The Time-Warp

jonathan bushJohn Gage, Sun Microsystem’s fifth employee and its former chief researcher, famously said “the network is the computer.” The majority of us experience this every day through interactions with a wide variety of highly-intelligent, super-connected networks including Facebook, which remembers our friends’ birthdays better than we do; ATM networks, which know instantly if we have the cash that matches our request; and the complex, yet seemingly simple interweaving of phone networks, which allows us to communicate smartphone-to-smartphone regardless of carrier. Sadly, healthcare struggles to grasp this important concept.

Earlier this month, I flew to Utah for a conference hosted by KLAS, a major healthcare research outfit, about interoperability. Interoperability is a clunky word that’s talked about endlessly in healthcare, but at its root is an important notion: health care information needs to flow freely. Interoperability means that important information isn’t stuck in proprietary enterprise software that a hospital spent millions of dollars buying years ago. Having this information in the right place at the right time equates to reduced risk of medical errors and makes the delivery of health services more efficient and less costly. I’m convinced more than ever the only way to free information from the silos where it’s currently stranded is for the industry to embrace connectedness by switching to cloud-based, open networks.

The goal is clear. Yet healthcare IT executives and those buying their products remain stuck in the old ways of thinking. In their minds, software is still the computer, and sunk costs keep it so. As such, health information is largely trapped on technology islands that are maintained at great expense onsite at hospitals across our country versus flowing across the care continuum via a universally available information network. Just how bad is the data jam? An Epocrates’ survey earlier this year of nearly 3,000 physicians found that only 14 percent of physicians can access usable electronic health information across all care delivery sites and six out of 10 doctors, even when in the same organization, aren’t effectively sharing information.

I was in Utah with key executives from the country’s leading health information technology providers – including Cerner, Epic, GE Healthcare, Meditech and others – to talk about how to improve interoperability and how to measure it. Yet the conversation among my colleagues and competitors quickly foundered. We’d gathered to talk about interoperability, yet no one believed that we could measure it because we all run different systems. Instead, the consensus was, we’d have to make do with a subjective survey of hospital chief information officers.

To me, this is the ultimate irony, and represents a complete failure to move into the modern era. I felt as if I were sitting in a room of record executives pushing back against the inevitability of the switch to digital music. I wanted to shout at all of us to wake up, and see how the old ways of doing business are being disrupted. Information will move online in healthcare, just as it has in other industries, including financial services, which has similar privacy concerns, whether health care providers and vendors like it or not.

There has been talk in Washington about requiring a certain level of information sharing among electronic health records companies. I believe, however, that a better solution will come from the market. One way to spur change is to try and gauge, quantitatively, how well information is flowing and make known the positive things happening as a result; rather than backing away from doing so because the effort is too hard or, perhaps, too embarrassing.

That’s why I decided that athenahealth would voluntarily report both our total automation rate (that is, the amount of information we are able to process electronically versus manually because of the web of connections we’ve built from our network to interact with the rest of care continuum), as well as the number of actual connections our network offers to those on it. Today, with more than 95,000 network connections with the likes of pharmacies, labs, hospitals, health information exchanges, other vendors, we are able to automate approximately 59 percent of our work. We believe that by disclosing these numbers – the only company to do so, that I know of – we can pressure others to do the same. And I believe strongly that, over time, as healthcare information moves online, those numbers can only go up and will become a competitive differentiator.

To be fair, things aren’t as bad as they were a few years ago. Some of the more egregious data-blocking practices, like the practice of charging exorbitant fees, are going by the wayside. Meanwhile, traditional software players are teaming up to create connected islands of information from which data can flow freely. athenahealth, is, in fact, building stronger connections to historically closed systems. Yet these moves to build information channels are akin to putting a Band-Aid on a broken arm: The entire enterprise software system in healthcare is broken.

With the cloud, the network really is the computer. We see this on athenaNet, which has 67,000 providers, serving nearly 70 million patients. Not only does everyone on the network benefit from access to the same information channels, but the connections also allow lessons to be learned and solved universally. For example, if a doctor in Minnesota gets a claim denied because of some medical coding requirement that she didn’t know about or because of a new insurer rule, that knowledge can be learned network-wide so that when those circumstances are recreated that denial won’t happen again.

What can be done? Committing to advancing connectedness is the first right step, but health care requires a paradigm shift related to the technology it uses and invests in. Health care executives must learn from other industries that the free flow of information is more than just some ethereal ideal. It must become a reality and we must measure it. Interoperability means that health information is unshackled so that it can go where it’s needed, securely and reliably, just as information does in other areas of our lives. Only then will we be able to say that we’re living up to the words that John Gage so presciently said so many years ago. “The network is the computer:” it’s time that health care meet the network.

