Victoria dumps HealthSMART e-health project


news The Victorian State Government has reportedly decided to walk away from its troubled central electronic health project HealthSMART, which has reached only a limited number of its goals over the past decade since it was initiated, despite soaking up several hundred million dollars worth of government funding.

The HealthSMART project was initiated under the Bracks Labor government back in 2003 with a pricetag of $323 million and a due date of 2007. In July 2010 it was reported that it had since had another $37 million pumped into it, with the due date gradually extending, and in November last year the Victorian Ombudsman reported that the project would cost another $243 million to complete.

The initiative was slated to replace the complex patchwork of e-health records systems used across Victoria’s public health sector with a series of more modern and standardised clinical IT applications, with major contracts having been signed with e-health vendors iSOFT (now part of CSC) and TrakHealth. However, as early as 2008, the Victorian Auditor-General had expressed significant concerns about whether HealthSMART could deliver as a project.

“The original HealthSMART budget, involving health agency co-funding capacity, was not realistic. Lack of certainty across health agencies about costs and funding sources have inevitably led to delays in implementation,” Victorian Auditor-General Des Pearson wrote in a media release in April 2008, following an audit of the project.

Pearson also found that targets for implementation were too ambitious. “Had there been more realistic estimates of the capability of the sector to implement technological change in a compressed period and a better appreciation of the poor state of information technology assets in health services, the Department of Human Services would have more effectively managed expectations around the timing of the roll-out of the strategy,” he said. At the time, Pearson said: “Despite these issues, HealthSMART still has the potential to fulfil the original vision of a patient-centric model of healthcare, supporting public sector health clinicians with knowledge and technology. However, to date, that vision has yet to be fully realised.”

Late last week a number of media outlets quoted Victorian Health Minister David Davis as stating that the State Government had now scrapped the project. The ABC quoted David as stating that it would now be up to individual hospitals to determine what their IT needs were in future. “We will no longer mandate particular health ICT projects,” he reportedly said. “There will be targeted approaches and we will work with health services in the forthcoming period.”

Similar e-health records problems have also been experienced in other states, with few government health departments in Australia having successfully tackling the issue so far. However, despite the problems, the Victorian branch of the Australian Medical Association, which represents doctors, last year called for the State Government to continue to invest in e-health initiatives. “Improved ICT will not solve all the problems in our health system, but these problems cannot be solved without improved ICT,” the AMA said at the time.

The AMA wanted the funding spent on an Apple iPad tablet for every doctor, which could display electronic drug charts, medication management systems and patient records, as well as funding specifically allocated to roll out medication management systems, build better interfaces between hospitals, general practitioners and aged care providers, and build wireless support in hospitals.
In addition, the group advised steady, recurrent funding could be spent on up to date computers for use by medical staff, remedying a lack of standardised software between hospital networks and providing for the replacement of “sub-standard hardware and software systems”. “To build these missing links in Victoria’s ICT systems, we need to ensure that there is adequate ongoing investment,” the AMA wrote. “AMA Victoria recommends recurrent funding of $60 million per annum to ensure an adequate level of ongoing investment.”

The news comes as the Victorian Government continues to suffer wide-ranging problems relating to ICT governance. In November last year, Victoria’s Ombudsman handed down one of the most damning assessments of public sector IT project governance in Australia’s history, noting total cost over-runs of $1.44 billion, extensive delays and a general failure to actually deliver on stated aims in 10 major IT projects carried out by the state over the past half-decade.

Firstly, let me say that it has been expected for some time that the Victorian Government would walk away from HealthSMART. With a decade and hundreds of millions of dollars to deliver the project and only a handful of implementations to show for it, this is truly a project which needs to be canned or at least substantially modified, and I think most people will applaud the move. I’ve been getting inside leaks and reading audit reports on HealthSMART for half a decade now and I’ve never been of the belief that it was really getting anywhere.

