Is there a plan B if eHMP fails?

[Ignacio Valdes]

Is there a plan B if eHMP fails? 

From my worms eye view as a heavy private sector VistA/CPRS user and vendor I have concerns about eHMP's viability for private sector. Reasons:

1) It's late. Try to say Aviva without snickering. 

2) What I hear of its architecture it is a beast.

3) 'We are doing just fine, thank-you' with refactoring and enhancing CPRS on our own. Despite its web/mobile limitations. We are setting our all-time patients seen in one day and weekly high records regularly now.

4) eHMP is almost certainly not built for private sector use in mind and will not have our many re-factoring and enhancements of CPRS (see 3).

and finally:

5) All I really wanted was a easily deployable web/mobile capable CPRS replacement that I can extend with 3). Instead all I got was this eHMP t-shirt. 

Kidding aside this is a serious issue that needs a solution. 

-- IV

[Steven Li]

We had the backbone of AViVA back in 2004 when we doing CPRS-R (resdesign) with the initial Desktop Manager tool, but the project ran into a Budget buzzsaw and the whole OIT centralization project put the project into mothballs until ~2011 when it was resurrected and demo'd at the last Camp CPRS/University in Las Vegas by Kevin and his team.

Like MyHealtheVet, a worthy internal project got vendorized and completely blew up and bogged down.  With each succeeding cycle of new contractor the code bloated and measurable progress diminished.

Sadly, this was seen across most significant projects this decade:

AviVa > HMP > eHMP/eVisTA -- bloated 

JANUS >  JLV  -- bloated

MyHealtheVet -- slow improvements since rollout with two entire new teams 

--Even custom vendors has had troubles with funding and moving forward after significant initial pilot projects:

Roger Baker's fav vendor -- Agilex developed a viable iPad reader for VistA back in 2012 only to stagnate for 2 years.  

EPIC has a lot of similarities in code from my understanding., but with its internal development team and commitment, EPIC has developed several viable mobile applications.

On Sat, Nov 19, 2016 at 8:51 PM RRichards <rmr...@gmail.com> wrote:

Aviva  (aka HMP 1.0)  was completed and ready to roll out enterprise-wide in 2012.

It was designed and developed by the VA  team that created  CPRS, with 20 years experience in CPRS/VISTA/MUMPS   (Kevin Meldrum and others). 

It was an excellent product and should have been rolled out - as-is - in 2012. 

However, the project was then taken away from the internal VA team and given to an external contractor with no  knowledge or expertise in  of CPRS code,  the CPRS RPCs,  or the VPR RPCs; nor  did the contractor  engage any of the original HMP developers.  

 Now, four years and 160M later, its architecture has become only more complex, less performant, and less deployable. 

The VA leadership of this project stated to the public last week at AMIA (American Medical Informatics Assoociation) that the anticipated roll out in  of eHMP will be April 2017.

eHMP  will support for each VAMC  site a total of 20 end users (i.e. it will support a total of less than 3000 end-users nationally).

Yes. It has some performance issues.

RR

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+u...@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Roger A. Maduro]

This would make for a good article for Open Health News. These are a series of issues and hard questions that need to be raised as the news administration transitions into key positions. Perhaps you guys can get together and put a rough draft that I can then work with.

My question at this point is whether the purpose of eHMP is to make VistA better, or to use hundreds of millions of dollars Congress allocated to VistA development, to instead create a transition to Cerner/Leidos/Accenture?

Roger

To unsubscribe, send email to Hardhats+unsubscribe@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+unsubscribe@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Roger A. Maduro
rama...@gmail.com

[Nancy Anthracite]

My thoughts on this are that eHMP could have a local cache of patient data on
each VistA instance in MUMPS file and that cache could contain all of the data
from other VistA sites that is not the one that is being used locally where
the patient is being seen and it could largely be fetched in advance.
Reminders, etc., could be written to access that cache in real time and both
the good things about eHMP could be used locally and having that data from the
Cache would enhance CPRS as well. The speed should be excellent. If eHMP
does not succeed, the code for collecting that Cache could still be used to
enhance CPRS. For the times that there absolutely has to be a look at all
patients at a national level, then the overarching eHMP with its cache could
be accessed. I have heard an estimate that only 7% of veterans are seen in
more than one facility so that should be doable and quick.

