EHRs: The hidden distraction in your doctor's office
spectrum.ieee.org
spectrum.ieee.org
which is fine until something comes up that you didn't anticipate and print out. Then you can a) fake it, end the visit and follow up with pt later after you've looked it up or b) log in and get the info
How do you have the pt's current med list? Does staff print it out after they've roomed the pt?
Also, how are you ordering test/procedure? Writing it down for staff to do later? Violates most org's "CPOE" policies. Otherwise pt leaves and your staff has to call to schedule later, including labs that maybe they could do before they leave.
You must have re-created a paper chart workflow in an EHR era which is only possible if your staff/org enables this for you
Most of us are just employed widgets in the health care factory, and don't have the pull to get staff to work with this kind of workflow
At least for the health maintenance stuff, I already know what needs to be done on that score before I even enter the room. If I have to grab something out of the record, like a result I wasn't expecting, I can quickly run back to the office (it's just around the corner) and come back.
So, no, no paper.
I can't remember all the details to a sufficient level that I feel comfortable that I'm not forgetting something
and how do you know the current vitals and medication list? When the pt tells you they saw Dr X for Y (that you didn't know about) do you not want to look at that in case it impacts your plan? I guess you go out and come back? If you rx a med that needs lab monitoring, did you memorize that too? What about trends in labs?
IDK, I need info while I'm seeing the pt
For the med list, I do know what they're taking, but my usual folks bring in their pill bottles anyway just so we can make sure they all match up. This is also useful because if I want them to discontinue something or change it, I'll write it on the bottle, and make the change in the MAR when we're done. We're not usually making massive med changes on any one visit.
If they saw someone in the interim, I'll have already seen it in the chart before I see them, and if it's not there, I'd have to order the record anyway so it doesn't matter. Most of the offices here are on Epic, so Care Everywhere will usually get their notes.
I think we just have different practice styles here.
I know of one practice that went all-in on the stuff. They had to re-hire their secretaries after their AI transcription recorded "this bone normal, no damage to this other area" but totally failed to mention that the first part of the sentence was "distal fracture to whatever", ultimately failing at it's most basic bloody function.
I'm pretty sure the founders are not doctors but tech industry types, who figured that there was some non-zero error rate and just like, collectively shrugged at the consequences.
What struck me over the years was the open hostility we faced from the staff. The admins would buy our product, then have us come do trainings. The clinicians seemed to resent every second of it and would just never use the tool.
Towards the end of my tenure there, a PM said to me “the last thing these people want is to have to learn yet another workflow”. Which is when the penny dropped for me that our tool was just one of a bazillion being force fed to these poor people. They want to spend their time with patients not a screen.
Despite it being the most mission driven I have ever felt about a product (we were literally trying to help cure cancer lol). I’ll never work in health care again. Like education, it’s a quagmire.
Remember: there's a lot of "health care" out there. Even if doctors resent EHRs, there's also drug discovery software, telehealth software, embedded software in medical devices, etc!
Project Ronin?
I suspect that people entering medicine do so to address human needs, and have very little interest in dealing with technology (or handling traditional paperwork for that matter). Couple that with a perception that pretty much anything digital being obsolete before it reaches market, and even more so when it can take upwards of a decade for the product to reach them, and you are left with a group of people who have nothing but dread about being stuck on a never ending treadmill that is outside their scope of interest and expertise.
Take that opinion with a grain of salt though. My background is in that other quagmire: education. I have seen some amazing tools developed over the years that were abandoned, so everyone had to move on. Worse yet, no replacement was created for most of those tools so everyone is back where they were before the revolution happened. (I'm thinking specifically of software used by teachers and administrative staff, but something similar can be said for software used to deliver the curriculum.)
I guess for future-proofing, the university moved to Blackboard. For a while, some courses were on Blackboard, others on CCNet.
We had a professor poll the class and ask which they preferred, and all 240 of us in unison said "CCNET!"
I still remember a quiz on Blackboard where the answer was something like "2" and it responded, sorry, the correct answer is 1.9999999999.
An accountant friend was just migrated to Workday(?) for their backend. Apparently whatever labyrinth configuration they have can only export 12,000 rows at a time. The official workaround they were given was to run reports in one week batches when a month of data is required. Previous solution could seemingly export unlimited amounts of data and time windows. A complete technical failure for which everyone should be ashamed.
Yet, the entire Workday chain of developers, PMs, management - all slapped their seal of approval on the product and pushed it out the door. Compiles? Good Enough.
https://open.substack.com/pub/healthapiguy/p/there-will-be-b...
