Bucks had CRS infection control issues for 18 months

14 Aug 2008

Buckinghamshire Hospitals NHS Trust was forced to develop workarounds for isolating MRSA patients after Cerner’s Millennium care records system was unable to cope with its superbug monitoring system for 18 months.

In minutes obtained by E-Health Insider under the Freedom of Information Act, the trust’s National Programme for IT in the NHS programme board says infection control flagging was a problem throughout 2007.

“The introduction of the care records system on the Wycombe and Amersham sites has continued to present some challenges for the infection control team.

“These have mainly focused around the inability to remove any MRSA flags entered in error, the disappearing of the flag on the bed board when a patient moves wards/bed spaces and the intermittent omission of a yellow star against flagged patients in PowerChart.

“Work has been undertaken across the domain with the help of the strategic health authority to agree a solution. A fix has been developed and implementation is planned for early April 2008,” the board reported.

In a statement, the trust told EHI: “The fix went in on 9 April. The benefit of this is that the problems we had had prior to this were resolved.”

The problem was spotted a few weeks after the system went live in September 2006. In October 2007, EHI revealed that the problems caused delays in potentially infectious patients being isolated.

Following the coverage, the board reported: “The historical MRSA infection alerts needed to be put onto the CRS system manually following ‘go live’.

“This took approximately six weeks to do, during which time there was a possibility that some MRSA positive patients may have slipped through undetected if medical notes were not available. There have been several other problems with the system which have resulted in a small number of patients not being isolated promptly.”

The trust was forced to resort to manual methods to ensure infection control assessments were accurately recorded and that those who needed to be kept in isolation were kept apart from other patients.

A fix was initially promised for the end of November. However, this date slipped so the developer could provide a fix suitable for the whole NHS; an approach the Buckinghamshire board did not appreciate.

“It is apparent that Mid and South Bucks and Nuffield Orthopaedic Centre use the MRSA flag to show historical and current cases. The fix that Fujitsu will bring in will means that workflow processes will need to change for these sites.

“Milton Keynes work differently, and want the fix. We will have problems with downstream reporting if the fix is applied,” the board noted.

The infection control flags in Millennium are only for MRSA, Hepatitis A and Hepatitis B. Buckinghamshire also wanted a flag for C.difficile, which they said “was of vital importance for the trust, and we would not be prepared to roll out to Stoke Mandeville Hospital without the ability to monitor this infection.”

However, the board paper reveals that NHS Connecting for Health refused to fund this enhancement. “A C.difficile enhancement has been turned down by CfH due to associated costs,” the board noted.

As well as the lack of infection control, the board expressed concerns about many other aspects of the system. These included an encounter slice tool not working correctly, impacting the 18 week wait; the NHS number being missing from letters; missing patients; patient lists not pulling properly; rescheduling proving difficult and maternity proving unfit for purpose.

As EHI reported recently, Milton Keynes healthcare community said its deployment “developed into an untenable situation which resulted in near melt down of the [acute] organisation.”

Buckinghamshire says the focus on its neighbours meant it received less attention for fixes to their CRS, which impacted the go-live date at Stoke Mandeville.

The hospital eventually went live with Millennium at the end of March 2008, some 18 months after the Wycombe and Amersham sites, following successful testing of the infection control fix.

Joe Fernandez

Reader's Comments

first learn to walk....

14 Aug 08 08:22

This exactly describes the problem. These systems were bought and specified by people who have never run a hospital and dont know whats involved. Its not good blaming the suppliers, they have produced what they were contracted to produce, and what they were contracted to produce.

Unfortunately, the contract is all about smartcards, spine connectivity and interoperable care records. These are great things to have (probably), but the first thing a hopsital system needs to do is to support the buisness of care to patients, and sadly CFH just assumed that any old hospital system would do that. I know, i was there when people were saying things like (and i quote) 'a hopstial is a hospital, right', and sadly i dont see SHAs being any better

Can central systems be made to work? Answer: Yes. But only when the clinical management stucture of the hosptials - the people who make sure patients get treated - own the contract and have a bottom up governance structure that works.

What is the problem with adding bugs to tracking systems?

14 Aug 08 20:31

What is the problem with adding bugs needing tracking? Not being in a hospital, I would have thought the requirements were to be able to attach a label to an individual patient in such a way that it would be displayed in all suitable situations - such as bed management as well as in the EPR - and to be linked to appropriate results and alerts. Why wasn't it possible just to add a new bug - clostridium difficile - without having to write a whole new problem - and what will happen when we get bird flu?

