Thanks for that Nathan,
Can I ask the community - does anybody know of functioning example of a large scale (whole of NZ size) example of an electronic heath record held together via APIs? It seems that we’re on this path.
Of the functioning examples, how many data ‘sources’ can they keep held together?
I’ve not any specific knowledge with respect to the “functioning”, but it does appear Finland is a reasonable proxy for what is proposed via Hira:
Suggest you check out HealthOne, which uses the FHIR API
Cheers
steve
Thanks Ryan - it certainly does seem similar to what Hira aims to achieve. They seem to be doing it quite well!
Thanks Steve - I’m not sure HealthOne qualifies as an Electronic Health Record. While it utilises FHIR APIs to gather and collate information (I think), it is essentially a viewing portal. Thus it is a valuable component of an EHR, but nowhere near the whole shebang.
In terms of clinical workflow, one must visit the HealthOne application / site to gather information and then move to another application / site to actually do the work which requires that said information. Having the information integrated into the clinical workflow is my conceptual understanding of what a ‘functional EHR’ is all about, and what I think that @Mat is alluding to.
My experience in the UK showed me several examples of attempts at this type of EHR, both in monolithic systems (e.g. Epic, Cerner) and best-of-breeds. It was often disappointing to see how limited the data flow was between applications even within the same platform. A universal problem was the sheer amount of work required to set up a durable and effective integration between applications, especially as the number of applications grows.
Hi Mathew
I believe Recare has the product to deliver what you/we are after, with live data.
They are a private company who already have a digital national health record for about 80% of the population, and are ready to go.
Apparently the MOH won’t recognise this, due to this being owned by a private company.
There has historically been a desire for the Ministry to build one big single system and millions poured in to this option.
Today I’m told the best result comes from a network effect, as it is too hard to get every practitioner, lab, hospital etc to use a single system.
Many in Community Pharmacy are enthusiastic to aggregate using a third party tool such as ReCare.
It just makes sense not to reinvent the wheel, and the cost would be negligible in comparison to building it internally.
I see similar projects elsewhere, with public/private partnerships, so I struggle to see why this should not work in Health.
I’d be interested to see what everyone else thinks about Recare and their offering.
Ngā mihi
Shane
Thanks @NathanK,
There are two main reasons I posted this question.
The little I can see of the Finnish example required significant investment and (I think) preparedness to host data by their ministry. They also seem to have reduced complexity by having single ‘products’ which do one thing… ie there’s one national PACS, one national meds dispensed system and ?a bunch of legacy stuff which I was unclear about how or what they used that data for.
So yeah- not quite health one.
Anybody who has experience working in the Finish system- or any other system which is a large scale interconnected ecosystem similar to the proposed solution- please speak up with your experience!!
Cheers
Mat
I tend to agree with your points behind 2 - and agree what I think I heard from Hira was specifically excluding a “central” solution.
I appreciate the idea of patient stories as the guiding principle - and it should be further appreciated patient success stories are tied to clinician success stories. What looks like success for my patients is when they are at their greatest acute needs, and depend on our instant access to data.
For example, a typical stressful situation for a family and for treating clinicians in my scope would be:
Patient born in Tauranga -
Transferred to Starship for cardiac issues -
Multiple visits and procedures in first two years -
Teenager now in Wellington -
Vacationing with family in Christchurch -
… now in my Emergency Department.
I struggle to imagine how a distributed “we’re all friends” situation allows for a rapid, usable synthesis of all these sources of patient data, versus a central authoritative entity. If we can ensure this patient’s story enables clinicians to help it have an optimal outcome, I think we’re on the right track.
Plus security issues, as mentioned.
I confess to not knowing much about ReCare, and a Google search hasn’t revealed much except for this (hidden behind a paywall):
Can anyone from @pharmacists enlighten us?
Thanks Shane @Kensingtonpharmacy for the headsup around ReCare. From what I can see, this seems like a connection for pharmacy related products - i.e. things like meds prescribed + dispensed, alerts (I note the vaccination prompts) - is that right?
