EPtalk by Dr. Jayne 3/12/26

Dr. Jayne Goes to Las Vegas

I’ve always taken a taxi when visiting Las Vegas. It has been years since I had to stand in a crazy long line, and today I figured out why. It’s because everyone is packed into the Uber/Lyft pickup area.

The kitschy neon artwork was the only good thing about the experience, which is in desperate need of a process improvement project. I was only there because I had an Uber voucher that was about to expire. I’ll be back in the taxi line next time for a cheaper and less stressful experience.

I spent some time Monday catching up with old friends at the HIMSS Native American & Indigenous Health Symposium. The clinicians in this space care deeply about their patients and the populations they serve, and are often working with minimal resources.

One session featured representatives from GDIT and Oracle Health. An attendee pointedly asked how the company will ensure that facilities that serve this population won’t experience the issues that some of the Veterans Administration hospitals have had. The answer was not reassuring. 

From there, I headed to the exhibit hall for a sneak peek at the setup process. I didn’t have an exhibitor badge, but no one challenged my entry. That was a big change from the recent ViVE conference where the door teams checked every badge at some entrances. 

 
CoverMyMeds has a swag machine in the lobby, and I received a pouch with some nail clippers that I’ll be donating when I get home. I was actually hoping for the hand sanitizer since I had forgotten to replenish my supply, but I was confident that I could find some in the exhibit hall when it opened.

I put my feet up for a bit and then was off to the opening reception. It was held once again at Caesar’s Forum, which is across the street from Caesar’s Palace. That created confusion among attendees and taxi drivers alike. 

It was a target-rich environment from a footwear perspective. I found myself also eyeing these embroidery and pearl-embellished jeans. I’m not sure who the team was with the matching Nikes, but they looked sharp. 

I had a chance to catch up with the incomparable Ross Martin, MD, who shared a great story about performing as Elvis for HIStalkapalooza 2013. He was even able to produce the highlight reel from the event on his phone, which was a nice treat. 

On the way back, I did a detour to the Bellagio Fountains. I was lucky enough to catch my favorite song, which made for a perfect cap to the evening.

The next couple of days were a whirlwind, with plenty of sessions and laps around the exhibit hall. It felt a lot like the old days of HIMSS. I saw fewer influencers and more people ready to do business than I did at ViVE. 

Some complained that HIMSS doesn’t include meals with registration, but those who ventured down to the Hall G lower level were rewarded with cookies, fruit, brownies, blondies, and the elusive Kouign Amann pastry. The lower level was dubbed “The Park” and also included the odd tree here and there, as well as a food court and a place to pet kittens.

At the Epic booth. I was pleased to see that their new AI solution offers sensible patient-facing information for a scenario where a patient asks if they can celebrate with bacon while in Las Vegas even though they have high cholesterol. As usual, the Epic booth had the most plush carpet in the exhibit hall. I didn’t know that I needed a coffee table that was embellished like the painting “The Starry Night” before I went there, but apparently, I need that in my life.

It’s great to see the next generation of healthcare tech professionals coming to see the big show. I enjoyed an impromptu demo from graduate student who was working on the TheraCare.ai platform.

It seemed like everyone was talking about AI or SaaS solutions, but I always enjoy visiting vendors that offer physical technology, such as waterproof keyboards, innovative crash carts and workstations, and communications devices.

During my booth crawl with Dr. Craig Joseph from Nordic Global, we scoped out Athena Security’s hospital visitor management system technology, which includes AI-powered concealed weapons detection. Having worked at hospitals that have had serious security incidents and even injury to staff, such solutions are unfortunately necessary, and it’s nice to see innovation.

The show floor had a couple of themed seating areas, one of them complete with a digital fireplace. For the zone that had the bean bag chairs, they were less occupied than the more traditional chairs, which might say something about the agility of the average HIMSS attendee and our willingness to risk being unable to arise from a soft surface. The main floor also included a puppy park, although I didn’t see any dogs during the times I passed by.

First Databank had a cool giveaway with its adult coloring book. 

The best booth slogan goes to connectivity vendor Digi International, which promised “The Ultimate Hookup.” CognomIQ was close behind with their offer of a chance to “win a prize that doesn’t suck,” which was funny since it was a Dyson vacuum.

The footwear game was strong this year. IMO Health brought their shoe and sock A-game as usual. I’m sure they are looking forward to next year, when HIMSS returns to their hometown of Chicago. 

American Messaging had light up shoes for the whole team.

The ever-dapper Jonathan Shivers of Relatient married form and function with oxfords and argyle. 

My spotters were calling in outstanding outfits from across the show, but I wasn’t fast enough to catch the woman in the hot-pink suit with matching shoes or the pair of gents who were wearing matching brocade dinner jackets.

Wednesday afternoon featured a number of in-booth happy hour events. Drinks were flowing as long as you were willing to have your badge scanned. I always wonder about the return on investment for those events, since a good number of the attendees aren’t decision makers or budget owners. I’m sure it falls into the category of all publicity being good publicity, but I can only imagine what the event services vendors charge for a happy hour service.

I had to head for the airport due to some obligations at home, so I’ll be missing Thursday morning’s session covering the future roadmap for the Centers for Medicare & Medicaid Services featuring Dr. Mehmet Oz. I’ll be interested to hear from those who attended and whether you found the content inspiring. There is much work to be done in the US healthcare space.

If you attended HIMSS, what was the highlight of the event? If you didn’t attend, why not, and where are you spending your budget instead? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 3/9/26

I enjoy working with students and residents, so I was excited to be asked to present on the topic of virtual nursing. I asked who would be co-presenting from the nursing side, and I was met with a blank stare.

I have experience with virtual nursing technology and implementation, but I’m not seeing hospitalized patients at the moment. I haven’t seen it through that particular lens. I also can’t provide the nursing point of view. I was eager to bring one of my nursing colleagues on board.

As with any presentation, we wanted to come up with a catchy title that would make people want to attend, especially given the scheduling constraints for our anticipated audience. My proposed title of “Virtual Nursing: A Promising Fix or an Expensive Band-Aid” was probably sassier than the organization would have liked, so we toned it down a bit. After creating just enough of an abstract to be able to start advertising the talk, we split up to start creating content.

If you have recently spent time scrolling through healthcare tech sites, you probably noticed that virtual nursing is having a moment. Every few weeks, a new health system announces a pilot program, a vendor rolls out a freshly branded platform, and a press release lands in my inbox proclaiming that the future of inpatient care has arrived.

I understand the appeal, given the genuine staffing crisis that continues to grind down nurses across this country. However, the literature is starting to show that virtual nursing isn’t going to be the ultimate solution to a problem that has been building for decades. A recent study published in JAMA Network Open might give hospital leaders a reason to pump the brakes.

Before we dive into the data, we need to note that “virtual nursing” has become one of those terms that gets used in so many different ways that it risks losing all meaning.

At its core, virtual nursing refers to registered nurses who deliver care to hospitalized patients remotely, using video and/or messaging platforms instead of being physically present on the unit. The virtual nurse is stationed offsite, sometimes in a central hub within the same health system, sometimes even at home. They interact with patients and bedside staff through a screen.

Workflows might be active, such as assisting in tasks related to patient admissions. This could include reconciling medication lists, performing screening instruments, and helping educate the patient and family about what to expect. Similar tasks might be performed before or during discharge.

Another common use case involves virtual nursing teams that function as high-tech sitters. Sometimes they monitor multiple patients on a split screen, while at other times they might use AI-powered tools to determine whether a patient is at risk of fall or injury.

Less common but growing applications include triage support, documentation assistance, mentoring of newer bedside nurses, specialty consultations, preoperative screenings, interpreter services, and even serving as a second witness for high-alert medication administration.

Intensive care units have historically been early adopters of using virtual nursing for rapid response oversight. What is newer is the expansion of virtual nursing onto medical / surgical floors of the hospital. Most of the current conversation and controversy is playing out there.

Virtual nursing is often seen as the solution for nursing shortages. Some quick web searches reveal turnover rates of registered nurses to be greater than 15%, with large numbers of nurses leaving the workforce or planning to do so within the next five years.

One of my best friends is a nurse. She is constantly being floated to other units in the hospital that are outside of her specialty. She is sometimes assigned to be a sitter, which although important for patient safety, is a misuse of her time and skills since she is 1:1 in a lower-skill environment rather than performing her usual duties with multiple patients. For her, it’s not an easy shift, but is professionally unsatisfying.

Her particular specialty is hands-on. We haven’t really talked about virtual nursing, but I will get her thoughts as I pull this presentation together.

Virtual nursing is also touted as a way to retain nurses who have musculoskeletal issues or other reasons to move away from bedside nursing. Moving to virtual lets them bring their clinical judgment to the game in new ways, including the uncanny ability of seasoned nurses to spot the patient who is about to go sideways before the numbers change.

Virtual nursing can also free up time and attention for nurses who are physically at the bedside. It distributes some of the administrative burdens to a remote nurse.

This brings us back to the study, which was a cross-sectional, mixed-methods study drawing on the 2024 Nurses4All survey. The final analytic sample included 880 registered nurses working on medical, surgical, and intermediate care units across 418 hospitals in 10 states. These nurses had reported that virtual nursing was being used by their hospital, which made them well positioned to comment on whether it was actually helping.

In addition to answering questions about virtual nursing’s impact, respondents were also invited to provide a free text response to the question, “Please share any positive or negative experiences you have had working with virtual nurses.”

The short version of the findings is that virtual nursing is producing mixed results, and the mixed results lean toward unimpressive.

The majority of nurses in the study (57%) reported that the use of virtual nurses did not reduce their workload at all. Of that group, 10% said that virtual nursing actually increased their workload. Of the 43% who reported workload reduction, only 8% said the reduction was meaningful.

The quality-of-care findings were more favorable, but still underwhelming. A little over half of respondents cited a positive impact on care quality, but only 11% said that the improvement was substantial. Nearly half of the nurses reported no impact on quality, and 4% said that virtual nursing negatively impacted quality. As the authors said in their conclusions, these are decidedly mixed findings.

The free text responses fell into six themes: virtual nursing as a staffing workaround, virtual nursing as an extra pair of eyes, safety risks and time delays, added work, patient distrust, and administrative help or hindrance.

The staffing workaround theme is the one that should give hospital leaders the most pause. Nurses described a pattern in which virtual nursing was being used not as a supplement to adequate bedside staffing, but as a substitute for it. Multiple respondents noted that management was counting virtual nurses in the staffing ratio, which effectively reduced the number of physical bodies on the floor.

One nurse made the point as directly as possible. She would rather give up the virtual nurse entirely in exchange for having another person on the unit who could physically intervene when a patient needs it. That is not a ringing endorsement.