They're all actively user hostile and add features admin think look nice but provide no real value for classes.
The only time I’ve experience interop in healthcare is due to actual organizations merging. That’s it. This entire space is filled with incompetence. Maybe providers will actually use the tools if they work consistently. Food for thought.
I’ve asked, why do you need all this again and the answer is usually “oh we have a new system” or “we need to know if anything changed” (but that’s not what the forms ask).
Patient time is worth 0 to the medical system.
Recently there's been a big push for TEFCA (Trusted Exchange Framework and Common Agreement) as the network-of-networks that federates all the different HIEs. It's been slow, but it's progressing.
As usual, the problem isn't really technological -- it's getting all these different stakeholders, with different business models, etc, to agree on how it should work.
But it was voluntary (for the organizations, not so much the staff). There was no need for government to shower pharmacies with money to adopt it because it paid for itself.
I'm sure a lot of the staff initially met it with the same hostility. Even in 2010 when I was more in the field, we still had staff where their only computer experience/use was at work and otherwise lived an offline life.
Can't say I saw a pharmacy that didn't have a computer since the early 90s in Canada (and my memory doesn't go before that). And before that, at least they used typewriters. Meanwhile my GP was all-paper well into the 2000s except for some billing stuff. God help anyone that had to read his notes. But sometimes you're reimbursed sufficiently that there is no driver to change workflows even if it would be economic.
Ontario Canada.
Meanwhile getting things to work is filing tickets followed by "oh gosh that's so complex!" and months of moron pitcrew showing up " to fix it" who can't fix anything and who seem to think it's just that we're dumbasses who can't figure out who to reboot a computer.
Honestly it's difficult to not grow the instant opinion that IT should just shut the fuck up and fire themselves. Who the hell do they even work for?
Management. Management whose goals and incentives don't align with yours. Or IT's.
If management cared about your experience and quality of life, then presumably they'd be riding IT to get shit fixed. They'd be providing staff and resources necessary to resolve the issues. They'd be consulting with the staff using the programs before buying/deploying them. They'd be consulting with IT before buying/deploying them. They're not because they don't care.
They went and bought some EHR system and an expensive support contract based pretty much entirely on price and/or how many golf games the vendor would pay for, dropped the steaming pile of turds into IT's lap, and had them implement it. They probably also told them to go ahead and integrate it with all the other systems in use that they sourced the same way.
Meanwhile, every time they've done a budget for the past 20 years they've cut IT just a bit more because it's a cost center so the lower you can get that on your spreadsheet, the better you look, so there's like two kids and a grumpy old balding guy who spends most of his day working on reports for audit and compliance and they're responsible for the entire hospital.
(At the hospital one friend worked at he was responsible for taking support tickets along-side the two other IT staff, working the on-call, and also _every single integration between systems in the hospital_. He wasn't a software developer or anything. He'd just started as purely help desk and seemed to have a vague idea how to write documentation and only cost like $35k/yr so he was clearly the best person to be responsible for communicating with all the vendors and making sure the EHR system could talk to the MRI machines.)
But don't worry, even if this comedy of errors somehow gets to a working state... when that contract's up for renewal, they're going to look at the price and if a better one comes along they'll do the same thing again. Same for every other system in use all of which will have a ripple effect across every other system.
Hey, at least you have job securi--what's that? The hospital was just bought by private equity and merged with another hospital and the entire IT department's laid off effective immediately?
Cardiology, radiology, endoscopy, and labor and delivery all have their own systems for their internal usage while releasing final results to Epic.
I don't object to the idea that these products are made for admins. It's a business and it needs to make money to survive. I object to the idea that making a product for sale to admins precludes making it at least usable for those who actually put in the data.
On the other hand, when a team tries to build their own tools, they quickly realize they have to build a ton of compliance and interop code they never wanted to touch in the first place. That’s why open source platforms that handle the core infrastructure, like Medplum, HAPI, or OpenEMR, can be such a good starting point. They get the team 90% of the way there, so they can focus on what really matters: building a great UI/UX for their users.
I don’t think providers truly want to go back to pen and paper, but they are looking for a better way. They can see the promise of what the solution could be, but they just haven't experienced it yet.
Disclaimer: I work for Medplum.
It turns out that concept is called "Headless EHR," and it's pretty new.[0] Medplum (that the parent comment mentions) is one of the companies in this space.
