Comment

Why the US has overtaken the NHS with its EMR

Emeritus Professor Angus Wallace, a leading figure in British orthopaedic surgery, academic research and the former chair of the National Osteoporosis Society, looks at why the NHS has fallen behind the US with the development of electronic medical records (EMRs).

The NHS in England should have excellent, comprehensive and widely available computer patient records. It interacts with and provides care for a total population in England of over 55 million, and it has no significant competitor. Thus it occupies a position of a monopoly, not only in provision, but also in enormous purchasing power. 

The NHS is currently endeavouring to introduce electronic records across the whole medical system – in primary care – general practice; secondary care – hospitals, and also in social services. The philosophy is that if health and social care are able to access up-to-date medical records for patients or clients, then care will be better “joined-up”, more effective and fewer mistakes will be made. 

However, this NHS process has been poorly managed, has taken a long time in coming and, in the meantime, the United States seem to have moved this forward more quickly and effectively. 

US electronic medical record 

As far back as 2002, the Cedars-Sinai Medical Center in Los Angeles, California, attempted to implement a new EMR system, but the $34m project failed. There were a number of factors that contributed. For example, the order entry system for drugs and laboratory procedures was more time consuming than doing the orders by hand. Not only was the new system more time consuming, but it also alerted physicians with numerous electronic reminders and alerts that they felt were excessive. Everyone learned from this experience and later systems were designed to be much more user-friendly. The US attitude also changed and they concluded that one large system may not necessarily be the best way forward. 

The US leaders realised that a number of steps were required in order for EMRs to be introduced and adopted. For instance, there needed to be a supportive environment, adequate training and resources, a clear direction, and engaged people. Strong leadership was also required to facilitate the changing environment. It was also really important to involve the clinical staff. Many hospitals used ‘physician champions’ – doctors who educated their peers on the benefits of EMRs. They also understood the importance of using quality department leaders to ensure that the EMR system was beneficial in providing quality care. 

However, the really major step forward was a recognition that hospitals had to have an incentive to invest in increasing EMR adoption, and the US decided to use both a carrot and a stick to do this. In order to stimulate national investment, the US Congress passed the ‘American Recovery and Reinvestment Act of 2009’ to save and create jobs almost immediately after the recession, but with secondary objectives of providing temporary relief programmes for those most affected by the recession and to invest in infrastructure, education, health, and renewable energy.

The ‘Health Information Technology for Economic and Clinical Health Act’ portion of this stimulus law provides payments for providers that show they have reached particular standards, and this led to more hospitals adopting the EMR. In 2014, three-quarters of US hospitals reported that they electronically exchanged health information with outside ambulatory providers or hospitals. This represented a 23% increase since 2013 and an 85% increase since 2008. 

UK electronic health record (EHR) & summary care record (SCR) 

The Department of Health (DH) and NHS England have also endeavoured to move forwards with electronic records and systems for patients. 

However, it is moving forwards in a much more complex way. This is highlighted by the lead organisations for the digitisation of health undergoing many transitions during the past 17 years.

Organisations involved in leading EHR and SCR 

In 1999 the NHS Information Authority (NHSIA) was established as an NHS special health authority by an Act of Parliament with its headquarters in Birmingham. Its aim was to bring together four NHS IT and Information bodies (NHS Telecoms, Family Health Service (FHS), NHS Centre for Coding and Classification (CCC), and NHS Information Management Group) to work together to deliver IT infrastructure and information solutions to the NHS in England. 

On 1 April 2005 (yes, April Fools’ Day!) the NHSIA was replaced by NHS Connecting for Health (CFH), which was part of the DH Informatics Directorate, with the role to maintain and develop the NHS national IT infrastructure. It adopted the responsibility of delivering the NHS National Programme for IT (NPfIT), an initiative by the DH to move the NHS in England towards a single, centrally-mandated electronic care record for patients and to connect 30,000 GPs to 300 hospitals, providing secure and audited access to these records by authorised health professionals. 

NHS CFH ceased to exist on 31 March 2013, and some projects and responsibilities were taken over by the Health and Social Care Information Centre (HSCIC) which is to be renamed NHS Digital this summer, in part, because of increasing dissatisfaction with the performance of it. So, in 17 years, four different names have been attached to the organisations which have been responsible for the development of the EHR. 

Electronic hurdles to overcome 

The challenges of developing the EHR and SCR should not be underestimated. On the one hand the concept is simple: create one electronic record for every person in the country. On the other hand it is necessary to cover the following important issues: accuracy of the records, legacy transformation (that is converting old paper records into electronic ones), confidentiality (encryption and data transfer), access of records to individuals or organisations, and last, but certainly not least,  the change in technology in the past and future. 

We can all be critical of how slow and cumbersome progress has been, but the big question that should now be asked is: “Do we have an organisation which is capable of delivering the EHR and SCR for England and the UK, or would it not be better to provide the guidelines and encourage the commercial sector to compete to provide the best products?” 

Valuable contributions from the royal medical colleges 

As a result of some far-thinking leaders, it has become clear that we are moving towards an EHR which has ‘information overload’ and so much irrelevant information that the important data – the patient’s actual medical problems and medical condition – get submerged under millions of bits of less relevant data. 