Nurses described the added work component through examples, such as having to correct documentation errors introduced by virtual nurses who lacked familiarity with the specific patient or the unit’s workflow. Others noted that by the time they had exchanged messages back and forth with a virtual nurse to address a concern, they could have simply handled it themselves in the first place. The overhead of coordination was, in some cases, consuming more time than the task being delegated.

The patient distrust findings deserve particular attention, because they highlight a reality that technology enthusiasts often underestimate. The patients who populate medical and surgical units are not usually digitally engaged, younger adults who are comfortable navigating a video interface while also managing acute illness, pain, and anxiety. They are frequently elderly, cognitively impaired, hard of hearing, or simply overwhelmed.

One nurse commented that patients treat virtual nurses like a commercial during their favorite show, ignoring them or trying to fast-forward them. That is blunt, but probably accurate in a meaningful subset of cases.

Another noted that virtual nursing only works well for patients who are cooperative, not in pain, and have all of their immediate needs met. That is a fairly narrow slice of the typical med-surg census.

To be fair to the technology, the study also identifies areas where virtual nursing provides genuine value, such as when a virtual nurse acts as a scribe for a bedside nurse who is performing physical tasks.

The authors ultimately concluded that virtual nursing might not be as much of a big win as hospitals expect, and using it to subvert staffing requirements is likely to create more problems than it solves. Virtual nursing is most beneficial when it is implemented purposefully with clear workflow definitions and adequate training for all involved.

I also see potential for work on the technology side. Nurses reported delays in messaging between virtual and bedside nurses, equipment failures, and camera  and sound issues. These are examples of failures in workflow design, equipment selection, and testing. Vendors in this space should be doubling down on creating tools that actually fit into the hospital unit rather than those that look great in a demo.

As all good study authors do, the team noted the need for additional research, including analysis of the technology in units that were not part of the study, or analysis of variation across hospitals.

For hospitals that are deploying virtual nursing, it’s a prime opportunity to involve nursing informatics and clinical informatics experts to ensure that solutions drive value through improved outcomes and staff satisfaction. If implemented thoughtfully, virtual nursing has real promise. But gaps exist between the promise and reality. Closing them will require more than buying a platform and pointing a camera at the patient’s hospital bed.

Is your organization using virtual nursing? Has it lived up to the sales pitch or caused more problems than it has solved? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 3/5/26

A health system that went live on a system-wide EHR five years ago with promises to upend its best-of-breed strategy is sending notices that it is finally migrating its laboratory systems. The lab migration project has been underway for quite a while, although it seems like communication to end users is just beginning.

The first communication explained the what and why of the project. It also outlined the communication cadence, so that people can be on the lookout for more information as the go-live gets closer.

I suspect that the team is planning to over-communicate for two main reasons. First, no one likes surprises. Second, many physicians, especially those who are community-based rather than being employed by one of the health system’s entities, likely assume that the project has already been completed and aren’t anticipating changes.

We will see how the communications unfold as go-live approaches. I’m sure that the finance folks will be glad to stop paying maintenance to multiple vendors.

The hot topic around the virtual physician lounge this week was an article in Nature Medicine that looks at how ChatGPT Health performed at triaging medical emergencies. ChatGPT Health, which launched in January of this year, was designed specifically to handle consumer-driven, health-related queries.

The authors submitted a set of 60 clinical vignettes across multiple clinical domains and conditions to the chatbot and to a panel of three physicians. The physicians triaged them based on clinical guidelines and their own expertise.

The tool underperformed. It failed to correctly identify one-third of non-urgent cases and nearly half of emergency cases. It recognized stroke and anaphylaxis as emergencies, but failed to refer the user to the emergency department for the life-threatening conditions of diabetic ketoacidosis and impending respiratory failure.

Other scenarios tested biases, such as when family or friends minimize a patient’s symptoms.

The ability to appropriately generate crisis intervention messages was unpredictable. Interventions appeared more often when cases discussed suicide generally than when the discussion included a description of a specific method of self-harm.

The authors conclude, “Our findings reveal missed high-risk emergencies and inconsistent activation of crisis safeguards, raising safety concerns that warrant prospective validation before consumer-scale deployment of artificial intelligence triage systems.”

User access to ChatGPT Health is limited by waitlist. Parent company OpenAI says it will make it widely available when it has finished validating its safety and reliability.

I’m surprised by the tool’s poor performance. Triage protocols have been available for many years and are commonly used by nursing staff in primary care offices. I wonder if the model was trained using any of those references or if those weren’t included because of intellectual property concerns.

We’re partway through the spring conference season, with ViVE in the rearview mirror and HIMSS on the horizon. A fair amount of alcohol flows at health tech conferences and it’s not just during the after-hours parties. It seems like happy hour events on the show floor are an expectation rather than an exception. A timely piece in The Harvard Gazette examines the effects of binge drinking on the digestive system.

Authors of the study, which was published in November, found that a single episode of binge drinking, which they defined as four drinks in a two-hour period for women or five for men, can make it harder for the small intestine to keep bacteria from entering the bloodstream. The research was performed in mice with the alcohol administered by gavage, which is the research equivalent of a beer funnel. I’m going to have a hard time keeping that imagery out of my mind next week when I see people hitting the cocktail circuit at HIMSS.

A fair amount of literature shows that younger generations are consuming less alcohol than older groups such as Gen X and Baby Boomers. Given the amount of alcohol-fueled bad behavior that I’ve seen during my time in the industry, that’s probably a good thing.

Speaking of HIMSS, I’ve got my trusty sneakers packed so I can stroll the exhibit hall in comfort, but I’m also taking my dancing shoes for after-hours sparkle. I’ll be making at least one trip to the Bellagio to catch the fountains after dark, since they are my favorite of the excesses found on the Las Vegas Strip.

I will be stopping by the booths of our sponsors (anonymously, of course) and looking for the best booth décor and of course footwear. Stay sharp during those booth shifts and save your cell phones for scanning badges, sharing party invitations, and emailing me your cute shoe pictures.

From Jimmy the Greek: “Re: AI tools in remote meeting platforms. My organization allows us to use them to create transcripts and summaries. It’s been helpful, but I literally laughed out loud when this turned up in a recent recap of a section that the AI tool titled ‘Product Staffing Woes and Teen Sleep’”: 

Robert and Susan discussed Susan’s staffing challenges, with Susan noting she was down to 1 3/4 of her intended six developers. Robert offered to support, but explained he couldn’t help directly. Robert then shared his personal experience with his teenage son‘s morning routine difficulties, leading to a discussion about teenage sleep patterns and morning habits. The conversation concluded with Robert introducing the topic of the next big project.

Pre-meeting small talk is common, especially in organizations where meetings don’t start on time. Seeing it memorialized highlights how much time is actually spent talking about topics that aren’t moving the organization forward. If you’re seeing a lot of these types of items in your summaries, it might make sense to disable the feature that automatically starts recording and transcription, and instead, manually start the process when you are ready to begin the meeting.

What’s the most amusing thing you’ve seen in an AI-generated meeting transcript? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 3/2/26

Clinical informatics is a broad subspecialty. Board certification requires being knowledgeable across a broad range of domains. The American Board of Preventive Medicine, which along with the American Board of Pathology can grant certification, distills it for its website:

Physicians who practice Clinical Informatics collaborate with other health care and information technology professionals to analyze, design, implement and evaluate information and communication systems that enhance individual and population health outcomes, improve patient care, and strengthen the clinician-patient relationship. Clinical Informaticians use their knowledge of patient care combined with their understanding of informatics concepts, methods, and tools to: assess information and knowledge needs of healthcare professionals and patients; characterize, evaluate, and refine clinical processes; develop, implement, and refine clinical decision support systems; and lead or participate in the procurement, customization, development, implementation, management, evaluation, and continuous improvement of clinical information systems.

The description doesn’t specifically describe patient-facing tools, but it does cover individual health outcomes, improving patient care, and building the patient-physician relationship. These goals are easier to accomplish when clinicians have tools at our disposal that help patients understand their own health situation and provide education and information.

Plenty of other entities are trying to grab our patients’ attention, which can lead to interesting conversations in the exam room as we work to counter medical misinformation or try to lead patients to consider evidence-based care plans.

I was surprised to see a study in Communications Biology this week that looked at direct-to-consumer testing. The results of tests that look at the microbiome of the digestive system varied dramatically among laboratory providers. The authors sent identical stool samples to the vendors, but each identified different bacterial levels, and only three of more than 1,200 bacteria were consistently identified across all the reports.

It wasn’t just variability between single samples that were sent to multiple facilities, but also among identical samples that were sent to the same facility. For one set of samples, the lab identified one submission as “unhealthy,” while two identical submissions were “healthy.”

The authors hoped to better understand the consistency and reliability of direct-to-consumer testing, which is not required to comply with the same level of regulations that traditional clinical laboratories must meet. Many of these tests fall under the category of “wellness” rather than being designed to diagnose a specific condition. Many physicians find the term “wellness” irritating because it has been used to hawk everything from unregulated botanical substances to jade eggs that are to be placed in the vagina to enhance sensuality (they are also an infection risk and may cause pelvic floor dysfunction, so those are a “no” when patients ask.) 

The authors found that bacteria in the genus Clostridium had the most variability in the reports. Three labs failed to detect it in one or more samples, and one reported it at five times the expected level.

The authors attribute the variability to different reference databases, reporting cutoffs, sample processing protocols, testing methods, and quality control standards. One of their goals was to make a point that just because a direct-to-consumer test is popular doesn’t mean that it is accurate, and that patients should understand the limited evidence that is behind such tests.

As a middle school science fair judge, it is an issue when three identical runs of the same experiment give different outcomes. For those who are curious, the paper details how exactly they prepared the identical specimens, all of which were obtained from a single donor.

Some direct-to-consumer tests get a lot of attention and often lead to patients arriving at the office of their primary care physician, asking us to treat something that isn’t actually a problem. I’ve seen multiple people bring in salivary hormone test panels that aren’t evidence-based and also allergy testing results that can be downright dangerous if not handled appropriately.

I enjoy working with patients who are engaged and want to take action, but these visits often lead to lengthy conversations that may not fit in the typical busy primary care schedule. Also, patients are almost universally unaware that at-home tests are not of the same level and quality as those that we would order in the office or during a virtual visit.

The authors call on the industry to take concrete steps to improve the transparency and interpretation of gut microbiome testing. These could also be applied to other types of testing. Specifically, they call on labs to address the idea of clinical validity and whether testing yields data showing correlation or causation with respect to a given health factor.

They also call for improved analytical performance in the testing process, maximizing both accuracy and precision. They go further to recommend that the industry work with testing companies and other stakeholders to create guidelines for testing, which would improve the validity of testing as well as the confidence of consumers who seek it.

Although patient-directed stool testing isn’t something you typically hear much about, research like this highlights some of the opportunities for clinical informatics experts to lend their knowledge to the task. We can help identify if a population-level issue  needs to be investigated, perform qualitative and quantitative research to understand the scope of the problem, support researchers as they seek data and information around the topic, and identify how the findings might be used to improve patient care.

