20 ms·
"Operations (not surgery) are so incredibly bad / incompetent in most healthcare settings that software frequently gets the blame for much deeper problems." "I
by hyponatremia121 3y ago
"Operations (not surgery) are so incredibly bad / incompetent in most healthcare settings that software frequently gets the blame for much deeper problems."
"In my experience doctors are a tremendous barrier to resolving problems in healthcare operations,"
I'm a hospital-based physician that works in a system with great operations and results. The physicians, nurses and other staff work amicably together. Management is reasonable/nice. The EMR, though is universally despised. No one likes it. It is a major factor in burnout. The UI/UX is inconsistent. There are slowdowns and outages daily. There is a well known lag in the appearance of text in text boxes after typing that seems to be variable. I've caught the EMR cancelling orders I placed on critically ill patients in the ICU more times than I can count. We have to actively protect the patients from the EMR. Healthcare workers aren't perfect but they are trying to do their best for very ill people in a high-risk setting and the EMR is well-known blocker. I long for the days of paper records because this is worse. Paper charts didn't go offline, have slowdowns, didn't lose orders, were easily located, and easy to enter data into.
- davak 3y agoHypoNa -- I think you made a throwaway account for this reply, but I would really appreciate continuing this conversation with you. My email is available under my profile.
- hyponatremia121 3y agoI know that devs that work on EMRs might read this thread. I'm pleading with you to take the EMR to the next level when it comes to UI/UX and reliability.
- duffpkg 3y agoI've managed large enterprise health organizations across multiple sites that use Epic. We built an adjunct system that is still used by some of them that papered over serious problems that doctors using them never knew wasn't only Epic. We defined, documented and tested real world workflows before we subjected patients to them. There are a lot of other ways to address those issues operationally. You have management, IT and operations problems, not software EMR vendor problems. No software is perfect, Epic less perfect than most. Your facility chose that system and decided how it was implemented in a way that is dangerous or potentially dangerous to patients. No software vendor can fix that. That's really the heart of my rants in response to this article. I built that EMR, it was/is called ClearHealth/WebVista/HealthCloud. Your facility would never have bought it because incompetent management will only buy Epic. We solved the problem by buying organizations and replacing the incompetent management.
- peteradio 3y agoAre you still involved in that EMR? It looks like the company was acquired a decade+ ago.
- duffpkg 3y agoWebvista and HealthCloud are still utilized by some facilities and maintained as an internal system. EMR as a standalone line was <10% of ClearHealth's business when it was acquired in 2017, I remained involved in some capacity until 2019. ClearHealth's main business was as medical/hospital management company.
- beretguy 3y agoI am such dev. It’s not up to us. It’s up to our management. We would love to engineer good quality EMR, but upper management wouldn’t let us.
- repelsteeltje 3y agoThat, my friend, sounds a lot like The Cat Ate My Source Code [The Pragmatic Programmer] Price, lead time, quality, (and scope) — pick any two. When the stakes are high, don't rely on your manager to pick quality — that is your responsibility. And if you can't convince your "boss" to give you enough time to deliver something that meets your bar, quit.
- hobs 3y agoAnd then we have about 90% of working sw devs quit tomorrow - it sounds great but people are not going to leave when they know the next place is basically the exact same.
- repelsteeltje 3y agoI beg to disagree. Although I've only worked for European and US companies, my experience is that even bossy bosses appreciate quality design and understand that lack of time, pressure and exploitation cause products to degrade. But pressure can only exist if there is an equally opposing force — you, me, us. If my recommendation to bluntly quit sounded harsh, I'm happy to tone that down. There is value in compromise, but just make sure your voice is heard, say no if you cannot / will not do it in such little time to the extent that you feel is required. Fighting for the time to do our job well is our struggle. Sometimes it's easy, more frequently it's hard, but when it's simply impossible — have your boss find someone else. Quality — at least internal quality — is primarily the responsibility of the software developer, not her/his boss or some QA department.
- hobs 3y agoI don't think it sounds harsh, and I have worked with dozens of tech companies and directly for a dozen or so - the only one that wasn't obsessed with cheating out on quality to make a buck was bilking his customers so much that he didn't care. At every phase of the conversation is management saying they have this hard deadline and I have been fired multiple times for being too vocal about the failures of process or technology, though mostly I just get ghosted for pushing back when things are unreasonable, at this point I exclusively go for visualizations, gantt charts, etc - to communicate that the timelines are not possible with the current investment in the team/tech/whatever. I just got laid off for (as far as I can tell) being the person who kept asking the question about why we were doing a thing that provided no technical or business value for a year, and in most of these cases some exec thinks they know best because they read some magazine equivalent that tells them this is the new hype revolution. Basically if you can find some highly skilled or highly profitable company that's concerned about losing their edge you might be able to do this, but your next boss might change the entire equation.
