Interoperability 3.0

Apple users all have that one Android friend.  I know I do.  Their Samsung Galaxy Quest Triple Wide Screen 2 is obnoxiously different.  Sure, your friend boasts about features like a better camera and longer battery life for half the price, but you know that it’s bulky and too often makes decibel-crushing clicks. But we all know that what really gets an Apple user's gears grinding is the inability to connect with Android users in the same way we connect with other Apple users—you can't send an Android a damn iMessage. The internal connectivity of your Apple ecosystem becomes moot. Your MacBook Pro is rendered useless for communication, unable to share cat videos, reddit links or a vital emoji response with your Android friends. Somehow watching Princess Waffle's latest catnip mishap alone in your room isn’t as fun. Technically speaking, why can't Androids receive iMessages? The answer is simple: the two systems don’t interoperate.

Longtime Apple users have witnessed the evolution of perhaps the most beautiful integration of platforms, hardware and cloud services on the consumer market.  Sure, the big Apple product release events have become a bit stagnant (“We shaved off 0.02mm of thickness!”) – but what I continue to be blown away by is the delightful cross-product integration that occurs so elegantly with Apple products.  My AirPods connect.  My iPhone syncs.  My AirDrop transfers.  I hold my breath for the day Apple releases the un-f&%kable printer: a device so over-engineered, it never drops connection, jams or runs out of ink (ball’s in your court, Apple). 

What’s interesting here, is the difference in interoperability between disparate systems vs. the same family of systems.  Apple devices auto sync, yet a simple text message cannot be sent (in 2020) between an Apple computer and an Android phone.  How can interoperability be so poor between some of the most advanced consumer products on the market?  Because no one has forced them to.  This same problem seeps into healthcare as we look across its fragmented landscape. 

In healthcare, the diversity of systems is staggering.  Different vendors (Epic, Cerner, Allscripts), different implementations or instances of the same vendor (Mt. Sinai, NYU Langone, Columbia Medical), and even different subsystems (ER, in-patient, rehab) don’t always connect. The interoperability challenges expound tremendously.  Thanks to legislation—(surprised? Don’t be… I’ll explain in just a second) interoperability is rapidly advancing.  While we may still be at iPhone-to-Android text status now, we are quickly moving to what we call here at Particle Health: Interoperability 3.0. 

Here’s how we think of the Three Interoperability Evolutions:

Interoperability 1.0: “The 1-to-1er”

In 2009,  Barack Obama signed the American Recovery and Reinvestment Act.  As part of this legislation, the HITECH Act was enacted, proposing a national network of digital medical records.  Additionally, the Meaningful Use program was established—requiring that providers begin to share information.  In addition, the Federal Government started to put their money toward true interoperability and began providing grants to fund State Health Information Exchanges (HIEs.)  Because of the money now flowing into the interoperability ecosystem, a new category of companies sprung up, touting “Integrations Services.”  These companies excelled in taking a vendor, (a Diabetes App, for example) and getting it plugged into one of their hospital customers.  This can be seen as 1-to-1 integration—where two discrete systems are directly plugged in to each other, even without any default interoperability. Some organizations have figured out really smart ways of “network-effecting” their integration services so that they own a proprietary integration protocol, further shoe-horning their place as a de facto integration option and charging a premium for requisite white glove integration services. While this method expertly solves problems for hospitals and other large systems, it doesn’t exactly solve the problem for organizations looking to build nationally, or outside of direct hospital sales cycles. 

Interoperability 2.0: “The Scraper”

Next came the Portal Scrapers.  Because of Meaningful Use, every provider had to at least offer a patient portal that folks could log in to to see their test results, immunizations and other relevant health info. This spawned a category of aggregation systems touting “interoperability” that required the patient to provide their credentials for each provider they were connected to. While this method works pretty well in industries like banking, it simply does not work that well in healthcare.  Patients have a hard time remembering all the places their health data hides and they definitely can’t remember their passwords to all those portals.  Even if they did, Patient Portals only hold a fraction of their data in the EMR, at least partially because many systems require the providers to actively release the data to the portal. This method has been implemented many different ways.  One implementation methodology called SMARTonFHIR, has continued to survive due to its capacity to share data with third parties that hospital/EMRs themselves might not really want to openly share with—like Life Insurance, For-Profit Research, and Law Firms, etc.  This hesitation to share with third parties highlights what can be seen as a battle between patient choice to share data and concerns around the privacy and security implications of ubiquitous access. It also eats into some EMR revenue streams customized for things like Life Insurance underwriting. Interestingly, this battle plays directly into the next evolution of Interoperability. 

Interoperability 3.0: “The Developer’s Era”

The industry is entering a new era of interoperability.  This new era is built on national networks & backed by new legislation—namely the 21st Century Cures Act, that was passed in 2016 and signed by Barack Obama right before he left office (interestingly bookending the HITRUST act from 2009.)  As part of the Cures Act, policy has been enacted to ensure interoperability and the free flow of data through Anti-Information Blocking provisions and the development of the Trusted Exchange Framework and Common Agreement (TEFCA).  These policies are a marriage of data exchange enforcement, data standards, incentives and patient rights, developing a framework for an agreement between the proverbial Apples and Androids of the healthcare world—the Epics and the athenahealths; the Mt. Sinais and the NYU Langones.  These systems are now required to exchange information with one another using defined requirements, with little wiggle room for interpretation.  And it doesn’t stop there—these policies demand Americans get full, unadulterated access and control over their data using APIs.  Interoperability 3.0 is that feeling when you haven’t finished an article on your laptop, so you push it to your phone to be finished later on the train.  It’s that feeling when your data on your phone automatically syncs to your new laptop that you just opened out of the box.  It’s that feeling when you download a telemedicine app and the doctor knows exactly what medications you’re on and what allergies you have without you having to to fill out that stupid form one more time.  

Interoperability 3.0 is the developer’s era.  All of a sudden, you just… have access to clean data from every EMR in the US and can use it to create.  One agreement, one query—fully parsed and normalized medical history.  This ability to build something new is akin to when Plaid opened up banking data to groups like Venmo, Mint and Robinhood and a sort of Hunger Games of FinTech innovation sequence took place—a FinTech Renaissance that championed the consumer. 

Interoperability 3.0 promises us in healthcare what Apple did so well in technology—it strengthens the pull with each newly connected service, device or solution.  And here is, perhaps the first, good news of 2020:  Interoperability 3.0 is right around the corner. In fact, it starts April 5th, 2021.

Come get to know Particle Health and see how providers, developers, payors and other stakeholders are connecting to a fully interoperable medical system, just by searching with a patient’s demographics.  We’ve got a sandbox chock-full of synthetic EMR data—jump in and let us know how we can help!