We can also configure the tools at our disposal to help identify which patients would benefit from such testing, configure clinical decision support systems based on new evidence, and automate the creation of treatment plans based on the results while delivering effective patient education along the way.

Many of my colleagues think that clinical informatics team members just build order sets and flowsheets in the EHR all day. They don’t necessarily have exposure to all the different types of healthcare technology we can employ and how it can have an impact on the patients and communities that we serve.

As more of us enter the field, we should be able to provide that kind of education and exposure to our specialty. Our colleagues should know what we can do, just as they know how a cardiologist or pulmonologist can contribute to the care team.

What’s the most interesting clinical informatics project your team has done? Has your group built any tools that address direct-to-consumer testing or management of those results? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 2/26/26

I appreciated Mr. H’s comments earlier in the week about the challenges that companies encounter when trying to justify the cost of ViVE and HIMSS exhibits via booked revenue.

As a CMIO, I am unlikely to do business with someone just because I saw their booth at a conference. In fact, having a poorly prepared or apathetic booth staff is probably worse than having no booth at all.

Years ago, my CIO had recommended that I follow up with a vendor whose rep he had spoken to at a high level. I arrived at the booth, identified myself, and said that my CIO had referred me to take a look. I received the conversational equivalent of a pat on the head, with an instruction to come back when my CIO could also participate. I guess they missed the title on my badge and didn’t understand that I was the one with the actual decision-making authority for clinical applications.

In preparing to attend ViVE, I spent too much time deciding what to wear given temperatures ranging from 50 to 80 degrees. And of course, figuring out which shoes to pack. I was grateful to have HIStalk’s Guide to ViVE document to help me scope out some visits with vendors that weren’t on my list.

I noticed that some savvy vendors didn’t list booth numbers, but instead provided a list of their executives who would be on site and instructions on how to book a meeting. I also appreciated those who highlighted members of their company who would be speaking, the planned topics, and where to find them. Those kinds of listings are more likely to catch my attention than a boring blurb about being a cloud-hosted SaaS platform just like everyone else.

In traveling to ViVE on Sunday morning, I was caught in the gap between the Department of Homeland Security saying that they would be suspending TSA PreCheck security lines and the subsequent reversal of the decision. I travel often and at generally the same time, so I recognize a lot of the TSA staffers that typically work PreCheck at my airport. They’re usually pretty chill, even during busy Monday rushes.

Going through the “regular” security line on Sunday, the agents seemed more stressed. Travelers were also certainly stressed. Many who usually go through PreCheck didn’t seem to know how to put their items in a bin or get through quickly. Fortunately, I had seen the announcement of closures before I went to bed on Saturday night and left early. Even so, the security line was extremely long. About an hour later, the PreCheck line was back open, providing relief to the chaotic main screening line.

I’ve been part of several startup companies, so I understand what it’s like to have to show up and work without pay while you’re waiting for your next funding check to clear. Several members of my family were without pay during the last government shutdown. It can be devastating for the average US family that isn’t positioned to absorb that financial challenge.

When traveling, remember that kindness costs nothing. If you have friends or neighbors who are being impacted by government shutdowns that seem to be our new normal, consider offering whatever support you can.

ViVE is an interesting conference due to its co-location with CHIME events. These tend to draw more CIOs, which can make for a higher proportion of conversations with attendees who not only have a budget for solutions, but also the authority to spend it. Still, a “see and be seen” element exists. 

I overheard a couple of folks talking about how they didn’t really have a plan for the conference. They weren’t sure why they were there, or how their attendance was adding value. They were, however, happy to have gotten away from the northeastern US before winter storms hit, and seemed to appreciate the California sunshine.

I also overheard someone who said that he was prospecting for his company that was less than a week old. As a seasoned buyer, I hope he’s not leading with that tidbit.

ViVE provides meal service for a portion of the conference. They had a large seating area, but it was crowded. Outside food truck options were available and popular, but my schedule sent me to the grab-and-go option more than I would have liked.

It was great to see old friends and meet new people. Monday was my busy night attending vendor events. The Healio AI launch party was seafood forward and seemed to have a good turnout. From there, I was off to the Supreme Communications event, which was casual but fun, and then to the Abridge soiree at the Ritz Carlton. I spotted quite a few CMIOs from top 20 health systems at that one.

The best party of the night by far was hosted by Evergreen Healthcare Partners and Fortified Health Security at the Grammy Museum. Attendees had access to an exhibit featuring Tejano music queen Selena. The menu choices were on point, particularly the mini salted caramel chocolate tarts. I had an early morning of work waiting for me in the Eastern time zone, so I was back at my hotel early.

Following my calls, I made my way to the convention center and attended a few sessions that seemed meatier than those that I encountered at HLTH in the fall.

I noticed several people who were wearing microphones even though they weren’t speaking. I wonder how much of their day they record, or maybe they just aren’t taking their microphone off between times they need it. It reminded me of the early days of Google Glass, when people had to wonder if they were being recorded. If you are one of those folks who always has a microphone at the ready, feel free to weigh in with your strategy.

The most eye-catching booth backdrop that I saw was this one from the Berwyn Group. It drew me into stopping in to hear their pitch, where I learned about how they support organizational population health efforts by ensuring the accuracy of information when patients are deceased. I hadn’t thought about that in detail, other than how it impacts me in primary care. The team was great to talk to and explained their business well, so if you’re in the market for a solution to support death audits, give them a look.

As always, IMO Health brought their footwear A game to the conference. On the last day of the show, I saw a woman who was walking to the show floor wearing flip flops at 8 a.m. I don’t know if that was her first choice, or whether it was need-based following less than stellar footwear selections earlier in the week, but kudos to her for sporting them proudly.

I have enjoyed my time in the industry and in seeing tools evolve. I had a chance to chat with the folks at Medicomp Systems, who showed me their generative AI capabilities and how new elements are working seamlessly with the Medicomp Quippe tool. I was glad to see that one of their demo personas named “Seymour Patients” continues to be alive and well, or at least as much as one can be in the virtual world.

Overall, it was a more productive week than I anticipated, which is always a nice surprise. Now I’m hoping for the best for my trip back to the East given the number of canceled and delayed flights and the amount of snow on the ground.

If you attended ViVE, how was it? What were your biggest takeaways? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 2/23/26

It’s clear that AI is here to stay. I’ve spent quite a bit of time looking at studies that seem to be either proving its value or dismissing it on the basis of inaccuracy and risk.

Healthcare people tend to look at it with a specific lens. I reached out to contacts in other industries to better understand how they are approaching it, and whether their professional organizations have produced policies or recommendations around its use.

The first person who responded to my query is in the field of law. The initial portion of his response addressed the high-profile problems with AI that have surfaced in the legal world. A number of cases involved attorneys who used AI to construct briefs, but failed to catch that the AI fabricated citations for cases that didn’t exist.

Similar to what we encounter in healthcare, issues exist with the content on which AI systems are trained. Attorney-client confidentiality must not be compromised by becoming part of a data set. Similar risks involve algorithmic bias and discrimination. Attorneys have been sanctioned for misusing AI, with some being fined for fictitious citations.

The legal community is discussing accountability for the use of AI. Ethics experts agree that attorneys are ultimately responsible for the accuracy of matters that are being handled in their name.

My attorney friend shared his opinion that even the best AI isn’t as good as some of his most seasoned paralegals and researchers. His firm tends to proceed with caution, although it does not have a formal policy on the use of the technology. He thinks about about using AI to create documents similarly to having a summer legal intern do it. He reads everything with a critical eye in case it misses the mark, just like interns sometimes do.

We chatted a bit about the idea that AI probably isn’t as good as a law student at the top of their class, but might be better than a student at the bottom of their class. This has parallels with medical education. It is different asking a fourth-year sub-intern to present a case than to ask a third-year student who is on their first clinical rotation to do the same.

We agreed that the idea of blind trust in AI is risky, especially when professional licensure is on the line.

The American Bar Association issued its first guidance on the ethics of AI use in 2024. It specifically noted the need to ensure that legal billings are appropriate for tasks that are conducted using generative AI tools.

The attorney in question is also a commercial pilot. He had a few things to say about the use of AI in the aviation space. Airlines have been using it for operations functions, including maintenance optimization and the modeling of passenger behaviors such as their likelihood to check bags or buy additional services and amenities. Consumer-facing AI includes support chatbots and booking and ticketing systems.

On the maintenance side, AI can help with troubleshooting complex airframes that generate sensor data. Mechanics also use it for maintenance documentation.

He mentioned incorporating AI into flight simulator systems. It uses real-world cases and events to create realistic emergency scenarios that might go beyond the experience of a human simulator operator or operational handbooks.

I must have posed my question at just the right time, because he mentioned a recent announcement about the US Air Force’s Flying Training Center of Excellence. It is developing an AI-based “Instructor Pilot GPT” that is designed to interact with students who are undergoing pilot training. The tool will be trained on flight manuals and aviation documentation. It will help student pilots assess their performance and will provide rapid access to reference procedures. Similar to the commercial side, they hope to use the technology in flight simulators.

The Air Force uses a closed training environment that contains documents such as military protocols, federal guidance, and flight-related publications. I chuckled when I read a quote from one of the people who is involved with the project, who referred to the subset of information as a “data pond.”

Another comment in the article sounded a lot like the conversations that we are having regularly in medical education. Students are on their phones using LLMs every day, so they will expect it as they move forward in training.

The article also notes important concerns that I hadn’t considered in healthcare, such as cybersecurity risks. What happens when your fighter jet GPT gets hacked and harmful information is injected? The same thing could happen to a healthcare system, which would provide the ultimate example of medical misinformation.

As far as professional organizations or regulations, the Federal Aviation Administration issued a formal notice on the use of generative AI tools and services in March 2025. The first page of the document highlights the need to ensure that generative AI use “is conducted in an ethical and responsible manner.”

The notice applies only to FAA’s employees and contractors, but it includes policy elements that are similar to what I see in hospitals and care delivery organizations. These include a requirement to request approval for using generative AI software, the ability to request support for specific use cases that have already been identified, and the need to ensure that AI tools that are found on the internet have been approved by the organization.

The FAA also cautions about the risks of AI infringing on intellectual property, the need to review AI-generated content for accuracy, the need to be transparent about where AI tools are being used, and the principle that it shouldn’t be used to “perform or facilitate illegal or malicious activities.”

I am waiting to hear back from contacts in other industries and will share if I receive compelling insights. If you or your organization does crossover work in areas other than healthcare, how are those industries tackling the use of generative AI? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 2/19/26

Clinical Informatics is a broad specialty. But depending on our job roles, we sometimes only get to work in a handful of its domains.

