The following article and letter appeared a few days ago in The Age.
Health 'myki' blows budget
DAVID ROOD
June 14, 2010
VICTORIAN hospitals have slammed the state government's trouble-plagued $323 million health technology system - dismissing its benefits as limited and accusing the government of putting hospitals at ''serious risk''.
A series of documents from the networks that run hospitals across Victoria reveal a litany of problems and dissatisfaction with the HealthSmart system, which is running four years late and $35 million over budget.
Health board minutes - seen by The Age - show hospitals being left to meet funding gaps for millions of dollars, with networks writing to Health Minister Daniel Andrews and his department to try to find money for ''hidden costs''.
The opposition and the Australian Medical Association seized on the state government's latest information technology failure, saying patients and doctors were losing out.
''This is the Victorian health system's myki,'' opposition health spokesman David Davis said.
HealthSmart aims to co-ordinate the different computer systems running in hospitals and bring in new programs such as electronic prescriptions to reduce medical errors.
But according to June 2009 documents from West Gippsland Health, the board decided to write to the Department of Human Services expressing its ''dissatisfaction with the system and raising their concerns that it poses a serious risk to the organisation''.
Western Health documents, also obtained by the opposition under freedom of information, show frustration at a $15 million shortfall in capital funding and recurring costs.
And in April last year, Bendigo Health warned of potentially insufficient money to fund the crucial clinical part of the information technology system.
More detail here:
And a letter the next day to the Age.
Nothing smart about this health system
DESCRIBING HealthSmart as ''myki'' is a misnomer (''Health 'myki' blows budget'', The Age, 14/6). Myki is an inconvenience compared with the dreadful implications of HealthSmart, which affects life and death decisions about patient care.
I have been trialling and analysing the HealthSmart hospital discharge summary electronic interface for general practitioners. A traditional and effective summary is one or two pages. HealthSmart generates an unmanageable 15 or 30-page report.
The formatting is amateurish and critical information about patient care can be buried 10 or 12 computer screens in. GP notations and actions are lost and unrecoverable because of incompatibility with software used by about 85 per cent of general practitioners.
It is already a shameful misappropriation of more than $300 million diverted from healthcare without pre-existing evidence of workability and effectiveness. This reflects poorly on the Department of Health, hospital network chief executives and boards.
Network bosses should be advocating for credible small-scale development with field trials to prove benefits and effectiveness. In other words, applying the same standards to themselves as they demand of healthcare professionals.
Dr Dennis Gration, Tecoma
On this basis I thought a little update was warranted. This, remember, is a seven year program which was to cost $360M.
The program has a good website with lots of information which is very good I have to say when compared with what we see from DoHA.
The site is here:
Of special interest to me was the apparently current timeline document.
This can be grabbed from here:
From all the documentation two things are pretty clear.
First the administrative and payroll side of the plan has got to its goals pretty convincingly –which is very good.
The Community Management Systems and the iPatient (from iSoft) patient administration systems have also gone well and implementation is largely complete.
The second and bad news seems to be that the clinical systems have been very considerable laggards in all this.
We now have the following:
POST-PROGRAM IMPLEMENTATIONS
Clinical Systems
Northern Health 2011
Western Health 2011
Royal Women’s 2011
Loddon-Mallee RHA 2011
Melbourne Health 2011
Southern Health 2011
PCMS
Eastern Health 2011
It seems what we have here is actually the list of applications that are really very late indeed and are so called ‘post-program implementations’ when clearly they were actually meant to be in program.
Over the years I have had a number of missives in unmarked envelopes from Victoria chatting on in horror about how the clinical systems are being done.
Two key points have been central inflexibility and excessive cost to the target organisations.
It seems there is still some way to go. I wonder will it be possible to do better on the following implementations?
David.
Postscript:
Just today, and after I had drafted the text above, a new article on the issue has appeared.
Computers could cause deaths, warn doctors
DAVID ROOD
June 22, 2010
THE Alfred hospital's computer system is so bad that its own doctors are warning it will inevitably lead to ''catastrophic, and perhaps fatal'' consequences for patients.