I thing some of the members of the Senate and House Committees on Veterans
Affairs don't know what a great thing VistA is and certainly other
Representatives don't. We need to tell them so they won't spend 10s of
billions of dollars replacing it. At the very least, please send this image
to your Representative and mention that Roger Baker things is would cost $16
billion to replace the top rated EHR with one that is less highly rated by its
users just because the Commission on Care thinks another EHR will have better
cost information, when the cost information in the private sector is not even
collected the same way. And do it NOW because this decision may be made in
January.

Reference for the screen capture is :

http://www.medscape.com/features/slideshow/public/ehr2016#page=9

--
Nancy Anthracite

[Steven Li]

Nancy,

The problem is Politics -- and in this case, the politics around vendor contracts.   EHR isn't a bullit or space part where changing vendors is easily accomplished and replicated.  There is a significant 'institutional/coding memory' that is lost with each change over.   Since the loss of internal devlopment projects from Class III to Class I programs, the incremental innovation on new projects dies after the initial Proof on concept and pilot.

The true benefit of JLV and future eHMP is the ability of data aggregation from multiple sources into a single GUI.  The concentration has been on the VA/DoD concentration, but the true benefit is across civilian sector and the VA/DoD  where the vast majority of veterans has some external care.     CPRS has Remote Data to other VistA instances, but doesn't integrate it and static.  eHMPs initial step is able to bring them all together, then bring in DoD and HL7 data, like EPIC and others for the full complete picture.  --- All sitting over the base VistA architecture.  

CPRS has its own warts (like improper utilization of reminders resulting in the dreaded reminder fatigue), but its strength (and weakness) is its complexity and need for consistency in program support over time -- hence it inherently not a good fit for the DoD to maintain internally with its cycling of personnel or a rotating vendor. 

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+u...@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+u...@googlegroups.com.

[Steven McPhelan]

If we do not want to speak from hearsay, perhaps Roger should make a formal FOIA request of the VA for the amount of money spent, the number of staff involved, and the time lines expended on the modules mentioned in this email.  Such a request should be as specific as possible.  There seems to be responders on this email that have information and dates that would assist Roger in being as specific as possible.

Steve

Washington spends money it isn’t authorized to spend, takes power it isn’t given, and ignores laws it is required to execute. - Gus Bilirakis, Rep. Florida

[Steven McPhelan]

I was thinking it would be interesting to see the numbers before the vendors got involved and afterwards.  Then it would be nice to have comments from those knowledgeable about the readiness of the products prior to going to vendorization.

[Nancy Anthracite]

FYI, Gus Bilirakis, who is quoted after your signature, is on the House
Veterans Affairs Committee.

--
Nancy Anthracite

[Brian Lord]

As the tech lead for the VistA development for eHMP your understanding of the eHMP system and the amount of work involved in building the system is flawed. IF you are only looking at what was done inside vista our first patch contained over 150 brand new routines 7 support patches, and more than 30 new protocols as well as currently more the 156 new RPC calls. 

THAT is just the VistA side, there are two other major components, the JDS which is a revolutionary concept and developed software, which is also developed in Cache because it is the only language that had the transactional performance needed to meet the demands of the VA. The GUI framework is a much more capable framework that allows for customization, at multiple levels and tighter security, at the same time many more features that streamline performance at all levels of the enterprise. 

I agree the transparancy has been less than I would desire, and make no mistake this is NOT a simple system, because the solution needed was not a simple solution. This is the worlds first Medical Exchange and it is an impressive engineering achievement to anyone that understands the details of what it is.

On Sunday, November 20, 2016 at 11:08:58 PM UTC-5, RRichards wrote:

Rough figures that I recall for VA to develop HMP to near-production state before it was vendorized, where the cost was over 6X.

I can ask a few of the VA folk involved to verify. 

All the vendor did was to (1)  re-write the Java to Javascript, and  (2) re-do the GUI using legacy DoD specifications  (i.e. the abandoned iEHR specification  which called for "portlets") .   This is all 'brain-dead' re-engineering; it created nothing new.

[Brian Lord]

So if someone has specific questions I'll be glad to answer them. I think part of the confusion is people may not understand at a core level what eHMP is. It APPEARS that it is nothing more than a CPRS replacement, but that is about as true as people believing CPRS is all there is to VistA.

What is eHMP? Simply put it is a multi-tiered system that allows the cached aggregation across an entire enterprise of all available medical information to be served up to any number of users in a flexible, adaptable format. 

