cradle to grave

Published: 21/03/2002, Volume II2, No. 5797 Page 11 13

Electronic patient records lie at the heart of the government's plans to modernise the health service. By 2005, health service organisations should have reached the target set for them in the government's Information for Health framework. This means by then they should have reached level three of the six-level model for EPRs.

Level three means that all patient administration details should be online as should other functions, including electronic clinical ordering, results reporting, prescribing and online communication between different health bodies.

But as this target draws near, a growing number of questions are being raised about almost every aspect of EPRs, from technical challenges to issues over ownership, confidentiality and access.

On a purely technical level, there are still plenty of challenges. Organisations have to decide whether to buy in an EPR system or build one themselves, using existing systems.Most are opting for something of a combined approach, looking to buy an EPR system that will help them pull data from existing systems. But pulling together patient information from different sources raises problems over data quality and incompatibility.

And even those parts of the health service well down the road towards EPRs are still struggling with questions of encryption and security, which are vital if patients are to feel their online records are safe and confidential.

'Incompatibility is the biggest challenge, ' says David Plein, IT administrator at an eight-GP practice in Kilburn, north west London.

Mr Plein's practice is now entirely paperless, with all documents scanned into the system. The PC-based system is based on the Read Code classifications, but also enables GPs to add as much free text as they need to patients' records.

Mr Plein is happy with the progress his practice has made over the past couple of years in getting its records online, but points out that because other GP systems are incompatible, information cannot yet be sent online.

'There is no GP-to-GP transfer - if one of our patients leaves, we have to print out their records and they have to be input all over again at the new GP practice, ' he says.

'This is a big question and one the industry is trying to work out, but I do not see anything happening for at least two years.'

Similarly, referrals from the Kilburn practice still cannot be made online, despite the practice being linked to the highly secure NHSnet, because of a lack of encryption software. This, too, is being worked on by the practice's IT supplier, but in the meantime, the practice is in the frustrating position of being an island of automation, with its own EPRs that have nowhere to go.

This is a major issue, which is why the emphasis in overall strategy has shifted from EPRs to EHRs - electronic health records - that would hold far more information and encompass all aspects of a patient's interaction with the health service.

Simon Goodwin, IT director at Cornwall NHS IT Service, one of two national demonstrator sites for developing EHRs, emphasises that with an EHR, the key technical challenge is integrating data from many different sources. 'There are issues over data incompatibility and quality, ' he says. 'For instance, some records may not always have an NHS number.'

It is necessary to combine data from systems that may use different levels of the Read Code classifications.

Creating an integrated framework is complex, adds Mr Goodwin. Cornwall is working with a number of suppliers to develop different parts of its systems. Its core information repository comes from IT supplier Sybase, which also provides the software to integrate data from different legacy systems. 'In some areas, we are at level six but in others we are not yet at level three, ' he says. 'We are working on the gaps.We are making good progress, but there is a lot still to do and one of the biggest obstacles is funding.'

Some organisations have made a clear decision to build, rather than buy in, EPR systems. Barts and the London trust, which launched its first stage of an EPR system last May, has opted for the former approach, building on its existing systems.

Barry Elliott, the trust's finance director, says this approach has involved a lot of development effort, but has enabled the work to move at a pace the trust could afford and resource. 'We are using web-based technology to create a front end to existing clinical information systems, so we do not have to replace legacy systems, 'Mr Elliott says.

'This has given our development staff a real sense of achievement and it is more manageable from an operational point of view: we can now proceed to replace clinical information systems at a pace we can manage in a sensible way.'

Mr Elliott acknowledges that this approach might not suit everyone, but says it has a lot to commend it, in enabling development work to proceed at a lower cost and with a lower risk.

Those trusts that opt to buy, rather than build on existing systems, may do so in the hope that buying in a system will be a quicker way to reach the Information for Health targets. This often involves major systems investment. Bolton Hospitals trust, for instance, went live last November with one of the largest EPR implementations in the UK, based on software from supplier iSOFT. The first phase of the rollout involved migrating 4 million patient episodes into the system.

