Inside e-health on EPRs and the Goldilocks question

In the first of his regular columns, Jon Hoeksma asks who has been trying EPR projects - and which countries have managed to get them just right?

One of the biggest questions when it comes to the development of shared electronic patient records is what size do they work best at? It’s the Goldilocks question: the answer is that you don't want them to be too big or too small, but just right.

Do it too big and you risk systems becoming unwieldy and struggling to get local buy in. Do it too small and you lose economies of scale and risk replacing islands of information with a series of archipelagos.

Unfortunately, the question of what will be just right depends on what you're trying to do and where you're starting from. A good rough rule of thumb, though, seems to be that it’s a good idea to match EPRs to natural communities in which people live, work and actually identify with.

Too big? Too small?

The scale at which EPR initiatives are attempted frequently has little to do with technology and more to do with politics, geography and history.

Across the industrialised world, countries and regions are trying to develop shared EPRs. In many places, these efforts are building on local initiatives in hospitals, GP practices and clinics that go back as much as 20 years.

Nevertheless, federated states like the US, Sweden and Germany have tended to try highly devolved approaches, while more centralised countries like England and France have tended to adopt national ones. 
“Across the industrialised world, countries and regions are trying to develop shared EPRs. In many places, these efforts are building on local initiatives in hospitals, GP practices and clinics that go back as much as 20 years.”

Try a region for size

One good example of an area that is building on past investments is Sweden's Kronberg county, which has developed and implemented an EPR system that enables a patient's records to be accessed and shared by anyone working anywhere in the county.

The system closely resembles what the NHS IT programme is meant to deliver in the form of the NHS Care Records. The big difference is that Kronborg’s EPR is in daily use, with new clinical modules being added each month.

Arguably, one of the keys to success in Kronberg, and in other Swedish counties such as Upsala and Stockholm, is that healthcare is a responsibility devolved to Sweden’s 20 counties. And in the case of heavily rural Kronberg - the 'moose capital' of Sweden - there is a total population of just 180,000 and a small, tightly knit healthcare system. Total cost: just a few million Euros.

Or a small country

Scotland, though, shows that a population of a few million can still be a manageable size to work with. NHS Scotland has made substantial progress in rolling out its emergency care summary – a project somewhat analogous to England's Summary Care Record (SCR), for which pilots have just begun.

The population of Scotland is around 5 million; total health IT spend is about £30m annually. And one of the biggest shared EPR success stories to date has been the Spanish region of Andulcia, population 7.4 million.
“NHS Scotland has made substantial progress in rolling out its emergency care summary – a project somewhat analogous to England's Summary Care Record (SCR), for which pilots have just begun.”

England also has some, local shared EPR projects underway. The biggest to date covers Hampshire and the Isle of Wight. Launched in 2006 as a local initiative, its Clinical Data Repository has 1.5 million patient records covering GP, hospital and single assessment process (SAP) episodes. The system is used by authorised health professionals across the county.

But when it came to the national NHS IT programme, the government decided that big was best. This held out the potential of huge economies of scale and of being able to award big enough contracts to attract heavyweight suppliers.

The Department of Health says that carrying out a national IT procurement saved a notional £4 billion. The five regional contracts it placed each cover about 10 million and £1 billion over ten years.

The downside is that the NHS IT programme has suffered significant delays and faced persistent criticism that it has failed to secure local buy-in. However, and significantly, it has adopted a model of implementing early systems into 'natural health communities' often consisting of one of more hospital trusts and surrounding community services.

Or a natural community

This might make it sound as though the best answer to the Goldilocks question is small countries, biggish regions or large counties. However, in the US, attempts by the Federal government to create regional health information organisations (RHIOs) to promote the development of EPRs have met with little success.

In part, this is because there is no clear economic model for RHIOs; but the communities they cover have also been described as being artificial. In contrast, closely integrated US healthcare networks – such as the Veteran's Association and Kaiser Permenante - have enjoyed far greater success in developing EPRs.

So the answer to the Goldilocks question seems to be small, but not too small and as big as you can manage without becoming unwieldy. But, above all, a natural constituency: either a coherent geographic population or a clearly defined community of professional users working within an integrated healthcare system; and ideally both. 

About the author:

Jon Hoeksma is technology journalist of the year, and the co-founder and editor of the E-Health Insider industry portal: www.e-health-insider.com