Many companies run legacy systems that cause them a lot of trouble when it’s time to expand. The original systems were designed to certain specs, which are now wildly out-of-date. So what do they do? Invest in an expensive new system and risk colossal failure? Or try to keep the existing system running with a few patches here and there?

This is essentially the problem facing the US government — on a huge scale. 

The Veterans Administration and the Department of Defense are coming under sharp criticism for delays in developing a joint, paperless system for handling veterans’ medical care. Under the sobriquet #EndtheVABacklog, veterans have brought national attention, ranging from news programs to Congressional hearings, to the delays, which average 273 days and are not expected to be cleared up before the end of 2015. Reportedly, 97% of VA claims are still on paper.

The VA backlog is not actually new; similar complaints were made last year, with similar figures and similar horror stories. “Last year, veterans filed more than 1.3 million claims, double the number in 2001,” wrote the New York Times in September, 2012. “Despite having added nearly 4,000 new workers since 2008, the agency did not keep pace, completing less than 80 percent of its inventory. This year, the agency has already completed more than one million claims for the third consecutive year. Yet it is still taking about eight months to process the average claim, two months longer than a decade ago. As of Monday, 890,000 pension and compensation claims were pending.”

What is new now is a recent revelation from the DoD and the VA that, rather than developing a joint system as they had planned since 2008 — the estimated cost of which had ballooned to as much as $12 billion — the two agencies were instead going to develop a common graphical user interface. This would enable the VA to continue using its award-winning, though 25-year-old, Veterans Health Information Systems and Technology Architecture (VistA) system. The DoD was going to use some system upon which it hadn’t yet decided but which might be the VistA system.

In response, the backlog has received a new swarm of Congressional and media attention. Organizations such as the Iraq and Afghanistan Veterans of America (IAVA) also met with a number of Congressional representatives this month. In addition, the Government Accountability Office did a study, “ELECTRONIC HEALTH RECORDS — Long History of Management Challenges Raises Concerns about VA’s and DOD’s New Approach to Sharing Health Information,” on the two agencies’ efforts.

While Congress criticized the DoD for not simply saying it would use the VistaA system, agency representatives pointed out that it would still be considered a new acquisition and that they lacked the finances to implement it. Moreover, because it’s not a commercial system, there isn’t really a support system for it. Finally, it is behind the state of the art in electronic health records technology.

The VA said it is working on a paperless system that is currently in 25 of the agency’s 56 regional offices and is expected to be in all of the offices by the end of this year. Offices that already have the system indicate that it significantly cuts the amount of time required.

The backlog problem isn’t entirely the fault of the two agencies. “Claims submitted by Post-9/11 Veterans include more than double the number of conditions claimed by Veterans of the Vietnam era,” writes the VA. “To put this in perspective, in 2009, VA processed claims for 2.7 million individual medical conditions. Last year, that number had grown to over 4 million.” In addition, veterans of previous wars are living longer and many are filing for benefits for the first time. Finally, the agency decided to expand the number of illnesses presumed to be linked with Agent Orange exposure, which added a great deal to the backlog.

Although members of Congress pontificated during the hearings that of course no expense should be spared to take care of our veterans, it’s not clear they would have felt the same had the agencies actually spent $12 billion to develop the system — particularly with the issue of sequestration hanging over the proceedings.

In addition, while the subject didn’t really come up during the hearings, there’s no guarantee that the $12 billion system would have worked or been on time. Using the principles of Agile development, one could argue that the simplest way to develop the program would be to take existing systems and find a way to have them work together, rather than to develop a new from-scratch system.

Certainly government has no end of software failure horror stories of its own, ranging from last December’s $1 billion Air Force failure — even using, as the DoD has suggested for the VA, commercial off-the-shelf software. Like the VA, the Air Force is also planning to meet its 2017 goals using its legacy systems. Similarly, the California Department of Motor Vehicles in January scrapped its IT Modernization System after spending $50 million on it.

“We ignore the biggest objectives, which are cost-effective ways of producing and delivering government services,” writes Gartner vice president Jerry Mechling in Governing, in an analysis of why government IT projects so often fail. “What we really want isn't better technology but better outcomes, such as reduced crime, increased academic achievement, improved health care and efficient tax collection.”

The VA and the DoD are facing some almost impossible decisions. What does seem clear, however, is that a lot more attention is going to be paid to the EHR issue and developing interoperable systems.

Related Posts