I’ve always enjoyed public health informatics work and being able to identify opportunities where we can make changes that help thousands of patients and families. Pregnancy continues to carry higher risks in the US than in other developed countries. I recently ran across the Baby2Home app that is designed as “a one-stop platform supporting perinatal families through an evidence-based collaborative care model.”

The tool includes mental health screening, access to stress management resources, and connectivity to care managers for support. It also offers the ability to track infant-related parameters such as feeding, sleeping, diaper changes, growth, and vaccinations.

Researchers tested the tool during a multi-year study that ended in 2025, with 642 first-time parents randomized to receive either typical postpartum care or typical care plus the app. Members of the intervention group had improved mental and physical health scores and were more confident in their parenting skills compared to those in the control group. The data was presented at the Society for Maternal-Fetal Medicine meeting on February 11. Although the tool is currently investigational, I found it compelling and will be watching to see what happens next.

Speaking of companies I’m following, I was delighted to see a recent LinkedIn post containing a video from the folks at CognomIQ. The peppy beat perfectly channels their call for organizations to “Drain that stagnant swamp of a 1990s data lake.” The post says, “We’re not mincing words or hiding behind flowery rhetoric.” They weren’t kidding, since they call out several prominent vendors by name.

The snappy chorus of “Healthcare data sucks, you can’t dress it up” had me rolling. So did, “We build the board a house of glass and pray the question’s never asked.” All of us have been there, but few are willing to become a lightning rod by saying it out loud. Props to the team that created this campaign. I’ll see you on the dance floor.

From Captain Incredulous: “Re: LinkedIn. In a moment of weakness, I accepted a LinkedIn request from a friend of a friend. Within 24 hours, my new connection emailed me at my work address. He asked me to introduce him to a well-known CEO in my network and advocate for a partnership meeting. He even went as far as to suggest a draft email for me to use. He has now sent three emails about this issue.”

The reader shared the email thread, and it is certainly presumptuous. Additionally, I found some irony that the reader failed to notice: the draft email included mentions of how the author’s company could help the CEO at his previous employer rather than his current one. Putting myself in the reader’s shoes and knowing the CEO in question, I would definitely mention it to him, if only for a chuckle.

My inbox is bursting with cold email outreach efforts asking to connect at ViVE next week. Colleagues are receiving similar messages from startups that are desperate to meet. Most use words like synergy, partnership, and collaboration. Of those in my inbox, many include the salutation “Hey.” I know ViVE is the hip cool cousin of the conference scene, but it still feels unprofessional to me.

My favorite request just said, “I will be attending VIBE and connecting with people across the healthcare space” without stating the requester’s company or why it might be relevant to me. The misspelling of the conference name captured my attention, but I’m still not going to book a meeting.

From Jimmy the Greek: “Re: AI answers that are obviously incorrect. Check out this thread, where I asked whether I should walk or drive to the car wash.” AI recommended walking if the user didn’t mind exercise and if the weather was decent. It suggested driving if short on time or if the route isn’t pedestrian friendly. It completely missed out on the fact that the car would not be at the car wash if the user walked. It confidently stated that “walking is the more elegant move,” unless the car wash was of a certain configuration. It concluded by asking the user to specify what kind of car wash was involved so it could “pick the smoothest plan.”

The average healthcare IT team consumes a large volume of caffeinated beverages, so this article in the Journal of the American Medical Association caught my eye. The authors investigated whether long-term consumption of coffee and tea is associated with dementia risk and cognitive function. The study was large, with 132,000 participants and up to 43 years of follow up. The findings showed that, “Greater consumption of caffeinated coffee and tea was associated with lower risk of dementia and modestly better cognitive function, with the most pronounced association at moderate intake levels.” No similar association was observed with decaffeinated coffee.

Study participants were healthcare workers. Females were drawn from the Nurses’ Health Study and males from the Health Professionals Follow-up Study. Data was collected every two to four years using a food frequency questionnaire.

As we all know, correlation does not imply causation. One should also be cautious about extrapolating these findings to non-healthcare workers since many of us have other behaviors that might not be typical. A shout-out to all the emergency department workers out there who have disordered eating habits, disrupted sleep, and fond memories of colleagues sneaking out through the ED doors to smoke cigarettes before returning upstairs to counsel patients about smoking cessation.

I’m a stickler for starting meetings on time to be respectful of those who are punctual. I’ve been fortunate to work in organizations that use the 25/55 meeting scheduling paradigm, which gives people five minutes to transition between calls or meetings. I’ve seen how it can help more meetings start on time.

Even without a back-to-back meeting schedule, some people are habitually late. During a recent discussion on meeting management, a colleague shared an article about people who arrive late and the causes. Although some people may be overscheduled or previous meetings might end late, there is also the phenomenon of “time blindness,” in which people are unable to identify how long an activity might take or to understand how much time has passed.

People might also arrive late if they don’t want to engage in pre-meeting banter. I’ll admit that I haven’t thought much about that. Starting on time reduces the available time for small talk, but it’s something to think about the next time I’m on someone else’s meeting and they’re “just waiting a few more minutes for people to arrive.”

How does your organization support on-time meetings? Are agendas and timekeepers a must or something only found on the wish list? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 2/16/26

I met up with some clinical informatics friends over the weekend. Our conversations focused on a few key themes. One was how much time we’ve spent in the field, as we realized that the most junior member of the group just hit the 20-year mark.

None of us set out to specifically look for technology-related roles, but each of us found our way to the field as we identified gaps in what was happening around us and stepped in. The most senior of the group got involved because he had an engineering background and saw what was going on elsewhere in the country with respect to electronic charting. He wanted to see his institution be a leader in the field rather than a follower, and worked with a big tech company to build a clinical repository for the organization.

It initially compiled data from just one hospital, but as more hospitals joined the system through mergers and acquisitions, the repository became more complex. The team that supported it needed clinical expertise to handle issues like normalization of laboratory values and standardization of test orders. He had some great stories about how the repository grew and became integrated with electronic medical records. It lasted for more than two decades before being retired in favor of a half-billion-dollar investment to move everyone to a single platform.

My colleague still refers to the systems clinicians use as EMRs. That led to a discussion of EMR versus EHR. One member admitted that he uses the terms EMR and EHR interchangeably because he keeps getting loaner computers from his IT department and hates having to go into all the different dictionaries to prevent the system from autocorrecting EHR to HER.

That admission led to a discussion about how the Microsoft Office suite and Office 365 applications handle such things, which bafflingly makes the setting device-specific rather than defaulting from the user profile. I’m no expert, but I know that it’s annoying every time I get a new laptop. I usually end up consulting Google because finding it in the application settings isn’t intuitive.

Quite a bit of back and forth ensued around the merits of EMR versus EHR. I was surprised by how passionate some of the people are about one or the other. We all agreed that “health record” is more comprehensive than “medical record.” One of the group felt that the latter sounded more serious since “health” is often linked with “wellness,” which often includes non-evidence-based and consumer-oriented services.

People pulled out their phones to look for articles for and against each term. I was surprised that the first response that popped up in my EMR versus EHR query was that “EMRs are mainly used by clinicians for diagnosis and treatment, while EHRs are designed to be shared and accessed by the patient. EMRs are less susceptible to cybersecurity issues, since they are not being shared with patients, but are securely managed by the practice.”

That got us rolling, since none of us has encountered a cybersecurity issue related to patient use, but we’ve seen plenty of times where trained employees and hospital medical staff fell victim to phishing schemes. An ASTP/ONC blog addresses the topic, but it’s from 2011. Some of its language is identical to what I found in that first response, which leads me to suspect that the vendor had done some copying and paraphrasing from ASTP.

Most of us agreed that now it’s kind of a stylistic thing and we aren’t bothered when vendor folks use the words interchangeably. I’ve worked with vendor organizations whose style guide spells out which term to use when referring to their products, but not everyone has one of those. I remember reaching out to Epic a few years ago to ask if it had an official position one way or the other and was told that it doesn’t. If that has changed, feel free to drop a comment and let me know the current state.

That conversation led us into a whole “words have meaning” discussion. That immediately drew me in because the industry is plagued by people who use words that don’t make sense. Maybe it’s a phrase they learned during a corporate training class, or perhaps they saw it in an article. When they come to me as a CMIO and start spouting words that don’t completely work together, it makes my attention go zinging off elsewhere.

One colleague, who is a doctor of osteopathic medicine, noted that nothing turns him off more than using “MD” as shorthand for “doctor.” Both have the same number of syllables, so it’s not like saying MD is faster or easier. It’s not worth it to use it in a way that alienates a subset of physicians.

One of the group brought up a recent position paper in Annals of Internal Medicine that addresses “The Ethical Significance of Names in Health Care.” It’s an analysis of the physician versus provider debate, referring to the latter term as contributing to “deprofessionalization.” The authors felt that their examination was unique because it looks at the situation from an ethics perspective.

I popped up the article while we were talking and was excited to see a Shakespeare quote in the first screenful of text. The article includes a review of the origins of care-related words, including patient, physician, doctor, and compassion.

One might have expected that the article would recommend simply not lumping physicians in with other kinds of heath care providers, but it went further to suggest that we get rid of the word “provider” entirely: “Language in health care has ethical and practical implications. Physicians should be referred to as physicians, not providers. Also, when describing professionals with varied credentials who care for patients, the terms clinicians or health care professionals, should be used.”

I’ve been a fan of the word clinician for a long time. It’s shorter than the recommended alternative. I plan to stick with it.

Not surprisingly, we stumbled into a discussion of clinical informatics versus medical informatics, and even a debate about informaticist versus informatician. There’s actually a paper from 2024 called “Informaticist or Informatician? A Literary Perspective”  that goes deep into the history of the two. My colleague quickly sent me a link.

It is a fascinating read. The authors close with a clinical informatics spin on a classic Shakespeare quote, which made me smile. They also received a chuckle with their line, “Whether you are an informaticist or an informatician, may you collaborate better than the Montagues and Capulets.”

Even though the conversation was all over the place, it’s always good to catch up with colleagues who have fought the same battles and who have made it through the same topsy-turvy changes within the industry. They are not only knowledgeable, but are generally a fun bunch. I feel privileged to have them on my phone-a-friend list when times get tough.

Is your company in EMR or EHR mode? What phrases, taglines, or buzzwords make you cringe? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 2/12/26

JAMA Network Open published an interesting Original Investigation last month that set out to answer the question: “Are greater levels of generative artificial intelligence (AI) use by US adults associated with greater levels of depressive symptoms?”

The authors surveyed 20,000 adults. Of those, 10% reported daily use of generative AI tools and 5% said they use it multiple times per day. Higher levels of AI use were linked to increased depressive symptoms, especially in certain age brackets. The authors note that additional research is needed to understand the nature of the association and differences in impact.

AI truly is everywhere, but I was surprised to see it on the new RealFood.gov website that replaced MyPlate.gov. Users who post a question are redirected to the Grok AI tool.

