At the end of last month, President Trump dismissed VA Secretary David Shulkin, MD, following weeks of speculation that he would do just that. In Shulkin’s place, Trump nominated his personal White House physician, Rear Admiral Ronny Jackson, a physician who’s never managed a large healthcare organization.
Shulkin’s dismissal left unsigned the contract with Cerner to replace the VA’s venerable VistA healthcare IT platform. It’s not that the Cerner deal is really threatened without Shulkin—VA watchers still expect that his replacement will sign the contract—it’s that no one can be sure when that might happen if the Senate balks at confirming Jackson due to his lack of organizational management experience and other budget and programmatic concerns.
This VA leadership lull provides an opportunity for reflection. Specifically, it’s worth asking while we have the time whether Cerner is the right path for the VA to take. While the decision may seem like a no-brainer to some, the VA’s situation is unique and arguably calls for a singular approach to both existing organizational issues and a major healthcare IT decision.
Specific to the Cerner decision, what should the new VA secretary consider?
– How the current VA staff feels about VistA: When Shulkin was dismissed, Healthcare IT News seized the opportunity to ask readers for their opinions. Among VistA users, a majority said they want to keep the existing system and make necessary changes. This is consistent with a separate inquiry Healthcare IT News conducted almost a year prior. In that article, the author references Medscape EHR reports from both 2014 and 2016 in which VistA was the top-ranked system. Interestingly, the lowest ranked system in the same survey was AHLTA, the DoD system currently being replaced with Cerner.
While no one argues that VistA doesn’t look outdated, it’s what the system can still accomplish that has users defending it. Advocates like that the system can be rapidly changed, that it supports up to 2,500 concurrent users, that is has a 99.5 percent uptime, that it gives clinicians a comprehensive view of patient information.
“… at least it was built more for the clinician than the biller,” said one VistA user on a Reddit forum. Indeed, the fact that VistA was built from the ground up as a clinical, not billing, tool segues right into a second consideration.
– The origins of VistA creation: In a fee-for-service healthcare system, the greatest priority for doctors after providing quality care is getting paid for it. That’s not a knock on doctors; it’s an acknowledgement of the American system. The VA, however, is not a fee-for-service system, so there was never an incentive to make billing the primary technical objective.
“The Hardhats’ (nickname for original system developers) key insight—and the reason VistA still has such dedicated fans today—was that the system would work well only if they brought doctors into the loop as they built their new tools,” writes Arthur Allen in a Politico VistA profile. “In fact, it would be best if doctors actually helped build them.”
Even if it is the less-attractive system, a situation the VA can fix with the right contractors, VistA is still clinically more effective than most alternatives. So which is the priority, beauty or efficacy?
– His obligation to responsibly use taxpayer funds: The initial agreement between VA and Cerner came in at $10 billion, then later ballooned to $16 billion. Anyone with even a cursory knowledge of large government IT projects knows costs will rise, and there is always the risk that the money will be spent and the project will fail. No such risk exists with a system that is already implemented like VistA. And, to be clear, the risk is significant. Similar projects embarked on by the Coast Guard and Epic collapsed. Many healthcare IT analysts have even predicted failure for Cerner before they even get started.
Why? Because there are several ways in which the VA’s Cerner project is risky. Sure, many call it a technological endeavor and evaluate it as such, but the human resources challenge is equally daunting if not more so. Many VA clinicians and staffers feel passionately about their mission and VistA, and they may not take kindly to learning a new system they don’t want or the way in which Cerner became the ultimate winner.
– The way the original Cerner deal came together: Yes, the legal challenges to Cerner’s sole-source contract with the VA were dismissed, but resentment will still linger. If I’m a long-serving VA clinician, a decision that didn’t really even consider VistA looks questionable or worse.
“No-bid contracts don’t just breed government profligacy, they also open the door to brazen acts of fraud and corruption,” said David Williams, president of the Taxpayer Protection Alliance, when the Cerner contract was announced.
As Williams points out, a no-bid deal gives Cerner more leverage than it would have in a competitive process. The fact that DoD went with Cerner gives the VA a precedent to lean on, but these wo systems will not be even remotely the same and will still rely on interoperability standards to facilitate communication, just as they would to connect two separate systems. Indeed, Shulkin signing the actual contract was delayed for months as the VA and Cerner ironed out interoperability details. Why was that necessary if the original argument for Cerner was mostly, “Because DoD is using it”?
– The future of healthcare IT: The stickler for healthcare IT moving forward is interoperability. The key to solving this challenge system-wide is not to make sure every clinician uses the same system. That’s just blatantly not the way to rectify a technological challenge. The choice of Cerner does not automatically make communication with DoD easier, it just provides the illusion that it will. What if the federal government were to upgrade VistA, make it available as a comprehensive IT platform and contribute to establishing standards that ensure interoperability? That sounds more like a healthcare IT future that can benefit all involved.
– What agency autonomy feels like: It’s not just that the development of VistA since the late 60s is a feel-good story and a counterweight to the idea that the federal government can do nothing right. Sure, VA has had problems, but VistA has not really been one of them. When VA made serious reform strides in the late 90s, VistA was the tool that made them possible. Moving forward, IT projects at VA will involve a third party whose primary interests are not providing the best possible care for veterans. More than just a technical change, moving away from VistA could be relinquishing a level of control VA will wish they could get back.
Any analysis of the decision to replace VistA is incomplete without viewing it in the context of an effort to privatize the VA, which Shulkin alleged shortly after departing the agency. If that truly is the goal, then the Cerner contract makes little sense. Why spend $16 billion on IT for an agency you’d like to shutter?
But then, why spend $16 billion on a roll of the dice when you already have a platform in place that works and clinicians endorse? Yes, VA has some challenges, but pretending that VistA is one of them just detracts from improving veteran care—the highest priority.