Behind the scenes of government contracts to build covid vaccination logistics and booking systems is a story of egregious waste and mistrust.
Let me show you how deep our digital health dysfunction rabbit hole goes.
When one of our largest and most successful healthcare booking engines, HotDoc, announced this week that it was moving into the hospital market, it confirmed the fourth major fail of a state government in building a suitable and robust vaccine booking and administration system in time to serve the vaccine rollout in their state.
If the federal government had not bungled vaccine supply in the manner it has, creating a situation where we have significantly extended timelines to where the bulk of vaccinations are still due to take place, the failure would probably have been far more apparent.
But whether any of the state governments end up getting their act together or not, the story of how the state and federal governments interacted to plan and contract an array of best in class global technology vendors and consultants to sort out their administration issues in delivering vaccines to the population will still stand as a significant example of government dysfunction and waste.
HotDoc expects their solution for the Austin hospital in Melbourne to be up and running within a week. It will include capacity to manage 1,000 appointments per day, a complete scheduling solution, eligibility screening and stock management for AZ and Pfizer.
A spokesperson for the group said that the solution would be fully transferrable for most hospital set-ups anywhere in the country and the group is already looking at delivering the solution to St Vincent’s hospital in NSW.
That HotDoc can deliver a solution within a week that meets most of the criteria of the tender by the Victorian state government for vaccine administration in one of its major hospitals, and that the tendered solution, awarded to a global provider of technology that is a household name, but is apparently still not ready, is now a common story now across most of the big states.
The themes of waste that run through the story (and most of the state-based tenders) are hard to reconcile, even in the panic-based atmosphere that COVID created for government.
Nearly every state saw fit to do their own separate tender for their own system despite the fact that the core of every system – a booking engine – was a requirement that was very similar between every tender and already available in Australia via several accomplished local technology vendors
Nearly every state included in their tender some spectacularly complex specifications, some for inventory management systems that have never been built anywhere in the world (for obvious reasons, as it turns out) and which should have been obviously over complicating the matter in context of the practical needs of vaccine administration and delivery in this country.
Every state and the federal government awarded contracts to major global technology vendors and consultants. Virtually no contracts were won by local vendors despite many of them having existing technologies that were already working across the country.
The federal government under the Department of Health (DoH) also sought to build its own national booking system and this was known to the states but the states didn’t trust the DoH and continued with their own tenders and builds.
So far only one state, South Australia, decided to use what the DoH eventually came up with, but only after their tender process had failed – probably because of how inordinately complex they had specified the build of their system.
The DoH had originally planned and tendered to do its own national system as well, also using a global vendor, but very late in the day it realised through talking to the local technology vendor community, which includes an array of very well established booking-engine providers already operating nationally with very well developed systems, that what they were planning would be impossible to implement in the timeframe they had given themselves.
The DoH defaulted very late in the day to relying on the existing network of commercial booking engines that serve the GP community and pharmacy already and tendering a national system out to local vendors. This system would service all those healthcare providers who did not already have a booking system and, presumably, any state that wanted to use the system instead of building their own separately.
The sum of all of this was that we had all our states and territories specifying and building their own vaccine administration systems for management by the state or territory once the federal government had provided them with supply, when it must have been apparent to many that 90 per cent or more of the requirements of any one of the states could be met by one system.
One default of such a mess was that there was never going to be a chance of either the federal government or any of the states actually being able to manage inventory, supply and bookings properly because after the federal government gave up and tendered out their system to an existing local vendor, none of the systems were capable of talking to another.
As a part of the problem, when the federal government defaulted to the existing GP and pharmacy booking-engine providers to conduct appointments for pharmacists and GPs, there was no provision for any of these systems to talk to either the Commonwealth Booking System (tendered by the DoH), or any of the state systems.
At some point of time around March it must have become apparent to the DoH that around the country we were building something like 20 new and disparate vaccine management systems, all via expensive new contracts with big global tech vendors, usually in partnership with a top tier consulting firm, all of which had, as their major core function, the functionality that was already being delivered by our local national commercial vendors.
As an example, the contract for the system in Western Australia was awarded to global customer relationship management vendor Salesforce, and consultant Deloitte. The contract was awarded for around $5 million.
When the time came to turn this system on, it wasn’t working well enough to use, so the WA government hastily rang around and asked local booking-engine vendor 1stGroup if they could help.
1stGroup had a solution up and running for the WA government within one week.
WA ran off this solution for the next few months until the global vendor solution was working.
According to a spokesperson for 1stGroup, their solution was developed off their existing tech stack within one week.
That 1stGroup could come up with something in one week is presumably because this group had been working on and developing systems for the local healthcare providers in the country for years and knew the ins and outs of the issues that would be faced.
The global vendor, although no doubt an accomplished operator in the right circumstances, was coming in cold to the problem. And it showed.
According to 1stGroup, their solution does everything the global vendor solution now does. Presumably WA had a contract to honour and so eventually returned to the global vendor solution (and paying that vendor for the contract).
The HotDoc announcement this week alerted us to the likely failure of Victoria’s global vendor solution – at least until now.
Why else would one of the major hospital networks in the state default to using a local vendor at the last minute (actually, way past the last minute)?
According to HotDoc, their solution, like 1stAvailables, will be up and running in one week. Again, presumably HotDoc is way in front of the curve of the global vendor, whose solution isn’t working, despite the Victorian tender specifying that they wanted the system early in the year.
The global vendor in the case of Victoria is Microsoft. This contract is worth $5.8 million.
Again, and like with Salesforce in WA, no one is saying Microsoft is a bad outfit for designing and implementing complex IT solutions. But clearly a large part of the solution being sought by this state government was sitting in front of their eyes in the form of what HotDoc does.