Or in short it is the worlds first fully functional Medical Exchange. There are three components, the medical source, which is primarily VistA, but there are DoD feeds, VLER, DAS, HDR, CDS, essentially any medical data can be consumed by the system. The GUI which is more of a framework, and allows for rapid prototyping, better security, and faster response times. Then the magic, the JSON Data Store, or JDS, which is a cached data collection of all the available data stored NOT in any form of DB but as raw JSON in a hierarchical UID subscript that allows for incredibly fast retrieval of data which then the requesting application processes to present the information in quick and understandable way.

We've entered the writeback phase the system is now taking orders and writing the information to VistA, and updating the cache in real time. So we are far beyond Plan B, it is already much more functional and faster than the JLV was.

It is a LOT of work to replace CPRS, CPRS does a great many things. The VA also does not want just a replacement, they want enhancements that have either taken too long or just plain cannot be done in CPRS.

I'm sure you all have questions. I would be glad to answer them. 

On Saturday, November 19, 2016 at 9:32:07 AM UTC-5, I, Valdes wrote:

[Roger A. Maduro]

Rafael,

More details on the "brain-dead" DoD re-engineering of VA code would be very useful. That is what I was getting at earlier. What I see is that from a technology standpoint, the VA has been taken over by the DoD and DoD contractors. Last phase of the takeover occurred recently when Lockheed Martin's consulting business was taken over by Leidos. That means that all Lockheed Martin contracts, and contractors at the VA are now part of Leidos, the prime contractor for the Military EHR together with Cerner and Accenture. And they seem to be making all the decisions.

This become clear in a recent HIMSS-National Capital Area chapter meeting in Washington (11/17). The meeting was a presentation on the future of VHA and Defense Health Agency interoperability and collaboration. The amazing thing is that there was not a single speaker from the VA/VHA in the panel. It was all either Accenture or DHA officials. All talking about VA technology strategy! One can't make this up, guys!

Here is the link to the meeting

Interoperability the Key to Data Sharing | National Capital Area Chapter of HIMSS

And here are the speakers:

Moderator:

• Dr. Wendell Ocasio, Managing Director, Chief Medical Officer, Accenture Federal Services
  

Panelists:

• BG. Gen. Jill Faris, U.S. Army's Assistant Surgeon General for Mobilization, Readiness and National Guard Affairs (Confirmed)
  
• Gail Kalbfleisch, Director, Federal Health Architecture at US Department of Health and Human Services (Confirmed)
  
• CAPT Hung Trinh, Director of Innovation and Engagement, DoD/VA Interagency Program Office (Confirmed)
  
• Lauren Thompson, PhD, Director, Defense Healthcare Management Systems, Office of the Under Secretary of Defense for Acquisition, Training, and Logistics, DOD/VA Interagency Program Office (IPO) Confirmed)
  
• Col. John Scott, Informatics Policy Director, Health Affairs, Department of Defense (Confirmed)

So who is really making technology policy at the VA and what are all those Leidos contractors really doing?

Roger

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+unsubscribe@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Roger A. Maduro]

Brian, 

Thanks. This is the best and most succinct explanation of eHMP I have every seen. At last June's OSEHRA Summit I asked David Waltman for a 500 word explanation I can give readers of Open Health News, and some clear, basic diagrams. I still have not received anything of the sort. Perhaps you can write an article based on what you just posted. 

--

--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+unsubscribe@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Brian Lord]

I will gladly get you something Roger. I'm sorry if communication has been terse. There has been some big shakeups at the VA level, and their new policies have reduced our staffing and required more work. In addition to I think there was a huge collective breath holding until the election, and now there is a scramble to figure out what the new priorities will be. 

I'll definitely get you some pics and 500 words or so, by the end of the week. If you don't hear from me Friday throw a rock at me. :-)

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+u...@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+u...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Brian Lord]

So I think it important that you know, Dr. Occasio was the medical director for Agilex, and then when they got bought by ASMR he moved to that company, and now that AFS has bought ASMR he works for them.

Wendell is one of the brightest people you will ever meet. As a medical officer he is the most astute technologist I've ever had the privilege of working with. His questions drill to the core of the problem and then force you to research something you are supposedly an expert in. As one of the Project drivers he pushed for the eHMP to try and use the latest, best, most capable solutions available. We prototyped and discarded dozens of technologies to make eHMP work, and work well within specs. We thought we were going to use MongoDB for the JDS, but it wouldn't scale the way we needed it to, We started off developing for Fyre, but moved to JSON, and lots of other improvements in our engineering. 

