Healthcare Data Interoperability and Workflow Interoperability: Four Questions (and Answers)

This post is prompted by this week’s Healthcare Leadership (#HCLDR) tweetchat. It’s at 8:30 EST on Tuesday. The following questions come from @JoeBabiain‘s tee-off post the Challenge of True Interoperability and Why It Matters. (By the way, please also check out my five-part series on Healthcare IT News this week: Achieving Task and Workflow Interoperability in Healthcare.)

T1: How urgent is the need for true healthcare data interoperability and why?

Even more urgent than data interoperability is workflow interoperability. The latter is a layer of interoperability above the former. Data interop is about getting messages from one system to another and having them mean the same thing in both systems. Workflow interop is about messages having the intended effect of the sender of the message. Much health IT investment and software development activity today is putting down a layer on top of legacy EHR and health IT systems. Much of this activity is about getting to workflow interoperability.

To some extent, data interoperability is a prerequisite for workflow interoperability, but not completely. Workflow interop can strategically compensate for problems at the data interoperability level. For example, workflow technology can escalate data interop problems for more intelligent automated handling or even human intervention. Consider this extreme example. Before data interoperability existed between EHR and health IT systems, some degree of workflow interoperability already existed. How is this possible? When a physician clicked a button to send a document to another organization, she or he did not care how this was accomplished, merely that was accomplished.

Before any data interop even existed, humans did the necessary work to achieve workflow interoperability. It was inefficient — involving copy machines, faxes, phone calls and sneaker net — but it was intelligent. Workflows were intelligent because the humans carrying them out were intelligent. They understood the purpose of the communications, so they tried to do what was necessary to achieve the intent of the communications. Today we have increasing data interoperability, but we’ve lost some of that intelligent workflow processing along the way. We need to marry together both data and workflow interoperability to get where we need to go.

Workflow interoperability essentially requires models of workflow and work, and their automated interpretation by workflow engines, AKA orchestration or process engines. Therefore workflow interoperability requires workflow technology.

T2: What experiences have you had with lack of interoperability?

My experience with lack of interoperability is that of a programmer working on EHR interoperability for a vendor. Typically, our EHR customer would approach us about interfacing with some source or recipient of patient data, such as clinical labs, e-prescribing, vaccine registries and such. So my experience was that of moving from a state of data and workflow non-interoperability to a state of data and workflow interoperability via use of interface engines, message parsers, and configuration of incoming and outgoing patient data workflows.

T3: Do you see incumbent providers willingly getting “on board” or will further market forces/regulation come into play?

First of all I’ll assume by “getting ‘on board'” you mean “Ready to participate or be included; amenable.”

I’m not sure if you mean for “providers” to mean clinicians, or providers of EHRs and interoperability solutions. However, it is an interesting question in all three respects. In all three cases, providers are essentially already “on board.” Healthcare interoperability is like motherhood, apple pie, and the American flag. Everyone is for healthcare interoperability. The main exception is that EHR vendors get a lot of flack because they are perceived to be “information blocking.”

While I am sure that there is some of this going on, on the whole, I don’t believe this is the main, root cause of healthcare’s “interoperability problem.” (I also know I am probably in the minority in this view.)

The contentious nature of healthcare interoperability, especially regarding blaming corporations and/or the government, reminds me of some discussions of why HIEs (Health Information Exchanges) aren’t as successful as everyone hoped. The reasonable, and accurate in my opinion, view has been that there’s been a lack of sustaining business models. Exactly the same point can be made about into why we’ve not better achieved healthcare data (and workflow) interoperability. If we can figure out how to better incent healthcare interoperability, then we’ll make better progress, goes an increasingly popular view.

Part of the reason we lack sustaining healthcare interoperability business models is that obsolete workflow-oblivious technology makes even the possibility of interoperability too expensive. This is the link between business models and technology models I’ve written about before.

The point I’d like to make here is that business models do not exist in a vacuum. They rely on technology models. Our current health IT infrastructure is notorious (in my mind, and in more-and-more other minds as well) “workflow-oblivious.” Without what academics called “process-aware” information systems, data and workflow interoperability are simply too difficult a problem to solve effectively and efficiently. In other words, a major reason healthcare interoperability has not been forthcoming, is that our fundamental health IT infrastructure lacks necessary architecture characteristics for transparent and flexible workflows within and among healthcare organizations. So, the healthcare interoperability problem still boils down to lack of sufficiently sophisticated workflow technology within and among healthcare data-exchanging partners.

T4: What can we as healthcare leaders do today to change the current state of interoperability? Can it be done?

Yes, workflow (and data) interoperability in healthcare can indeed be accomplished. Open discussion forums such as the weekly #HCLDR (Healthcare Leadership) tweetchat are extraordinarily important to getting to true workflow interoperability in healthcare.

I am sure most readers of this post are familiar with the healthcare Triple Aim (and related Quadruple Aim). I have a Healthcare Workflow Triple Aim, which I wrote about in my Health IT Workflow Silo post in HL7Standards.

  1. Educate healthcare leaders, clinicians, health IT, and healthcare social media influencers about healthcare workflow and workflow technology.
  2. Highlight healthcare workflow and workflow interoperability successes, in terms of healthcare organizations, IT vendors, and stakeholders.
  3. Recruit the best minds in workflow technology, from both inside and outside of healthcare, into accelerating use of process-aware technologies to facilitate true workflow interoperability.

Are we making progress in regards to the Healthcare Workflow Triple Aim?


I see plenty of evidence that all three legs of the Workflow Triple Aim are materializing into existence and lots of wonderful synergies are occurring between them! (Boy is that a mixed and mangled metaphor!)

My evidence? Too much to go into here, but I’ll leave you with this mornings’ Today’s Thought… 🙂

… which I wrote before I knew I would be writing this post.

Anyway, I look forward to tomorrow evening’s #HCLDR discussion of interoperability!

Leave a Reply

Your email address will not be published. Required fields are marked *