HotDoc employs more than 100 people in Victoria and is one of the most attractive and successful healthtech companies in the country. It wouldn’t have been a bad bet, even if it was just a backup bet.
We can guess why each state decided to do their own solution separately: they didn’t trust the Commonwealth to come up with something that worked and didn’t want to be in a situation where their constituents could blame them for something they hadn’t controlled. And they weren’t about to work with another state.
It’s classic “if it’s not built here” mentality combined with the significant dysfunction that often accompanies a federated system of government. But you’d think in this day and age, given the enormous waste and expense, that some of the state governments might have worked their way around the issue a bit better.
Apparently not.
But why were local solutions and vendors overlooked so comprehensively?
In part, this may have been because some of the state government tenders for their solutions were very poorly thought out. They combined in the same tender detailed and often wish-list requirements for a sophisticated inventory management system for tracking vaccine from source to every single point of distribution, with a vaccine booking and administration system.
The two requirements, although obviously related, are worlds apart and putting them together as core requirements pretty much ruled out every local vendor who already have the vaccine booking and administration component down pat, because the inventory management requirements were massively complex, possibly even uneconomic, to build. For obvious reasons, such inventory management systems for vaccinations didn’t exist anywhere in the world. Building them would be highly risky for a local vendor.
Ironically, given none of the state systems are working yet, or at the very least working to their original and fantastical specifications for vaccine inventory and logistics, such systems still don’t exist anywhere in the world. And not even the big global vendors being paid a lot have managed to crack the requirements put to them.
If you look at the detailed specification for the Victorian specification, two things stand out. Most of the specification is for vaccine and administration booking, the technology and capability for which already exists with most of our local vendors, and that part of the specification which concerns cold chain logistics and inventory management is reasonably outlandish. It is specified to a level that would never have been feasible.
To give you a taste, here are some of those specifications:
- capability to support cold chain management including but not limited to capturing data from temperature sensors throughout the supply chain and providing appropriate alerts.
- integration with third-party vendors / online ordering systems
- ability to scan and track receipt and consumption of vaccine details down to the vial level, including additional features to support safe and efficient use of multi-dose vials
None of these specifications were even vaguely achievable in the context of how the federal government had specified its vaccine roadmap in March. Nearly half of the work was going to be done by GPs, and they already had booking systems, and integrating every one of them to talk to a master state-based or even federal system was at all times fanciful after the roadmap came out.
Here is the whole specification if you’re interested. You’ll note most of it is to do with booking and administration – which the existing vendors all can do.
In NSW it appears that something very similar to WA and Victoria has unfolded. Without a working centralised state system, some LHDs have abandoned the state solution – at least for now – and gone with a local vendor to sort out their problems.
In the case of one major city-based LHD, the vendor and the LHD have indicated that the solution they have come up with could be used by most hospitals and mass-vaccination hubs in the country because it is a scalable cloud-based solution. So far it hasn’t been used outside of this LHD, however.
The theme of local vendors being able to develop solutions rapidly that are scalable, cloud based and applicable across most of the country is a common one. 1st Group, HotDoc, HealthEngine and the vendor in the NSW example above (Five Faces – ironically a Queensland-based vendor) all say they can have their solutions up and running in most locations around the country within a week.
Yet we have none of the major global vendor systems working in any of the states yet outside of WA, and we know that delivery was problematic, expensive and delivered late.
The Victorian and WA contracts were both worth around $5 million. If we amortise that across every state and territory, each of which went for a different solution to largely the same problem using expensive global vendors and consultants, it looks as though between the states and territories, various LHDs and the federal government, we may have spent something north of $50 million across the country on supply chain management and vaccine booking and administration systems, the majority of which are still not working.
If you consider that most of the state-based contracts were for what HotDoc, 1stGroup, HealthEngine and a few other clever local vendors have been doing for years now and have been able to deliver on demand on an interim basis within a week or so via cloud solutions when asked by desperate local vaccine providers, then you might easily imagine that the vaccine booking and administration needs of every state and territory for their hospitals and mass vaccination hubs could probably have been delivered by one of these local vendors across the nation for something under $5 million, and possibly less.
For safety’s sake you might have even awarded two contracts to two local vendors so you had back-up.
Notably, much of SA outside of GP practices has adopted the HealthEngine solution built for the Commonwealth (the Commonwealth Booking Platform CBP), which is free to anyone in the country to use, so long as you don’t already use one of the commercial booking systems. The CBP contract was awarded for something like $3.8 million. Presumably, if its working in SA, it probably could be working in every other state, and rather than spending more than $50 million we might have spent just $4 million instead.
Amid all these state contracts, the federal government contracted Accenture and Salesforce to deliver a master logistics and inventory management system for the whole country. We know now that such a system couldn’t work even it had been built because none of the state systems are working so it would have no data to gather on that front, and of the booking systems that do work, the GP- and pharmacy-based commercial systems aren’t integrated with any Commonwealth system.
If we didn’t already suspect that this major Commonwealth contract has been largely a waste of money to date, then the mayhem we witness most days, of name calling between the states and the Commonwealth, over vaccine supply and uncertainty surely suggests it has.
And if we think that maybe all this waste will somehow get sorted before our supplies of Pfizer vaccine arrive on our shores to save the day?
Even though some groups such as Austin Health in Victoria and selected LHDs around the country have gone off piste with local vendors to sort out something their state government had failed to provide them, and now have working systems, most of the state-based contracts with the large multi-nationals still have to be honoured.
So, if and when they do come online eventually, as occurred in WA with Salesforce and Deloitte, the local vendor will then be asked to stand aside.
There’s a lesson in here somewhere for our state and federal governments in how they collaborate (or not) to deliver a better experience for doctors and patients.
Without wasting lots of time and monolithic sums of money.