[0]: https://healthapiguy.substack.com/p/to-ehr-or-not-to-ehr
Disclaimer: I know a number of people who work for Epic. ;-)
And you may end up re-creating bad workflows instead of updating to better ones.
Anyone who has ever looked at an EHR/EPIC screen, can tell you that the 1990s Web called, and wants its tables and frames back.
In fact, one doctor I went to, still ran Windows 95 (in 2009), because they didn't want to deal with new interfaces.
Engineers are notoriously unsympathetic to usability and simple GUIs, but I have found them to be an absolute gold mine, if you want people to actually use your product. Apple and Google are trillion-dollar companies, now, mainly because of their simple, usable UX.
As the other comment pointed out, it's a balance. Simple is not the same as user-friendly, but they live on the same street.
Doctors routinely deal with concepts that would confound me, but they are often quite technophobic, when it comes to computers. I have a friend that's a really skilled anesthesiologist, and is constantly asking me the most basic questions about his iPhone.
Complex interfaces can be trained, but the magic is to have an interface that can be explored. If you train someone on rote, then they go to pieces, when anything changes.
However, if you give them an interface that doesn't penalize them for exploring, and has clear, unambiguous affordances, they can easily adapt to things like updates, and they won't force you to have to maintain an ancient UX.
But designing that kind of UX is quite difficult, which is why so few people do it.
I maintain my emacs config
the problem is if someone changes something, that immediately impacts my efficiency which slows me down, then the patient's are pissed, and the administrators are too (which is ironic since they're the ones who signed off on the change)
It has to be rote, no time for exploring
No big deal.
Not really my wheelhouse.
HTML forms are a metaphor for literal paper forms. They don't have to be complex. One of the forms in the EHR system I am familiar with uses a stick figure layout. So if you are making notes on the left leg you just type it in next to the left leg. I don't see how this is difficult.
Meanwhile I can't figure out how to get my iPhone to show me what photos I took in the park by my house and every setting change involves consulting a web search or LLM.
On one hand you have massive GUIs spanning the whole screen containing hundreds of controls. On the other you have airy GUIs with more empty space than actual content and every time you want something you have to open 3 layers of menus to find it.
Both are wrong. The correct thing is to find a balance. The balance depends on the usecase as well as the users.
This is what makes it hard. You can't just code up an app and throw is over the fence. You have to actually engage with the users, watch them perform their work, even try it yourself. You have to understand what is important and what is a distraction. You have to understand when these things chance. And you have to understand that beginner users evolve into experts all the while you have new beginner users coming in.
Hospitals were among the first to get "computers", I'm talking the big mainframes and such that used to be popular in big institutions & universities. On these systems many hospitals each individually hired programmers to construct custom databases for their record keeping. While most have by now have transitioned into a more standardized structure, like HL7, the original sin has carried forward enormously bizarre data structures that make you wonder if the designers were deliberately trying to sabotage the possibility of good software in the industry. I can't think of a better example of why you should never design by committee.
Yet in parallel to all this, capturing medical data is already hard. Doctors are most comfortable just writing notes freehand, recording the patients current state, notable observations, treatments and so on. When modeling this it becomes very tricky because you basically need a proper medical background and be a good at data modeling / programming. This kind of person is basically a unicorn in the industry everyone wants but can never get.
Consider, just for a moment, all the complexities that come with dealing with the thousands of different units and their conversions within the industry. Some doctors don't even use the same units for certain measurements, entirely out of personal preference. Then remember that measurements are the easiest part of the system to model, even what should be the simplest part of the entire thing is hard. Also yes, you will have to re-write all this from scratch, there is no special library or open source software to help. Everytime someone makes tools for this they keep it proprietary.
But that's just the tip of the iceberg, to really get an idea of what I mean, just look at HL7. It's basically a data format that is like a cursed csv with about 5 layers of deliminators for nested entries, since all hospitals like to be super special, the specification tries to be "flexible", so what exactly these characters are is not actually standardized! It wasn't enough for HL7 to just be a data model, they needed to violate a few OSI layers and interlace it with the transport protocol too!
So in essence you must establish a bizarre handshake on top of tcp to learn what the hospitals super special configuration of the standard is, the very syntax itself! Worse, 90% of it is the same for all hospitals but the 10% that isn't is entirely unpredictable!
Then you have the actual data model itself, like demographics, lab records and so on. They change the specification every few years! You need to support it all since this committee of monsters don't seem to care much about the migration path! All the changes they make seem pretty arbitrary to me but what do I know?
