Forget submarines, Defence dives deep on digital health

The Department of Defence has started work on a digital health system for our entire defence force that looks like it would make a pretty neat blueprint for the future of interoperability in Australian healthcare. 

Last week the Department of Defence (DoD) finally gave its blessing to letting the public officially know about a healthcare project that is very big: JP2060 phase 4 (they should probably come up with a sexier name), a $299 million health knowledge management system that, if developed as specified, should be a blueprint for how healthcare interoperability could one day work in Australia. 

It’s not as big as a fleet of nuclear subs, sure – $116 billion by the time one hits the water in 20 years or so, ouch – but in the context of Australian healthcare, and in particular the idea of an interoperable system that seamlessly shares data between all points of a healthcare system compass (tertiary, primary, allied, pharmacy, even dental), it is very big and interesting.  

A few key points around the project: 

• More than 95% of work will be delivered by Australian industry, including seven locally owned and operated medical software groups, with the centrepiece being a system originally developed as a ground-up cloud system for GPs in Australia: MediRecords. 
• The system will be both entirely cloud based, linking every point of care from primary to dental to allied and hospital, but able to operate offline functionally when needed (as you probably are going to need at some point in defence). 
• The solution being developed will help in the everyday management of ADF personnel within and without Australia – 85,000 personnel, including 58 primary care clinics – but will also be deployed at point of injury through the evacuation chain to rehabilitation and recovery. 
• The solution will record, store, aggregate and analyse health data for the entire ADF population, unifying primary and occupational care with emergency and hospital care data in a manner that Australian governments have only aspired to thus far as far as broader population health management is concerned 

Possibly the key outtake of the contract is that it is a firm pointer to the rest of Australian healthcare with reference to where the future of interoperability and care delivery is heading. 

Despite the political woes that come with the portfolio, of all government departments, the most forward planning and strategic thinking is the ADF. When they do technology, they will usually do the best in an attempt to future proof themselves. 

Why buy a diesel sub when the future strategic environment clearly points to nuclear being the required option, even if you have no nuclear capability? Why deploy a healthcare solution that isn’t interoperable in the manner that is optimised via FHIR, the cloud and fast evolving standards for seamlessly and securely sharing healthcare data in real time over the web?  

Same principle. 

When finished, the project will offer all of us an example of a near-perfect closed ecosystem sharing of data between key points of a healthcare system delivery compass: hospital, primary care, allied care, dental and pharmacy.  

Of course, the defence system won’t have to worry about the complications of funding and payments, state versus federal politics, or the need to stay connected to certain legacy systems for various reasons. But it will offer us a unique insight into what our entire system could be one day if we manage to get alignment of intent across state and federal health departments. 

But aren’t defence notorious for stuffing projects up? 

With such a forward-leaning project, might it suffer the same fate as the submarine project and get mired in complexity, politics and a lack of local technical capability? 

Interestingly for Australia, this is not likely, in this particular defence project, for a few reasons. 

The first is that the project is being run by US-based integrator Leidos, a group that is already the major (and successful) integrator for defence force healthcare systems in the US and, coming out of the US, has a lot more experience in an environment where cloud-based interoperability and the roll-out of key web- and cloud-based standards across healthcare, such as FHIR, is already common. 

Secondly, the local technology groups selected for the project are “tip of the spear”, and capable as far as cloud healthcare is concerned. I’m the executive director of one of them, so I’m reasonably close to what is going on. Leidos is a very serious global integrator with lots of serious people and it simply would not partner with companies that do not have the capability for a project it is managing.  

As some indication, on a recent trip to the US to present the project to key Leidos staff involved in US contracts, I saw that quite a bit of interest was expressed in whether aspects of the MediRecords technology stack could be transplanted for certain US-based projects being run by the group. 

The full list of local suppliers includes Alcidion Group, Ascention, Coviu Global, Fred IT Group, Nous Group, Philips Electronics Australia, Precision Medical and Titanium Solutions Australia.  

If this is the future, is the past now defined? 

The centrepiece of the project is the MediRecords cloud patient-management system, which will supply the primary care module, a patient admissions module and a patient portal.  

MediRecords was originally developed with a view to being the Xero of the healthcare market for patient-management systems in Australia, but when it was launched more than five years ago, it suffered significantly from having to backwards integrate from its modern cloud-sharing architecture to so many legacy server-based systems in order to meet the expectations of doctors.  

Today, both primary-care and specialist patient-management systems remain mired in the past of on-premise server-bound, to the point where government contracts will often specify integrations for patient-management systems that do not talk to the cloud properly.  

A recent tender for the supply of deidentified patient data to the government via PHNs and GP practices actually specified that only legacy PMS systems could be used. This is a great way to keep your healthcare system anchored in the past. 

Our current key patient-management systems include Best Practice and Medical Director for general practice and Genie for specialists. All these vendors have either developed a cloud version of their product (Medical Director and Genie) or are saying they are deep in development of their cloud versions (Best Practice). 

But without other key parts of the healthcare system moving to cloud at the same time, the practicality and cost for the patient-management vendors of moving their customers over isn’t there. There are just too many integrations these systems are still talking too – payments, patient bookings, secure messaging, referrals, data capture and download, et al – that are still using legacy technology. 

And that the major access points to all our doctors in Australia are legacy-bound, server-based systems means that anyone building new applications that require doctor or patient access are usually building their integrations in a legacy manner. Even though we think of the patient booking systems as modern and largely web based, their integrations to the patient-management vendors lock their technology in the past to some degree. 

Interestingly, both Medical Director and Best Practice were in consortiums that tendered for JP2060 and lost out to Leidos and MediRecords. That each lost out to the much less visible MediRecords outfit isn’t necessarily that MediRecords was the best cloud technology offering, but that the integrator is probably as important in selection as the partner vendors and their technology. But at the start Leidos, which was always frontrunner in the integration stakes given their US track record, probably had their pick of PMS vendors to partner with, and MediRecords was the vendor they ended up going with for the core of the project. 

Another indicator is probably that Alcidion, Coviu and Fred IT are clearly the leading local vendors for the technology roles they are playing in the project.  

After MediRecords, a key role in JP2060 will be played by Alcidion’s knowledge-management products, in tying together all the data being generated by the project. 

Seamless and meaningful data sharing are a key part of the project, and is a key issue for Australian healthcare departments, which currently rely on disjointed de-identified patient data feeds from hospitals, PHNs and other sources, all of which are dragging data out of systems that code the data differently from a clinical perspective.  

Twenty-seven of the country’s 31 PHNs have formed an alliance to manage data out of one modern service that they have developed out of Western Australia. However, from a population health perspective, this group still faces the huge issue of how they centralise and share this data, how they equalise data sourced from different PMS systems that code data differently, and whether the data, which is effectively hacked out of general practice patient-management systems, is really cohesive enough to create meaningful population models from. 

The new ADF system will be the first closed health ecosystem in Australia with all data coded via SNOMED, and shared seamlessly by every key health service in the system, and which will have an Alcidion AI technology overlay eventually applied to the data. 

Declaration of Interest 
Jeremy Knibbs is the publisher of The Medical Republic and a technology and business writer for the group. Among some other interests in healthcare start-ups, he is a non-executive director of MediRecords, which is mentioned in this article.