Well done! Can I just say that for many of us who are officially retired, we are still engaged in…
EPtalk by Dr. Jayne 8/8/24
I like to stay close to my family medicine roots, so was excited to provide some local locum tenens coverage for a friend who was taking a much-anticipated vacation. Her practice uses an EHR that I’ve used to deliver patient care in the past, so I was confident in my ability to step in without a lot of retraining. Since I’m a new user on this particular system, I expected it would be a bit slow, especially since I wouldn’t have any medication favorites built and wouldn’t be terribly familiar with documentation templates in her particular system.
What I didn’t expect, however, was finding a defect that was recently injected into the system as part of an upgrade. In the prescribing system, the pick list that would normally be used to select how many times per day a medication should be taken had been morphed into something virtually unusable.
For an ambulatory medical practice, one would expect all of the “by mouth” options to be at the top of the list for easy selection. Instead, I was greeted by all the intravenous selections, followed by rectal and other options. The list also wasn’t responding correctly to keyboard inputs if I tried to do a type-ahead search, which meant that I had to scroll (and scroll and scroll) to write prescriptions. Needless to say, it was less than optimal for patient care.
I mentioned it to the office manager and she gave me explanation of it being upgrade related, so I can only hazard a guess at how many thousands of users are having their time wasted by this bug. It doesn’t impact physicians who primarily prescribe using a favorites list, so I guess I know what I’ll be doing this evening to make tomorrow less painful. Many newly trained physicians enter practice in July and August, so I bet I’m not the only one.
I also had the opportunity to attend a continuing education webinar this week. I was particularly excited about the session because one of my former medical students was presenting on an important clinical topic. I’ve presented on hundreds of webinars over the last decade, hosted by major academic institutions, medical societies, technology vendors, state health departments, and volunteer organizations. The best ones conduct a practice session or at least distribute a set of ground rules to explain how presenters should interact with each other and with the audience. When organizations don’t do this, sometimes they get lucky and everything goes well. In this situation they didn’t get lucky, however, as one of my friend’s co-presenters apparently didn’t get the memo to turn her camera off when she was not presenting.
Since there were only three panelists, their camera feeds were front and center. I’m assuming that her co-presenter was multitasking and looking at something humorous based on her facial expressions. Unfortunately, those expressions were occurring at a particularly sensitive point in the discussion that made it appear that she was laughing while serious patient harms were being reviewed. I’d like to assume that this was just an oversight on her part, and that she didn’t mean to be disrespectful, but either way it’s bad form. I hope someone at her organization recommends that she review the recording so she can see how she was projecting herself to the world.
I wish I would have run across this article earlier in the Summer Olympics hype cycle. Some of the parents of competitors were outfitted with heart rate monitors to see how their anxiety manifested as their athletes competed. NBC displayed data from the father of one of the US gymnasts during one of her routines. One would think that it would be enough to just display the facial expressions of loved ones since you can see every second of anxiety or amazement broadcast to the world already. Commenters on the article felt that displaying heart rate data was a bad idea, using words such as invasive, creepy, and unsportsmanlike to describe the practice.
Another article that I ran across this week detailed a physician who is accused of behaving badly by making over $1.5 million in personal charges on his business-issued credit card. The physician pleaded not guilty to the charges, with his attorney stating that “the funds he used were not stolen funds.” The card was used for $115,000 in cash advances, $176,000 in pet care, $348,000 in personal travel, $109,000 in gym memberships and personal training, $52,000 in catering, and $46,000 in tuition payments for his family. A savvy commenter called out the fact that he spent more on pet care than he did on his children.
The amazing thing about the situation is that the charges occurred over a seven-year period before being caught in an audit. According to the article, his institution is the only state-run hospital in New York City. One would think that being a public institution would make for stronger accounting controls. The physician is scheduled to appear in court again at the end of September.
Speaking of September: I discovered this week that the spelling and grammar checks in Microsoft Word will not catch “September 31” as something you shouldn’t type. It’s something I’ll be manually watching for in the future.
I wrote earlier in the week about the evolution of language and how that might impact large language models. I was excited to see this article about the forces changing language on a daily basis and that teenage girls are a major driver. It should be noted that the article is from Australia, which has its own unique linguistic offerings. Some of the experts interviewed in the article note that young women drive changes faster than young men and that this isn’t a new phenomenon – it has been studied extensively, including reviews of letters written from the 1400s to the 1600s. The fact that social media connects people from different regions and countries is also driving rapid change. One expert encourages people to place themselves close to a group of teenage girls to listen to how they communicate as a representation of where language is headed. I’ll be looking at my interactions with various community youth groups differently moving forward.
What do you think about changes in language and how they might be driven by social media or other societal forces? Leave a comment or email me.
Email Dr. Jayne.
