Curbside Consult with Dr. Jayne 1/16/23

As a CMIO, one of my primary responsibilities is to make sure the EHR is configured in a way that makes it easy for clinicians to do the right thing. This involves everything from determining the content and display order of an order set to creating documentation templates and workflows that make sense for a given specialty, subspecialty, or particular type of visit.

In a large healthcare organization, managing this content can be complex. It can seem like we never have enough money, time, or personnel to do everything we want to do. We have to juggle priorities and manage conflicting requests from teams that might be in conflict with organizational priorities. Some days are easier than others, but when the going gets tough I’m glad that I have my “village” of fellow CMIOs that I can reach out to for advice.

During a recent call, one of them brought up this study that was recently published in the Journal of the American Medical Informatics Association. The title was eye-catching: “Behavioral ‘nudges’ in the electronic health record to reduce waste and misuse: 3 interventions.” The authors, working with the EHR team at an academic medical center, identified three workflows that might be driving users towards medical errors, waste, and misuse. They modified the system to try to nudge providers towards high-quality outcomes. but with varying degrees of success.

They had a couple of strategies for how they updated the EHR. “By changing the direction of these nudges – in one case, via making the less appropriate order more difficult to find and use; in the second case, by making the more frequently desired imaging easier to find; and in the final case, by presenting an easy to find alternative – we attempted to nudge providers toward reduced waste and misuse.”

The first situation dealt with a blood test. There were several variations of the test available and having an alphabetical order display that placed the least-desirable option higher on the list was likely contributing to erroneous orders. The modification removed the less-appropriate option, replacing it with an order panel that included educational content to help the provider make a better choice, including pre-checking the more desired test.

The second situation addressed the issue of providers erroneously ordering a CT scan of the abdomen when it was more likely that they wanted to order a CT of the abdomen and pelvis. The researchers assumed that alphabetical placement was an issue here as well. They reordered the list to place the more desired option higher in the list.

In the third situation, the authors looked at prescriptions of benzodiazepines that are given to help patients with anxiety during medical procedures. Prior to the intervention, the default quantity for the medication order in the EHR was what one would prescribe for a patient who was taking the medication on a routine basis rather than just taking it before a procedure. This led to prescriptions for more pills than would be appropriate for the situation. The team created a new order that made it clear that the intent was for pre-procedure use. It dispenses two pills with no refills and includes an additional comment that it is to be used as needed for anxiety prior to a procedure.

The authors noted some challenges in determining how effective the nudges were. For the anxiety prescription, there was a very short baseline, so it was difficult to determine the level of improvement. They also commented that the benefits of changes to the system have to be balanced against the cost of implementing them. There was a fairly dramatic difference in the time needed to create each solution: six hours for the blood test, three hours for the imaging order, and 16 hours for the anxiety medication order.

The changes were presented to end users as part of general educational guidance that is released with monthly EHR updates. In my experience the uptake of monthly update documentation can be variable, so there’s a good chance that some users simply stumbled upon the changes in the system. It would be interesting to look at how different specialties interacted with the new orders. For example, whether they made more of a difference among physicians in a specialty that interacted with the orders at a higher frequency than those who ordered the tests less frequently.

In the article’s discussion, I was interested to learn that “as compared to interruptive alerts, nudges in the EHR literature have not been as well described.” That’s an interesting point, because alerts that interrupt the workflow have become general annoyances for many clinicians, where nudges can be embedded in the design to the point where users might not even perceive them as having been deliberately placed. I wasn’t aware of the “Nudge” group at the University of Pennsylvania, but I’ll definitely be keeping an eye out for writeups of their work.

I also hadn’t thought of some of the work I recently incorporated into my own EHR as being nudges, but in hindsight, they are. I got the idea from a presentation I saw from one of the nation’s premier children’s hospitals and extrapolated a piece of it to the work that our clinicians do. It hasn’t been live long enough for me to know how well it’s been received, but I’m looking forward to finding out.

Another interesting dynamic to explore would be whether there were any specific complaints from end users about the incorporation of the nudges. For items that appear in a list, changing the order or removing an item can interfere with muscle memory and will feel bothersome to those who had adapted to finding the right choice in the list in their own way. It can take time for those users to re-adapt to the new presentation. For items that appear as part of a search, changing those can be less bothersome.

Since the study was done at University of California, San Francisco (UCSF) Health, I’d be interested to hear from anyone who was on the team responsible for the changes or from end users who experienced it.

What user-facing nudges or interventions are you working on for 2023? Leave a comment or email me.

Email Dr. Jayne.