The Health Informatics Unit at the Royal College of Physicians has acted as the focus for clinicians coming together to work out what is the most important data to collect and include at discharge during a patient’s admission to hospital, and what information is needed by the hospital when the GP refers a patient into hospital for a medical assessment. Providing a common standard structure for information to be communicated along a patient journey helps avoid duplication and improve continuity of care.  

Also, by recording information in a standardised way at the point of care it can be extracted and used for clinical audit and research, reducing the burden on clinicians of separate data collections. The reports that they have produced have been universally applauded and supported by the Academy of Medical Royal Colleges and Professional Record Standards Body (PRSB). The standards are recommended in national policy: ‘Personalised Health and Care 2020 Using data and technology to transform outcomes for patients and citizens.  A framework for action’, and it is anticipated that they will be recommended as the standard structure for electronic discharge summaries in the NHS Standard Contract 2017. 

GP summary care record 

Although it is claimed that currently 94% of NHS England patients have had an SCR created (increasing on a weekly basis!), many patients struggle to get copies of their SCR through their GP. It is also clear that the use of the SCR across organisations – hospitals, pharmacies and even between GPs when patients transfer their practice – is still not happening as originally hoped. 

This problem has been compounded by multiple commercial companies providing their individual SCRs for GPs with a lack of interchangeability, and there is an ongoing significant problem of legacy data produced by software companies which have either pulled out of the market or disappeared. 

Confidentiality of patient records 

The HSCIC has had an unfortunate record in communicating with patients and clients about the confidentiality of their records. Patients should not be coerced into agreeing to share their information, but should be provided with the confidence to agree to a sharing process. 

It is in everyone’s interest for patient and client information to be shared between organisations, but reassuring patients that their information will remain confidential has been a challenge when even those close to the systems are aware that it may be too easy for some health workers and social care workers to access that information. The systems are in place to monitor who is accessing records through NHS Smartcards and PIN numbers, but breaches of security are occurring and it appears that the monitoring of those breaches is still a problem. 

Perhaps a mechanism of direct patient involvement at the time of sharing would help gain the patient’s confidence, with the patient carrying their unique access number with them on a Smartcard with PIN protection in the same way as we use our debit cards every day. The north of England seems to be leading the way, with its North of England Sharing Agreements providing an open approach and reassurance for organisations, patients and clients. 

On 4 May the Guardian reported in an article – ‘Google given access to healthcare data of up to 1.6 million patients’ – that the artificial intelligence firm DeepMind, owned by Google, was provided with non-anonymised patient information as part of an agreement with Royal Free London NHS FT. This included detailed information about people who are HIV-positive as well as details of drug overdoses, abortions and other confidential patient data from the past five years. Even when data is anonymised it is remarkably easy, with today’s software, to de-anonymise it as highlighted by Paul Ohm’s article on ‘Broken Promises of Privacy: Responding to the Surprising Failure of Anonymization’, published in 2010. 

Repeating errors of the past 

I believe the problems we are seeing in the move towards a digitised health record and social services record is a consequence of the huge complexity of the health service, and the interference from the centre and the government currently in progress. 

A good example was the introduction of NHS Choose and Book in 2005. The initial concept of this new system was that it should be an electronic outpatient booking system only – and this could have been developed relatively easily and had a high chance of success. Unfortunately, Alan Milburn (health secretary 1999-2003) and John Reid (health secretary 2003-2005) decided they wanted the system to not only be an electronic booking system, but also to be a distribution service encouraging competition between hospitals, thus making the system doubly complicated. As a consequence, Choose and Book failed to deliver and eventually was converted to an NHS eReferral Service in 2014. 

Again the politicians and the DH are creating problems with the digitisation of health records by interfering inappropriately, and forcing an agenda of pushing hard for cross-linking between NHS digital records and social services digital records, which, although beneficial in an ideal world, is creating even more problems when the EHR, the SCR and the social services electronic records are not yet fit for purpose. 

The way forward 

The NHS’s track record on moving forward digitisation in the health service is not good. A more proactive relationship with industry is being developed. There is more likely to be better progress in future if hospitals focus on developing a system which works for them and for that system then to be rolled out across the NHS in future years. Although this seems to be what was happening with the introduction of the e-Hospital System, developed by Epic and launched by Cambridge University Hospitals NHS FT (Addenbrooke’s) in October 2014, it had all the failings of the introduction of the EPR experienced by Cedars-Sinai Medical Center in Los Angeles in 2002. Unfortunately, Addenbrooke’s was alleged to have over spent on its system by Monitor – an example of a hospital trying hard to innovate and then being held back by a government-dictated regulator. 

We need to do what the US did – use a carrot and a stick approach – and we must reduce the amount of over-regulation which is currently strangling the NHS, and we must produce appropriate financial incentives. We should also take note of how the professional record standards can provide the glue that enable hospitals to focus on systems that work for them, but enable interoperability between these systems.

Documents developed by the Royal Medical Colleges with the Health Informatics Unit 

1. Generic medical record keeping standards and associated templates for auditing clinical records (2009)
2. Standards for the clinical structure and content of patient records (2013)
3. Professional guidance on the structure and content of ambulance records (2014)
4. Standards for an Electronic Discharge Summary (in preparation)