Rescue plan for Royal Free

21 Nov 2008

A 90-day rescue plan is underway at The Royal Free Hampstead NHS Trust to try and fix a catalogue of 22 major problems with the Cerner Millennium Care Records System installed by BT.

Problems caused by the system have so far cost the trust £7.2m, disrupted the delivery of care and led to delays in patients getting treatment, affecting its ability to meet national access targets.

Barts and the London NHS Trust has experienced similar problems, and E-Health Insider has been told that a similar recovery plan is underway at the trust. BT, the local service provider for London, announced in October that it would suspend all future CRS deployment plans.

In a November board paper, Andrew Wray, the Royal Free’s chief executive, says: “By early September it became apparent that the trust could not continue with the system as supplied.”

Having alerted BT Health, the Department of Health, the London Programme for IT, NHS London and the National Programme for IT in the NHS, a series of high-level visits to the trust took place.

This led to the development of a turn-around programme, titled “business as usual”, that was launched on 6 October. The rescue plan is based on a 90 day work programme to address each of the outstanding issues, many of which were supposedly fixed before the Royal Free took the Cerner system.

Work is being carried out by “a full system and support team on site, working on the Royal Free build of the software directly and thus changing the London Programme model of one build appropriate for all trusts.”

By moving to directly re-writing and reconfiguring the Royal Free’s version of Millennium, BT, LPfIT and NPfIT appear to given up on the concept of a single version of the software to be used across London, providing the basis for interoperability across the capital. The standardised solution approach is central to the way local service provider contracts were specified.

With all further deployments of Cerner Millennium on hold in the capital, all of BT Health’s and LPfIT’s attention now appears to be on trying to sort out the problems at the Royal Free and Barts and the London.

Convincing future trusts that they can take the system without lurching into crisis will be a challenge for the future. Hammersmith Hospital, St Mary’s Hospital and St George’s Healthcare NHS Trust were all meant to have taken the system by now.

Royal Free was the third trust in London to have taken the Cerner Millennium CRS from BT. But it was the first to receive the London Configuration 1 (LC1) version.

In the November board paper, Wray also says: “In line with other Cerner implementation sites issues have resulted, including problems associated with data entry, system processing, data management and reporting despite assurances pre-go-live to the contrary.”

“This resulted in a significant impact in relation to waiting list management and patient booking processes with activity for both outpatients and inpatients reducing during the months of July and August 2008.”

As well as the list of fixes required work would also be carried out on “enhancements” to the Millennium CRS system, “to address operational and performance issues raised by managers and clinicians”.

The report ends on an upbeat note: To date, progress appears to be good. The trusts implementation team are working with increasing confidence that the problems the RFH has encountered with the system can be overcome.”

Jon Hoeksma

Reader's Comments

"Rescue plan" ?

21 Nov 08 11:26

Sounds more like a disaster recovery programme than a rescue plan. For the time being sticking with "business as usual" would appear to be excellent advice for any Trust still considering implementing this system!

Who is in charge here?

21 Nov 08 13:01

>>Having alerted BT Health, the Department of Health, the London Programme for IT, NHS London and the National Programme for IT in the NHS, a series of high-level visits to the trust took place<<

Does anyone else think this restates the core problem with NPfIT rather than a step toward the solution?

The dialogue should be directly between expert users in the Trust and the software supplier!

is this a typewriter, a system, or a medical device? That is the question.

radonco@live.com

22 Nov 08 03:32

Jon,

You are due compliments for this thorough report. Is this culprit instrument not a medical device considering its vital function, the problems with which "disrupted the delivery of care and led to delays in patients getting treatment..."? Indeed, it meets the definition of a medical device (MHRA Bulletin No. 17, amended September 2008):

"The definition of a medical device is: any instrument, apparatus, appliance, material or other article, whether used alone or in combination, including the software necessary for its proper application intended by the manufacturer to be used for human beings for the purpose of: - diagnosis, prevention, monitoring, treatment or alleviation of disease,..."

Thus, any adverse human events from the disruption of medical care and delays in treatment brought about by this apparatus ought be reported to MHRA and MDA at its website URL mhra.gov.uk

One cannot imagine that the morbidity and mortality of patients so affected was zero, but then again, these are difficult outcomes to admit had happened from a care altering experiment.

Indeed, there does not appear to be much difference between a pacemaker with defective software that results in harm to patients and this care records device with software flawed with "22 major problems" in need of "enhancements...to address operational and performance issues raised by...clinicians", with one exception. A defective pacemaker has adverse impact on fewer patients.

Respectfully submitted,

Propensity

Contract Underperformance?

24 Nov 08 08:47

Evry Trust that has taken the Cerner Millennium solution (including Fujitsu) has had major problems with reporting.

This was a known issue, 18 months before the Royal Free went live.

As they were assured that all these matters had been resolved before going live -and they had not, does this not only constitute under performance of the contract as specified and agreed, signed off in writing etc but basic fraudulent behavior?

BT/Cerner delivered a product that they basically lied about; why Have I not seen any efforts whatsoever to re-coup the lost revenue from the solution providers?

No penalties; no resignations, no firings - has this project now gone so far down the realms of the sublime that the concerned players have come to expect this outcome regardless?

As one of the previous victim areas of the same stunt -we were not surprised in the slightest -just impressed at the savvy of the Royal Free in costing lost revenue so fast -our lot have still not done that, because....... the reporting tools do not work!!

Ta-daaaaa!

Who is missing

26 Nov 08 09:30

Interesting report, not a great surprise to hear though. One thing does error spring to mind. Royal Free are 4th Hospital in London to go live.

Everyone seems to focus on Barts and the London in the R0 deployment phase and overlooks Barnet and Chase farm and Queen Mary Sidcup!

