It doesn’t take long for technologies to outgrow their usefulness. IT systems need constant tweaking, updating, or replacing, or else they become an inefficient drag on the organization. Of course, that’s much easier said than done – especially in the public sector, where funding decisions and approval processes trail far behind the tech cycle.
Take the recent Department of Veterans Affairs scandal. While organizational culture issues took center-stage, the agency’s antiquated systems for personnel management, appointment booking, and medical histories probably made things worse. And the delay in processing veterans’ records was exacerbated by the Department of Defense and VA using incompatible systems for medical records. But as former Air Force CIO Bill Lord told me, updating IT doesn’t match an agency’s central mission in terms of priorities. Before retiring in 2012, it was a balancing act he knew well. For the last three years of his career, he was responsible for the $6 billion portion of the Air Force budget dedicated to IT, where his main responsibilities ranged from Air Force cyber activity to IT upgrades to career field management. Below is an edited version of our conversation.
Give us a sense of the challenge of keeping up with technology as a military CIO.
Part of the day-to-day existence of the CIO is figuring out how to integrate modern technology. In the Air Force, it’s like changing an aircraft’s engine while it’s flying – no downtime, no interruption of service. We also had to plan the budget cycle into everything we did. So larger projects had a long lead-time, which conflicted with the rapidly changing tech cycle.
One of the things that IT departments are after, and I think it could be related to the Veterans Administration, is efficient use of those budgets. In my estimation, the Air Force was spending almost 70% of our budget on legacy systems – that is, maintaining software applications that were written decades ago. Potentially, the VA has hundreds of applications that are ancient. Modernization of some of those legacy systems could be a solution to free up more resources to fulfill the primary mission, which is health care for veterans.
So why has VA relied on these antiquated applications for personnel management, for appointment booking, for keeping track of medical histories, and so on, for so long?
I think it’s probably the ability to get IT dollars. It’s a budget decision, which depends on what the principle operation is. I know in the case of the Air Force, the priority is to fund flying and the modernization of flying, as opposed to funding the modernization of IT systems supporting flying. I can only assume that’s analogous to how the VA’s primary operation is healthcare, not IT operations. So CIOs always have to balance supporting current operations and enabling those supporting operations to be more efficient.
I can imagine that balancing act is frustrating.
It can be. There’s technology that can perhaps help solve these problems within the VA, and there could be IT systems in the VA that could be modernized for efficiency and cost savings. But this involves investing money into areas that are sometimes the last to be considered. That’s the bane of the existence of the CIO.
So how do you get prioritized higher in the pile? Tell leaders, “Here is the return on investment if you invest in this technology or this application or in this modernization of our ancient system.” But this involves throwing away the old system and programs that don’t work, which is always very hard to do in government.
Why is that so difficult?
The concept of sunk cost, which really is not one that you should consider, but one you have to answer to with the press. It’s difficult to face that pressure and try to fix the problem at another great expense. It has to be decided at the CIO level.
For example, the Expeditionary Combat Support System [a resource planning software project undertaken by the Air Force between 2005 and 2012] just didn’t work. The program was about five years in the making, and almost $40 million had been spent, but it took several years before the problem was escalated to the level of the Secretary of the Air Force and the Chief of Staff. They finally tossed it last year. It was the absolute right decision, but the ECSS was supposed to eliminate about 300 legacy systems, so when the program was cancelled, those legacy systems now have to be sustained and updated all independently – potentially costing millions a year.
Going back to the subject of the VA scandal, can you explain the IT connection – or lack thereof – between the Department of Defense and VA?
As all military personnel on active duty transition into the VA medical health care system upon retirement or medical discharge, all of their medical records have be transferred over to VA. But today, the two systems are not compatible. The Air Force system of records used a different format than the VA’s system. The Office of the Secretary of Defense’s CIO is working with the medical CIO community to align those various (Army, Air Force, Navy, and USMC) systems with the VA’s.
They’ve spent a lot of money doing it, and it’s not yet completed. So you could have potentially had an IT solution for folks who were waiting for appointments because they were waiting for records. And therefore, some backlog may have possibly been minimized.
How long has this been a problem? And how long has this effort to align the systems been going on?
The OSD CIO has been working with the VA for a couple of years on this problem, but I don’t know how far along they are. While there is a good bit of digitizing today’s records involved, you also have to use the same system for digitization so that data can easily flow between the two departments. There are contractors gathering money to write software that will not only connect to the two, but also make the two medical record systems look the same.
HIPPA also forced the Air Force to closely scrutinize our networks so we could protect personal health care data being transmitted between DoD health care providers and [other] medical professionals. We had to ensure the data was in a format that was not easily compromised. That’s what began this modernization of records between the two of us, because we had to do it for HIPPA. They fall under the same HIPA requirements, the DoD and the VA.
What new IT and management systems would you recommend to make the VA run more efficiently, to cut costs, and to prevent future problems?
In my Air Force experience, the software systems that are the oldest had the most to gain by modernization. The tools that EvolveWare provided us eliminated a lot of dead code and updated the business rules, which were often very outdated. And metrics need to be produced automatically, by embedded software, not PowerPoint slides created by humans. If the electronic system that the VA uses for appointments could also create the metrics like wait times, then you rely on the system to produce it and avoid human interaction – who’s reporting the wait times – which is maybe where some of this went awry.
So by making the system more efficient with steady metrics, you get rid of the human element, and data wouldn’t be so vulnerable to corruption.
Right. And I don’t think it’s hard to get systems to produce those kinds of metrics. But once it’s designed, metrics have to be defined by the senior leader. That way you have configuration control and governance of software that’s not left up to one or two people in any department. Then there has to be a culture of adhering to them collectively and not under any cover of darkness, so that you can avoid the manipulation of data. And there’s also a culture change of having to believe your instrumentation, which is sometimes problematic. Luckily for me, one of the things they teach Air Force pilots is to believe your instruments.
What should the VA’s priorities be now to bounce back?
From an IT standpoint, I think it’s looking at how they can become more efficient, so funding can be spent on healthcare and not on legacy IT systems. It’s convincing leadership of the necessity to do that. And the way to do that is to have a process that looks at everything you’re doing in terms of IT – medical systems, appointment and record systems, personnel information, and even to the way they manage their fleet of vehicles.
Then leaders need to assess whether to kill those systems, update them, or get new ones. We need to accelerate the work we’re doing between the VA and DoD. They need to review everything that we do in the IT business and what governance decisions need to be made. Do we fund or not fund? How do we prioritize that? And then how do we convince senior leadership that it’s the right thing to do?