I'm still only scraping the surface here but my exposure has been limited to what I do, which was processing all this from the perspective of a medical device that only needed to deal with a subset. When I imagine the struggle one would have with actually dealing with the entire thing holistically I feel empathy and a desire to never have their job.
It's like building a house on top of an active volcano. Any illusion I had that my medical records could be used for anything other than basic notes for another doctor to read have long since shattered, because clearly that's how all of this mess is actually being used in practice.
Oh and don't forget HIPPA! Even when you roll up your sleeves and try to fix the problem, you learn you aren't even allowed to thanks to the governments overbearing regulations against using medical data for things that could help society. Wish they just made it a crime for insurance companies to use instead of whatever this is.
The fact any of this works at all is a fucking miracle honestly.
Interoperability is also one of those "holy grail" things that is really hard.
This is actually not unique to healthcare (see others above posting about Learning Management Systems and Workday). As a result, most enterprise software across verticals is similarly dated. Some research here: https://open.substack.com/pub/healthapiguy/p/there-will-be-b...
To doctors these things are just tools, tools that they want to be able to pick up and put down 100 times a day without having to think about it. A good tool can be operated mostly on muscle memory and needs to remain static 99% of the time. Imagine if the tools a mechanic or carpenter used changed in form and function all the time. Last year they used a right handed circular saw, next year they are forced to use a left handed worm saw. Or imagine a framer picking up his hammer he has used for the last 10+ years and going to give it a swing and missing his mark, only to find out last night his boss took his old hammer and replaced it with one 2 ounces lighter and 2 inches longer and his boss refused to give his old one back. Or a guy digging through his toolbox to pull out a lesser used item like helicoils that he knows is in a medium sized yellow box, wasting tons of time looking and possibly going for a different and less ideal solution, only to find out later the helicoils were at some point put placed into a small sized blue box instead because someone else decided the old box was a bit too big and wasting space.
I started to think this after seeing the bills from multiple visits, where it's often broken out, in detail, what they had done. It's probably not as bad as that, there probably is some record-keeping happening in there. But considering how overworked most doctors are in the public health system, and how little time is commonly allocated for each patient, it can feel a bit like you didn't actually interact with a human doctor.
The first big step towards untangling the gordian knot in my book is pivoting the industry to a capitated payment model, so compensation doesn’t require tying everything back to CPT & ICD codes, or tracking super-anal quality metrics for CMS. Once you make that jump, it’s pretty easy to imagine a lightweight note-taking + file-hosting platform that lets providers document summary data on a patient profile wiki-style, and attach notes to the profile in any arbitrary format using customizable templates. (I’m basically picturing a mashup of Google Drive + Wiki.js + Obsidian features.)
Handling meds, orders, and referrals in a cross-platform way is starting to become a solved problem with FHIR. Tack on modular plugins for each of the above to the patient profile page, and you’ve reinvented the functionality of an EMR in a way that sucks far less and lets users get shit done.
https://www.newyorker.com/magazine/2018/11/12/why-doctors-ha...
https://web.archive.org/web/20250104014248/https://www.newyo...
The fun part is about 4/5 of the way in and starts with
> Some people are pushing back. Neil R. Malhotra is a boyish, energetic, forty-three-year-old neurosurgeon who has made his mark at the University of Pennsylvania as something of a tinkerer. He has a knack for tackling difficult medical problems. In the past year alone, he has published papers on rebuilding spinal disks using tissue engineering, on a better way to teach residents how to repair cerebral aneurysms, and on which spinal-surgery techniques have the lowest level of blood loss. When his hospital’s new electronic-medical-record system arrived, he immediately decided to see if he could hack the system.
I'd elaborate but it wouldn't be good for my mental health
edit: I'll give one example: my org can't even implement single-sign-on even though it's essentially all MS
When any country mentioned hits the population of a small or medium US state, let us know how it goes.
> Canada, China, India, and Japan also have EHR system initiatives in place at varying levels of maturity.
Apparently the author could not care less. Apparently even the WHO could not care less, given the linked document tells us nothing.
As always, it's the US versus the world, and the world is a giant nothingburger, save some flyover countries in Europe that could be part of Greater Germany or Greater Russia for all anyone cares. How is the UK, Germany, France, Russia, or China doing? Oh...
> The United Kingdom was hoping to be a global leader in adopting interoperable health information systems, but a disastrous implementation of its National Programme for IT ended in 2011 after nine years and more than £10 billion.