One of many issues

15 Aug 08 08:25

Unfortunately this is one of many, many issues with Millennium, so named because it preceded the new century -thus showing its age.

The software simply does not cope with multiple sites and the Uk was the first implementation of such (source: Fujitsu manager) and the biggest implementation they ever undertook. With Fujitsu acting as the inflexible front-end and Cerner never in contact with the clients, to the point that they never even surveyed the issues pertaining to each hospital, the system ended up with so many 'work arounds' or bodges in plain English that the cost of putting them all right was the key failure in the contract re-negotiation.

One of many issues

15 Aug 08 08:30

Unfortunately this is one of many, many issues with Millennium, so named because it preceded the new century -thus showing its age.

In my experienced the software simply does not appear able to cope with multiple sites and the Uk was the first implementation of such (source: Fujitsu manager) and the biggest implementation they ever undertook. With Fujitsu acting as the inflexible front-end and Cerner never in contact with the clients, to the point that they never even surveyed the issues pertaining to each hospital, the system ended up with so many 'work-arounds', or IMHO bodges in plain English, that the cost of putting them all right was the key failure in the contract re-negotiation.

EDITOR.......... BTW, you've missed a trick -try asking about the mess with follow up patient appointments and the 'fixes' that have quadrupled the time to make them since the software was first shown 3 years ago. Thats what makes the queues go out the door and across the car park.........

Rubbish!

15 Aug 08 19:39

"Its not good blaming the suppliers, they have produced what they were contracted to produce, and what they were contracted to produce. "

IMHO this is utter rubbish. Instead of regurgitating opinion dressed as fact, have a look at OBS2 in the EHI archive against which Fujitsu (and CSC for that matter) were contracted to deliver and contrast that with the less than wonderful systems delivered so far.

Was the NHS was sold a pu?. In my view the suppliers shouldn't have been paid a brass farthing for what's been delivered so far - Granger said they'd be paid on delivery and the vast majority of us are still waiting.

It is also worth reading the Misrepresentation Act 1967 and the Trades Descriptions Act 1968! (I'll bet Jon edits out this last bit!!)

merely trivial teething problems

18 Aug 08 06:29

perhaps they need to go back to color coded sticky tape?

Cepi

with all due respect....

19 Aug 08 10:31

...i do know the OBS documents rather well, which is why i say they dont even attempt to describe what a PAS is supposed to do. If you dont beleive me, try doing a search for DSCN, which describe the key reporting requirements that the NHS are required to meet. Let alone any operational response time like how long it should take to actually book a patient.

I repeat: the problem is the contract not the supplier

hepatitis alphabet

Matthew.Grove@northumbria-healthcare.nhs.uk

22 Aug 08 14:48

Hmpf. Suspect the flags are actually for Hepatitis B and Hepatitis C, not A ... A is a relatively benign diarrheal illness whilst C is a nasty blood borne infection that can ultimately kill.

Nevertheless - the point made earlier is well taken. These days C Diff is more of a problem for my trust than MRSA; we have no idea what the next five relevant bugs will be, and we need to be able to add new flags as they come along.

Chain of blame

26 Aug 08 17:13

The OBS was based on (as the initials stand for) an Output Based Specification. Thus, those of us used to writing the detail were prevented because we had to leave it to the supplier to define how they would achieve the technical bit. This was because the supplies advice (external consultants doubtless employed at huge rates) thought this would provide better value. Ahem.

In contrast, our Operational Requirements documents from the previous generation of procurements are keeping current systems going. These did mention DSCNs, within the price, and lots of other things that our LSPs claimed were not included.

The contract did not aim to deliver the OBS in full anyhow, since the contractors response fell short of the requirements, in documented ways. And we subsequently learned of many other interpretations, or items glossed over in the various rushed reworkings.

I do blame the suppliers, because having offered half a system at a supposedly knock down price, they then appeared to try consistently to squeeze all the useful features into renegotiated extras, outside the contracted price.

And in the South, this was done to the extent that the base delivery became worthless, and eventually the parties split.

It took a long time to learn that the base R0 system that we thought we were all getting (successfully running at Newnham and Homiton), was actually a very early freeze of thier initial implementation, with negligable transfer of the various developments they locally made go get the system going.