The EMR, and AI: Are they good? Pose risks? Both?

When I worked with residents in the hospital, the electronic medical record, EMR, was relatively new. We were fortunate to work in a hospital that invested heavily in a good, well-regarded EMR, and spent quite a bit on training the doctors to use it. In the end, almost all the “stakeholders” agreed on which one was best, and the hospital bought it, and had us trained. Good for them.

The EMR wasn’t perfect though. In addition, the hospital didn’t buy all the parts. EMRs come in modules, some necessary, some elective, especially back then. It was clear that the hospital had prioritized the modules for billing, and especially for maximizing billing. Also, anything that the subspecialists who earned the hospital lots of money wanted. Other modules, particularly those that would enhance primary care, were more rudimentary or absent. Some of the things that many of us thought would be easily facilitated by a computerized database and looked forward to having were not available. Surely, once everyone is loaded into the computer, it should be simple to print out a list of all the patients with diabetes assigned to a particular doctor! That would really help us to track them, contact them, make sure they didn’t fall through any cracks. Whoops, sorry, we didn’t buy that module. The maximization of potential billing, on the other hand, was not only there but required many different screens to be filled out, effectively transferring work to physicians from someone else.* And it was inconsistent in how it treated health risks. For example, tobacco use had a who series of questions, including information the patient themselves probably forgot about how much, when, etc., but there was only one on whether they drank alcohol.

There were many things that the EMR did make easier, though, including writing long notes in the chart, since people didn’t have to write by hand. Like the Word® program I am using, and most other computer programs, cut-and-paste became easy and routine. Residents’ notes got longer because they could cut-and-paste yesterday’s note and (hopefully) update it. But sometimes they might forget the update part; it could be embarrassing if yesterday’s note said “surgery tomorrow” and it still said it in today’s note, even though the surgery had occurred that morning! The EMR also facilitated making notes longer by importing all the lab results and radiology reports. This is important information, but it is also available elsewhere (i.e., in the lab and radiology sections). A simple “Radiology exams normal” or whatever they showed would have been much better than cutting and pasting the whole report, as well as briefer. Better because it would have required the resident to read it, make an assessment (“it’s normal”, or “it shows a tumor”) and write that. It would have required thinking. Not to say that they didn’t think, but a summary in their own words would have demonstrated that in a way that cut-and-paste couldn’t.

But the biggest problem with the EMR is the amount of time that it takes to complete, especially in outpatient clinic settings, and especially for primary care clinicians who usually have a wider variety of issues to address and less money to hire others (scribes, sometimes nurses or even NPs or PAs) to do their documentation for them. It is not uncommon for primary care physicians to spend more time documenting in the EMR, frequently at home at night**, than seeing the patient! And in the inpatient setting, hospitals hire nurses to comb charts looking for ways to “upcode”, to charge more.*** (A part of the ongoing contest between providers and insurers to see who can hit the other up for more (except when they have been vertically integrated, more common in outpatient settings, see Vertical Integration saves money. And CVS and its competitors use that to line their pockets, not provide healthcare, May 21, 2026). And potentially costing the patient more, if the insurer refuses to pay it all.

And now we have AI. Or AI is having us. The debate on AI, on whether it will create a great new world or a “Brave New World” à la Huxley, rages on, now with the Pope getting involved with a new 42,000 word encyclical. AI is happening, will continue to happen, and will continue to have effects, many untoward, and some of those resolving – but not necessarily in ways that are good for people. And there are many different people, not just in the US but in the world. Recent commentaries have suggested the benefit would be greatest for the well-off and well-educated (well, almost all things do), although what seem to be “regular” people are using it to bolster their “home brewed lawsuits” and clogging up courts (good or bad?)

I know a lot of doctors who are thrilled about AI, and see it as a vehicle for reversing, or at least slowing, the constant drain on their time that comes from more documentation being required for billing, for insurers, and even for government regulations, in some ways a counter-weight to the EMR. They have apps that record the entire encounter, and then AI drafts a progress note that covers all the essential information in the conversation for both clinical and legal/billing purposes. Then the clinician reviews, augments, and corrects the AI-generated note. Hopefully. That is a danger. AI (as well as clinicians, it should be noted) can make mistakes, and provide incorrect information. With people, we know who to blame. However, recent experiences with friends and family encounters with the health care system suggests that once something gets into the medical record, especially a digital one (indeed, all digital data collection), it is there forever and efforts to correct it do not always take.

