You wouldn’t confuse your baseball card collection with the Microsoft Access database management system you use to manage it. Or confuse your patients with your EHR. But people, especially in healthcare, confuse workflows with workflow management systems all the time. I’ve discussed this important distinction before (EMR Workflow Systems vs. EHR Workflow Management Systems). But I think it is worth revisiting. Confusing workflow with workflow technology is what gives rise to the notion that introducing process-aware information systems into healthcare and health IT is mere “tweaking” of workflow.
— Dan Munro (@danmunro) August 26, 2014
Last week I moderated one of the weekly #HITsm tweet chats (Fridays, 12-1 EST Thank You To All Who Tweeted About Healthcare Workflow, Wearables, BPM, and Flowcharting!). It was well attended, 40 percent above average. In the introductory post I wrote for the HL7Standards blog I emphasized the distinction. Here is the shorter version of that.
Workflow is what actually happens when work is done. It is a series of steps, or tasks, that consume resources (money, time, effort, attention), and achieve one or more goals. Virtually all purposeful activity involves workflow.
Workflow technology, on the other hand, has some sort of model of workflow. This is model is executed or consulted, in conjunction with human users, when they do their jobs. These executable process models are at the heart of what distinguishes healthcare workflow technology from generic healthcare information technology. All information systems “affect” workflow (that is, influences workflow, for good or ill). But healthcare workflow technology “effects” workflows (that is, drives, makes it so, in the Captain Picard, Star Trek sense).
Why do I harp upon workflow technology all the time? Because it is the next generation of application architecture heading down the pike toward health IT, which is about a half a generation or more behind other industries. Sometimes folks say, “So what! Fix the incentives and the tech will fix itself.” I agree we need to fix healthcare incentives (whatever that means, I hear lots of strenuous debate about that particular topic). But even if we fix the incentives, billions of dollars have cemented frozen healthcare workflows into place. Similar to, by analogy, early human proclivities (such as eating until overfull in the presence of food) plaguing us this modern day, a decade of frozen health IT workflow will resist pressures to change for the better, if the better even presents itself.
Furthermore, regardless of which incentive regime we finally impose, it won’t work unless we have true workflow technology to make it work. It doesn’t matter, free market vs. socialized medicine, Meaningful Use-driven software development vs. Meaningful Use-Be-Gone-driven software development. We will have invested so much in workflow-oblivious, workflow-frozen healthcare information systems, that (A) they’ll be too expensive to change, and (B) we won’t have enough will or resources left to change them. We need to begin making our health IT systems more process-aware, now.
The sentiments of the previous paragraph are why I wrote Fixing Our Health IT Mess: Are Business Models or Technology Models to Blame?, one of my most popular (or at least most tweeted) blog posts.
Moving to executable models of work is not merely tweaking workflows. It is moving to a more flexible, effective, efficient, transparent and systematically improvable substrate in which to realize whatever systems of healthcare incentives we ultimately move to. And I can guarantee you one thing. We won’t get them right the first, second, or even third time. That’s why we need to create systems of health information management in which workflows can be more easily “tweaked.”