Ewan Davis – Director of Woodcote Consulting shares his thoughts on VistA
VistA is the open source electronic health record (EHR) system developed and used in the American Veteran’s Health Administration (VA). It is widely acknowledged that over recent years the VA has transformed itself from one of the worst healthcare provider to probably the best in the USA with its use of VistA rightly credited as on of the key enablers of this transformation.
This has success has spurred the development of WorldVistA and The Campaign for NHS VistA I’ve been aware of VistA for may years and have followed its’ progress with interest, but had not been convinced that it had much to offer in a NHS context. However, over recent months, mainly through my work with HANDI I’ve been lucky enough to meet some of those promoting VistA in the UK, from the VA and from WorldVista and am now persuaded that bringing VistA to the UK could help to enable transformation in the NHS.
My vision for VistA in the NHS varies somewhat from that of The Campaign for NHS Vista, who as I understand it see the ultimate goal being VistA as the single IT system for the NHS, but I hope a vision which is compatible with the direction of travel of the VistA community.
My earlier lack of enthusiasm for VistA flowed from a view that it was simply “a better way of doing the wrong thing”. The wrong thing because the idea of a single system across the whole of the NHS is neither achievable or desirable and a better way, because I believe in this part of the domain open source is a better way.
My view that “single system across the whole of the NHS is neither achievable or desirable” has not changed, but I am now convinced that VistA can be used widely in the NHS to facilitate transformation through front-line engagement and that VistA could be a key component in a open ecosystem for health and care.
In this blog I want to make a few observations about the VA and VistA, explain why I think VistA is great and how I think we should deploy it in the NHS.
Before the 1990’s the VA which delivers health care to 8.6 million American ex-servicemen and their families was heavily criticized for providing poor quality care. The VA responded to this with a transformation programme in the late 1990’s which saw massive increase in quality with costs held constant despite a greater than 30% cost increase in the US healthcare system as a whole over the same period. Today the VA provides some of the best care available and provides one of the few models for sustainable health care in the USA (much to the embarrassment of the American right) See Philip Longman’s inspiring book: “Best Care Anywhere: Why VA Health Care Is Better Than Yours”
There where a number of factors supporting for this transformation:
- Strong and inspirational clinical leadership focused on improving quality, patient safety, and efficiency
- Strong links with academic medicine
- A major shift to primary care
- A lack of perverse financial incentives for both the VA and it clinicians
- VistA which provide an integrated EHR to support the processes of care and mobilised information to support the planning and delivery of care.
The important point to note is that while VistA was an important enabler of the transformation it was just that, an enabler.
As the only significantly publicly run health provider in the USA the VA is probably the most NHS like entity in the US healthcare system but it is none the less very different, much smaller (less than 1/6 the size of the English NHS) and is a single, if complex entity. This compares to the NHS which consists of 900+ separate legal entities (excluding GPs, dentists, community pharmacist and opticians) with increasing autonomy as many become Foundation Trusts. It also important to acknowledge that despite the criticisms levied against it the NHS is actually a high performing health system in the global context, nothing like the basket case the VA was prior to its transformation, and the NHS has well developed primary care supported by world leading IT.
In my view VistA’s strength is not that it is a single system, but that it is a systems built from the ground up with development that has been clinically led from the front-line and which because of its open source model has managed to get strong engagement from front-line staff and been able to benefit from improvement and extension from a community of users and developers (particularly users who are also developers) which is now global.
VistA is a venerable product with is basic architecture created in the late 1970’s and its seminal development occurring during a period when the perceived wisdom in healthcare IT was to try and build single enterprise wide systems. Compared to most other attempts to do this VistA was spectacularly successful, but at this level VistA operates in a 20th century paradigm providing, as I said earlier, “a better way to do the wrong thing”.
However, what I now understand, is that because of its open source model and the passionate commitment of its community VistA is capable of, and indeed already is, metamorphosing into something that can provide a core component of the healthcare IT ecosystems that we have to create to take healthcare IT into the 21st century.
VistA offers us two things:
- Software honed by real use and front-line input over many years that we know works in a number of different healthcare systems.
