Posted by: Tony Shannon | September 18, 2013

Transatlantic Thoughts.. A VistA .. to/from the NHS

As a keen supporter of the move to improve healthcare with better information technology and open source in particular, I’ve recently returned from Washington DC where I attended the 2nd OSEHRA (Open Source Electronic Health Record Agent) conference. As I’ve recently been involved in OS EHR developments in the NHS, I was honoured to be invited to speak from an NHS perspective of the moves in the NHS towards Open Source in Health IT & plans for “NHS VistA” in particular. The time spent at OSEHRA was well spent, where it was great to meet others involved in leading the charge of healthcare improvement.

The OSEHRA Foundation is working to foster the development of an open source Electronic Health Record – something I believe will be key to the many improvements that are needed in healthcare in the 21st Century. The OSEHRA Foundation has been born from the foresight and vision of the US Veterans Administration who have gifted their leading VistA EHR to the world, for both the betterment of the VA Healthcare system and indeed the greater good.  They made this move towards establishing an open source ecosystem, as a deliberate next step forward for their much heralded VistA EHR, which is acknowledged as a leading EHR that underpinned the VHA transformation in the 1990’s.

The OSEHRA/NHS/VistA presentation I was able to deliver gave me a chance to explain my own time in Informatics from when I began back in 2000 in Washington DC, through my time in the NHS National Programme for IT, working with the international openEHR Foundation and back more recently as in CMIO/CCIO role in the NHS in Leeds.  Over that time I’ve been observing developments in Health IT on both sides of the Atlantic and its plain to see that there is now much common ground between these two camps.

Thanks to the leading efforts of the VA and OSEHRA and since the demise of the NHS National Programme for IT,  there has most recently been an enlightened shift towards the greater role of open source to positively disrupt healthcare by the leadership of the NHS in England, so it is fair to say that the intersection of these leading edges of OSEHRA and NHS England is of real signficance to the world of healthcare.

While I’ll shortly explain the areas of difference between the OS communities in both the UK and the NHS there are as you might expect some common themes that span the Atlantic including

  • Clinical Leadership is understood as key (OSEHRA is chaired by Dr James Peake, MD  and NHS England is moving to support the NHS CCIO movement).
  • Healthcare is best understood as an ecosystem that needs evolutionary change
  • Recognition of the key value of open source as a postively disruptive force to enable this change.


Open Source in Healthcare.. The challenge of bridging the transatlantic gap.

Through the efforts of the VHA,  the US has advantages in terms of open source Health IT in terms of VistA as the premier OS EHR in the planet.  While the NHS also has early movers in OS, there has been relatively little effort in terms of learning from VistA to date.
The UK has the advantage as being quite advanced in its effectively computerised primary care which has advanced their thinking wrt semantic interoperability, the GP2GP efforts being a case in point, while early innovators within the NHS OS community are moving towards adopting the openEHR industry “standard”.
I’m most interested in the intersection of these open source and open standards areas,  as per my own work within the Leeds NHS where we have 3 key projects that address this intersection (Leeds Clinical Portal/Leeds Care Record/Leeds Innovation Lab).

Certainly there appears to be fertile ground to build on success of  VistA, while aligning with related EHR progress that has been made in the UK, Europe and elsewhere.  To grossly simplify the roadmap of VistA development while exploring where this fertile ground may lie, I might explain a view of the current VistA community effort at 3 levels.

1)      Vista Refresh –Modernisation of the current stack (eg the work towards US Meaningful Use standards as part of the leading work of Oroville Hospital.)

2)      VistA Refactoring – this is a more indepth and medium term piece  of work, aligning the varying VistA codebases towards a common OSEHRA VistA version, again with the same current stack (CPRS, FileMan, M etc)

3)      VistA ReEngineering- work towards a modern web UI, amove towards web services and a service oriented architecture, plus some deep thinking on the way forward for Mumps the language/database.

To migrate VistA into the NHS will of course require more of this work, so one might imagine that the starting point for NHS VistA would be at least begin with the refactored OSEHRA VistA effort.  My sense is that given the current state of the open source community in the NHS, the interest and potential of an NHS VistA platform lies in exploring and collaborating with the deeper VistA reengineering effort. This has been challenging territory for the VA & DOD with their recent iEHR effort hitting difficult ground. Yet it is this challenging area where I believe that OSEHRA/NHS community collaboration may be most fruitful for all parties..

Ways forward for VistA in the NHS

Following the Campaign for NHS VistA, the NHS England Safer Hospital, Safer Wards initiative now highlights that there is clear interest in exploring the potential of VistA in the NHS. Certainly the strength of the VistA community in the US suggests that those experienced in VistA can help to make this work in an NHS Trust..with the right clinical leadership, healthcare improvement culture and the right technical team it could be done.

However if these VistA deployments are done as yet more standalone enterprise deployments in isolation from the already active NHS efforts towards open source and open interoperability standards, it will be an opportunity lost. So herein lies the real challenge for NHS VistA, not to make a success of 1 or more deployments, but rather to help underpin the bottom up grass root movement that has begun in the NHS towards an open source platform for the transformation of healthcare. This long-awaited platform needs to engage frontline clinicians, the SME market and the wide range of healthcare organisations across the NHS alike. Accepting this presents a significant opportunity for both the NHS and OSEHRA, lets now look at 3 related challenges.

3 Key Challenges

Challenge # 1: Community & Code
Challenge # 1 is to engage both the open minded clinical and developer community from the NHS in this NHS VistA effort.  So the work should look to leverage the energy and effort of the NHS CCIO movement, existing NHS OS pioneers, plus the NHS HackDay and HANDIHealth movements etc.