My point is, Wendell is not a company shill, he is absolutely not dictating anything to the VA. I personally know his only driving force, and it is one we hear and live with every day, is to build the best medical system that can be built, build it as quickly as it can be done, and to ensure it meets the needs of the Vets it needs to serve. That and only that drives him. I've been on the wrong end of his ire more often than I care to mention, but he is always a problem solver and interested in fixing the problem, not making himself look good, or doing something he thinks doesn't meet these goals. 

The VA is in turmoil. Why, I have ideas, but don't really know. Their processes are changing, their priorities are different, their rules are shifting, we are still trying to understand how this will affect our project, but also how the VA will look moving forward.

IF you have questions I am certain Dr. Occasio would be glad to try and explain. The bottom line would be, what are your questions? 

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+u...@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+u...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Concerned Citizen]

Per Nancy, VistA is top rated. AHLTA is tied for last. Lets have the guy who architected AHLTA build a new frontend for VistA. It almost smells like the goal is to create a toxic environment with a failed application that encourages a move to a COTS EMR. 

The last I heard, eHMP didn't even have enough servers to be viable. Talk about DOA...

It is all politics and money. Having the CSO move from ASMR to VA leadership probably didn't hurt. 

On Saturday, November 19, 2016 at 9:32:07 AM UTC-5, I, Valdes wrote:

[Concerned Citizen]

Thanks for posting that picture Nancy!!!

VistA is top rated. AHLTA is dead last. Lets have the guy who architected AHLTA build a new frontend for VistA. It almost smells like the goal is to create a toxic environment with a failed application that encourages a move to a COTS EMR. 

[petercyli]

What do you mean by "eHMP didn't even have enough servers to be viable"?  It seems you are implying some scalability issue.  Please be specific.

Thanks.  

[Steven Li]

Haven't been in VA for 2 years now, but my experience with JLV is that the system underestimated the load demand and the system kept hanging somewhere -- don't know if it was the server or the interface on the data pulls, but it was a constant issue.

eHMP certainly has the legs.  It was initially to site side-by-side with CPRS as an enhancement and then was pressed into a replacement with the failure of the DoD/VA joint venture.  I'm hoping that it will do well enough, but I'm skeptic about VA's intentions as the leadership has largely been replaced with vendor friendly decision makers.  

IF eHMP fails, the most ideal backup plan IMHO:   Emulate Kaiser.

---COTS EPIC with MOU on codeshare development --  best of both worlds

   VA gets to keep its  strong developed infrastructure of programmers, CACs, and informaticists. 

   EPIC gets to cherry pick the new codes, repackage them and sell to the public.   

   VA Staff+Private sector+Veterans all get an instant boost of streamlined information exchange that's already developed and the #2 EHR on Medscape.  

Not bad Plan B if the powers to be were wanting it.

--
--
http://groups.google.com/group/Hardhats
To unsubscribe, send email to Hardhats+u...@googlegroups.com

---
You received this message because you are subscribed to the Google Groups "Hardhats" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hardhats+u...@googlegroups.com.

[Kevin Toppenberg]

Brian,

It is great to hear from you.  Hope you haven't left the discussion.  

Your post gives me hope.  I think that if the VA were to try to make a CPRS-in-a-web-browser as the ultimate goal of the project, it would be doomed to failure.  That is because CPRS grew up over many decades, and it will likely take its replacement decades to mature into it's final form.  And during the transition from system A to system B, the new-comer would come up short in comparison for far too many years.

But if, instead, the new system offers a greatly enhanced scope of medical data that can be accessed during patient care, then users will flock to it.  Yes, they might drop back to the older system for the odd task that the new system doesn't support yet.  But they want the new stuff because only there can get the latest data.  

Thanks for the summary, and best wishes.

Kevin T

On Monday, November 21, 2016 at 9:12:48 AM UTC-5, Brian Lord wrote:

So if someone has specific questions I'll be glad to answer them. I think part of the confusion is people may not understand at a core level what eHMP is. It APPEARS that it is nothing more than a CPRS replacement, but that is about as true as people believing CPRS is all there is to VistA.