Jonathan Bush is the CEO of athenahealth.

Categories: Uncategorized

Tagged as: ,

12 replies »

  1. So among our goals, there’s interop, there’s absence of data blocking, and there’s broader “data rights.” The 21C Cures Act language defined interoperability, in part, as the absence of data blocking — so they are related — and the Trust IT Act effectively makes both interop and lack of info blocking prerequisites to continued marketing and use of an EHR.

    As far as criminalizing limitations on full health data access goes, John, I think that’s a reach … But I get that the perspective is an outgrowth of framing health care as a right, and the fact that health care can’t be fully accessed without full and free access to health data. As you know, I agree with you in principle: access to health data is very important. However, in our legal system, healthcare isn’t really a “right” and withholding health data isn’t going to be criminalized any time soon. (I imagine someone may seek to creatively jam withholding of health data into a False Claims Act sort of claim; such a claim would be buttressed if Medicare CoPs included a “patient must get unrestricted access to data” requirement. MU has the VDT requirement, but it’s been cut back.)

  2. Dr. Palmer,

    I would tend to disagree with the premise that physicians only rarely need information from an outside (“non-local”) database. I work in a semi-rural emergency department, but we are only about 30 miles from a metropolitan center with several tertiary facilities, including a large teaching hospital.

    Frequently we see patients in our ED that were just discharged from one of these facilities, or at least are actively followed by their specialty physicians. Having at least SOME access to a discharge summary, a CT or echo report, or basic labs would often make our job much easier. It would also significantly reduce the redundant studies needed in our ED, reducing cost, time, and invasiveness to the patient.

    I agree this is not on every patient. But it is more than a rare occurrence. Like all ER docs, when a patient is asked at triage about their medical history and their active medications, and their response is “it’s all in ya’lls computer” – it makes my freakin head explode.

  3. [ Brain lock ] I used interoperability when what I meant to say was unrestricted access to data. My brain finished the sentence by adding interoperability. (Just tack interoperability onto the end any sentence and you’re ready to attend an industry conference). Interoperability obviously counts, but thinking on a more fundamental level re: data rights. As in, legislation that defines withholding access to health data as a crime.

    Very poorly phrased on my part …

    Agree with your analysis

  4. Another time warp: I use the historical parallel of the Bretton Woods agreement. When the US had the last viable navy near the end of WWII, Europe expected that the US would create a Pax of high tariffs on trade. Instead, they agreed to protect international trade at their own cost. The following increase in trade helped rebuild Europe and, in turn, helped drive the U.S. economy. When will these EHR vendors realize they have the power to create incredibly valuable exports? Sadly, only when their current customers demand better access to remote data (imports).

    Mind you, I’m no free-trade true-believer (I believe there must be better labor rights on all sides to realize full fairness), so it only works in certain contexts. Likewise, we need a “free-trade” data system, but we also need better-defined patient rights to access that data and the ability to control how it is used. Much like labor, the value derives from the patient, and we should receive move of that benefits that come.

  5. John — There have been two bills introduced in the current session addressing interoperability. The 21st Century Cures Act (had a good run in the House, dead in the water in the Senate – see details here: http://healthblawg.com/2015/07/whither-interoperability-century.html) and the Trust IT Act introduced recently in the Senate: http://www.cassidy.senate.gov/newsroom/press-releases/cassidy-whitehouse-introduce-bipartisan-legislation-to-improve-health-it-systems. Given the current level of dysfunction in Congress, though, I wouldn’t hold my breath waiting for either approach to pass.

  6. Charlie Kenney said ” interoperability — because it is so powerfully linked to the needs of physicians and patients — is the future”

    and John Irvine said “Given that it is widely understood that lack of access to health data causes preventable harm and blocks the progress of science”

    BUT Dr. Palmer said “I am not so sure that the medical need is _that_great to have interoperability.”

    As a patient, I am with Dr. Palmer….I don’t want the cost and the privacy risk of the quest for EHR interoperability. As a public policy observer I am completely unconvinced the whole EHR effort isn’t a costly boondoggle that bears more similarity to the Y2K world end scenarios that cost hundreds of billions of wasted funds that would have been best used elsewhere.