- namdnay 3y agoIf the new EMR (or any other IT system) doesn’t look exactly like the old one, everyone will drag their feet switching to it
- bombcar 3y agoThis is the underlying issue for anything where software isn’t the primary goal. Nobody wants to change their workflow that they’ve trained over years, no matter how absolutely broken and absurd it might be.
- duffpkg 3y agoThis sounds a lot like Epic. Respectfully I don't understand how your facility can face those issues with a system like that and also be considered to have great operations and results. With Epic it is possible to workaround a lot of notorious problems but management has to understand and have in place the operational capacity to do it. If a system is allowed to persist in an organization that may have or actually did result in patient deaths in the ICU and it was up to me I would see the facility shuttered until that got sorted out. I also want to be clear that doctors are not by any means the only obstacle to resolving workflow problems. You being put in a position to protect patients from a business process of the facility that may accidentally result in their death is the literal definition of incredibly bad / incompetent operations. That software didn't magically appear and become responsible for ICU ordering magically on it's own. In this instance whatever insane people have been given responsibility for that implementation are responsible. After consideration, this comment from a doctor, in a nutshell encapsulates so much of what is wrong. A doctor on the front lines alluding to staff burnout, thinking that people almost or actually dieing because orders are being cancelled inappropriately in the ICU is somehow "a software problem", while simulatenously praising the management, operations and results of their institution. Welcome to healthcare.
- rendaw 3y agoAre you saying that the cancelled orders are due to a management decision/policy, and not a software bug? Or that EMR wasn't intended for managing orders and the decision to use it in such a way was wrong? Asking as someone who's mostly unfamiliar with this. Edit - I'm trying to make sense of this thread, and it seems like this might be a reasonable summary: There's good and bad software (epic). The fact that management chooses epic instead of good software makes it a management problem. @hyponatremia121 says that EMR software (epic) is terrible, which doesn't contradict the above. @hyponatremia121 pleading for EMR devs to focus on UX might be misplaced since their only experience is epic and not better EMR software. @duffpkg saying that epic shouldn't be blamed here seems overreaching. If epic were good/hard to misuse this conversation wouldn't happen in the first place. But there's always well marketed terrible software, and I agree organizations that aren't able to avoid this do have organizational problems.
- DANmode 3y agoSo you and your colleagues are protecting the patient (thank you). Who's been protecting this EMR? (Who likes using it?, who doesn't lose their job by continuing to choose that vendor?, who gets kickbacks from that vendor?)
- kakoni 3y agoIn Finland the tender was between Cerner and Epic. Because our largest hospital district wanted to buy American eletronic medical record system. I guess the biggest influencer was Kaiser Permanante (=they are world class and are using Epic, so it has to be best system in the world)
- Zanfa 3y agoI'm assuming the hospitals in Norway & Finland used some form of a public tender where the cheapest solution wins. I don't think you need malice or corruption to explain it, just good intentions.
- peteradio 3y agoThere's malice on the sell side. If they can't integrate a trial period then its not worth attempting. Its dumb terminal work not like its actually expensive just very very lucrative because its run like a mafia.
- Eddy_Viscosity2 3y agoExactly, when selling a big software system like this to the government, the contractor will very deliberately ensure that it follows the requirements exactly (because requirements are never specific enough) so that the government will then have to go back to them for fixes and upgrades forever. If they see a poorly worded requirement that they can implement as-is knowing it will cause a problem, they celebrate because its a future revenue source. The only way I can see to avoid this is for a government to have its own dedicated software developers who make these types of applications and maintain them. Preferably open-source so that other governments can use them as well. The incentives change and they'd probably save a ton of money.
- complex1314 3y agoOne of the biggest scandals in Norwegian health care at the moment is a botched transition to Epic in one of the biggest university hospitals. Doctor dissatisfaction has gone to the roof at the point where 50% of the doctors are considering quitting. https://www.nrk.no/trondelag/70-leger-soker-aktivt-ny-jobb-ved-st.-olavs-hospital-etter-innforingen-av-helseplattformen-1.16348026 https://www.nrk.no/trondelag/70-leger-soker-aktivt-ny-jobb-v...
- sl-1 3y agoSame thing going on in Finland [1]. 1. https://fi-m-wikipedia-org.translate.goog/?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp&_x_tr_hist=true#Kritiikki https://fi-m-wikipedia-org.translate.goog/?_x_tr_sl=auto&_x_...
- sergioisidoro 3y agoYour link is broken. I believe you wanted to link the Apotti wiki page https://fi.m.wikipedia.org/wiki/Apotti_(potilastietoj%C3%A4rjestelm%C3%A4) https://fi.m.wikipedia.org/wiki/Apotti_(potilastietoj%C3%A4r...