I took advantage of a preloaded sample question that addresses the challenges of eating non-processed foods when the closest food source is a gas station and an individual is on a fixed income. The answer also recommends that users consider food stamps, food pantries, and other community-based options.

Some of our government health leaders are promoting animal fats as healthier options so I decided to poke the AI bear and asked whether animal fats cause heart disease. Grok sided with existing data that suggests that higher consumption of meat, dairy, and eggs is associated with increased risk of cardiovascular disease. Score one for Grok in keeping with established mainstream science.

I asked Grok what happened to MyPlate. I was impressed that it clearly identified those who are responsible for the new approach, none of whom are physicians, nutritionists, or dietitians. Grok noted criticisms of the “return to a pyramid format that was abandoned partly because it was harder for people to interpret compared to the simple plate model.”

It also offered a follow-up question that compared the old guidelines to the new. It also identified critics, including Harvard Nutrition Source, Stanford, and public health groups, and their concerns, such as the scientific process that was employed.

As a family physician with a keen interest in the literature, I’m concerned at how this paradigm minimizes fiber. If flipping the guidelines were a scientific study, I’m not sure it would make it past Institutional Review Board approval. So many of us feel like we’re living in an unregulated science experiment.

I asked Grok its thoughts on the links between red meat, fiber, and colon cancer. It gave me a deep dive into exactly how red meat contributes to colorectal cancer risk and the protective nature of fiber. At least I have some facts in my head for the next patient who comes in asking about their diet, because I won’t be recommending the new guidelines without extensive discussion of context and validation.

Amazon One Medical has introduced a new beta feature to help patients navigate their lab results. The Health Insights functionality, which is included in the One Medical membership, provides commentary on 50 standard blood work results while grouping tests together by health domains such as cardiovascular and immune function. Users complete a questionnaire, then the tool generates a wellness score and offers evidence-based lifestyle recommendations with scientific references. I’m not a subscriber, but I would be interested in hearing from anyone who has had a chance to check it out.

Given the number of health systems and care delivery organizations in the US, it seems like someone is always merging, acquiring, or separating. The M Health Fairview brand will be retired from a subset of hospitals in 2027 as part of a new agreement between the University of Minnesota and Fairview Health Services. The deal, which was approved by the university’ board of regents on January 30, shifts the partnership from a joint clinical enterprise to an academic affiliation.

Healthcare administrative types may be familiar with the nuances of those structures, but I doubt that the majority of patients who live in the communities that the organizations serve will understand what the shift means.

The groups will still work together with regard to the University of Minnesota Medical Center. However, details of the agreements that impact the respective physician groups are not yet public. In short, the university will maintain control of academic and research functions, while Fairview will be responsible for hospital operations.

Relations between the two organizations have been strained for a number of years. The university opposed a merger between Fairview and Sanford Health, after which Fairview opposed the university’s push to merge with Essentia Health. The Minnesota Attorney General became involved, triggering facilitation and mediation efforts. The new agreement will be in force for 10 years compared previous partnership’s 30 years.

Speaking of branding, Texas Health Resources has inked a deal to buy naming rights to Texas Health Mansfield Stadium. It includes “prominent branding throughout the stadium, a refreshed logo and a new digital presence… while also laying the foundation for a wide-ranging collaboration focused on community health initiatives.”

The hospital president and CEO indicated that the facility “will serve as the central wellness hub for the entire Mansfield community.” Given the fact that stadiums are typically closed and locked when teams are not in play, it will be interesting to see how it becomes a wellness hub. 

The article mentioned that Texas Health will be the facility’s “Official Health Partner” and “will collaborate on initiatives designed to promote healthier lifestyles, including community programming and enhanced food and beverage offerings that emphasize more nutritious options.” Hide the jumbo nachos and the foot-long corn dog, y’all.

The ASTP meeting kicked off earlier this week, with a focus on the intersection of health IT and the priorities of the US Department of Health and Human Services. Main stage sessions include titles such as “Health Technology for Transparency and Affordability” and “Making America Healthy Again through Technology + Care.” Breakouts covered standards, data exchange, and of course information blocking.

I’m interested to hear from anyone who attended, and in particular, from people who attended previous meetings. How was the attendee mix compared to the past? Were topics handled similarly? What was the overall mood? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 2/9/26

Most of us have figured out by now that it’s difficult to spend a day without using some kind of AI-powered tool. I was a fairly early adopter of the Amazon Echo devices after receiving one for a gift. I used them until recently to control some of the lighting in my home. (RIP, Wemo smart home devices.)

I mostly use the Alexa assistant to get quick answers to straightforward questions, such as “What is the temperature?” and “What is the forecast?” so I can do a little planning before I drag myself out of bed in the morning. In the kitchen, I primarily use it for timers or to play music while cooking. I haven’t been impressed by the “skills” that it offers, however.

A couple of weeks ago, I started receiving teasers to upgrade my device to new voices and personas. I held out since I didn’t want to be a beta user. Ultimately, I gave in and was pleased to find a somewhat sassy voice that is officially described as “grounded” or “easygoing.” It reminded me of one of my favorite audiobook narrators, so I decided to give it a try.

Over the last few days, I’ve noticed some quirks. I’m not sure if it’s specific to the voice I selected or if something is going on with the cloud services, but Alexa started giving me more information than I was asking for. Instead of simply giving me the current temperature and the forecast high, it added commentary like “it’s going to be a great day” or something similar.

Then I noticed it providing information that seemed disordered. For example, telling me that the current temperature was 38F with a forecast high of 47F, but that it currently feels like 44F, which just doesn’t make sense. It also tells me that tomorrow’s forecast high will be in the 40s when the Weather Channel thinks it will be a dozen degrees higher.

Tonight, I was thinking about some travel plans and asked Alexa what the correct time zone is for Nashville since I can never remember and was multitasking. Alexa confidently told me that “Nashville is in the Central Time Zone,” but went on to offer information that I didn’t ask for and told me the time. Since I’m on Central time, I was surprised that it was wrong.

I was curious to see what Alexa would say if I called out an incorrect answer. It replied, “You’re right, I should have been more specific. Nashville, Tennessee is indeed in the Central Time Zone.” I had to specifically ask the time and it finally answered correctly.

It’s one thing for a system to provide inaccurate information in response to a question, but it’s another to offer incorrect information that wasn’t even asked for.

I’ve seen some positioning for virtual assistants, including Amazon Alexa and Google Assistant, as general purpose tools that can help the elderly age in place and manage daily routines. They are also supposed to be helpful for reducing social isolation and providing voice-activated medication reminders. What happens, however, when those tools don’t do the right thing? What happens when the tools are confused about what time zone they’re in and it leads to a patient taking medications more than once? In that situation, a simple non-AI alarm app might be more reliable and provide greater safety for patients.

Later in the day, I found an email from Amazon listing how “Early Access” customers made Alexa+ better, including such items as being more responsive during chats, a better sense of when you want to engage, and that it “adapts to your vibe” by learning and adjusting to the user’s communication style.

Honestly, I’m not impressed. As soon as I get some free time, I’m going to experiment with some of the other voices to see if they’re as problematic as the one I selected or if the entire system is just not meeting my needs any more.

Meanwhile, I’m starting to make a list of all the grossly inaccurate responses that I receive from AI tools. I recently read a novel that was based on a true story and asked an AI-powered search tool what happened to some of the main characters later in their lives. The answers should have been straightforward, since the characters I asked about were part of a World War II effort to project works from the National Gallery of Art by storing them at the Biltmore Estate in Asheville, North Carolina.

Instead of providing facts, I got some wild speculation about the Gallery’s director, David Finley, which required visits to a couple of primary sources to fact check. As an upside for the next time I need a random nugget of obscure information, I now know that 40 cubic feet of Finley’s personal papers and artifacts now reside in the Gallery archives, including dried flowers, a cigarette case and lighter, and postcards from a honeymoon in Greece.

The Super Bowl is a big deal in the US. A lot of watercooler conversations happened at the end of the week around whether people had plans for the big event and which team they might be supporting. I was one of the people who don’t really have a connection to either team, but was really rooting for the combined squad of US Air Force and US Navy pilots who were slated to perform the pre-game flyover.

I have to say it did not disappoint. The seven-ship formation led by the B-1 was on point. Those of us with aviation geek tendencies knew there was more to come, and the US Navy livestream made our day as the camera panned back to catch the second B-1 approaching in full afterburner.

Having spent my career in medicine, I appreciate the fact that a seemingly short display like this is actually the end result of hundreds of hours of research, planning, and practice. It’s like one of those domino transplant surgeries where all the organs have to make it to the right patients in the right city at the right time with all the associated facilities and staff preparation.

The flyover planes originated in South Dakota and California. They were supported by refueling aircraft from Ohio, with everyone gathering nearby for the final maneuver. Each aircraft is supported by teams of maintainers who are in turn supported by other disciplines. Everyone is essential, much like in a hospital. Thousands of hours of training and education are behind each person’s ability to do their job when called upon. The pilots’ fist bump in the cockpit following the flyover was charming and I can only imagine how excited their families were to see that. (Photo taken from US Navy livestream).

If you partied for the Super Bowl, what was your favorite snack? Did your event include any heart healthy options, or was it all about Buffalo wings and pizza? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 2/5/26

It’s been a couple of years since I’ve written much about digital therapeutics. The number of vendors in the space is small, and the market got quiet after Pear Therapeutics, the one I knew best, declared bankruptcy in 2023.

Digital therapeutics require a prescription and are regulated by the FDA. At the end of 2025, the US Food and Drug Administration announced a pilot program to encourage the use of these solutions. The Technology-Enabled Meaningful Patient Outcomes (TEMPO) model began accepting statements of interest in January. Under the model, a subset of medical professionals can prescribe digital therapeutics before they are officially approved by the FDA, with the resulting real-world data being used to potentially support their clearance.

One reason these tools failed to gain traction was the reimbursement landscape, which left developers unable to build a sustainable financial model. Some companies pivoted into the direct-to-consumer space and marketed their tools as wellness apps to avoid regulation.

The TEMPO pilot was developed by the FDA Center for Devices and Radiological Health (CDRH). It is limited to prescribers who participate in the CMS Center for Medicare and Medicaid Innovation model for Advancing Chronic Care with Effective, Scalable Solutions (ACCESS). If you’re a providers who meets the acronym test and is planning to prescribe digital therapeutics, feel free to drop us a note.

A recent article in JAMA Oncology looked at the MyLungHealth tool, which can help identify patients who are eligible for lung cancer screening. The procedure is underused, with about 16% of eligible patients receiving the screening. The trial described in the writeup, which was conducted at the University of Utah and NYU Langone Health, showed how digital tools can help close gaps in care.