However, setting that aside, I think the following phrase, published by the ABC last week, should strike horror into the hearts of every public sector IT worker in Victoria (quoting Health Minister David Davis):

“He says it will be up to individual hospitals to work out what suits their own computer needs in future.”

It is clear that HealthSMART’s centralised approach was the wrong one. But let me say this in the strongest possible terms: An approach which simply lets every hospital decide how to run its own IT systems is also the wrong one, and represents the opposite end of two bad extremes.

If every Victorian hospital is literally left, over the next half-decade and beyond, completely to its own devices, with no oversight, to implement their own IT systems and upgrades, what will result is a colossal spaghetti mess of gargantuan proportions which will prove a nightmare to untangle when the state eventually seeks (again) to standardise its e-health platforms statewide.

In an organisation as large as Victoria’s Department of Health, you simply cannot leave individual units such as hospitals to their own devices when it comes to IT. This will result in long-term disaster. Hospitals will be unable to share records between each other’s systems. Every time a medical professional of any stripe joins a new hospital, they will need to be trained in new systems — different from their old workplace. Standardised reporting of the whole sector to the upper reaches of government will become a bad joke. And as various hospital-specific IT initiatives fail, there will be no overarching strategy and resources to fall back on.

I say again: It turned out that a completely centralised e-health strategy was not the answer to health IT problems in Victoria’s health public sector. But neither is stepping away from those problems entirely and letting each hospital solve their own problems. That will result in an even greater tragedy down the track — mark my words.

I think much of the difficulty in e-health at the moment is related to the relative immaturity of the technology platforms available in this area, coupled with a similar immaturity when it comes to the integration of those platforms in clinical environments.

Hospitals are complex environments with complicated, highly-specific business processes, and the feedback I have received repeatedly over time from many in the health IT sector is that often the available software platforms (iSOFT, TrakHEALTH, Cerner and so on, to name a few) are not quite mature and flexible enough yet for the task at hand. And when you try and implement these platforms in hospital environments, the medical workers are often highly reluctant to change their work practices to accommodate the new systems. So you often get a worse of both worlds scenario — where the technology can’t quite adapt right to the needs of its users, and where the users themselves won’t quite adapt right to the needs of the system.

These are problems which every government health jurisdiction in Australia is grappling with, right now — and, no doubt, many private sector health organisations. And these are tough problems. But walking away from them entirely, as Victoria’s State Government appears to be doing, is simply not the right answer. That represents an abrogation of responsibility which Victorians will pay a high price for in the long term. I would say I’m glad that I don’t live in Victoria — but then, NSW isn’t doing much better in this area.

Image credit: Cathy Kaplan, royalty free


  1. “but then, NSW isn’t doing much better in this area.”

    +1 Our public hospital here is one of the worst in Australia and it’s clear when you’re there in emergency, half the delay stems from the IT systems not operating as they should or giving them the wrong information.

    I truly believe the NBN will be go a long way to alleviating this problem of eHealth implementation- Is it possible that centralised eHealth won’t work in Vic, not only because the system are immature as yet, but also that the systems can’t cope with the transfer speeds? A virtualised environment transported by the unconstrained speeds of the NBN may go some considerable way towards improving these still fledgling system IMHO. (VERY humble opinion- I don’t work in Health OR IT, but I have had alot of exposure to and experience with virtualised IT environments)

    • “I truly believe the NBN will be go a long way to alleviating this problem of eHealth implementation”

      In my opinion the NBN will have little to no impact on this type of implementation.

      • Fair enough. Like I said, only my uninformed opinion.

        No question something needs to be done though. It’s ridiculous that the Health system, one of our most critical (and lifesaving- literally) services is constantly at the mercy of poorly implemented technology when we’re always promised technology will save lives and increase efficiency.

  2. I wonder what they are going to do with what they have setup? There are 3 datacentres built for Disaster Recovery, and one of them was in the middle of nowhere – so far away that, unless you used Telstra Fibre, 20km of trenching was required. There are thousands of server machines setup – I hope they will be re-deployed.

