Workflow Usability Principles for Health IT Tools: EHRs, Apps, Platforms

This post is prompted by [#HITsm Chat 11.6.15] Usability Principles for Health IT Tools. I’ll be simulcasting using Periscope during the #HITsm chat and will post link here, once it exists! :) (Unlike Blab, Periscopes can be scheduled yet.)

As usual, I interpret the questions within a healthcare workflow technology context.

Relative to workflow usability principles, take a look at my 2009 EHR/EMR Usability: Natural, Consistent, Relevant, Supportive, Flexible Workflow.

Topic 1: What problems in health have you seen that seem to be caused by, or at least exacerbated by, poor Web or mobile interface design?

Non-adoption due to inefficiency and ineffectiveness.

Topic 2: Which EHR vendors and health-related Web sites making effective use of modern, interactive Web and mobile interfaces? Examples?

Ironically, some of the pre-Meaningful Use EHRs had more modern user interfaces! This was because there was great variety and experimentation. The one I know best won the first three consecutive HIMSS Davies Awards (2003-2005). On small screens observing Fitts and Hicks laws is essential. I wrote about that in The Cognitive Psychology of EHR/EMR Usability and Workflow.

I’ve not heard much praise for any Meaningful Use certified EHR with reasonable-size user base. My further impression is that users take health-related websites pretty much for granted, even when they work well. For example, folks are so used to quickly searching, finding, and using health info, that they really do take it for granted, which perhaps is a form of praise. In the mobile health app space, there are some reference apps that get high marks. But there are no app-based EHRs I know of that have anywhere near the functionality of top desktop EHRs. And where they exist, I suspect user will find their workflows nearly as complex.

I think on needs to point out that comparing an app to a desktop EHR is a bit like comparing one of the simpler screens of that EHR to the entire EHR. In other words, don’t over estimate modern web/mobile UIs.

Heads down high productivity data and order entry workflows require chaining screens, apps, web pages, etc. And this is exactly why current systems, mobile, web or otherwise, do terribly. Why? Because there is not inter-task, inter-screen, inter-app infrastructure with a model of the work or workflow to execute, consult, and interpret user actions.

By the way, check out No More Isolation: Why Apps Cooperate More (love “app thrashing”!) and my Twitter conversation with its author (click date to expand conversation).

Topic 3: What do you think is holding back vendors from doing as well as good e-commerce sites?

Complexity of data and workflow. What’s the solution? Software that can model data and workflow. Most current health IT software basically models data and hardcodes workflows. By modeling the data we can change the kinds of data and relations among data to fit the needs of a particular domain. However, we can’t do the same for the workflows.

On a typical e-commerce site you authenticate, search, read, browse, purchase. The workflow is simple and the same no matter what you buy. In an EHR workflows are complicated and different from each other. When I use to design and support EHR workflows, a single workflow might consist of a couple dozen or more screens. I think the most complicated workflow was something like 30-40 steps and screens in a pediatric EHR for in an adoption clinic. The large number of tasks was due to children who had literally never been seen by a pediatrician, so there was a lot of catch-up work.

Topic 4: Are consultations with clinicians during EHR and app design sufficient to take clinician needs and workflows into account? What more could developers do?

Consultations with clinicians during app design may be sufficient. Apps have much more simple workflows than EHRs, which can almost be thought of as collections of dozens of apps working together against a common database (which today’s apps don’t do).

In contrast, consultations with clinicians during EHR design are usually insufficient. There are too many different possible workflows. There is no way for all of that knowledge to pass from the real world through the clinician’s brain through the programmers brain and into the collection of supported EHR workflows. This is why EHR should be based on workflow platforms. On a workflow platform not only can clinicians help design individual workflows (divide and conquer) but approximately correct workflows can be pushed out to the field and then tweaked by users in an iterative cycle of workflow improvement (see BPM life cycle).

Use workflow tech so user can change workflows after implementation w/o having to go back to programmers and recompile, test, reinstall, etc.

See my Citizen-Soldier, Citizen-Developer, User-Programmer, Physician-Informaticist for more on this topic.

Topic 5: Which open source solutions in health IT are making significant inroads in the industry? What gaps should the open source communities fill in health IT?

There are not yet a lot of open source workflow tools for use in healthcare (tho see first link below for a list). Keep in mind they are just the beginning, so there is not a lot out there to compare to yet. However I think we need to distinguish among open source vs open data versus open workflow. There are closed source proprietary workflow platforms, which nonetheless support relatively open workflows. Not only is are the workflows open to inspection, these workflows can sometimes also even be exported and re-imported into other workflow platforms.

@wareFLO On Periscope!


This entry was posted in social-media. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.

Post a Comment

Your email is never published nor shared. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

You can add images to your comment by clicking here.