Having explored under the hood of VistA, I am now clear that there is definite value in the underpinning technology, ie Mumps. However that is not the orthodox view in the UK where it is perceived as an old, dated technology. In my opinion this relates to the poor way in which the Mumps community has communicated and explained the value of the technology.
To better explain the past and future of Mumps in Healthcare,  I would draw your attention to the writings of the UK based Rob Tweed who works on the leading edge of VistA. Based on his deep analysis of Mumps and its community he has made a compelling case and some great tools available for the way forward; 1) Mumps the language *can* be replaced by a.n.other language (NB As it happens Javascript may be a very good fit)  while 2) Mumps the Database has an excellent fit with healthcare data which has been very poorly communicated to date. His excellent paper on a “Universal NOSQL Engine” should be required reading for anyone trying to understand the technology that underpins VistA.

Challenge # 2: Integration & Interoperability
It will be easily understood that the NHS is not a greenfield site for health IT… the NHS has plenty of existing (and inoperable systems), with a rich history of (primary) healthcare computing.  In essence a rip and replace approach to the adoption of an enterprise system was at the heart of the difficulties of NHS NPfIT . VistA in its current form may be best understood as a large enterprise wide system, which has advantages of course, but such systems always have boundaries … so lets hope the push for NHS VistA is not just about large but isolated deployments, but rather part of an active push for advancing interoperability across the NHS.

Challenge #3- Usability & the Move to the Web
The existing NHS OS community is a savvy one, keen to showcase developments at the leading edge of agile software development (e.g. openEyes & Wardware).  As many of the these efforts use web based technology (with all the benefits that now brings), the limitations of the Delphi and “Roll and Scroll” UI elements of VistA become apparent and may lessen its appeal.  Equally as NHS OS community seeks an effective underpinning platform, so the move to ready VistA for the web and mobile devices (via EWDLite, HMP etc) should be made now as part of the NHS VistA effort.


3 Key Opportunities

From these challenges arise related opportunities, which I hope would interest both those promoting VistA in the NHS, as well as those already active in the OS community in the NHS.  If any of this work is to emulate the historical roots of VistA towards a NHSVistA Gold Version, it should be done along clinically led, user centred, agile development lines, so I might suggest the following areas of focus..

Suggestion #1 Common Interest at the UI layer
To engage clinicians in EHR work, the most effective way to do so in my experience is at the UI level. So engaging NHS clinicians in an effort to refine the VistA UI for the NHS world should be good place to start.  It is acknowledged that while the CPRS UI of VistA is liked and very effective , it is fair to say that it is ripe for improvement esp wrt bringing it into the web and mobile age. (eg the work on Janus, HealthBoard) The NHS already has work done on a NHS Common User Interface, the  OS movement in England has already embraced the web world (ref Leeds Clinical Portal, WardWare, openEyes), so exploration in that common space appears an important challenge that should be tackled fast. My own experience of a 10 day User Centred/Agile Development project with Rob Tweeds EWD (inc  M) framework highlighted the productive nature of this VistA ready stack.

Suggestion # 2 Common Interest at the Service Layer:
At a layer below the UI there is common ground across the NHS & VistA community towards the development of an open service oriented architecture across healthcare. Some related items: Within England the leading work by the RCP HIU has indicated a clinically led way forward for the structure of the health record in England. These are currently a set of standards that have yet to be showcased in any open source EHR and so the NHS Vista development is a prime opportunity to do so. The clinical overlap between the RCP Core Clinical Headings and VHA Virtual Patient Record work has the potential to be a useful steer to the evolution of related technical services.. which are currently spread across VA MDWS/ VA VSA/openMDWS/VistA Expertise Network Web Services/WorldVistA SMART .. all of which overlap with a real interest in the NHS on a set of openNHS APIs..

Suggestion # 3 Common Interest in the Challenge of Data Models & Persistence
To the uninitiated it is difficult to spot the generic patterns that underpin the VistA codebase, especially in terms of clinical content. My impression is that across the M language/database there are deep patterns that underpin the flexibility of the VistA platform.  The difficulty of their current state that as of now these are nowhere near as easy for clinicians to visualise, comprehend, review and improve as internationally leading approaches such as openEHR archetypes (which are gaining ground in the NHS  in Leeds Care Record, WardWare, UCL Moorfields openEyes etc(Explore the International openEHR CKM for a flavour) . My instinct is that the pioneering work on VistA FMQL may be well placed to help spot these patterns and work well with the openEHR archetype methodology to address the challenge of clinical data models.
At a deeper technical level, the value of Mumps as the ideal database for healthcare needs to be better communicated across the NHS, esp to those now exploring the optimal means of healthcare data persistence – with changes in the database landscape emerging in the form of new NoSQL technology (e.g. MongoDB) in the age of Big Data.

As the NHS starts to explore VistA , that journey may begin with some NHS Trusts looking to deploy VistA  on an NHS patch, but if this is done in isolation from the rest of the NHS OS community it would be an opportunity lost. More importantly, both OSEHRA from the US and the NHS in England now acknowledge the need for clinically led,  agile development of an OS EHR platform to transform healthcare in the 21st Century.
Lets hope that the OS EHR community of both sides of the divide can now look to better understand each other, collaborate and stimulate a wave of transatlantic innovation.

PS: The forthcoming VistA Expo conference in Seattle (Oct 22-25) should be a great place to learn more from the VistA Community for anyone fortunate to be able to attend.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s