Lung cancer remains a leading cause of cancer deaths around the world. Screening is recommended for adults aged 50 to 80 years who have at least a 20 pack-year smoking history and who currently smoke or have quit within the previous 15 years. Barriers to screening include inaccurate or missing tobacco use history in patient records, missed opportunities to order screening, and lack of patient awareness.

The tool includes a patient education component with both videos and text-based content. Clinicians are alerted when patients engage. The study had 30,000 participants. Patients who received an intervention using a patient-facing tool integrated with the EHR patient portal completed more screening tests.

The authors noted that the end point of having a study ordered was a limiting factor. They encourage more research to look at strategies to ensure that patients complete the recommended CT scan. Props to them for also noting the need to test this approach in other care delivery settings, such as community-based primary care offices, to make sure that the findings are generalizable and to maximize impact.

They also noted the need to adapt the approach to address the needs of underserved patients, especially since patient portal use was required and rates of such use can be variable across demographic groups.

Most of the healthcare leaders who I talk to are trying to trim their budgets due to declining reimbursement and continued cost pressures. Vizient recently released data on healthcare expenditures and I was surprised to see that pharmacy costs are no longer the fastest growing expense category. Facilities and IT lead again, with IT hardware and software at a 5.66% inflation rate. IT services are close behind at 4.5%, with facilities management at 4.13%.

Other interesting tidbits: with the rise in medication use to treat obesity, bariatric surgery volumes are down 20%. The inflation rate for laboratory services is predicted to be less than 2%, which surprised me given the continued evolution of testing platforms and multi-result panels.

I work with a physician who is vocal about the tools we have for patient care. He is outspoken why AI is causing the downfall of civilization. He collects examples where AI tools have been wrong, specifically in situations where patient harm could have resulted.

His message of the week includes an example of uploading an image to identify a mushroom that a hypothetical patient might have eaten. The tool incorrectly identified it as being safe to consume, when in fact it was quite toxic. I’ve never been a mushroom hunter, but I’ve worked at a poison control center, so I hope that mycophiles and foragers are using multiple sources to confirm edibility before they sample their finds.

I appreciate his point of view and the fact that he provides interesting examples that make us think. But we’re not going to put this particular genie back in the bottle anytime soon.

One of my colleagues who is more accepting of AI told me about something called Moltbot, which apparently underwent a renaming in the time it took me to find time to research it. Now called OpenClaw, it’s an AI agent that goes beyond chatting and starts taking action. The solution is seeing rapid adoption given the fact that it’s free and runs locally. The tool can run using either ChatGPT or Claude models and can be assigned a vibe to embody as it goes about its work, which might involve executing commands or making changes to files.

The writeup in Scientific American had me chuckling as it noted that the tool “follows almost any order like a well-paid mercenary.” I’m curious about its potential, but leery of some of the risks as far as privacy and access. If you’ve given it a try, drop us a line.

One of our local care delivery organizations is looking to rebrand. I’m a little surprised because it has had no significant mergers or acquisitions that would indicate a responsible use of funds or a need to avoid confusion. It seems like more of a vanity project since the organizations have already been linked for decades.

They are apparently doing marketing outreach to local physicians, asking their opinions on logo and color combinations to see which have the most impact or best represent the partnership. I’m not sure if they’re also reaching out to patients for their opinions, but I would be curious to see how those might differ from those of the physician community.

Bottom line, however, is that this makes me a little angry. The organization’s cheapskate tactics have negatively impacted patient care in recent years. I wish they would spend the money on issues that directly impact patient care and improve the health of the community versus trying to look better than their competitors.

Has your organization been through a rebranding effort? Did it deliver the outcomes it promised or was it not worth the cost and effort? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 2/2/26

I’ve been doing a bit more clinical work lately because of how the flu season has played out in my community. Rates of Influenza A have been rising over the last several weeks, but we were cautiously optimistic when we started to see a small decline in flu-like symptoms.

However, the flu season decided to deliver a classic one-two punch, because influenza B is now on the rise. Looking at the statistics, this year’s flu season is one of the worst in the last decade as far as hospitalizations for pediatric patients. Our local hospitals are swamped. Hospitals are boarding patients in the emergency department for a prolonged times because they lack staffed beds elsewhere in the hospital. With float pools exhausted, nurses are being reassigned to units that are outside of their core area of expertise.

Hospitals can be full of overly rigid policies and procedures, so I was surprised to learn that one of my colleagues was hired by a local hospital and fast-tracked through their credentialing process in under two weeks. Although it’s great to see that when there’s a will there’s a way, it raises the question of why hospitals can take up to 120 days to credential providers under normal processes.

Putting on my process improvement hat, I wonder whether the process contains steps that are less critical than assumed, and perhaps those steps are skipped during fast-track credentialing. Alternatively, pieces of the process might be able to be expedited at an additional cost that hospitals are not usually willing to pay.

Either way, I was glad to see her get back into the trenches quickly. Having a physician on the sidelines when they are willing and able to work is a loss to community’s patients.

For those of us that work for multiple care delivery organizations or who work infrequently, a fair amount of anxiety can be created when you decide to pick up a shift. When you’re a PRN or as-needed staffer, you are theoretically supposed to keep up with changes to the organization’s policies and procedures. You are also expected to be aware of any changes that have been made within the electronic health record or other tools.

An organization that I work with makes this easier for clinicians. They have a high level of maturity around their EHR governance processes and it’s rare for them to deliver updates more than once a month unless something has gone wrong. Their documentation is great. I typically store all of their update emails in a folder and read through them before I go back on shift so that the changes are fresh in my mind. I arrived at this process after trying a “read as you go” approach that wasn’t as productive.

Another facility where I’ve worked at makes a hash of this with a far less robust process. Instead of sending a single monthly email with release notes that follow a standard format, every builder who is working on a change creates their own messaging without any overarching review. Sometimes the descriptions of changes and fixes are vague, making it challenging to figure out whether they will affect everyone or if they even apply to your department.

For this facility, I still store everything in a folder and refresh my knowledge before reporting. Regardless of how well I try to read and comprehend, the first hour or two of my shift feels like being in a carnival fun house, with all kinds of surprises popping out at you.

One of my favorite organizations to cover is a direct primary care practice. The practice is not a Covered Entity under HIPAA and doesn’t do any third-party billing, their EHR is remarkably simple, and updates to the system are few and far between. The platform they use is remarkably patient-centric. Documentation is a breeze since you’re focused on documenting the clinical encounter rather than meeting billing and documentation guidelines.

The practice has templated the visit notes to have three areas of focus. The “Short Term” section is like a traditional SOAP note and captures issues that are addressed during the encounter. This might occur in person, by phone, or via video visit. A “Patient Progress” section captures the bigger picture of chronic or recurrent conditions.

I like the patient progress nomenclature. It feels more positive than the traditional problem list even though it’s doing the same thing in capturing whether a given issue is improving, worsening, or remaining stable. That section also includes tools to help visualize and close care gaps, monitor preventive services, and track procedures or orders that are due in the next month, quarter, or year.

I can still go to traditional problem list or past medical history or social history sections in the chart. But it’s nice to have things pulled directly into the note where you can see them and understand how they might connect with today’s issues without having to click around.

The last section is simply called “Horizon.” It’s a bit of a catch-all for everything that doesn’t fit into the other two sections, but it includes information that helps the clinician chart a broad course with the patient.

Rather than just having demographic and family history information, it graphically illustrates the patient’s support system. It includes information on their cultural beliefs and practices as well as their general preferences in care. You can go here to figure out whether the patient is motivated to make lifestyle choices or prefers medications to address issues. It’s also where you can see notes on their living will and healthcare directives as well as the nature of any end-of-life care discussions.

Patient portal messages are blissfully absent in this practice. Patients use a secure texting platform to communicate directly with the physician or their coverage. Those interactions are added to the chart at the end of the conversation.

It’s elegant in its simplicity, but it works, primarily because the physician has a smaller patient panel than most insurance-based practices in the area.

I always get whiplash when I go from covering this practice to working in a setting where I’m incentivized to see as many patients as quickly as possible. Still, it reminds me of what it must have been like to be an old-timey physician who really got to know their patients.

For those of you who work with different care delivery organizations, what are some of the most striking differences you see? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 1/29/26

The Journal of the American Medical Association published a research letter this week that looks at how authors are disclosing their use of AI when preparing submissions to professional journals. The JAMA Network has required such disclosures since August 2023. The authors reviewed the data to better understand how AI is being used and disclosed.

Papers in which AI use was declared increased from 1.7% to 6%. Common uses were creating drafts, searching the literature, editing language, developing statistical models, and evaluating data. AI use was more likely in Viewpoints and Letters to the Editor submissions than in Original Investigations.

The paper concludes that without a standard for confirming AI use, it’s difficult to know if authors are underreporting. They add that the results may show a greater need for journals to confirm how authors are using AI and whether it’s appropriate and accurate.

Clinician burnout continues to be a major focus for care delivery and professional organizations. One of the top symptoms that I hear about from colleagues is their inability to disconnect in the digital age. Physicians feel that they need to check their inboxes for patient results and respond to portal messages during off hours to avoid having them piling up.

A new article in the Journal of Medical Systems describes a randomized controlled trial around Reducing Work-Related Screen-Time in Healthcare Workers During Leisure Time (REDUCE SCREEN). Researchers used a straightforward intervention to examine whether a link exists between clinician wellbeing and the use of work-related apps on personal devices. A cohort of 800 physicians, residents, and nurses was divided into a control group and one whose members were instructed to take specific steps to reduce after-hours work, such as using out-of-office notifications and removing work apps from personal devices.

They found that after a scheduled weekend off, those in the intervention group had double the reported reduction in stress compared to those who weren’t instructed to make changes in device use. The intervention group also had an overall reduction in screen time compared to the control group. The study was limited by the fact that one-third of participants failed to complete the post-weekend assessment.

The authors plan additional research to look at interventions that force disengagement from work during non-scheduled hours to see if they are linked not only to less stress, but to improved productivity during working hours.

From Home Care: “Re: AI solutions. My daughter’s college is working on AI solutions that could help individuals with cognitive decline live independently longer. This seems like a much better use of AI than some of the options currently out there.”

The article covers a project that brought computer scientists together with occupational therapists to create an AI assistant to help solve this problem. The team captured videos of patients with and without cognitive decline performing a specific task, then created models to identify cognitive sequencing errors during task completion. The system is cheekily named CHEF (Cognitive Human Error Detection Framework) as it looked at the executive functions needed to prepare oatmeal on a stove.

While a camera captured the subject’s movements, occupational therapy students also provided cues about safety concerns or other errors. The system’s vision-language model integrates videos along with text and images to identify both obvious errors and those that are difficult to detect. The team states, “This is an excellent example of applying the cutting-edge AI to a vital health problem with tremendous public health impact.”

As a family physician who has had many difficult conversations about aging patients who are struggling to remain independent, this is some of the most exciting AI-related work that I’ve seen in recent memory. I hope these types of solutions are a reality by the time I might need them.