A scathing letter from medical staff to The Alfred's management, obtained by The Age, warns that the hospital's electronic medical record system is a ''disaster'', with surgeons forced to compete with nursing staff and anaesthetists to access computer terminals.
In a litany of complaints detailed by the hospital's senior medical staff association, doctors claimed they were unable to look at more than one patient record at a time, with some staff urging a return to paper records.
''A number of my colleagues have taken the opinion that the current arrangements and systems are compromising patient safety and that it will only be a matter of time before we see catastrophic and perhaps fatal outcomes arising directly from the issues,'' the June 18 letter from staff association chairman Howard Machlin stated.
The Australian Medical Association said the information technology problem was widespread, with some hospitals woeful at providing basic access to computers.
In the letter, Alfred staff also complained that the software system left doctors looking at a computer screen for information on ''why the patient is sitting in front of them rather than actually looking at, or talking to, the patient''.
Dr Machlin stated that the number of computer terminals and the speed of the system at the Alfred was inadequate, and a particular problem in operating theatres.
A great deal more here:
All this does is confirm just how inflexibility in a clinical program can lead to very sad and frustrating outcomes.
D.
And early last year we received a HealthSmart Briefing at which the Acting Director of OHIS said "We haven’t deployed the clinical system - but we hope to remedy that within a few months” and "We will implement clinicals with a state-wide footprint to 10 health agencies by the end of 2010”.
ReplyDelete15 months later and we are no further advanced, in fact we keep falling further and further behind and no-one seems to have any idea about how we can get out of the great big hole we have been dropped into.
What precentage of hospitals' budget is invested in IT - 5% maybe?
ReplyDeleteNo wonder surgeons are forced to compete with nursing staff and anaesthetists to access computer terminals.
Just remember David, the Healthsmart website is produced by Healthsmart, perhaps not so rosy on the ground.
ReplyDeleteAgreed, benefit of the doubt in play! Clinicals which I care about clearly not going well!
ReplyDeleteDavid.
Well they have given themselves a few handicaps in the Clinical space.
ReplyDeleteFor a start they have adopted a monolithic, one size fits all state wide footprint, and in order to have a chance of implementation on such a wide scale the system functionality has been limited to the extent that it will probably please few clinicians.
And from the outset thay have hobbled the program by introducing the use of iPM as a companion PAS, when Cerner already had that capability. This has added unnecessary complexity to the implementation and created a situation where ward staff will have to use two disparate systems to achieve normal ward tasks. It has also necessitated the creation of complex interface mapping tables, which are going to become error prone and expensive to maintain in the years to come.
To add to the fragmentation and complexity, (and more complex interfacing) Hospitals are looking to source supplementary system such as Emergency from other vendors.
Not a very integrated or adaptive approach and one which is unlikly to match the aspirations of innovative clinicians.
The AS-IS Cerner Millennium solution is working well in 1000s of client sites all over the world. Inefficiencies are expected when HealthSMART Clinical Systems is a customised/cut-down version of the Millennium. The average Clinical Systems budget of a 500 bed hospital in the US is about 30 million. Whereas the whole State of Victoria allocated 363 million. We are getting what we are paying for...
ReplyDeleteThe user above is incorrect in comparing the $363m HealthSMART budget against the 500 bed $30m in the US.
ReplyDeleteThe US comparison is the cost for a Clinical system component (Cerner Millennium). The HealthSMART figure is for a multitude of programs (see the HealthSMART website) including PAS, Financial systems and the cherry on the cake - the Clinical System.
The Victorian Clinical system cost for the whole state is therefore only a component piece of the $363m.
I don't know the exact figures but if you dig a little deeper you'll find the Victorian state is trying to get a statewide Clinical System for about the same price as a single 500 bed hospital in the US.
In this I completely agree - You do get what you pay for!!!