Clarification

26 Nov 08 12:14

In response to the previous post, The Royal Free Hampstead NHS Trust is the third NHS trust in London to go live with Cerner Millennium, as provided by BT under the NHS National Programme for IT.

It is, however, the first to take the London Configuration 1 (LC1) solution, developed for use in London under NPfIT.

In total, four acute trusts in London now have Cerner, but Barnet and Chase Farm pre-dated the NPfIT programme, though it has been now been brought under the aegis of local service provider BT. This means BT can correctly claim it is responsibel for four Cerner deployments, three of which it has carried out.

Finally, and very pedantically, it could also be correctly said that BT has done four acute trust deployments in London, by double counting Queen Mary's Sidcup. So good they did it twice - first with IDX and then Cerner.

Jon Hoeksma, editor E-Health Insider.

Correction re BCF

26 Nov 08 14:21

Although BCF had Cerner previously, the R0 Millennium version was installed (by BT) as part of the National Programme and consitituted one of the three contracted R0 deliveries.

Correction

26 Nov 08 15:00

Jon,

Just to prove I'm an avid reader, I noticed the comment on the Royal Free was slightly off the mark.

BCF was deployed by BT as part of NPFIT in July 2007. Since then we have implemented Cerner at QMS, Barts and the London and the Royal Free - four in total.

We are also managing the Cerner systems at Newham and Homerton which as you know were installed outside the National Programme.

Hope this helps.

Regards

Avid EHI reader

homerton vs royal

26 Nov 08 18:21

Hi. I'm very new to all this, but I'm trawling and came across one report that stated how well Homerton had done on the roll out of their It. It's the same as the Royal free. Is it the programme that is basically flawed or the variences in implementation? Is it the fact that Homerton was done outside the NPfIT? Regards, Jo

Re: Homerton vs Royal

28 Nov 08 07:50

Jo,

You are abosultely correct the Homerton (like the Wirral before them) successfully deployed Cerner, without the clincal and operational issues that have plagued the CFH implentations.

I would have thought that the conclusion from this was blindingly obvious, but when I posted to this effect a couple of days ago, I was censored by Jon

Comment 2 The Missing Link

28 Nov 08 09:54

IMHO having worked with Accenture, Fujitsu and bumped into BT during the NPfIT Programme And having only seen one implementation work ... I think the poster of Comment 2 is correct. The work with SystmOne was heading for failure until the project teams in the East/NE Clusters managed to open dialogue with TPP directly despite the best endevours of the LSP. (the response from TPP was pleasingly constructive)

A lot of the issues with Cerner might have been resolved if a similar dialogue could have been facilitated. Reporting needs could have been negociated, some of the "wrong" questions (eg "Do you have any Child Protection concerns?" What idiot is going to say No to that??) rewriten or left out completely.

You never know RiO might benefit too

Cerner's not the issue...

28 Nov 08 10:11

Hi, being stuck in the mire, I can tell you the reason Homerton went so successfully, is because they bought it direct from Cerner. The reason BCF, RFH & QMS are so up the proverbial without a paddle is we have to fight the middle management layer of BT. It is BT & their mercenary attitude, and lack of care & understanding of the way the NHS works that is the major cause of the problems... Regardless of the problems inherent in the software itself, and the arguements as to whether the NHS should've been forced into buying an 'off the shelf' system designed for an American Health Care model. The main issue faced by Trusts is BT, and their lack of knowledge (about their product, and about the NHS), lack of professionalism, and outsourcing business model. Pity the Southern Cluster, if BT is imposed upon them as their LSP: you have been warned!

Cerner's not to blame

28 Nov 08 11:03

I agree with the previous comment...Cerner is a half decent product and Homerton University NHST proves that by not dealing with an LSP.

Unfotunately, IHMO it is all about the bottom line...how much did BT make by winning the London contract!?

It's a shame that it has come to this. It almost looks like BT have left Cerner in the dark as to how bad the problems really are at Royal Free and "dragged" them into this 90 day plan.

Kudos to Andrew Way for having the testicular fortitude to say that the system is not fit for purpose. However, we are weary that this is the tip of the iceberg unless someone at the top (DoH/CfH) makes changes sooner rather than later.

(post edited by EHI)

Expectations for a 'half decent solution'

28 Nov 08 14:21

Hi Going into the 21st century we should be expecting something more than a 'half decent solution' for an integrated patient care record. L.

Lets Damn it to death anyway ?

29 Nov 08 10:47

Why does everyone assume that QMS & BCF are up the creek like the RFH & BLT? Could it be that the unpalatable truth (for some over-exited individuals) is that they are getting on and using the system without too much fuss and bother.

As a PAS replacement the complex workflows and training overhead make Cerner a challenge. But once clinical functionality is layered into the mix its underlying value as a CRS becomes evident.

Blame culture

02 Dec 08 14:14

Ah, sigh and here we are again blaming the software or the LSP for the lack of deployment. Having worked for in the NHS and for both suppliers and LSPs in the South, London and the North West the biggest challenge is delivering the contract set out in NPfIT to the trust themselves. It is apparent that the requirements the suppliers build to are outdated and do not suit all trusts It is also apparent that the NHS would never be ready to accept the changing work processes within the timescales suggested. So one could say the blame is fairly spread. The suppliers do not always build to requirements, the LSP’s have tight timescales to adhere to and the NHS tardy That being said, however, there is no reason why the programme should fail if the NHS really wanted it to succeed. The principles are worthy enough, a central database (surely at least as secure as local paper stores or local IT services, despite the views of some groups) that can be accessed by all. The potential benefits are obvious to clinicians and administrators. Instead of supplier bashing a rethink as to what is required and how to deliver it within the principles of NPfIT by all parties might be better use of time. The NHS should then “sign up” and be held as accountable as everyone else if it fails.