HIMSS has announced that the keynote speaker for the upcoming meeting will be actor Jeremy Renner. The announcement promises “a thoughtful look at the intersection of determination, care, and innovation and the impact they can have when people come together in moments that matter most.” Those who register before Friday, January 30 have a chance to win an opportunity to meet him personally.

I did something that I haven’t done in a very long time today. I wrote a paper check to pay for a medical bill. The entire process was frustrating. I received a patient portal message that told me that I had a bill, but I wasn’t able to log in. I thought it was an expired password, but I could access the portal from a different link.

It turns out that the practice operates as two separate entities. They use the same EHR, but each practice has its own patient portal. Going back to the portal that I could access, I saw the billing statement with the header for the other entity.

Clicking the payment link took me to a “page not found” error, so I typed the link manually, with the same outcome. I repeated this process the next day, thinking that maybe it was a site outage, and had the same result. 

I called the number on the bill. They told me that they can’t take payments over the phone, so I was off to find the checkbook. If providers want to be paid in a timely manner, they need to make sure that their systems are working to make it easy for patients to pay.

I received two separate mailings from that practice today. The first was a check, which I assume was mailed by their billing service, that refunded me for an overage for the patient co-insurance portion of a procedure that I had last month. The second was a letter from the practice of the physician who performed the procedure featuring red “Second Notice” stickers to remind me that I was overdue to have the procedure and that they would make no further attempts to schedule it. This right here is US healthcare at its finest.

The American Academy of Pediatrics released its own childhood vaccination schedule this week, breaking with the Centers for Disease Control and Prevention on vaccine guidance. States are also issuing their own guidance or joining coalitions to discuss common recommendations.

The EHR where I practice most often continues to display legacy recommendations, and I haven’t heard of any plans to update them. I’m not sure if that’s because the work to do so wasn’t slotted into the IT build budget or if facility leadership is making a statement. Some days it’s refreshing to be outside the circle of decision- making, after having done it for so long.

How is your organization approaching the task of updating vaccine recommendations in your EHR? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 1/26/26

Significant portions of the US are experiencing arctic temperatures and significant snowfall this weekend. As the storm approached my area, I touched base with nursing staff at several hospitals to see how they were ensuring adequate staffing despite deteriorating road conditions.

They generally offered options for staff to sleep on campus, but approached the situation in drastically different ways. One hospital enticed nurses to sleep on campus to guarantee attendance, paid a retention bonus for the time between shifts, and provided meals Another sent a text message that was less than welcoming, treating those who planned to stay at the hospital as a burden by telling them to bring their own bed linens and towels. I’m betting that employee satisfaction differs between those facilities.

Speaking of things that didn’t resonate well during the storm, the marketing folks at Starbucks should reconsider their tactics during winter storms. While the National Weather Service was issuing advisories and our city and state public safety officials were urging people to stay off the roads, Starbucks was blowing up my phone with discount drink offers.

It seems like it would be easy to suppress those promotions in area codes where people shouldn’t be on the roads, whether they’re customers or employees. People who have storm-belt area codes might live elsewhere in the US, but I would guess that they are in the minority. Better yet, come up with a promo code that people can enable that becomes active in three or four days, when they start to tunnel out and are looking for a treat. My city is still focusing on clearing interstates and critical roads, so I will be staying put for a while.

We became skilled at pivoting to virtual meetings during the COVID pandemic, so I was surprised to see some meetings cancel off of my schedule even though they could have been held as web meetings or even as old-school conference calls. I could understand this for small organizations that might have let their virtual meeting subscriptions lapse, but these cancellations involve larger organizations that routinely have at least one or two people on video due to travel constraints.

Childcare issues could be at play due to school closures, but one of the only bright spots of the pandemic was getting to virtually meet the families and pets of my co-workers.

In last week’s Healthcare AI News, Mr. H mentioned the growing concerns that we are on the cusp of seeing AI-related malpractice lawsuits. Frankly, I’m surprised that we’re not already there, given how I see some of my colleagues using AI tools.

Quite a few knowledgeable clinicians, including clinical informaticists and AI researchers, understand the limits of AI. But large numbers of people are overly trusting of the content they see coming out of LLMs.

I’ve seen people cut and paste content containing obvious errors directly from a non-clinical AI tool into the EHR. I’ve also seen people operate wildly outside their scope of practice based on the ability to quickly access information that may or may not be accurate. Unfortunately, these are the situations where people don’t know what they don’t know, and LLMs can be extremely convincing even when they are wrong.

As an example, I recently saw a patient who was accompanied by a physician family member. The family member had a predetermined outcome that they wanted to achieve during the visit. They apparently thought that paying an $80 co-pay entitled them to see a physician who would suspend their professional knowledge and judgment and do the electronic equivalent of whipping out a prescription pad and ordering what they wanted.

I explained the clinical situation, the evidence-based recommendations, what I saw on the patient’s exam, what I had gathered from their history, and why I believed that the requested medication wasn’t appropriate in that scenario. The family member began arguing with me and was showing me his phone with his previous searches on the topic as a way to prove his point. Especially given that his specialty training wasn’t even close to the body system in question, he wasn’t aware that the articles being cited were only tangentially related to the diagnosis.

Fortunately, I’ve spent the last couple of decades working with patients who bring their internet research to the visit. I’m pretty good at educating while arriving at a plan of care that is mutually acceptable. However, I don’t have a lot of experience arguing with a peer who is putting blind trust in the output of a generative AI tool, so it was new territory.

I used my emergency department-mandated de-escalation training, so we managed to make it through the visit once one of the other family members in the room made the physician family member leave. With situations like this on the daily, it’s no wonder that clinicians have lost the joy in medicine. Having to argue with AI-generated errors when a patient’s health is at stake is something that none of us signed up for.

Mr. H also mentioned ECRI’s annual list of technology hazards, and I was gratified to see one of my soapbox issues in the number two position. “Unpreparedness for a ‘Digital Darkness’ Event” is a fancy way to say that an organization isn’t ready for an unplanned downtime. Maybe making it sound more exciting will convince people that they need to do something to get ready.

We should all know that cyberattacks are a “when” situation rather than an “if” these days, and that network or vendor outages are entirely possible. For clinicians who have always been dependent on the tools and safeguards that are built into the EHR, having to work without those can be frightening. It’s one thing to not have calculators or references at your disposal, but not being able to see the overall picture of what’s going on in the intensive care unit at full capacity is something else entirely.

Those of us who practiced in the olden days remember the large paper ICU progress notes that were the size of a poster board, but could fold up to fit in a standard medical chart. With just a glance, we could quickly figure out what was going on with a patient and formulate the best questions to ask during shift change.

The availability of electronic dashboards and monitoring suites has rewired those parts of my brain, but I bet that mental model is still in there somewhere and I could access it in a pinch. We need to remember that soon there will be more clinicians who have never seen that kind of paper documentation than those who have, and adjust our downtime preparations accordingly.

Are you prepared for a digital darkness event? Have you experienced any outages due to snowmageddon? Is your hospital treating staff who have to stay overnight in the facility like a blessing or a burden? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 1/22/26

The American Board of Preventive Medicine is notifying candidates that they have successfully passed the Clinical Informatics board certification exam. The certifications are retroactive to January 1, 2026. Congratulations to all the new Diplomates, and welcome to yet another continuing certification process that will have you asking yourself why you decided to become double-boarded.

From Straight A Student: “Re: online registration form for a training course that I completed recently. Prompts were in a ‘are there any’ format that asked about mobility restrictions or food allergies. A dropdown choice list appeared to be pre-populated with ‘none.’ My answer was ‘none’ for all of them, so I tried to just submit the form, which popped me back to the top with no feedback. The course vendor responded to my help desk ticket to say that the dropdown requires choosing ‘none’ and people miss that all the time.”

These sorts of Process Improvement 101 issues drive end users batty. The time wasted by users and the help desk adds up.

I wonder if user acceptance testing was done, since it should have been caught. Sometimes teams give the users detailed testing instructions outside of the application, such as “click here, then choose that,” which makes it impossible to determine how they will interact with the workflow. I also wonder if they are analyzing call volume to to identify ongoing issues. Then, has the help desk team reported the issue to development and asked for an update?

It feels like it would be more efficient to change the default to “please select from the list” or “choose a response.” Or, to add a page instruction telling users what to do.

I have been in countless conversations about the safety of healthcare AI solutions. I’m always interested in how the risks and benefits are portrayed to patients and other non-clinical, non-tech individuals. Mr. H mentioned a preliminary report by the VA Office of Inspector General that found that the Veterans Health Administration had some gaps in AI chatbot oversight. The story was also picked up by military-focused Task & Purpose, which ran its own version. 

Risks that were highlighted for the general audience included “producing misinformation, privacy violations, and bias, and that the systems had been put in place without review by the VA’s own patient safety experts.” I didn’t see mention of concerns that were noted by other publications, such as whether lags exist in providing current information for the LLMs to use.

An article commenter shared their physician assistant’s thoughts that “the AI is egregiously wrong 90% of the time, so he doesn’t bother with it.” Based on my own experiences with clinical-focused and consumer-focused AI solutions, that’s probably a significant exaggeration. I wonder if the user would benefit from additional education on prompt construction or effective use of AI tools.

The VA providers who I’ve talked to locally are happy with the AI solutions that are available to them. They are looking forward to continued expansion of their capabilities, such as helping craft more readable medical information for patients. If you’re a VA user, feel free to chime in. We can keep your comments anonymous.

I’m still in my New Year’s inbox cleanup extravaganza, and found an article about Hackensack Meridian Health’s canine-powered cancer detection program. The health system partnered with startup SpotitEarly for a clinical trial that examines the ability of trained dogs to detect cancer via patients’ breath samples. The goal is to validate the technique as a noninvasive cancer detection approach that might be more attractive to patients who are unwilling or unable to complete traditional screening recommendations.

The test is conducted by having patients breathe into a mask-like device for several minutes, followed by the dogs sniffing the devices. The dogs are trained to recognize odor signatures in the exhaled volatile organic compounds that can be associated with cancer. The dogs indicate detection by sitting next to a sample.

We know AI has to be involved somehow, and indeed it is. The company is using AI tools to document and analyze the behaviors of the dogs based on behavioral and physiological data.

SpotitEarly has been in the US market since May 2025, although it was founded in 2020. Previous studies of the technique found that the test was 94% accurate for detecting lung, colorectal, breast, and prostate cancers. If any readers are involved in the study, I’d love to hear about the “best boys” and “good girls” that are doing the sniffing and whether they prefer belly scratches or having their ears rubbed. My medical school had some public-facing research animals and they were the most amazing companions when they retired, resulting in a years-long waitlist for adoption opportunities.