And, back to the residents copying their notes rather than creating original ones, it is comparable (if more high-stakes than) to students using AI to write their papers. It allows the appearance of creation and completion without the thinking required to learn to do the job right. Of course, AI advocates argue that AI learns to think more reliably than do people. Maybe this is not a scary idea. A recent opinion piece in the New York Times by Dr. Helen Ouyang suggests that AI (ChatGPT, in this case) gives good, well-researched medical information, and, more important, is accessible to answer questions when the doctor isn’t. The author notes that ”Of course, as a doctor, I know when to question the chatbot and when to ignore it. Many other patients don’t.” That’s right, and that’s a concern. Most of us who have used AI know that it isn’t always right, but if it’s a topic we don’t know about, we don’t know.

The other thing that Dr. Ouyang liked about ChatGPT was, ironically, its personality, since “I had always assumed the ‘human side’ of medicine was the part A.I. couldn’t touch.” The AI was unflaggingly positive, upbeat and encouraging, and never got irritated about repeated or “stupid” questions. People miss this when dealing with – people. While some doctors, like other people, are not, by nature, always warm, positive or supportive, the circumstances in which they work, the pressure from their employers (see several previous pieces, recently Why is it so hard to get medical care? And what should we do about it?, March 15, 2026, and The problem with the US healthcare 'system': THE INSATIABLE PURSUIT OF EVER MORE MONEY BY CORPORATIONS AND WALL ST., Feb 25, 2026). We should also remember, that while being nice, and friendly, and supportive is usually good, it is also a strategy for gaining your trust that has been misused by bad actors throughout history. And AI never gets tired of doing it, never wants to go home, never misses its kids, and doesn’t have to worry about spending as much time completing the EMR as it did seeing you! (see Does AI communicate better than real doctors? If so, why is that?, Nov 20, 2025).

So, I guess that the jury is not in on AI, or its most effective and reliable and accurate utilization. When it is, it will probably be too late to change it.

  

*This is only one example of work that has been transferred to the primary user. I have long made my own travel arrangements, and like it because I know what I want, but it takes a lot of my time.

**Another example of work transferred to the clinician, at the expense of their family.

***See this piece for a clear example of widespread and profound upcoding: https://healthcareuncovered.substack.com/p/government-watchdog-agency-finds

Does AI communicate better than real doctors? If so, why is that?

The New York Times recently ran an article titled “Empathetic, Available, Cheap: When A.I. Offers What Doctors Don’t”, which should be very concerning to the medical profession as it emphasizes three things that they are often not. But probably won’t concern the real decision makers in healthcare – the corporate owners, “health systems”, insurance companies, and private equity. After all, their concern is solely making money, and they are doing just fine, thank you.

The article indicates that AI seems to be responsive to and nice to people, and seems to show respect, concern, and empathy; “seems to” is important, because these are computer programs, not people, and they don’t have any feelings. Nonetheless, people feel better when they are addressed with respect, concern, and compassion. Even if it is programmed and not real. The truth is that doctors and other actual people do not always do so, for a variety of reasons. And they don’t even have the chance to if the patient cannot contact them, which is so common as to be routine these days

For many years, I told medical students that, while they had worked very hard to master the language of medicine, learning idioms, jargon, eponyms, and acronyms so they could fit in and impress their seniors, residents and attending physicians, regular people would not understand them if they spoke like that. They had to be able to translate that back into their first language, English (or whatever their vernacular was). This is an important skill, for without it people (“patients”) won’t understand what you are saying, and won’t know what is going on with them. And that is important. It takes effort, and it takes intentionality – you must want the person to understand what you are saying. That’s is true even if what you are telling them is bad news, something that will make them upset or unhappy.

I thought about this after a recent conversation with a couple of current medical students. I made the points above, about the importance of communicating in a way people can understand, and observed that, in fact, often people did not understand. This was based on, among other things, the number of times I had to try to explain to my patients, as a family doctor, what their specialist was saying. And the number of times I had to try to figure out, as a family member or friend, what my family member or friend’s doctor had been telling them that led them come away with what seemed to be an incorrect understanding of the situation. I have even said “If you assume that no one ever understands anything their doctor tells them, you will be correct a distressing percent of the time”.

The students agreed, but when they gave examples from their experience, I became more concerned.

A surgeon I worked with was unable to get all of the cancer out, but when telling the patient used all kinds of technical and unfamiliar terms, like ‘clean margins’. It was like they were trying to not lie, but to obfuscate what they were saying by talking in words and phrases that were technically true but not meaningful to the patient. I was left, after the surgeon had gone, to try to respond to the patient who asked me ‘What did they just say?’”

