Diabetic VistA – The First Amputation

Scott_shreeve_smallMany people in healthcare got their first look at Scott Shreeve through his work with
Medsphere, the open source software company based on VistA, the information system developed by a team of programmers and doctors at the VA in the early 1970s. After a series of early successes that drew national attention, the project ran into problems when the founders clashed with management over how best to apply the open source model to the company’s business. With that experience now safely in the rear view window behind him, Scott returns to the limelight with Crossover Healthcare, a new venture with a Health 2.0 focus. Today he returns to his open source roots with an analysis of the news that the VA intends to replace a piece of the existing VistA system with a product made by a private vendor.

Diabetes (dī-ă-bētēz) n.

  1. Diabetes
    mellitus is a condition in which the pancreas no longer produces enough
    insulin or when
    cells stop responding to the insulin that is produced
  2. As a result, the glucose in the blood cannot be absorbed into the cells of the body.
  3. The symptoms include frequent urination, lethargy, excessive thirst, and hunger.
  4. The treatment includes changes in diet, oral medications, and in some cases, daily injections of insulin.

During
my medical training, I had the opportunity to rotate through several VA
facilities. The VA is an incredible place to train because of the
varied types and degrees of disease you see within this unique patient
population. More profound were the individual patients, these grizzled
veterans of war who have served their country faithfully and
heroically. They were typically very grateful for our services as
students and residents, and were the most complicit but also the least
compliant of patients.

Diabetes
was rampant within the veteran community, and I was able to see the
literal ravages of the disease during training. Diabetes is an
insidious, unrelenting disease that takes its victim piece by piece.
While the blood stream is awash in glucose, there is no mechanism to
get the vital energy source into the cells creating a Tantalus-like
starvation. Given that the effects are silent and painless over years,
many diabetics remain untreated until eventually the vascular
strangulation leads to necrosis of tissue. The signs were pathoneumonic
– the black, leathery, and advancing patches of dead tissue. The only
treatment for this is amputation. Starting from the toes and advancing
up to the knees and beyond, diabetes destroyed its victim piece by
piece, and it usually starts with a single toe amputation.

In an unprecedented announcement yesterday, Cerner has signed a 9-year deal with the VA
to replace the VistA Laboratory Information System (LIS) at 150
hospitals and 800 clinics. This is both a highly provocative and highly
concerning announcement– a double edged sword with deep cutting
implications. Which side of the sword you will feel will depend on your
perspective.

Having
sold VistA commercially for several years, the LIS was certainly one of
the challenges we needed to overcome in every sale. The VistA LIS is
actually one of the first parts of the system that was created in the
early 80’s. Several key figures, including George Timson, were involved
in writing the code, which unbelievably has changed very little since
that time. I distinctly remember sitting with George during one of our
early implementations which required some modifications to the code,
and him remarking along with one of his famous Mozart laden
profanities, “this is the same code I wrote 20 years ago.” (As an
aside – It is actually quite amazing that code written 20 years ago
could remain serviceable given all the dynamic changes in requirements,
features, and functionality over that time frame. In fact, can you
think of other software programs written that long ago that are as
widely distributed and remain functional in today’s computing
environment?)

So
what happened? How could the VA allow a critical, integral pat of VistA
languish for more than a decade? What are the implications of the
decision to wholesale components of VistA to proprietary software
vendors like Cerner? What does this mean to the other aging parts of
the system (Pharmacy, Radiology, etc) as well as some of the new, still
functional innovations (BCMA, CPRS, etc)?

The
“good news” is that this might lead to an entirely new frontier of
integration with leading commercial software. Since Cerner is the most
popular LIS in the market (and a very good one by the way),
many hospitals have Cerner already installed and so this could
potential decrease the impedance of VistA sales. In addition, perhaps
the collective weight of the VA and Cerner working to integrate the
public domain VistA code with the proprietary Cerner code will yield
new interoperability magic.

Unfortunately, I doubt it.

I
fear this is the first amputation in a long and steady surgical removal
of VistA from the VA. Piece by piece, subsystem by subsystem, the VA
appears to be looking to take a best of breed approach. All the
beautiful and inherent advantages of a single, integrated software
solution get thrown out the window as a patchwork of best of breed
solutions gets thrown into the mix.

For
Cerner, this is huge. No only is it a huge land grab, but it is more
than a critical beach head for wholesale replacement of VistA. Cerner
now is positioned to wipe out VistA Radiology, VistA Pharmacy, Vista
Registration, and ultimately the entire VistA clinical suite. Not only
is Joe Public going to lose his several billion dollar investment in
the largest and most successful implementation EHR to date, but he is
going to be paying even more in the future for all those Cerner
licenses.  Double edged sword, indeed!

This leaves me with two questions:

First,
what could have been? Had the VA taken a leadership position on this,
by developing the software in a collaborative, open, and transparent
way they could have saved the taxpayer and themselves literally
billions of dollars. They would have posted their requirements, they
would have engaged both internal and paid external developers to
continually enhance and build the LIS and other VistA subsystems. Had
the VA involved the DOD and the IHS, and had the various departments
sought to leverage their “collective intelligence” (and collective
budgets), they could have developed a unified framework which would
have allowed the parties to “reduce, reuse, and recycle” software
across departmental lines while still address the individual and unique
requirements of each. It is ironic how this proprietary philosophy even
kills departments required to operate in the public domain.

Second,
is there an open source community that can respond to this? Is there
developer capacity, talent, and organization within the community that
can take the soon to be defunct laboratory code and build from it? Are
there financial or any other type of incentives or rewards out there
for anyone to actually do this? Is there some customer organization or
corporate entity out there who believes they can take a “good enough”
solution and actually make it market competitive?

Re-reading
the definition of diabetes above, I am struck by what an apt metaphor
it is to describe this recent news. Given the prior possibilities for
such a different future, it is painful to watch a Diabetic VistA
getting its first toe amputation.

Leave a Reply

Your email address will not be published. Required fields are marked *