I was really hoping for examples of whole of system healthcare held together by APIs in a way similar to what we are proposing - e.g. everything from the blood pressure recordings of inpatients through to mental health outpatient clinic notes and well child records – i.e. something of the sort of breadth we seem to hope to achieve with HIRA, but which I have only heard of being doe via monolithic installations thusfar.
I’ll keep looking into the Finnish example…
Cheers to everybody for their replies!
Mat
@Kensingtonpharmacy and @NathanK - as I understand it, current use primarily re: medicines/pharmacy related systems (ala community pharmacy dispensing and GP data) but think possibility for further wider data linkages. Have seen some of the proposed benefits re; running queries on medicines-related population health intelligence and data analytics, appropriateness of meds use etc.
Happy new year everyone.
J
Hi Nathan/Mathew/Jerome
Nathan the product has it’s base in Primary Care, with GP’s and Community Clinical Pharmacy, but as Jerome mentioned it has much wider capabilities, including, as far as I can see, a live single patient record, with live medicine reconciliation.
For further, more comprehensive and accurate information the best person to consult would be Grant Bai.
My understanding is that Recare can meet the majority of the “Wishlist” of Primary and Secondary care clinicians, and if supported by MoH and the sector, would do so.
I’ll leave it to Grant to fill you in on all the applications and future possibilities of Recare in NZ.
Wow! That would be a jackpot from my perspective! Reconciliation is core to reducing medication errors and has been well researcher-ed and described. However, I am unaware of a product that can reconcile and then update two separate data-sets (e.g., primary care EHR, and pharmacy EHR) . . this would be the ultimate push-pull interoperability . . . amazing, if possible.
If that is possible, then classification/problem list, patient goals (unstructured!), and, medicine alerts, would be other vital components to be able to reconcile over time.
In my GP world, the closest example we have is ‘GP2GP’ which works best when transferring an EHR between two places using the same software (e.g., MedTech). It’s got great ‘push’ functionality, where imported data is allocated to the right part of the EHR (e.g., unstructured daily consultation notes, meds, classifications, vitals, etc), and the person importing can then do manual reconciliation . . . it’s still not perfect, but far better than duplicating data-entry from a scanned 50 page PDF which used to be the case!!!
We are progressing, I’m confident 
As @NathanK highlighted, HealthOne is not an EHR, nor is our Midland Clinical Portal . . . both are what the Americans would call ‘Health Information Exchanges (HIEs)’ . . these are great and awesome steps for information sharing, but the ‘exchange’ is a misnomer as it’s read-only with ‘pull’ but no ‘push’ interoperability. On the ground, they are limited by clinicians having to click around and forgetting passwords.
Hi Nathan.
At time of writing HealthOne is a HIE
(Definition (Electronic health information exchange (HIE) allows doctors, nurses, pharmacists, other health care providers and patients to appropriately access and securely share a patient’s vital medical information electronically—improving the speed, quality, safety and cost of patient care)
Indeed H1 is on FHIR with a FHIR store for data (especially with the newer PMS’s like Evo and Indici the API’s are FHIR)
I find workflow is a difficult thing to define as with the existing systems there are many workflows supported by many different clinicians.
One of the issues with PMS migrations is that one HCP’s workflow is a joy when a PMS supports it but a curse to another HCP who may have a completely different workflow which is less supported.
After tuning my practice to Medech workflow for the best part of 40years the dispensing with a consultation slip and making the HCP invoice via the software is still a challenge for me after a year of part time GP Indici experience. Others will like it I am sure.
Electronic lab results that are not in printed format came a cropper initially when introduced at CDHB yrs ago because a printed lab result was used in all sorts of workflows by HCP’s
If one simply defined a single workflow in a new build PMS then all HCP’s would have to be retrained in that workflow and with an average age of GPs being 52yrs (52–63yrs comprising 38 percent of all GP’s) I can see that being a problem. One GP in five has a work expectation of 4yrs odd.
