How Do We Get From Clickity-Click Checklists to Great EHR Workflow?

Periodically I “repatriate” lengthy comments I’ve made to other blog posts. Especially when a comment lacks a permalink, so I can’t refer to it on Twitter. Today’s opportunity presented itself in the form of online discussion of Why A Patient’s Story Matters More Than A Computer Checklist.

The stimulus for this repatriation was the following…

(The original, embedded tweets append this post.)

Here are my comments on the HIStalk blog about a workshop, sponsored by the ONC and NIST, on EHR usability.

Dr. Mostashari noted that the number one complaint he hears is that EHRs are driving users nuts with all the “clickity, clickity, click”. He emphasized that, ideally, an EHR is not just a documentation system, but, rather, a “workflow engine”. He emphasized this point: “It’s a workflow engine! And if workflow is working against users, making them do things that don’t make sense…That’s the number one challenge for EHRs to fulfill their full potential!” (Disclaimer: I added the quotes for emphasis. While it may not be word-for-word accurate, it is a very close paraphrase of what he said.)

As a long-time proponent of EMR workflow management systems and EHR business management systems, relying on workflow engines executing user-customizable workflow definitions, I couldn’t agree more.

User-centered design applied to EHRs with workflow engines and definitions will more quickly achieve more usable workflow than EHRs without workflow engines and definitions. In a later presentation, EHR event logs were mentioned as a potential source of data about user behavior. Workflow engines create even more detailed event logs full of time-stamped who-what-why-where-when-how information that can be used to spot usability problems. Techniques such as process mining should be brought to bear on these logs to improve workflow usability. What are the tasks that are taking too long (physicians staying late to complete records), aren’t being completed (potentially threatening patient safety), or are being completed incorrectly (also potentially threatening patient safety) or redundantly? (wasting financial resources)

Even if insights into the root causes of problems with EHR workflow usability can be teased out of EHR event logs and other data (such as from usability testing), without malleable workflow, there is no way to turn insight into improvement without expensive rewriting, recompiling, retesting, redeploying and (in some cases) retraining too. Many EHRs are customizable in many respects, but their workflows are not nearly as customizable as those of workflow management systems and BPM suites, from which healthcare IT should beg, borrow, or steal ideas and technology.

So, if the number one problem with EHRs is usability, and the number one problem with EHR usability is workflow, then Dr. Mostashari is right. EHRs are, or should be (and can be!) true workflow engines helping EHRs achieve their full potential.

P.S. I see considerable evidence for diffusion of BPM ideas and technology into healthcare IT. However, in comparison to hot topics such as mobile, social, and cloud, they haven’t burst upon the scene. I hope that current intense interest in improving EHR workflow usability will accelerate mainstream appreciation for the virtues of what BPM researchers call “process-aware information systems” or PAISs.

[later comment, responding to request for more information]

There’re lots of BPM vendors. Quite a few BPM vendor directories too.

Tough to isolate is the subset explicitly or obviously relevant to healthcare. Only a minority of BPM vendors have that content online, though the amount is steadily growing.

I use to tweet links at the intersection of EHRs, HIT, BPM, workflow, usability, safety and productivity … At http://ehr.bz I archive the best links (minus unrelated science, most popular culture, and all bad jokes). There’s over a thousand links [CW: now edging toward 2000!] to material at the interface between healthcare and BPM.

If you search for “BPM” within the webpage, there are a couple hundred instances. Many of these links lead to material about, and links to, specific BPM vendors. I also link to web content about EHR problems (workflow, usability, safety, productivity) that I feel might be solved by BPM tech. I’m thinking of creating a similar, more vendor-oriented, directory, as I’m beginning to find it hard to keep track all the potential players myself! Other folks may find it useful: Stay tuned!

[CW: Since then, I created the POW!HIT! directory, for People and Organizations improving Workflow with Health IT.]

For a general overview of BPM, BPM suites and relevance to health IT and EHRs I could do worse than cite my own blog posts on the topic.

EMRs and EHRs Need to Solve “The BPM Problem”: Why Not Use BPM to Help Do So?

Well Understood, Consistently Executed, Adaptively Resilient, and Systematically Improvable EHR Workflow

Hope you may find some of the above helpful.

–Chuck

Embedded tweets:

Ugh. Misinformation/ignorance. None of these "requirements" are . Problem is E&M and design/config

— Farzad Mostashari ()

yes but if design/config is so bad that most MDs dislike EHR interactions thats a very big problem in general and esp for MU

— Niall Brennan ()

yes, but misdiagnosing the problem as "government EHRS regulations" leads to wrong prescription, and missed opportunities

— Farzad Mostashari ()