EPtalk by Dr. Jayne 9/7/23

I enjoy giving back to medical schools and residency programs in my area, so I’m looking forward to teaching an upcoming class on clinical decision support and how clinicians use it in the real world. I had barely started culling research articles and building my slide deck when this New England Journal of Medicine article virtually dropped in my lap.

The authors found that although there has been a lot of attention to ensure that the clinical decision support tools themselves are safe, there hasn’t been enough attention to making sure that clinicians understand how they work or how best to use them in clinical settings. Over the last two decades, I’ve seen plenty of clinicians disregard even the most basic clinical decision support interventions, so the article definitely had my attention. 

The authors suggest that medical education curricula should include content on how clinicians should best incorporate algorithm-generated information into their own decision making, and to discuss the use of clinical decision support tools with patients. Trainees should also be taught to consider how different patient data inputs might change the output of a tool, as well as how to evaluate whether a tool might be missing key elements needed to appropriately evaluate a given patient.

The authors propose that learners should engage in practice-based training scenarios where they use an algorithm, discuss its predictive value, understand how it is using different data points, and understand its overall effectiveness. The article also included a brief history of clinical decision support tools, which have been around since the 1970s, and I’ll be using some of their examples in my presentation.

Speaking of presentations, I recently attended one delivered by some “experts” who advertised a session to help “technology people” better understand how to work with physician end users. There were some interesting statistics thrown out, many without citations. One that seemed very specific was the claim that the average physician can only type 22 words per minute. I decided to hit the medical literature and see what I could find, with scholarly articles putting physicians more in the range of 30 words per minute. That’s a pretty significant difference and I would caution folks that it’s not a good idea to throw that kind of data out without citations when you’re speaking to a group of technical experts.

I think they also failed to understand that there might be clinical informaticists in the audience or that some of the things they were saying about clinician end users were offensive. Comments such as “you have to dumb it down when you’re working with physicians” are a poor idea, especially when those physicians are on the webinar.

Another strategy they advised when working with physicians was that of creating special roles and/or titles for physicians to get them to be EHR ambassadors to their peers, particularly without additional compensation. In recommending that a physician be named something along the lines of Regional Site Director of EHR Adoption, they went on to say, “Titles are cheap. We give them away like candy.” They had some other ridiculous things to say, and I was glad I had a cocktail in hand to help make it through the session. Another clueless move: they failed to recognize that often EHR analysts, implementation specialists, and other “technology people” tend to be fiercely loyal to their end users, so I’m sure the speakers didn’t make many friends with those folks, either.

A friend recently mentioned his experience with EqualityMD, which is a technology platform dedicated to helping LGBTQ+ patients who are seeking healthcare providers who have been recognized for their delivery of respectful and affirming healthcare. The system gathers patient information, including biometric data, race, pronouns, gender identity, sexual orientation, ethnicity, and languages spoken. It also gathers information about patient preferences for their care teams and ultimately suggests physicians and other providers who might be a good match. Patients can then schedule with and engage providers through the platform. My friend has had a good experience with the solution and feels like the company is “the real deal,” so I’ll be interested in following them over the coming months.

Sometimes I get my news from HIStalk just like everyone else. Earlier this week, Mr. H mentioned Air Force physician complaints about issues where adolescent patients can’t access their records via patient portal. As someone who has been dealing with the challenges of privacy and confidentiality issues faced by healthcare providers who see adolescents for the last eight years, this didn’t seem too off the mark to me. Mr. H is always great about providing citations so I followed his handy link to the original article in Stars and Stripes.

Apparently one of the Air Force’s chief medical information officers told a gathering of people, “So there’s this weird dead zone as of right now with the portal. With children 13 to 17, you can’t see their medical information and your kid cannot create a login because they’re not 18 yet.”

Although this may sound wild to some, it’s actually the way that a large number of healthcare organizations in the US have to handle adolescent confidentiality and privacy due to clashes between state laws and EHR functionality. It’s certainly not unique to Oracle Health. Most of the EHR vendors have a long way to go as far as being able to allow teens to access their data online, but prevent parents from seeing information that might cause harm to the patient. Some systems are better than others at allowing different parts of the record to be parsed for parent view versus patient view. However, the fear of legal issues seems to be a driving force for many provider organizations who are deciding how to configure their EHRs and patient portals.

There have been quite a few presentations on this at conferences, ranging from the AMIA Annual Symposium to various EHR vendor user groups. I’d love to have policymakers sit down with adolescent patients, their parents, their guardians, and their care teams. That’s the only way they’re ever going to understand the complex situations around which they are trying to create rules which seem like a good idea, but ultimately backfire on many of the involved parties. Until then, or until EHR vendors step up, we will all be stuck with the current complex web of proxy access, parental rights, and different ages of medical consent not only across state lines but across medical conditions as well.

How is your organization handling adolescent privacy and confidentiality with respect to access in the patient portal? Leave a comment or email me.

Email Dr. Jayne.