed for Te Whatu Ora
so we could define it - and use that as the repo which would leave people to build a form that integrates?
You overestimate our caring!! But it is a strategic choice.
Would it help to have an eHealth Forum GitHub repo - would sidestep it nicely and provide an obvious Wild West space.
But you really need to open up the official one too. And I need to open up bits of the Forum at some point (is a lot of work re privacy and protecting it adequately)
We can have a private repo - but then non-employees of Te Whatu Ora canāt see.
A public repo is fine but is subject to scrutiny of the public also I think we might not be clear on what our commercial partners would think of this.
can see this rapidly being better being the enemy of good, and likely why the vendors and others before us have failed.
I would also not limit us to sharing forms, or even care if we shared the form itself. Knowing who is doing what across the country, how others have approached and thought about a problem is just as valuable.
We can argue that the new operational model and ethos will remove this, but we know itās going to take time and sometimes, just sometimes - you have to be the change you want to see.
letās start small, see who wants to share what they are doing - but more importantly - have a place where people can just ask and see who answers
we can show by doing - and offer up a few things as local examples of how we have done things
I will endeavour to produce a list of forms/content that is in that private github repo so people can see what there is. There are probably some national ones that we should put in there from screening programmes etc. Will cast about.
Well, there is a more solid one here - but this is designed for a super deeper integration where a dedicated Forum actually powers maintaining a complex existing codebase:
This has the potential to power a living repository on a grand scale - where we might be in a few years if this concept really takes flight.
Pulling Back a Bit - what goes first?
Now, there is a fundamental flaw with a repository-first solution for this: Accessibility.
GitHub and other code repositories are totally familiar for devs. However, only a small number of particularly geeky clinical and IT support staff have dipped their toes into it. The learning curve is steep, and there is a whole ecosystem of language and conventions that is quite impenetrable for some time (bitter personal experience).
Therefore, 90% of the digital health community would probably be excluded (at least initially). That is not okay.
However, using a community-first approach (i.e. online forum) with links into repositories - that is doable and accessible. People would be able to jump in as deep as they need / want. And there doesnāt need to be a grand repository - the online forum can connect several personal / unofficial / official / private / public all together.
yeah I think the conversation part, forum for ideation, and an understanding of who has done what already would be most beneficial at this time.
ā¦I was chatting with Matt H recently about the work Vidhya and others had done in meds and how great it would be to see something similar in the referrals space.
If we could use pelvic mesh to drive not only a transitional state but also target conversationsā¦
Imagine a world of expressed core clinical guidance for referrals, consistent desired business process (with feedback loops, handle declines, notifications, escalations), business data standards, and interoperability patterns (the business view not technical - describing the patterns we want to enable).
Then we can fork the technology viewā¦ integration standards, tech enablers, etc.
Iām rambling but this is where my passion is atā¦
I think it would be great to have a github/discourse integration. Like you say, most people would just keep up to date with the forum side of things but the GitHub helps keep the technical stuff organised properly.
This is an area of interest of mine as well. I work in the tech space for HealthPathways where clinical guidance for referrals is one of our core pillars. One of the goals Iām working towards is a smoother flow from clinical guidance to actioning referrals so interoperability is a big focus.
Implementing a national integrated EMR would solve a fair chunk of this problem. In QLD we digitised over 2000 forms into the EMR and did away with the paper forms. As each site came onto the system there was a process to check their current forms against the digitised forms in the EMR to see if any additional forms where required or updates. By the end of the implementation the majority of forms had been digitised and standardised resulting in very few local forms being required in all the major areas such as ED, theatre, inpatients, outpatients, maternity, community, allied health etc.
Thatās similar to what a few hospitals have undertaken already here; I believe. Excluding Te Toka Tumai; we are lucky to have computers . It would be good to hear from people what theyāve learnt since that process occurred, and how would they do it if given the chance again. I feel Iāve heard the sentiment of āletās do more than just digitise what we already haveā.
have a personal GitHub profile!!!
. It would be good to hear from people what theyāve learnt since that process occurred, and how would they do it if given the chance again. I feel Iāve heard the sentiment of āletās do more than just digitise what we already haveā.