BOB WACHTER, MD

The policy known as Meaningful Use was designed to ensure that clinicians and hospitals actually used the computers they bought with the help of government subsidies. In the last few months, though, it has become clear that the policy is failing. Moreover, the federal office that administers it is losing leaders faster than American Idol is losing viewers.

Because I believe that Meaningful Use is now doing more harm than good, I see these events as positive developments. To understand why, we need to review the history of federal health IT policy, including the historical accident that gave birth to Meaningful Use.

I date the start of the modern era of health IT to January 20, 2004 when, in his State of the Union address, President George W. Bush made it a national goal to wire the U.S. healthcare system. A few months later, he created the Office of the National Coordinator for Health Information Technology (ONC), and gave it a budget of $42 million to get the ball rolling.

The first “health IT czar,” David Brailer, focused on convening stakeholders, banging the drum for computerization, and creating standards for health IT. The seemingly arcane matter of standards turns out to be crucial, since only through a common language and protocols could computer systems have any shot at sharing data with one another (“interoperability,” in IT-speak). This is not a new issue in the world of technology: a protocol known as TCP/IP was central to the success of the Internet. And standards are why your light bulbs and electrical plugs fit into their respective sockets when you bring them home from the hardware store.

Brailer did what he could with $42 million, but – when you think about trying to change the course of the $3 trillion dollar a year U.S. healthcare system – there was only so much he could do. Within five years, however, the ONC’s budget received an injection of new resources, and not a small one: from $42 million to $30 billion. The story of how that happened is an amazing blend of happenstance and opportunism.

In 2008, when Congress passed a $700 billion stimulus package to rescue the economy, it chose to spend the money on “shovel-ready” projects – those whose plans were on the drawing board. Tucked among the highway construction and railroad repair jobs that were funded was a project of a different sort: $30 billion to help the American healthcare system go digital.

Policymakers, concerned that doctors and hospitals might buy the computers with federal money and not use them, attached a very big string to the money: a set of criteria that IT vendors and those buying IT systems needed to meet to quality for the federal bucks. These criteria became known as Meaningful Use.

Let’s take a deep breath and review the state of the universe, circa 2009 – particularly in light of what we know today: that Meaningful Use has become the most controversial, even vilified, policy initiative in the health IT world, perhaps in all of health policy (okay, maybe second to Observation Status and the SGR). In 2009, very few people would have argued that it was a good idea to create a detailed set of government regulations dictating how doctors and hospitals should build and use their electronic health records. But that is precisely what MU has done. Some slopes are, in fact, slippery.

And yet, putting myself in the place of the 2009 decision-makers, I don’t see any villains, or even any particularly egregious blunders. It’s just that things have gone off the rails, which is why we now need to change course.

Okay, back to 2009. The first question: Was it a good idea to use federal money to promote health IT? My answer is yes. In 2008 only about 10 percent of hospitals and doctors’ offices had electronic health records. As long as Congress was spreading $700 billion of federal fertilizer around to stimulate the economy, why not use some of it to rectify this market failure? I think the health IT incentives were sound policy.

Second question: Did we need a set of standards to accompany these incentives? Here, too, my answer is yes. There had been earlier initiatives, mostly by private insurers, in which doctors were given “free” computers and simply put them on their shelves. That, of course, would have been scandalous when scaled up to federal size. So Meaningful Use, as a policy, made sense.

The third question reflects the common complaint that the federal incentives drove the purchase of “immature IT systems” – and that we should have waited until the systems were more mature. Here, too, I’m unpersuaded. A program with a longer timeline would likely not have met the shovel-ready requirement. Moreover, the vendors had been working on their EHRs for decades (Epic was founded in 1979); a couple of years’ delay wouldn’t have gotten the systems any closer to perfection. The only way that health IT was going to get better was to implement the best systems and improve them, guided by insights born of real-life experience.

So, given these facts on the ground in 2009, I believe the policy decisions were sensible. And for a while, everything went pretty well. Meaningful Use Stage 1, implemented in 2010-12, consisted of achievable standards designed to ensure that EHRs were being used effectively. But it was not so prescriptive as to stand in the way of the primary goal, namely, wiring healthcare. Adoption rates soared and MU ensured that the computers were being used.

With Meaningful Use Stage 2 (2012-present), things went sour. The standards became far more aggressive, veering far more deeply into the weeds of clinical practice. MU now dictated how doctors should give out handouts to their patients (they must be prompted by the computer). It held doctors and hospitals responsible for ensuring that patients viewed and transmitted their data to third parties (most patients had no idea how to do this). It forced EHRs to meet onerous disability access requirements. All of these are noble goals, but all are bells and whistles – the kinds of changes you make after you’ve nailed the basics of getting the darned machines to work safely and efficiently. I spent a morning in June watching Christine Sinsky, a primary care doctor in Dubuque, Iowa and an expert in practice redesign, struggle to survive the regulations. While the ONC’s goals were laudable, she said, meeting the MU requirements had become “like [solving] some riddle or puzzle. Life is hard enough. Why are we making it so much harder?”