It must be noted that the Alfred IT article and the HealthSMART IT article are about issues being experienced by competing IT models.
ReplyDeleteThey can be compared in that they are both trying to address similar IT problems but we must make sure we do not link one with the other as they are very different.
The HealthSMART 'state' model is based around utilising common programs, architecture and configurations across all 'participating' sites (These can be found at the HS website).
All 'participating' hospital trusts are over time brought up to the same level of IT infrastructure across the board.
HealthSMART and it's successes/failures must be evaluated against what it is attempting to achieve, and the resources allocated to it.
I am lead to believe that historically the Alfred management chose not to participate with the HealthSMART program and as such are a model of a hospital trust going it on their own. As such they must be evaluated against what they have achieved/ failed against the resources allocated to them.
I believe HealthSMART and the Alfred do use some similar software however they would have separate contracts with Vendors, and different versions and configurations of the software so cannot be directly compared.
As such the two provide a good basis for comparison of a 'State' driven IT program vs. a hospital driven program and which provides the maximum benefits for the best price.
It would seem from the articles that both are currently failing end users.
I suggest they "both are currently failing end users" because they are both based on the same core system. And that raises the fundamental question of whether the Cerner system should have been adopted by the State's HealthSmart project in the first place.
ReplyDeleteIt had already be selected by the Alfred well prior to HealthSmart. The individual heavily involved in that selection at the Alfred then slid across to become Director of the state's health IT project (HealthSmart) and somehow her choice of Cerner for the Alfred was vindicated when it became the system of choice for Victoria. !!
I am confused as to why we are having so much trouble implementing the Cerner Clinical system in Victoria.
ReplyDeleteI have talked to multiple colleagues and associates who have seen the Cerner Millennium architecture in place outside of Victoria both overseas(US, UK) and also in NSW. All said the system worked well, and their opinions were generally positive about it.
I mean thinking about it, if it was such a bad system why would it be so globally popular and widespread.
I can appreciate country differences but NSW seems to be running the system OK. What is causing such problems in Victoria?
The Alfred also used an alternative PAS to that offered in the Cerner product. The same flawed architecture was replicated in healthsmart. Cerner may well have had a hope as an integrated solution, but now it is only a bit player in a very fragmented solution and it is difficult to see how it can now present a comprehensive, patient centric view of a patient's information.
ReplyDeletehealthSMART....a very interesting can of worms. The theory of standardising software between hospitals is certainly a good one, but the size and cost of the project was grossly under estimated. This has resulted in;
ReplyDelete-funding issues for the healthsmart project
-stupid statewide footprints to save money
-poor decision on the applications stack chosen as funding wasn't available to purchase the best in each category.
For some healthcare institutions which had no systems in place prior the healthsmart offerings were great. It was also a big win where large hospitals without a specific system in place received a new healthsmart system - i.e PACS system installed in a large hospital which did not have a proper PACS system prior. The issues have all been with existing large hospitals which admittedly had outdated systems requiring replacement and being forced to adopt healthsmart PAS/clinicals/PACS systems.
IMHO - With the amount of money spent all up so far on cerner and iPM - and what needs to be spent to bring everything up to speed, then all the support required to be paid down the track; i'm sure we could have setup our own software development team of highly experienced healthcare systems developers and designed and wrote our own PAS/clinicals applications of a much better calibre. We would have then had flexibility to make customisations and changes, as well as design the applications properly. What baffles me is if you look at PAS/clinicals applications they are not complicated pieces of software at all - yet vendors still manage to make a mess of it.
Guess we will have to all grin and bear it though.
Both ipm and cerner are pretty old systems, no real multi-tier architecture for proper flexibility in system delivery and maintenance (cerner may just qualify), no web capabilities, reliance on Citrix for distribution. Very old world.
ReplyDeleteAnd a one size fits all approach, which flies in the face of the problems to be addressed - i.e. clinical information workflows on the ward are very much a social construct and require a flexible system that can be quickly moulded to fulfil clinician requirements. Not easy under the healthsmart model