- danielovichdk 3y agoDenmark transitioned to Epic some years ago. It was a major frustration on many levels. https://www.sciencedirect.com/science/article/abs/pii/S1386505622001824 https://www.sciencedirect.com/science/article/abs/pii/S13865...
- briffle 3y agoEPIC is flashy, has a gorgous campus in Madison, WI, and has lots of marketing. but at its core, it runs on MUMPS as both a language and database. I know many people that have worked on their integration teams, and could not imagine having to deal with what they do. https://en.wikipedia.org/wiki/MUMPS https://en.wikipedia.org/wiki/MUMPS
- karmajunkie 3y ago
- submain 3y agoAs someone who worked at a fortune 500 company making such EMR software: There's no incentive to make the UI or workflows better. They don't pay the bills. Software is sold to the suits during dinners and baseball games, not doctors or nurses. Besides, a great portion of the development is outsourced chasing lower costs. The code reviews were so bad that a coworker used to joke that "we'd get more stuff done if we just fired the overseas team". The biggest and most well funded dev team was the one that worked on Revenue Cycle. I quit a few years ago and haven't looked back.
- elevaet 3y agoThis is so broken and depressing. I had a tour of the EMR system being used in my jurisdiction and it was shocking how bad the UI/UX was. The scenario you describe is exactly what I imagined. As a software developer I feel so motivated to fix it, I know a small team could do a better job. But I don't even know where to begin to try to enter that industry.
- nothercastle 3y agoYou don’t. It’s a regulated industry that regulated competition out
- karmajunkie 3y ago100% this. EHRs almost universally suffer from usability problems. people actually love having their medical records available. They hate the UX they have to go through to see them. The source of the problem is that the people deciding which EHR to use are almost never the people using it. it’s ultimately the CFO or CEO who’s ultimately going to make the call, and they do it based on metrics like price tag or how many accreditation checkboxes it fills, not on user satisfaction scores. Until those user satisfaction numbers make it into CMS guidelines for reimbursement, the situation’s never going to change.
- hyponatremia121 3y ago"The source of the problem is that the people deciding which EHR to use are almost never the people using it. " Yes, this. We are a Cerner (now Oracle, yaaaaa) site, not Epic. The Cerner folks I've met in smallish meetings are nice/great on the front lines. However, the deal that our CEO/CFO/COO/CMIO made with Cerner was years ago and, per rumor, involved large financial penalties for early cancellation. Subsequent people in those roles have chosen to uphold the contract. The local management and healthcare teams in my particular facility are great but I take a dim view of our c-suite as do the majority of the work force. Things are reasonable locally for most people, I think, except for the EHR. So, I would plead with leadership teams of EMRs, EHRs, healthcare systems, and shareholders of these entities to do the right thing and give healthcare workers working in high-risk settings the proper tools they need. You are currently failing us. No one wants unreliable, difficult software installed in critical settings such as ICUs. I've submitted tickets, emailed leadership and had conversations with the CMIO. If there is improvement it is so slow that I cannot discern it. In anecdotal conversations over the years I've observed that the vast majority of healthcare workers hate all EMRs but give Epic the nod because it is the least worst, and for some, reasonable. I have met a few Epic users that are enthusiastic about the software but that is after they arrived at our Cerner site and were in shock. @duffpkg - my original comment was under yours and I wasn't intending to malign your project. I'm not familiar with it but do like that it is open source. Respect to the developers that work on EHRs. I recognize you are held down by management. I'm perplexed by UX/UI problems in EMRs. This must cost more to fix than I suspect. It seems to be the easiest of the issues from my layperson/healthcareworker/tech-enthusiast viewpoint. One more thing to add: an effect of bad EHR is healthcare worker burnout. Bad EHR is a known contributor. Burned out healthcare workers are at increased risk to make medical errors. Bad EHR -> burnout -> errors -> adverse outcomes. This is a well known flow to us in the clinical setting that we attempt to mitigate. The executive teams of orgs that produce or purchase EHRs are never held accountable for bad outcomes even though, to my view, they share responsibility. Edit: government regulators (CMS, etc) also share in the blame for bad EHR and outcomes.
- elevaet 3y agoThis aligns with everything I've heard from multiple nurses and doctors in my circle who work with the EMR here.
- Justsignedup 3y agoThe classical "the executives bought software because it ticked checkboxes, and they'll never be the boots on the ground using it"
- cmiles74 3y agoI'm not seeing substantial investment back in the EHR, wherever the money is going it's not being spent on improving the product. Epic remains a thick desktop app served over Citrix sessions, their web-based workflows still haven't materialized.
- xcrunner529 3y agoAre you running Epic? As a patient it has been amazing to use and have everything available. My doctor uses it effectively and it’s quite cool she can do all prescription management etc through it. Now even virtual meetings.