This July, Karen DeSalvo, the director of ONC, told me that her office was looking to scale back the MU regulations. Jacob Reider, ONC’s deputy director, using a delicious euphemism, also conceded that the MU Stage 2 requirements were overly “enthusiastic.” While I appreciated the forthrightness of the ONC leaders, I wondered whether they would achieve their goal. After all, scaling back is not among the core competencies of government bureaucracies.

In the past month, both DeSalvo and Reider left ONC. DeSalvo is now the acting Assistant Secretary of Health, taking a leading role in the Ebola response (after announcing she was leaving ONC, the Department of Health and Human Services clarified that she’ll still be involved in its policy decisions, though will no longer have day-to-day management responsibility). Reider simply resigned. In the past six months, in fact, more than half of ONC’s senior personnel – its chief scientist, chief nursing officer, chief privacy officer, and director of consumer eHealth – have jumped ship.

Why? The last of the $30 billion will be spent by the end of this year. While it would appear that ONC will lose its power of the purse, it’s not that simple – the plan has been that, starting next year, rather than receiving a bonus for implementing an EHR that met MU standards, doctors and hospitals would begin seeing Medicare cuts if they didn’t meet them. But with Medicare already slashing payments through a variety of other mechanisms (value-based purchasing, no pay for errors, readmission penalties and the like), many people think that it won’t have the stomach to penalize hospitals and doctors for failing to meet the increasingly unpopular MU standards.

Unpopular is an understatement – the Meaningful Use program has clearly lost the hearts and minds of clinicians and CIOs. As of this month, only two percent of eligible physicians and about one in six hospitals have successfully attested to MU2 requirements. Even former supporters have taken to calling the program Meaningless Abuse. Not good.

In the face of all these challenges, ONC appears adrift, stripped of its resources as it tries to administer a failing program. It’s no surprise that its leaders are rushing the exits.

What should become of ONC and Meaningful Use? The key thing to remember is that MU was an accidental program – one that never would have happened had the economy not tanked in 2008. So rather than trying to salvage it by tinkering around its edges, it is time to rethink the whole shebang. In this, I agree with John Halamka, the Chief Information Officer at Beth Israel Deaconess Hospital and the chair of several HIT policy committees, who believes that ONC should “declare victory” and markedly pare back Meaningful Use.

Declaring victory would not be unreasonable. Against the primary goal of wiring the American healthcare system, ONC’s program worked – the number of hospitals and doctors’ offices with functioning EHRs skyrocketed from 10 percent in 2008 to approximately 70 percent today. The health IT market is far more vibrant than ever before. Even Silicon Valley – which has always given healthcare a cold shoulder – has now joined in the fun, with major health IT initiatives at Apple, Google, Salesforce, Microsoft, and in garages all over San Francisco.

Rather than continuing to push highly prescriptive standards that get in the way of innovation and consume most of the bandwidth of health IT vendors and delivery organizations, MU Stage 3 should focus on promoting interoperability, and little else. Last month, an expert panel presented ONC with a reasonable set of recommendations calling for standardized, publicly available application programming interfaces (APIs), the EHR version of standardized light sockets. This change would allow EHRs to communicate with each other and developers to write apps that could link to the large systems like those built by Epic and Cerner. Promoting this kind of interoperability would be a judicious role for a smaller, less muscle-bound ONC, and for MU Stage 3.

Ending the prescriptiveness of MU doesn’t mean abandoning the goal of using EHRs to improve healthcare. Now that the vast majority of U.S. hospitals have EHRs, the stage is set to promote the outcomes we care about through Medicare’s existing programs – without micromanaging the technology. When Medicare publicly reports adherence to evidence-based practices, hospitals with health IT systems will install decision support to meet those standards. When Medicare penalizes hospitals for excess readmissions, hospitals will create electronic links to primary care clinics and nursing homes. When Medicare ties patient satisfaction to hospital payments, healthcare system will offer their patients access to laboratory results, x-rays, and on-line scheduling, to say nothing of email and telemedicine access to clinicians. When ACOs live or die based on their efficient use of resources, they will implement computer systems that help them conserve resources. It is the outcomes we care about, and hospitals and doctors should be free to use whatever IT tools (or other non-IT strategies) to achieve those outcomes. That’s the best path forward.

By the end of this decade, I believe we will look back on the 2009-2014 era and see government intervention – particularly the $30 billion incentives and the early years of Meaningful Use – as having helped transform medicine, finally, into a digital industry. As our IT systems get better and our processes and culture adapt, this transformation will end up improving patient care and, eventually, saving money, notwithstanding our rocky start.

So today, even as we struggle with Meaningful Use, let’s praise the government for knowing when to intervene. Let’s hope that tomorrow, we can praise it for doing something far harder: knowing when to stop.

* * *

Note to THCB readers: Bob has spent the last eight months writing his book, “The Digital Doctor: Hope, Hype, and Harm at the Dawn of Medicine’s Computer Age.” It is done (and the early feedback has been gratifying) and will be published by McGraw-Hill in the spring. (You can pre-order it now on Amazon if you’d like.)