- A model for development and continual improvement that allows meaningful engagement with front-line staff (particularly clinicians) and which enables the resources of a global community to be applied for the benefit of all.
What we need to move forward is an open ecosystem in which apps designed to meet the specific needs of end users can be orchestrated to work together and can interoperate to share information and manage the processes of care. I believe that as a result of work going on in and around the VistA community that it will evolve to become a key component of this platform while simultaneously being able to meet the needs of those NHS organisations whose current requirement is for a good quality traditional EHR.
VistA will require localisation to make it suitable for the NHS, but would seem to already be closer to NHS requirements than many commercial product emanating from the USA. My experience with commercial suppliers is that they wildly underestimate the localisation effort, and I suspect the same will be true for VistA, certainly an evaluation by the Scottish NHS some years ago rejected VistA as it was then on the grounds of localisation costs, but things have moved on and given the support of an enthusiastic UK community this does not seem an insurmountable barrier and with such localisation VistA would immediately be able to provide a much better solution for NHS organisations than many closed-source alternatives.
However, what makes VistA really attractive to me are other things:
- The open-source approach to development and implementation which engages and empowers clinicians and other frontline staff
- The ability to share improvements from and with the growing global VistA community
- Work being done in and around VistA to enable it to provide a platform for apps and take advantage of the the latest web and cloud technologies
My ambitions for VistA are not the same as many in the NHS VistA community. I’d like to see VistA (as an EHR) widely used in the NHS, but I don’t want to see it as the one systems for the NHS. I am also interested in how VistA can provide a core part of an open ecosystem for health and care IT but I don’t want it to be The Platform for the NHS, just part of it. However, I believe my views about the next steps are broadly aligned with others and I want to work with them and am happy to see where this leads us with an open mind.
I want to see more plurality both in systems and platforms. I want to encourage cooperation so we can explore how we can get things working together and I’m particularly interested in working with those things that have an open model: VistA, SMART Platforms, OpenEHR, Open Clinical, etc, but also with those vendors with proprietary models where they are willing to open up their systems, particularly those of a pragmatic bent like the many existing vendors that work with IHE to demonstrate real world interoperability between commercial systems. However, I also want to see competition, particularly in the app space so entities with different business models can compete to provide niche functionality and the best user experience confident that the underlying ecosystems will enable their apps to interoperate and be orchestrated to share data and work together.
I also want see some plurality in the platform with competing suppliers of EHRs and PHRs (so patients can decide who they trust to hold their data) and multiple providers of knowledge and information that apps can consume, allowing users to decide who the trust and which business models they are comfortable with.
We also need to address the weaknesses in VistA:
- Its core is built using 1970’s technology (MUMPS) and even it most ardent users describe it source code as “labyrinthine”
- Its user interface looks increasingly jaded
- As VistA is used more widely in many different health care systems it becoming increasing difficult to handle source control and configuration management so that improvements from all parts of the community can be used by all others who might benefit from them.
However, I believe the above can addressed and indeed work is already underway.
The first two points can be dealt with by wrapping VistA’s core code in a service wrapper to provide an API that allows new user interfaces to be built outside the core using whatever technology is available to deliver the best possible user experience. As most development occurs in the user interface layer most developers will have no need to work with the core code, will be insulated from its complexity and can use whatever tools they wish to develop user interface components. Tools such as EWD and SMART are already available to support this approach and although SMART is currently read-only it is of particular interest as it provides a standard layer for apps that allows the same code to work with any SMART enabled EHR.
The third point requires a more sophisticated approach to configuration management and source control, (more info from OSEHRA) fortunately the issue is recognised and work has already been commissioned to address it, but it’s not an easy problem (details here).
So in conclusion I have a different vision for VistA in the NHS than the NHS Campaign for VistA. However, we both want to see an NHS version of VistA and we both want to see it widely implemented. So let not worry about where we hope and think this should lead. Looking at the NHS Campaign for VistA’s 8 point plan http://nhsvista.net/our-ambition/ The first six steps align broadly with what I want to see happen so lets get on an do it and see where it leads us which I suspect might be a different than any of us currently expect.