Having had experience in at least 2 journeys to find the “perfect” PMS I kind of wished I had dodged those bullets.
So is H1 a PMS? Certainly it has encounters, prescribing, dispensing, classifications, allergies and health measurements and may have a future to add those data directly so for now it is a HIE but for the future maybe the jury is out.
HealthOne certainly makes whole of system work in the South Island a whole lot safer, more efficient, more effective and independent of location.
If you’re looking for examples of truly distributed EHRs - i.e. whereby data is acquired in real time from various source systems - then there are very few, if any, genuine examples to be found. NZ has certainly made the first step with the NHI being the ‘source of truth’ for some (but not all) patient demographic data. Hopefully, this will be extended to NZePS (for all prescribing and dispensing data), NIS for all immunisation data, test data CDRs for Pathology results, etc., etc. This is certainly the approach proposed for Hira - however, this is not merely a technical challenge, but also requires a cultural change whereby our EHR systems no longer function as independent silos that maintain copies of data originating from other systems. One of the many lessons from over 10 years of GP2GP is how quickly data loses fidelity once it’s passed between different systems; a recent example is COVID-19 immunisation records which look very different in GP2GP transfers from what’s originally passed from the COVID Immunisation Register.
@emily.gill @NathanK
HealthOne (HCS) is in SOME cases a ‘view’ of the patint record - but, when necessary, provides (roles based security) single-sign-on to the applications sitting behind. This is a charistic of ‘portals’ in that they provide the appearance of a single system from a multitude of BOB systems.
Data for H1 / HCS could equally be provide from something like Orion’s Health’s Amadeus (HIE implemented in several countries).
Hey Folks
Adding a couple of new comments here given the feedback and some of the research I have done.
“All in all, the information systems for patients are underdeveloped
in terms of their ability to monitor quality and performance, and they are
scattered across different operators”
https://apps.who.int/iris/handle/10665/327538
So - kinda good, kinda not.
Maybe we aren’t doing too bad here, eh? It certainly highlights that the risk of failure is high, and that it is wise to go for incremental (but steady) change without hard deadlines.
Or a completely different approach, such as Scotland’s. I’m not entirely sure how well that is really going though, and the new update to it seems to moderate what was once a radical strategy (seperation of applications from data in a single platform adhering to open clinical data standards)
HealthOne uses an approach where it provides a single view of the patient EHR. It does utilise other specilist application in some case. In those cases, it utilises single signon (no extra logins) and is in patient contect (where appropriate). Upon entering data into that external system, the entered data is almost immediatly availalbe in HealthOne. When I left Canterbury, we had some two hundred forms that allowed for data to be directly input into HCS/HealthOne.
If you want a true electronic data record, for use with analytics - the HIE model is well established. OrionHealth’s Amadeus has been sucessfully implemented in many locations - specially in the USA and Middle East. HCS/HealthOne/MCS can sit directly on that Amadeus platform,
I Just cant fathom that a piece of software that is clinician designed, based on a strong privacy framework, and has one click multi PMS access in patient context from GP, and is part of Orion’s hospital concerto, that is used over 200,000 times a month in the SI isn’t what the North Island (or NZ if you care to think big) wants.
It just defeats me 
There must be something wrong. For goodness sake it will enable whole of system care during omicron.
yep I know I am biased but heavens to Betsy is anyone listening!!!
Hi All
Hope everyone is keeping well.
From my understanding Hira is not trying to be an EHR. It is trying to provide federated (distributed) data services, identity management (who am I) and consent services (I allow you to do this) for trusted parties to consume the data and to create a virtual patient record as needed. It is about a tiered architecture that seperates data and function (function being what we get from an EMR product). The thinking is you can allow local innovation and an apps economy if you get the data and exchange right.
A company in Slovenia blogs a lot of about this concept and coins the term ‘Postmodern EMR’. Here is a link- note that the company’s product is based on OpenEHR standards which we aren’t trying to do in Hira.
I will find some other references and post shortly.
Thanks
Anna-Marie