What is eHMP? Simply put it is a multi-tiered system that allows the cached aggregation across an entire enterprise of all available medical information to be served up to any number of users in a flexible, adaptable format. 

Or in short it is the worlds first fully functional Medical Exchange.

-snip-

[Nancy Anthracite]

At an estimated $16 billion to replace VistA , move all of the Veteran data to
a new EHR and add all of the extra data fields necessary to make the COTS
system support Veterans and replace all of the applications that another EHR
does not have that VistA has for Veterans and the VA, do you honestly think
Epic is worth it as the SECOND best EHR according to Mescape.? I don't think
it is worth it even if eHMP fails.

Even if eHMP does fail, there will still be some valuable artifacts that come
out of the project - most significantly the patient data Cache technology that
can be leveraged.

Then there is the question if the the move to Epic would be successful.

--
Nancy Anthracite

[Steven Li]

In a perfect world, true.  VA is not that world.  During my years there I've seen VistA develop from VT100 green screen through eHMP and everything in between.  CPRS' guts with 40 versions later are essentially the same GUI since Windows 3.11.

The reality is that eHMP was initially designed to by an add-on system and now pressed into service as a replacement after a five year failed joint DoD/VA venture.  It may accomplish its mission if given the resources; but I believe Congress (and now the VA decision makers not favorable to internal development) won't have the patience needed to see this through and seek a vendor solution.

IMHO, EPIC would be the ideal solution as precedence has been set on maintaining an independent developing structure that VA has and stepping up to a full functioning EHR that includes HIE, mobile integration, and comprehensive communication and scheduling suite.  

[Nancy Anthracite]

And that is worth $16 Billion?

--
Nancy Anthracite

[ivaldes]

Question 1: The code in CPRS classic that governs whether the Encounter form pops up or not is very complex 'spaghetti'.  What is its equivalent in eHMP? How is this complexity handled in eHMP?

Question 2: Orchestration among patient scheduling for schedulers and CPRS classic is very important for efficiently seeing patients. What is the user interface equivalent in eHMP? Does it work now? 

Question 3: Is a Surescripts client capability currently implemented in eHMP? 

Question 4: The Team list capability in classic CPRS is only partially built out (no connections with Encounter, all users or only self can view a list granularity, no Next or Previous cycling capability). Has it been fully built out in eHMP with these items?

Question 5: In eHMP clinical Notes is there mixed typeface, bold, italic, graphic embedding capability? Or is it legacy ASCII only?

[r...@rcresearch.us]

What we could do with $16 Billion Dollars. CHCS-1 only cost $2 Billion
and some change to do 200 hospitals around the world and the customer
changed platforms at least 3 times. PDP-11 >> Vax >> Alpha >> SCO Unix
on Commodity Servers in various configurations. What we could do with a
fraction of $16 Billion.

[Nancy Anthracite]

The CPRS team has wanted to do more to improve CPRS for a long time. If some
of the money that they seem willing to put into putting into a COTS product
could be put into beefing up the in-house teams and letting them do more of
what they and the users want, that would help during the transition, if there
ever is one, but I don't think that if eHMP fails that that should mean the
death nel for VistA. It does not make sense to me to spend $16 B because
there is not web based interface. Not to mention that we still have the
option to buy the Javascript based CPRS if they need a web interface for some
circumstances so badly.

I think it would save money to bring eHMP in house if that proves viable. I
think it likely that eHMP will need to be run with each instance with a local
cache in addition to having a national eHMP for use for some limited
circumstances, but that will remain to be seen.

--
Nancy Anthracite

[Steven Li]

That $16Billion is an old number thrown around by Roger Baker to justify open source usage a few years ago.  The most recent contract bid for DoD was $9Billion and Congress is going to use that as a number to throw around.

I'm all for the next Gen in-house CPRS and believe the cadre and spirit that created CPRS can be replicated again IF GIVEN THE RESOURCES and ability to bring people together.  

Sadly, I do not believe the climate will support such.  Last year Congress knee capped VistA funding and all the political winds are blowing the COTS direction ever more strongly this year and their Plan B is COTS.   I sincerely hope it's NOT another Cerner-BigBlue - ALHTA style choice; and the Plan B would be a compromise that keeps long term talent in house instead of a pure vendor system -- which in recent products have had marginal cycle enhancements after the initial gains.  

[ivaldes]

Second chances are what life is all about.