Whenever I raise the matter of a bug (or misconfiguration, or anything gone wrong), and the response is “this is how this situation came to be”? I get wary.
My experience has shown me, that the primary topic needs to be how to correct the situation. That needs to be Agenda Item The First! How, and When, and By Whom.
It’s often interesting how the situation came to be. Sometimes it identifies political, economic, or process barriers that will have to be overcome. It can inform the solution, in other words. But it’s never the primary issue.
As such, talking about How This Situation Came to Be, needs to be several items down the list of things to discuss. It’s supportive rather than primary.
If ‘How This Situation Came to Be’ is the very first response? It too often indicates that the speaker is entirely disengaged from providing solutions. It’s a red flag that “no help will be found here”. And even if they do help? There will be lots of foot-dragging, excuses, and meaningless or ineffective activities placed on the To Do list.
Weirdly though, it doesn’t always mean that. There are plenty of people in the IT world who are detail people. They tend to organize their thoughts and work from the bottom up. From details into increasingly higher-order levels of complexity and organization. Such people can give off apparently negative vibes without meaning to. Detail people can easily lapse to discussing the easier parts of of an issue first (exactly the category How This Situation Came to Be falls into).
Speaking in defense of all the people whose reflexively go to root cause when something goes wrong, the challenge with fixing the problem without knowing how it came to be is that you sometimes/often/usually need to know why something went wrong before you can know how to fix it. If a patient comes in and says “doctor my arm hurts” there is no way to know how to treat the patient unless you ask a few questions first. Did the patient fall against a hard surface and break a bone? Lift something heavy and pull a muscle? Sleep funny and pinch a nerve? Are they having a heart attack?
Your comment about giving off negative vibes me wonder if I know you irl, I got nearly identical feedback from a manager many years ago who said to me “I know what you’re doing and how you think but when you do that out loud other people are hearing you say ‘no'”. Its something I work really hard on, because its completely counter to my nature — I need to talk through problems before I can find the path to a solution (literally, I wander around my apartment talking to myself all day long) but when I do it in front of an audience, especially the muckitymucks on the business side, they think I’m making excuses for why I don’t wanna or aren’t gonna fix XYZ when really I’m pulling all the threads to find the most expedient solution that won’t make things worse now or in the future.
For the record, I’m a detail person! I’m not an executive and don’t get many opportunities to organize a team and hand off the grunt work to others.
Still, my comments stand. I’ve been “the user” too, reporting issues to the responsible people. Once you’ve been stonewalled a couple of times, you get the message. No help will be forthcoming. Their idea of a response is solely talking about “how this problem came to be”, like that’s sufficient!
Maybe this will help. I’ve developed communications skills to overcome my natural habits.
First is, let the user talk. Interruptions irritate them and often they need to get something off their chest.
Next is, your earliest verbal contributions matter. Resist the temptation to deep-dive into troubleshooting! Any or all of these approaches are positive:
1). Say something generally supportive, like “That must be awful!”, or “How can you work like that?”;
2). Questions should be focused to clarify the user story, and be answerable quickly and easily. Ideally, with Yes or No responses, but a short response should be all the user needs to do;
3). Sometimes, if the user is skittish? You need to provide an all encompassing reassurance blanket. Tell them “I understand. This is unacceptable system behaviour and we’ll get to the bottom of it.” Whatever you say, you must communicate that you are on-board with fixing the problem and it’s a priority for you now too. It doesn’t need to be your top or only priority, but it’s a priority.
4). Next stage? You can often break through the client’s frustration by involving them in the solution somehow. Literally anything will do here! Ask them to show you the problem. Ask them what they think the solution ought to involve or look like. Ask them if anyone else is having the problem. But once you ask them for something, and again it can be almost anything, they are now involved in a more positive way. And it backs up your “I’m interested and paying attention” message.
5). At this juncture, you can branch out your technical response. Want to deep-dive? Go for it! Want to discuss ‘how this problem came to be’? Knock yourself out. One helpful technique I sometimes use is to think aloud, and announce that you are thinking aloud. it’s very powerful to say things like, “You know, if we could get to Destination X, that would be a great solution, don’t you think?” Showing openness to varying solutions, and that your mind isn’t made up yet, can be an incredibly strong move.
So, that’s it. My 2 cent’s worth!
Gee, I use an EMR that’s behaving a lot like the one you describe. I had favorites saved before the upgrade, and they used to display if I hit the yellow favorite star first. They don’t do that now, but if I free text a couple of letters of the favorite, it then comes up. But yes, for a new entry you pretty much have to free text the frequency. Interestingly, it seems to pick up that information and calculate the number correctly a lot of the time, not always, so there may have been an undocumented method to this madness. That said, I used to say this EMR ran like a late 60s GM car built on a hungover Monday morning, but lately I say it was built by Boeing.