Health service IT suppliers are clearly keen not to miss the buying opportunities that the 2005 target represents, and have spent a lot of time and development effort in putting together software modules to meet demand.

But suppliers are also warning that if trusts leave their buying plans too late, they may be in for a disappointment.

Sean Brennan, head of healthcare strategy at Northgate Information Solutions, says many health service organisations are opting for a combination of buy and build and are looking for suppliers who can integrate existing systems into an EPR system providing functionality that will take them to the level-three target.His company has developed its EPR model based on software used in France and Germany, which it felt was more suitable for adaptation to the UK market than some of the software used in the US.

'There is no single, clear strategy for trusts because it depends where they already are, 'Mr Brennan points out.

'Most trusts already have something in place and most are looking at building on those existing systems. The frustration is that although there is an overall target, each individual trust is getting to that target in its own timeframe, so there is a potential danger in having all 280-odd trusts waiting until the last minute.'

Mr Brennan says this would present IT suppliers with a major problem. 'Most trusts are nowhere near level three and, while they could still make the target if they started tomorrow, that could then swamp the suppliers, ' he says.

Mr Brennan is also concerned that with the whole specification and procurement process still taking a long time in most trusts, they could begin buying in EPR systems, but end up with their requirements having moved on by the time the system is actually in place. If there are too many trusts clamouring at the last minute for EPR systems in the run-up to 2005, Mr Brennan says suppliers will be forced to cherry-pick. 'That means some trusts will then struggle to get in any system. It is frustrating.'

The technical challenge of pulling together incompatible data into a single EPR or EHR is one issue that was highlighted critically last year in a report on one of the NHS EPR demonstrator sites, the Hadfield Medical Centre in Derbyshire.

1In their evaluation, Samina Munir of Salford University and Dr Ruth Boaden of the Manchester school of management stressed that EPR pilot schemes should not be used to test the technical weaknesses of systems.

2'The technical issues are not insignificant and undoubtedly affected the outcome of this project, ' says their report. 'In future, more attention should be given to technical issues before patients are allowed access - they should not be used as guinea pigs to highlight and resolve technical issues.'

The evaluation also highlighted issues of confidentiality, and access. Confidentiality, though important, was not a primary concern of the patients. Their main finding was that patients want to be more informed about their healthcare and welcome openness from their GP.

They also want all health professionals to have access to EPRs if required, in order to provide a more integrated health service and minimise repetition of information.Also, patients want to be able to access their EPRs from anywhere - being able to access records while on holiday, for instance, was seen as particularly useful.

In Cornwall, Mr Goodwin sees the most fundamental issue as that of patient consent.He and his team have been working with clinicians from all disciplines on how to create and implement an integrated patient record for primary and secondary care that will provide all necessary information, at the appropriate point, when delivering care.

One of the difficulties is working out, for instance, what information should be available when a patient needs emergency care.

Another key issue is what information from GP level should be included in an EHR, and who should authorise that information to be included.

'There is no clear way forward in terms of consent, ' says Mr Goodwin. 'We are working on a pragmatic basis, allowing patients to be asked if they are happy to allow information to be included in an EHR that will be seen by other people, but we need some sort of national guidelines and a workable solution.

'It really depends on whether you see the health service as a single healthcare system, and someone has to grasp the nettle.'

REFERENCES

1Munir S, Boaden R. Patient Access to Patient Records at the Hadfield Medical Centre.

www. nhsia. nhs. uk/erdip/arch ive/documents/hadf/hadf63. doc NHS Information Authority website, 2001 [cited 11 March 2002].

2Eversheds. Patient Access to Electronic Patient Records at the Hadfield Medical Centre, Evaluation Report.

www. nhsia. nhs. uk/erdip/arch ive/documents/hadf/hadf61. doc NHS Information Authority website, 2001 [cited 11 March 2002].