Based on some of the other email traffic in my inbox, quite a few physicians made a New Year’s resolution to look for different employment. Several of them seem to think that informatics is something that you can just jump into because you are “techy” without any formal training or experience.

Some startups will hire clinicians in this situation, but I always encourage people to consider formal coursework to better understand the informatics landscape. I’m a big fan of the courses offered by the American Medical Informatics Association. The virtual courses are convenient, and the in-person ones are great for networking with colleagues working in the field.

A number of highly qualified clinical informatics physicians have recently been displaced from EHR vendors and health systems, so it seems that as long as mergers continue, the job market will remain challenging.

Are you looking to make a career change in 2026, and if so, how are you approaching it? Leave a comment or email me.

Email Dr. Jayne.

Curbside Consult with Dr. Jayne 1/19/26

Based on the contents of my inbox, it feels like everyone is talking about recent research from Michigan Medicine on emoji use in the electronic health record. The research letter was published in JAMA Network Open last week. It examined 218 million notes belonging to 1.6 million patients. Researchers found that emoji use was higher than previous studies that looked at clinical texting tools. 

The authors identified 372 emojis within 4,162 notes that were created during the last five years. Of those, 35% were patient portal messages to patients, followed by telephone messages at 28%, encounter summaries at 15%, progress notes at 14%, and patient instructions at 6%. The University of Michigan patient portal doesn’t support patients adding emojis to communications.

The smiling face with smiling eyes was used 1,772 times, with communications emojis such as the telephone receiver and calendar appearing 544 and 429 times, respectively.

The article contains an illustration of the 50 most commonly used emojis along with their official names. Just skimming through them, I’m not sure that I would come up with names or descriptions that matched their official titles.

Take the “briefcase,” for example. Could people be using it because it looks like an old-timey doctor’s bag? I can’t remember the last time I saw someone carrying a briefcase that looked like the emoji. Even looking at the most used one, the smiling face with smiling eyes, I would describe that one more as blushing than having smiling eyes. I also would not have correctly described “beaming face with smiling eyes.”

Some of them were new to me, including “busts in silhouette” and “bar of soap.” The latter got me thinking about how many people actually see or use bar soap these days, given the popularity of liquid soap and body wash products. Similarly, how long will it be before people no longer identify a “telephone receiver” as such?

I wondered about the context for some of the emojis that were used, such as the “P button,” the “small blue diamond,” and the “round pushpin,” and how they might be used in medical communications. The most concerning to me was actually the least used, the “police car light.”

Researchers note the risk of confusion in using emoji to communicate, especially in older patients. While most emoji use occurred among tweens and teens, patients in their 70s had the second highest usage. The authors call for organizations to develop guidelines to promote clear communication and professionalism in clinical communications. I once encountered someone who used a particular emoji extensively before discovering that it wasn’t a Hershey’s Kiss, so I agree with the concern.

The authors go on to note that measuring emoji use is just the beginning, and that future investigation should look at how emoji “might affect patient understanding, trust, and outcomes – and explore whether these playful digital symbols offer new opportunities or pose unintended challenges in electronic health record communication.”

One of my close physician friends sent me a link to a Facebook post about the article. It had some pretty funny comments about which commonly used emojis were missing from the study, along with those questioning whether the AI tools clinicians are using to write messages were responsible for the addition of emojis. A couple of commenters thought the research was frivolous, but those sentiments were countered by others who were clearly concerned with the potential impact on patients.

Another colleague with ties to Michigan Medicine said that emoji use in the medical record was prohibited, although he wasn’t able to find the specific policy. He said that he remembered a conversation with risk management where it was discussed, however, and that there were significant concerns about the meaning of symbols within the context of the legal medical record. Although the policy could have been changed, I’m wondering whether some clinicians still haven’t fully internalized that the patient portal is part of the legal medical record.

He said he’s not opposed to their use, especially with pediatric or teen patients with whom clinicians are trying to build rapport. Still, he advises residents that if deleting the emoji changes the meaning of the message, either the emoji shouldn’t be used, or it should be supplemented by actual words.

I was curious about the previous research that looked at clinical text messages. In 2023, clinicians from Indiana University School of Medicine looked at the content of messages that were sent by hospitalists who used a secure messaging platform during 2020 and 2021. Messages with emojis were identified, as well as those with more old-school emoticons.

The authors found that the majority of the emojis and emoticons “functioned emotively, that is, conveyed the internal state of the sender” where others “served to open, maintain, or close communication.” The authors also noted that “no evidence was identified that they caused confusion or were seen as inappropriate.” They concluded that “these results suggest that concerns about the professionalism of emoji and emoticon use may be unwarranted.”

I believe that differences exist in how clinicians communicate with each other compared to how we communicate with patients. In the former, we are more likely to use medical abbreviations or jargon. With the latter, we should be using terms that are more clearly understood by patients. In my experience with peer review, communications with patients are typically held to a higher standard.

It will be interesting to see what kinds of guidelines or policies organizations come up with as far as regulating the use of emojis in patient communications and charting. I reached out to medical staff leadership at the facilities where I’m affiliated, and none of them recalled this topic coming previously.

I found citations for a half dozen other articles that looked at the content of clinical text messages among hospitalists and other members of the clinical team, as well as norms for emoji use. I didn’t have time to go down that particular rabbit hole this weekend, but I would be interested to hear from readers that have strong opinions on emoji use or those who have been involved in this type of research.

Do you use emojis in patient-facing communications? If so, how do you use them? If not, what do you think about the practice? Leave a comment or email me.

Email Dr. Jayne.

EPtalk by Dr. Jayne 1/15/26

Plenty of people have been asking me for my thoughts about last week’s announcement of OpenAI for Healthcare.

Models that are tuned to physician needs and that have been through robust clinical testing certainly offer advantages. The incorporation of the organization’s internal documents via SharePoint and other platforms is also attractive.

I recently chatted with a friend who is both a physician and an attorney about the impacts of such integrated solutions on the medicolegal landscape.

In the current state, with many physicians playing the “bring your own AI” game and using various solutions on their phones, no connection exists between those queries and the legal medical record. However, an enterprise platform that ties it all together and specifically encourages the use of patient data and PHI adds an additional layer of complexity to medicolegal investigations.

It won’t just be about the EHR and its audit log. It will involve all the potentially related queries that may have been entered and acted upon by the care team. We’re starting to see some legal activity around physicians who based their decisions on inappropriate AI-generated information. This is an area to watch.

I also wonder about the ability for hospital policies to negatively influence access to information by clinicians. For example, if you work in a hospital that restricts certain procedures or medications for religious reasons, how will those limitations shape the responses when those prohibited treatments might be the right answer for a patient?

This could evolve to include a bedside component for patients. They could ask questions about their care plan while hospitalized. However, they might learn that their care is limited by their choice of facility.

My conference BFF Craig Joseph, MD recently wrote that healthcare is betting on the wrong AI instead of looking at solutions that actually improve clinical outcomes. He cites a study from the University of Southern California that found that physical robots outperformed chatbots in reducing psychiatric distress. He goes on to talk about how the brain perceives interactions when there is a physical presence compared to a virtual one and about the benefits of emotional experience in delivering care.

It made me think of my own experiences with physical therapy. It’s an advantage having your friendly (or not so friendly) physical therapist right there urging you to push yourself compared to a therapy bot at home that is less perceptive when you’re slacking off.

The robots used in the study looked fairly low-tech and had crochet covers, reminding me a bit of the cats in Disney’s “Lady and the Tramp.” For a tech industry that focuses on flashy products, these wouldn’t even be on the radar. I agree with Dr. Joseph that sometimes low tech is best. Maybe we’ll have to make that the focus of our next conference booth crawl.

Speaking of low tech, I was talking with a couple of physician friends recently about the Oura ring as a potential adjunct to addressing sleep issues. One colleague swears by his, although the actions that he has taken based on the ring’s sleep data are the things that every family physician recommends for sleep issues: consistent routine around and time for sleep, adjusting environmental conditions, appropriate timing of meals, and keeping a basic sleep diary to identify triggers.

My other colleague proposed a decidedly low-tech approach: sleeping with a stuffed animal. He pulled out his phone to share a Wirecutter blog from last year that addressed the tactic. It cites several scholars and their comments on the practice, including notes on how it might help adults shift from a state of cognitive arousal to the more relaxed mindset required for sleep.

The blog notes the lack of literature on adults sleeping with stuffed animals, but I bet if we threw some AI into the mix, people would be eager to study it. Maybe those crochet cats can work the night shift as well as having a day job.

From Night Nurse: “Re: my annual refresher training. Passing pre-tests exempted us from that section. This was one of the questions. What kind of world are we in where this is considered an appropriate question?”

I have unfortunately seen some bad behavior from healthcare providers during my career, so I agree that we should be screening for people who have thoughts like this. I don’t think a bold annual training question is the way to pick them up. Even in a written survey, I would probably recommend a more subtle approach to identify those who have such sentiments. I’ve done a fair amount of work writing test questions and I wonder what the hospital’s item writers were thinking with this one.

From Tech Traveler: “Re: swearing. I’m a medical device representative and read your blog to keep up with healthcare tech topics so I can commiserate with the physicians I call on. I’m in and out of operating rooms and physician lounges all day and notice that there’s a certain amount of swearing that goes on among physicians, but it seems to vary by specialty and age as well as by topic. I’ve joked about doing a research project to explain the phenomenon, but it looks like researchers beat me to the punch.”

The article notes that although swearing is “often dismissed as socially inappropriate,” it has been linked to increased physical performance through state disinhibition. That is a psychological state in which individuals are less likely to restrain their behavior. The authors propose that this leads to flow, confidence, and focus, with those who swear being able to perform better on strength and endurance tasks than those who used neutral words.

They note that “these effects have potential implications for athletic performance, rehabilitation, and contexts requiring courage or assertiveness. As such, swearing may represent a low-cost, widely accessible psychological intervention to help individuals “not hold back” when peak performance is needed.”

Another one of the practices where I receive care has finally given in to the private equity company that has been pursuing it for the past couple of years. The physician mentioned this at a recent visit and shared the behind-the-scenes story. She has been struggling since she opened a second location, but has been keeping her head above water through the availability of same-day dermatology appointments, which turned local primary care doctors into a loyal referral base.

We’ve all been impressed by her ability to fit people in. Who doesn’t love being able to have a patient’s suspicious lesion removed in a timely fashion? Before she opened, patients often waited months for appointments.

Although she offers some cosmetic dermatology services, the practice is heavily skewed towards medical dermatology. She shared that automatic payer downcoding has been financially devastating. Her attempts to promote the more lucrative cosmetic treatments, which are typically cash pay, couldn’t compete with local med spas that run coupon specials. She decided to give in with five years to retirement. We’ll see how well that same-day availability holds up with private equity operations leaders at the helm.

If your care providers have been acquired by private equity, what changes have you noticed? Leave a comment or email me.

Email Dr. Jayne.