    I only has a passing involvement, potentially offering carrier services to connect the sites to the CBD. Based on how long that process took, I’m guessing its the slowness of bureaucracy that has stopped this project, and increased the costs.

    To Seven_Tech – this was a “cloud” setup – where the hospitals connect to datacentres off-site. All the hospitals already have fibre, or access to fibre – just like Victorian Schools. This is why the NBN will have very very little impact on health services. I’d even suggest, for a corporate SLA grade service that the current costs the Victorian Schools pay for fibre is much, much less than under the NBN.

    Too bad for those telco carriers that signed 12 month contracts with them, based on 3 year amortized costs. They’ll be hurting a bit – a few hundred grand down the drain for fibre builds.

    • Jason- Fair enough. As I said, my involvement and opinion base was purely from the little I’ve read of eHealth in general. Which is to say not very much and also little in VIC.

      I’m glad they have fibre already and are trying the cloud solution- this seems increasingly the way to go in large concurrently accessed systems. It seems amazing to me though that we can have such vast business cloud systems that work quite well (not perfectly by any means) and yet our public hospitals are languishing with systems that don’t meet their basic needs. Very unfortunate.

      Sounds like, if that system Dan mentions in the US worked as well as they said, we should look immediately into that. Problem here of course, as was shown with Kev’s try with eHealth, state politics ruins any hope of cooperation.

      And people wonder why I hate politics….

  3. Holy crap that is a lot of money wasted, but no where near the estimated AU$11bn the NHS wated in the UK delivering, about as much.

    I wish I could find the doco I saw on Discovery about a hospital in the USA that bascially just employed a bunch of coders and re-wrote the entire hospital’s ICT systems from scratch to deliver a totally integrated e-health network. It cost something like $80m to implement, including a bucket load of new hardware, and delivered 100% of the project goals in under 3 years.

    • “It cost something like $80m to implement, including a bucket load of new hardware, and delivered 100% of the project goals in under 3 years.”

      Sounds like a pipe dream in terms of Australian tech projects ;)

  4. I know I’m naive and certainly not technical. But it seems to me that the answer is to not go too big too soon.
    Is it possible to choose one only smallish/medium hospital and write the software to do all of the functions required. Stick the results on the remote database. Once that system is fully functioning, tested and implemented, go for a second one and set them both up to amalgamate records and be able to view the data. Here’s the really naive bit. Once you have two done, isn’t it just a matter of scaling the system up to do more?

  5. Whilst there is no surprise in the failure of HealthSmart I agree that we cannot go to a fully decentralized system. Australia is going through the learning curve of how to implement health IT, and this is never an easy activity. However the patience to complete it should lead to better more efficient healthcare.

  6. Hmmm … I’m in London this week … where they know all about how to screw up big ICT projects in healthcare.

    The whole sorry saga is deeply troubling from the perspective of how to more effectively and sustainably balance autonomy for individual hospitals/groups of hospitals with the need for system-wide information sharing.

    We seem to persist with bull-in-china-shop swings of the pendulum from one extreme to the other …”one-size-fit-all” for ten years (failure!) then “let a 1000 flowers bloom” for 10 years (chaos!). Governments just seem to be the wrong mechanism for dealing with this as they just can’t help themselves from being attracted to overly-simplistic, short term, solutions to complex problems … and then anyway incompetently delivering on both centralised services and effective coordination of decentralised activity because they can’t focus on anything difficult for more than a few years … Gasp!

    The big consulting firms and thier “laptop goggles” have a lot to answer for as perveyors of overly simplistic and ultimately impractical centralisation strategies. “The medice is good but the patient keeps dying” … really?

    On balance, I think that empowering autonomy at the hospital/cluster level is a more resilient and adaptive approach – particularly with the growing emergence of cloud computing services which can deliver massive economies of scale and effective shared services via hospitals making individual choices based on standards and evolving critical mass of solution adoption. Socialism has turned out to be a failed exeriment in both economics and in whole-of-government ICT. Give me a capitalist market any day … and the skills to know how to buy well and safely.