No doubt when the US gets the standards and apps done, the rest of the world will magically start working too. All the billions spent and the world piggybacks and gives nothing back, save, quite amusingly, China. As always.
I don't know "how it goes" but Poland has the population of a large US state.
> Ohio has long been a national leader in EHR adoption, with nearly 5,000 primary care physicians signed up through the Ohio Health Information Partnership—more than any other state as of around 2011.[1] Cincinnati-based HealthBridge operates one of the largest and most robust regional Health Information Exchanges (HIEs) in the U.S., servicing over 30 hospitals and 7,500 physicians across multiple states.[1]
> In Ohio, a qualitative 2022 study surveyed provider and leadership perspectives on interoperability, finding high adoption rates: 96% of Medicaid‑PI‑eligible providers and hospitals had adopted EHR systems; non‑eligible providers reported adoption at 72%.[2] Epic Systems dominates the state as the top EHR vendor—used by 37% of Medicaid‑PI recipients and over 56% of other providers; smaller practices more often use NextGen, eClinicalWorks, etc.[2]
> The 2021 Illinois Health IT Survey, based on 175 respondents representing ~3,800 providers, shows 100% EHR adoption among respondents—up from 61% in 2011.[3] Participation in an HIE rose from 32% in 2016 to 51% in 2021.[3]
> For Illinois, key barriers reported: lack of provider Direct message addresses, reluctance of referring providers to accept messages (58%), and vendor cost constraints (46%).[3] Top reported improvements: decreased medication errors (64%), improved throughput (60%), and better reporting and referrals (60% and 57%).[3] The most difficult challenge: meeting program objectives (37%), followed by implementation cost and time (22%).[3]
Overall chat bot indicates Poland has unique patient IDs so no record duplication compared to poor US implementations, high interop within P1 compared to poor interop between US vendors, and good patient data access compared to poor implementation by US vendors. Chat bot gave little about burnout but mentioned Polish and US AI developments under way. I would assume there's poor interop between Poland and other EU states, likely much worse than the US IMHO. Not really any mention of other topics like clinician workflow, burnout, and productivity re Poland.
[1] https://www.healthpolicyohio.org/files/publications/hitprime...
[2] https://pmc.ncbi.nlm.nih.gov/articles/PMC10007006/
[3] https://hfs.illinois.gov/content/dam/soi/en/web/hfs/sitecoll...
On a city we have several places controlled by the same entity, and they use an integrated EHR, so that a doctor who sees a patient at the emergency department has access to it's full history from the tertiary center, but at the same time the major tertiary/quaternary hospital isn't managed by that same entity and doesn't use the same EHR system, so we can't share information digitally. To make things worse, one system is made in Flash and all computers need to have an outdated Chrome version with the Flash plugin to run it. The other system is made in Java and some form of custom frontend framework, which works ok until it doesn't.
Expanding on this other system made in Java, it's a federal hospital, and we have other internal systems which doesn't communicate with this main EHR, so for example emitting radiology requests need us to copy paste information from two systems (like address, contact numbers), and on top of that those systems aren't connected to the national patient registry, and daily I have residents redoing requests to merge the information, otherwise the requests are made invalid.
I haven't touched on payments, imagine that each health insurance plan have different billings and we need to adapt the reality of what we did to what code better pays and input that in the system, so in practice the records are tailor fitted for each payment system, the actual procedure descriptions change, and we need to remember all that when billing and when treating the patient.
Add on top of that system outage and unreliability, and I haven't even touched much on the UI, which sometimes loses input text data or sometimes we have to input in certain fields order or else the system crashes, or the fact that the tabindex isn't set on all fields and we need to click with the mouse to go to a field.
Personally I've made a simple system for my private practice, while it doesn't have all the functionality, at least I'm the one to blame for it's particularities. I'm still exploring how to better input the clinical data, and I'm starting to think that general systems doesn't work. Each specialty has specific routines which need to be accommodated in the system, be it structured forms, be it clinical image input with annotations and commentary. The field is huge, and we're looking at how to design UX for immediate input and for later review, which sometimes are at odds (for example, a single textarea is easy to input, but how do we parse that data and present a timeline of clinical signs for example?).
I guess we need a Linux of the EHR, something which we can iterate on. I've looked into open source projects, but I don't know if the field is entrenched in inherent complexity or we're all trying to model too generic abstractions on top so that a small team of developers can't comprehend the system.
I should publish some code instead of rambling, but as the field is covered in regulations, I fear not even a code license can disclaim legal obligations.