Obviously, this should not be the job of the medical student, but of the surgeon. And while it is tempting to say, “Well, they’re surgeons; communication is not their strength” (and while, as a family doctor, I like to think we are better at it), most or all doctors are guilty of this sometimes. (It is also true that it is even harder when you have to acknowledge that the bad news may, in fact, be the result of something you did wrong, but this is a separate area.)

I have recently had experience with close family members who had complications during procedures. One, during an endoscopy, had their blood oxygen level drop and had to have a breathing treatment afterwards, receiving a new diagnosis of asthma. This was upsetting, but at least they were told everything. Another, in a much more concerning episode, had major lung surgery. After the surgery, they had terrible, persistent pain which was not adequately treated. Several months later, visiting another doctor (not the surgeon), they were told that their oxygen level had also dropped severely, as a result of having a pneumothorax, a serious, potentially dangerous condition where air gets into the chest cavity and can partially collapse the lung. More relevant, it can be terribly painful. This might explain why the nurses, following their pain-management algorithms, did not give the patient sufficient pain medication. It is still not clear if they were told their patient had a pneumothorax, but it is definitely clear that the patient, my family member, was not told. They should, of course, have been.

There are a lot of potential problems with AI providing people medical information, some of which are discussed in theTimes article. For one thing, it could be wrong. It doesn’t really know you, and part of the reason that you are consulting the medical AI (or real clinician) is that you don’t actually know either exactly what is wrong with you, or how to put it in terms that will get you the correct answer to your question even if the AI is capable of getting the correct answer. Of course, sadly, the same can be true of real doctors, especially when you don’t actually speak to them; the article leads with the story of a person who wanted advice on how to increase the protein in their diet, and received generic – and unhelpful – answers from the physician on line (presumably a “patient portal”). For all we know, they could have been AI produced.

It would be much better – some of us would say essential – for doctors to communicate fully and honestly with their patients, using language that they can understand, even when the news is not good. And for them to be there, being, well, patient, while their patient tries to formulate questions, and answer them. But there are a lot of reasons that they don’t, or can’t.

A part of it may be that they are poor communicators, or uninterested in having their patients understand everything, especially if it could be embarrassing or take a lot of time. But AI doesn’t have that problem. It is not paid by the patient, and it has no set number of people it has to see in a given amount of time the way that real clinicians do. These actual clinicians often work in hamster-wheel conditions (time spent not only seeing patients but having to do electronic charting aimed at maximizing profit via upcoding as much as possible) which are not the fault of the doctor, but of their employers who are interested in “throughput” to make as much money as possible. Saliently, procedures are relatively well reimbursed but spending the time necessary to talk to a person to be sure that they completely understand what is going on is not. Of course, this is also part of the reason that there are fewer students entering primary care and more are entering better-paid procedure-based specialties.

Having a health care system that valued, and paid for, communication would be good. It would have to start with a system designed to maximize the health of our people, not corporate profit. Yes, there would still be some doctors who communicated poorly, and even made poor medical decisions, but those could be dealt with as individuals, rather than having them intrinsically encouraged by the system.

Doctors could and should do better, and maybe there is a place for AI. But there is no place for profit in healthcare.

Growth, Progress, Drugs, and Health Care: All in pursuit of profit

There are times that I worry that I might come across as a Luddite, opposed to new drugs, new technology, progress and change, because I am often critical about how these changes are happening, and also frequently have dampened enthusiasm and expectation for the probability that they will be successful in achieving their stated goals. For the record, I am not against either change or progress, although the definition of “progress” is a loaded one; progress is only good to the extent that it makes people’s lives better. Of course, if it does not it can still be considered “change”, but may well be regress.  Here I exhibit my values: it has to make people’s lives better, and not just those of a few people, but essentially everyone.

I have recently been re-reading John Nichols’ novel “The Magic Journey”, originally published in 1978. While Nichols’ most well-known book is probably “The Milagro Beanfield War”, published in 1974 and made into a film with Ruben Blades in 1988, “The Magic Journey” is his magnum opus. It documents the transformation of Chamisaville, a fictional town that is a thinly-disguised Taos, from a 400-year old subsistence agrarian economy with little cash changing hands, to a “modern” town. A major way this occurs is through the hiring of young people to build things, paying them salaries, allowing them to buy stuff, going into debt, and selling off their families’ land. It is a bit more complex than that, because the “Anglo Axis” controlling this change does far more direct and clearly evil things to move the process along, but the change, the progress (as it were), occurs. It both changes and does not change the life of the people who live there. Yes, they (or many of them) may now have cars and refrigerators and central heating (and the debt that goes with it), but they mostly all still remain poor, and on the edge of desperately poor, often teetering and sometimes falling. Meanwhile, the positive joys of their lives, the music and singing and storytelling and culture fade with each generation, except those that are commercialized for sale to tourists. The Anglo Axis does well, of course, and the one consistency in their decision making is how to make more money and, in order to continue to do so, cement their control and hold on power at every level.

