It is currently Fri Mar 29, 2024 1:59 am


All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 3 posts ] 
Author Message
 Post subject: VistA SDLC Model
PostPosted: Sun May 01, 2011 10:04 pm 

Joined: Mon Apr 25, 2011 8:03 pm
Posts: 3
Real Name: Rodney H. Kay
Began Programming in MUMPS: 23 Aug 1992
I have been asked to put together my thoughts on software life cycle development as it pertains to open-source electronic medical records; specifically, those that are aberrations, manipulations, and otherwise mutations of the software from the Department of Veterans Affairs, (VistA), so here goes.

VistA, in its fully supported stage (early 1990s) followed what could best be called the Waterfall life cycle development process moving through requirements, to design and architecture, development and coding, quality assurance and software testing (or verification as it was called), implementation, and maintenance and support. This is a fairly straightforward and rigorous process and each “package” (radiology, pharmacy, lab, medicine) was released after it had been verified as a distinct “version” of the package. Once the package was released, the process moved back up-stream with changes in design that had been gathered over the time it took to execute coding, verification, and implementation. These three steps could sometimes take a year or longer, but some were pushed out in as little as six months.

Somewhere around the mid- to late-90s two things happened. The powers that were decided to make their mark on the VA by switching from VistA to a conglomerate of COTS, or Commercial, Off The Shelf software that would be more modern than VistA and more efficient. After all, VistA had been around since the 1970’s, either in the underground, with the Hardhats, or elsewhere. The second thing that happened was the scare of Y2K. What was wonderful about the Y2K scare was that we realized all the dates in VistA were based on 1700, so the year 1999 in VistA was the three-digit 299. All that happened on 1/1/2000 was the year progressed to 300. We still worked long and hard identifying and correcting all the COTS stuff that was hanging on to VistA, but VistA was solid. This (along with the price tag) put an end to the COTS crowd for the time being.

The bad thing that happened during this time is that the best and the brightest of the VistA developers recognized the warning sign and took their skills and talents elsewhere. When the “powers that were” turned around and wanted to start up-parking and re-developing packages, there were not enough developers there to take on the job. This led to a couple of new discoveries in the life-cycle development. The first was that the medical profession had moved far from the present state of VistA, and the second was the futility of trying to catch up. The developers left did the only thing they could and re-invented the SDLC without realizing the significance of the process. It was just what had to be done.












The culled down waterfall model allowed “patches” to existing software to be released quickly and efficiently. However, these released patches solve a specific problem or two at the most. This patch was not a fix, just a patch. The model evolved over time to match the fluid model of medical advances. The realization that the status quo could not solve the problems led to the creation of the term “status fluxus”.

Status fluxus is the model that says you will never find a “fix” – it is not possible. Medicine moves so fast that even with a culled down model before you got the coding complete, the problem was history and a new crop of problems were there to solve. We reached the point where “not good enough” was the innovation of the eHealth industry. We met the need of non-consumers by solving problems, one at a time, quickly and efficiently, and cost effectively with maximized ROI for the consumer. Status fluxus involves the medical professionals in the software design process, by using their input to design the next patch to the software. The ROI comes in because there is never a new “version” to buy, just ongoing patches in the status fluxus waterfall.

Over the years, VistA has been adopted under the Freedom of Information Act by countries and industries all over the globe. MUMPS became Caché, VistA became OpenVistA and vxVista and many others. The Health and Human Services department released a smaller version as the “free” EHR solution for everyone. Of course, nothing about EHRs, no matter what flavor, is plug and play, but that is a story for another time.


Attachments:
File comment: Graphic of Status Fluxus
SF_SDLC.jpg
SF_SDLC.jpg [ 18.42 KiB | Viewed 9937 times ]
Top
Offline Profile  
 
 Post subject: Re: VistA SDLC Model
PostPosted: Thu May 05, 2011 11:33 am 
User avatar

Joined: Fri Nov 05, 2010 11:20 am
Posts: 16
Location: Portland, Oregon
Real Name: Martin Mendelson
Began Programming in MUMPS: 15 Feb 1983
Rodney, thanks for your take on the SDLC - I find it useful to know more about what I'm dealing with on a daily basis - i.e., kids.

My conclusion, based on several months of installing patches, is that the current method works up to a point: the point at which interdependencies become more complex than even the developers of the patches seem to realize. For instance, when I run into a problem where a patch references a global entry that does not presently exist in our system, I'm usually at a loss to know how that entry should have been set.

If there is a place to find a record of all the interdependencies I don't know of it. Such a record, preferably in graphical form, would be an invaluable aid.

Cheers.

Marty Mendelson


Top
Offline Profile  
 
 Post subject: Re: VistA SDLC Model
PostPosted: Thu May 05, 2011 8:04 pm 

Joined: Mon Apr 25, 2011 8:03 pm
Posts: 3
Real Name: Rodney H. Kay
Began Programming in MUMPS: 23 Aug 1992
Marty; that has always been a problem - the interdependencies. Some time ago we started a project to "re-do" VistA in a more logical way, so that interdependences were replaced with portals and access nodes. Needless to say, we did not get very far - the attitude at the time was, "if it ain't broke, don't fix it".

As far as a graphical listing of all the interdepences, I do not know if one exists, but if it does, then the folks at World Vista (http://www.worldvista.org) would be the one's to know about it.


Top
Offline Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 3 posts ] 

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 31 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Theme created StylerBB.net