Today, in reading over some industry news, my eyes settled on an advertising headline that gave me pause: “Is Middleware The Next Interoperability Miracle?” Now, I have to admit a couple things: 1) that vendors have to pitch the latest white paper with all the fervor they can command, and 2) that it never hurts to provoke conversation with a strong assertion. But seeing a professional advertisement include the word “miracle” — an expostulatory term which you might use to sell dishwashers — still took me back a bit.
And then I began to think about what I had seen. I wondered whether it will really take a miracle to achieve health data interoperability sometime in our lifetime. I asked myself whether health IT insiders like you, dear readers, are actually that discouraged. And I wondered if any vendor truly believes that they can produce such a miracle, if indeed one is needed.
First, let’s ask ourselves about whether we need a Hail Mary pass or even a miracle to salvage industry hopes for data interoperability. I’m afraid that in my view, the answer is quite possibly yes. In saying this, I’m assuming that interoperability must arrive soon to meet our current needs, at least within the next several years.
Unfortunately, nothing I’ve seen suggests that we can realistically achieve robust interoperability within the next say, 5 to 10 years, despite all appearances to the contrary. I know some readers may disagree with me, but as I see it the combination of technical and behavioral obstacles to interoperability are just too profound to be addressed in a timely manner.
Okay, then, on to whether health IT rank and file are so burned out on interoperability efforts that they just want the problem taken off of their hands. If they did, I would certainly sympathize, as the forces in play here are beyond the control of any individual IT staffer, consultant, hospital or health system. The forces holding back interoperability are interwoven with technical, financial, policy and operational issues which can’t be addressed without a high level of cooperation between competing entities — and perhaps not even then.
So, back to where we started. Headline aside, does the vendor in question or any other truly believe that they can engineer a solution to such an intractable problem, conquer world interoperability issues and grow richer than Scrooge McDuck? Probably not. Interoperability is a set of behaviors as much as a technology, and I doubt even the cockiest startup thinks it can capture that many hearts and minds.
Ultimately, though, whoever wrote that headline is probably keying into something real. While the people doing the hard work of attempting health data sharing aren’t exactly desperate, I think there’s a growing sense that we’re running out of time to get this thing done. Obviously, other than artificial ones imposed by laws and regulations, we aren’t facing any actual deadline, but things can’t go on like this forever.
In fact, I’d argue that if we don’t create a useful interoperability model soon, a window of opportunity for doing so will be lost for quite some time. After all, we can’t keep spending on this infrastructure if it’s never going to offer a payback.
The cold reality is that eventually, the data sharing system we have — such as it is — will fall apart of its own weight, as organizations simply stop paying for their part of it. So while we might not need a miracle as such, being granted one wouldn’t hurt. If this effort fails us, who knows when we’ll have the time and money to try again.
Its dead already.
ONC and CMS blew this entire program.
Cert EHR and all the 2400 pages yearly rule changes killed innovation. Put a few Boeing/Lockheeds in the driver’s seats (Cerner/Epic). Policy markets ALWAYS fail. And those non front line “experts” could not resist the temptation to see if they could get free data entry from MDs. They failed on multiple levels. Anyone with an ounce of database programming experience knows you need primary keys to match IDs. We do not have unique IDs for patients. That will instantly puts a huge wrench in any interop. Second, how do we get BACK information that is wrong, wrong diagnosis, wrong side, wrong patient, if its blasted out into the world incorrectly, how do we correct that information after it leaves our EHR and ends up in 1000 others. Also, there are very little to no standards on the data fields and exchange (hence FHIR, etc) but those are immature and not robust.
Finally we have an enormous security problem. We are hit with a right left punch of Hey doc, push your information out everywhere, we want it flowing, but don’t lose one byte or we will fine you a zillion bucks. Stuxnet anyone?
Finally, CMS Congress and ONC have lost the hearts and minds of physicians. I am on the front lines. I see it every day. I am the CMIO of our hospital. MACRA has put the nail in the coffin. It will fail just like MU and PQRS. Fail, in that we will have low participate rates, high penalties and disenfranchised MDs. It will also set back interop, EHR improvements, cost reduction, patient safety, patient sat and patient care at least another decade as MU and PQRS did.
My suggestion. The US gov get OUT of the exam room. No more EHR cert, no more data entry requirements, reporting, and chasing buzzwords like value based care. Let the market work. It was working and building slowly prior to HITECH. But the forced implementation lead to this disaster we have now. We need to retool. We need IT vendors to work directly with MDs to get the info we need. FIRST locally then interop later. YES it will be slow, but deliberate and MUCH better than now.
meltoots,
And if government chooses not to get out of the exam room (which I think is the most likely scenario), what then?
Exactly what you have now. A super complicated non functional system with no useful interop, with burned out MDs taking care of you. Don’t underestimate the damage that has occurred with these programs.
Your comments on both the technical and behavior pieces are spot on. Most of the problems with interoperability stem from policy, behavior, and governance – not necessarily technology (though legacy technologies keeping pace with new interoperability standards can be a challenge).
meltoots,
Would repeal of these programs take us back to what it was like? Or is the damage done and it will be hard to heal?