    Yes, someday EHR’s will provide net benefit as digitization has in so many arenas (airline ticketing, banking, financial brokerage, and shopping….to name a few). But in all these successful arenas the experimentation and innovation were privately funded by the brave companies that were highly motivated to risk the investment of dollars and corporate reputation….and they resulted in designs that consumers flocked to them for ease of use and yes, cost savings. That is not the case in today’s EHRs’….which are being mandated (crammed) into use, creating problems and costs for doctors and patients.

  7. The cloud is not a relevant differentiator for EHRs. In places like the UK National Health System where the bureaucracy controls most everything, the health information system fails for reasons that have nothing to do with networking.

    The problem with EHRs, cloud or not, is that they are not focused on the human patient and our human caregivers. EHRs are a resource planning tool for corporations and they contribute to a lack of transparency of quality and cost in a desperate effort to maintain the status quo of wasting $1 Trillion per year. If you’re the platform for waste of $1T the last thing you want is transparency and truly networked patient-directed care system.

    The JASON task force reports are very sophisticated about networking. They describe a Public API that would enable a transparent and patient-directed care system. Now who can we find to fund and politically support that?

  8. Good, Bobby
    Well, I guess if real-time transmission was not a constraint, we could have all kinds of data formats going through a sort of bank of reconciliation, like the world trade sysrem, where all the currencies are reconciled. Or we could just open up everyone’s different networks with telnet or web browsers. But it seems to me, a non-expert, that the more facile the interoperability, the easier the hackability. Do we have, in fact, any example of interoperability of such magnitude in the world? I guess we are talking about LAN to LANs; and no standard formats. Wow, how could the ONC even think this was possible?

  9. A reasonable person might ask this question:

    Given that it is widely understood that lack of access to health data causes preventable harm and blocks the progress of science, why does Congress not pass a law requiring interoperability?

    Has any legislator ever even introduced such legislation?

    I don’t think so.

    To me, this is Big Tobacco refusing to acknowledge the inevitable and fighting a delaying action for as many years as it can until it is finally forced to take action, allowing the extraction of as many dollars as possible

    Until we acknowledge that blocking access to data causes as much harm as blocking access to an emergency room or a drug or a life-saving operation, this will continue.

  10. The patient is a train. He is going through stations. The EHR/EMR is a three dimensional record (a magneric field or a reflective facet on an optical disc) of what the patient was like at one of these stations. An observer at the next station may or may not care what the patient was like at an old station. He may want to record new observations. The patient is constantly changing as he travels.

    My point is that I am not so sure that the medical need is _that_great to have interoperability. Ask your doctor friends how often they need to search other non-local databases. In thousands of clinical meetings, it seems rare to me that we desperately needed, say, another hospital’s records on a patient. Sometimes we do, but I don’t think it is a massive defect on our present systems.

    Perhaps the need for interoperability is for government studies or research or public health or insurance or actuarial data.

  11. I continue to call it “interoperababble.” If ONC would require a Standard Data Dictionary (“Standard Data,” as Dr. Jerome Carter puts it) as a condition of EHR certification, we might have a shot. Think of that, by way of analogy, as the type O blood of “data.”

    Per my blog:


    One. That’s what the word “Standard” means — er, should mean. To the extent that you have a plethora of contending “standards” around a single topic, you effectively have none. You have simply a no-value-add “standards promulgation” blindered busywork industry frenetically shoveling sand in the Health IT gears under the illusory guise of doing something goalworthy.

    One. Then stand back and watch the private HIT market work its creative, innovative, utilitarian magic in terms of features, functionality, and usability. Let a Thousand RDBMS Schema and Workflow Logic Paths Bloom. Let a Thousand Certified Health IT Systems compete to survive on customer value (including, most importantly, seamless patient data interchange for that most important customer). You need not specify by federal regulation (other than regs pertaining to ePHI security and privacy) any additional substantive “regulation” of the “means” for achieving the ends that we all agree are necessary and desirable…


  12. This is one of the most compelling pieces about health care I have read in a while. The notion that huge stores of useful data are stuck in silos — by design, no less — is harmful to the ability of physicians and other caregivers to do their jobs and, thereby, harmful to patients. How do we achieve the Triple Aim without a free flow of patiented-centered information? I wonder whether there is something of an historic parallel here. Not so many decades ago technology leader such as Digital and Wang failed to see the democratization of computing as embodied by Apple and others. The belief among many people at Digital and Wang was that computing was purely a business thing and there would never been a need for a computer in the home and that interconnectedness of machines was unnecessary. Are dominant EHR cos today making a somewhat similar mistake? Failing to recognize that interoperability — because it is so powerfully linked to the needs of physicians and patients — is the future?