The story of Chamisaville is a story of capitalism – of its triumphs, bringing progress, and its dark, dark side, institutionalizing a continuing oppression and repression. And, in this, it is a microcosm of the US, and much of the world. The health care industry is a big part of that world, and it is run by the same folks for the same reasons. To wit: make as much money as possible, regardless of who (else) gets hurt. And to press forward with the marketing campaign – health is good for you, right? – so that those who control it continue to make money. In capitalism, certainly US capitalism, “progress” is one of the two most commonly used vector words, along with “growth”. Both, the story goes, imply “better”, but this is not always the case.

We know – or should know by now – that “growth” can be terribly bad. With 7 billion people on the planet, most of them in desperate poverty we also have desperate inequality (in India, for example 9 people have as much wealth as the bottom 50% of the population – that’s 600,000,000 people!), we have limited resources. And we spend an enormous percentage of them on war, which continues to compete with the climate crisis generated by our dependence on fossil fuels to see which will be the first to wipe out life on earth, not just our health. The mantra that growth will solve all our problems is not only wrong, it is by now backward.

Yet, in health care, we continue to strive to make progress, and it is not always bad. But one does not have to be a Luddite to note the risks and contradictions. Novartis has just released a new gene therapy, Zolgesma® that treats spinal muscular atrophy, a terrible genetic disease. That is a good thing. But it costs $2.1 million. That is an unbelievable thing. Should it be made available to the children who need it? Yes. Who pays? Not the individuals (unless they are at least multi-millionaires). The insurers? Yes, but then it gets spread to all those with that insurance. The society? Like a national health insurance plan? Sure, but who decides that Novartis should get $2 million for it? This is a big deal. It isn’t that children with this disease should not be treated, but it is the fact that LOTS and LOTS of other people with more prosaic diseases could be treated for that money.

How to allocate resources will always be a difficult decision for health care. More for lower cost services that benefit more people or more for higher cost services than are critical to a few is an ongoing decision made in nations throughout the world because all resources are limited. However, a few things are clear (to me) that should guide these decisions. The benefit of the many should take precedence over the benefit of the few, even if the few are very rich (and I am not saying that the families of SMA children are). This is easy to say, but becomes more difficult when the benefit to the few is great and the benefit to the many is smaller. Or when you are one of the few. Quality of life is important; in the US and some other rich countries, enormous amounts are spent on high-technology care at the end of life. Yes, sometimes it is difficult to know when the end of life will be exactly, but it is often clear that what is being done is protracting existence without quality or hope of improvement. And, in this context, huge profits for the drug manufacturers is not something that should be built into the equation.

It is not only end-of-life care that skews high-technology. Investment in whiz-bang stuff is always, somehow, sexier than providing the care that we already know how to do, already is relatively cost effective, and is not brand new. We need to keep this in mind as we develop, invest in, propagate, and utilize new technologies. Luke Miner’s Op-Ed piece in the New York Times, ”For a longer, healthier life, share your data” argues that the Health Insurance Portability and Accountability Act of 1996 (HIPAA) too extensively restricts data sharing among providers and especially to researchers, who could use this “big data” to identify epidemiologic patterns and link them to genetic profiles within populations, enhancing  both the likelihood of diagnosis and the opportunity to develop treatments. In addition to HIPAA, his big emphasis is on artificial intelligence (AI), and how it could compile and sift through this data to achieve results never before possible. He suggests that, as long as HIPAA prevents sharing this data without your permission, that you give your permission.

This is not necessarily a bad idea, but can never be divorced from the MOST core concept: Who will control that data, and perhaps even more important, who will PROFIT from it. The evidence is absolutely clear: there is nothing so destructive, so evil, so heinous that some people will not do if it makes them enough money. This is lesson #1.

Lesson #2 is that before, or at least while, we invest billions in genetic drugs, personalized medicine, artificial intelligence, high-tech gizmos that will help some people, or maybe (no guarantee!) in some future help a lot of people, we have to be able to disseminate the well-know and cost-effective treatments that we have. Millions die daily from preventable (see: vaccines) or conditions treatable with things we have available to us.

The real magic, the real whiz-bang, the real excitement, will be in ensuring the widest possible implementation of what we already have to care for all the people who need it.