    There is no ‘silver bullet’ … but we have learned some things about the importance of the hypocratic oath, “First do no harm”. Massive centralised ICT projects in government services, unfortunately, do a great deal of harm because they frustrate the local autonomy and ownership which is the basis of innovation and motivation in the public sector … and because they are usually ‘beyond the wit’ of man or woman in terms of their complexity … and hence fail.

    Somehow we need to learn from all of this and embrace an approach to ICT which empowers individual agencies to do what they think is effective and efficient within a framework of sensible and pragmatic standards and shared experiences to minimise waste and promote synergies. This is the essence of the CIO role … it is not about ‘power’ and ‘mandate’ … it is about ‘leadership’ and ‘facilitation’.

  7. Inevitable. We’ve been living with these eHealth “revolutions” (domestic and abroad) for a decade now and the debate has always been how much will be blown by the governments involved and how long it will be before someone finally stands up and pulls the plug. There is also a common thread to each one: they are biased towards larger enterprises (usually foreign owned) at the exclusion of local vendors (often with best of breed products). The local pollies and public sector IT hierachy have been there just to make absolutely sure this is the selection process involved..

    And for the record – this change to allow individual hospitals/groups to do their own thing is actually a real bonus for our operations if truly applied. So – way to go new Victorian government!

  8. “The AMA wanted the funding spent on an Apple iPad tablet for every doctor”

    When customer demand brand names rather than functionality your in trouble from the start.

    • Glenn,

      when Australians say “tablet”, they mean “iPad”. The iPad has a market share of something like 90% in Australia, it is simply the best product on the market, and I would consider it absurd for the medical profession to request anything else.


      • Whilst the iPad is a great data viewing device, I have to say it does have limitations as a data entry tool in healthcare. A couple of years ago we did some work with a number of hospitals on a design known as the Mobile Clinical Assistant, basically a Windows tablet specifically designed for healthcare use. At that stage battery life was an issue, but it had a lot of extra functionality that made it more useful for use with hospital EMR applications.
        The world has moved on, the iPad changed the world of tablets, but I suspect the limitations of the iPad still exist in the hospital space.

  9. Coming from a pharmacy background, having seen some of the old systems in action, I’d say it’d be hard to change mainly because each specialty is so individual. In hospitals the legacy systems in use are fairly specific to each department; pharmacy for example tends to use an old linux based CLI application to deal with dispensing and PBS claims submissions, yet many hospitals have different “tweaks” to the program from being able to do certain reports to cosmetic changes such as different shortcut codes. And then there the other hospitals who decided to jump over to another system altogether and manually report out from their own closed system.

    I’d like to suggest that hospitals come together and ask each of the profession representative bodies (such as the Society of Hospital Pharmacists of Australia or the AMA for example) to come up with a unified set of data standards. This comes from seeing the unified medication charts that have been devised; if these professions can settle on a unified format on what information they want and where they want, maybe we could then build a bunch of interfaces for a unified back end data management system. IMO eHealth hasn’t taken off because there hasn’t been any set baselines on what information is expected by professionals. If the programmers knew what each profession wanted, the back end data shuffle would be a lot easier; the user interface could be customised per institution, which would partially solve the workflow issues as many health professionals dislike newfangled systems and bemoan the inability to do day to day tasks because the procedure has changed (or whatever reason they wish), whilst allowing some standardisation and migration of data into a unified system.

    TL:DR analogy: Each profession has a “hammer” to assist in day to day tasks; some hospitals have “hammers” that are larger, others have “hammers” that have multiple heads and others have multiple “hammers” for different situations. What eHealth has been trying to do is to make a ‘universal’ hammer that does everything that existing hammers can do. The people making the new “universal hammer” are having trouble making this new “hammer” because everyone wants and/or expects something different from their “hammer(s)”.

    Of course this is all personal opinion of someone who had been inside working with some of these systems before.

Comments are closed.