Despite the perceived benefits of electronically linking a patient's medical records, significant barriers must be overcome before the plan becomes a reality, as Mary Hawking explains

It may sound ideal to create detailed records by electronically linking all the files relating to the care a particular patient has received in their local NHS organisations, but on closer inspection there are a number of significant problems that need to be addressed if the plan is to work as intended.

One of the major objectives of the national IT programme and Connecting for Health was to introduce a summary care record and detailed care records, which are due to be introduced without piloting or discussion. As the definition of a detailed care record is unclear, I will discuss the model of a single shared electronic patient record.

Ideally, this would mean all the records of a patient's care are kept in one place, nothing is missing, the record is available at the time and place the patient is seen, it is available for audit and service management purposes, and it is always in the same format. With benefits like these, what are we waiting for? But despite the perceived benefits, life is not so simple. There are risks and problems to overcome and the end result may not be worth the effort, fit for purpose or value for money.

Data standards

From a GP's perspective, if you merge a GP's records with those of another organisation, how do you ensure the same standards of quality are maintained, and therefore maintain the record's usefulness and safety, when several different people will be entering data into the combined record?

In the model being implemented in the North Midlands and East cluster - involving GP, community and other primary care organisations - the record is basically a GP record - involving all the coding and alerts needed for general practice and the quality and outcomes framework, with community and other organisations such as child health and therapies using it on an equal basis.

Data can only be amended by the organisation that entered it. If someone in a different organisation thinks a diagnosis is erroneous or something needs changing - such as a prescription - the only way to change it is to send a request to the organisation that made the entry in the first place. But what happens in this scenario if the patient has been discharged from the workload or moved to a different area and there is no longer any legitimate relationship with the patient?

Until now, community trusts have not used electronic patient records and healthcare professionals in the community have had no training in this type of structured record, the read codes, or the implications for the GP record of entering erroneous or uncertain coded diagnoses. There are huge training and safety implications here, which will involve managers as well as clinicians. Then there is prescribing, which is always a tricky and risky area.

If a prescription, like any other bit of data, can only be changed by the organisation that made it, what happens when a heart failure specialist nurse (an independent prescriber) needs to change a patient's medication and cancel something prescribed by the GP or secondary care? Who is legally liable if something goes wrong and the patient takes both medications with unfortunate consequences?

Patient rights

What about the patient's right to refuse to have a shared record under the records guarantee? In the summary care record early adopter pilot areas, patients were leafleted and informed that they could opt out completely and not have a record at all. How would this work in a scenario involving single electronic records or detailed care records when there is no other record to link with? Which procedures are in place for informing patients of the change in the way their confidential records are being or will be held, and their options for opting out?

Further questions only add to the concerns. How do you manage a local health economy when some practices have migrated to differing GP systems for electronic records, such as the GP systems of choice? Further down the road, how would you change if the system is not fit for purpose or one sector wants to change or innovate? How do you manage the information governance framework - especially the bits about keeping personal information confidential and the high-risk area of prescribing?

When I started looking at the subject and asked the National Patient Safety Agency to put it on the IT risk register, I was thinking it was a GP problem. I now realise that it is a managerial problem as well.