His Doctors Were Stumped, Then He Took Over
nytimes.com
nytimes.com
https://news.ycombinator.com/item?id=12266170
My wife has since passed away unfortunately. I really need to write this all up at some point to close it off.
Anyway, I've come to understand that most doctors look at the scientific evidence indirectly.
They often have an excellent ability to follow a decision tree and work through the differential diagnosis. However, that decision tree is based on (a) observations and (b) what other doctors have tried. There's no attempt to look at the underlying mechanism to find novel solutions. So new basic research (e.g. on mTOR and VEGF as in the article) will have no effect on practice until it is proven in trials or until enough mavericks (like the patient in the article) have created a precedent.
It does make sense in terms of avoiding cost and harm -- biology is very complicated -- but I do think this conservative bias goes too far in cases where the treatments have been already proven safe in other contexts, or in cases where the alternative is already bleak.
They are, as you say, trained to arrive at a diagnosis to the best of their ability and provide treatment. A challenging diagnosis, however, is a fully-fledged scientific investigation. This is where clinical researchers (MD PhD) can really shine, but it's hard to apply this to individuals; the typical direction follows funding first, which may not be of much help to any given patient. At any rate, this approach ultimately does not scale.
My hope is that new tools (accessible databases, better data collection, and a dash of ML) will get more patients in front of the right pair of eyes more often. The situation with life-threatening diseases is bad enough, but there are also huge opportunities for improved quality of life from vague, nagging, or intermittent illness diagnosis.
I think the issue here is not that physicians "are not investigators" but more-so that the financial realities are in no way aligned with dedicating massive amounts of time to any given patient. Physicians generally bill around $300/hour in cash only clinics, so I think it's a poor allocation of resources to expect them to dedicate 100s of hours to a more obscure patient. PhDs are dirt cheap, better to hand the odd case study off to them and let those with clinical skills focus on being clinicians to the more typical patient.
I'm a unicorn. No one knows what to do with my diseases. I've been told, with absolute certainty, for 25+ years that my worst one will eventually go away.
Finding effective treatments are more the result of drunken sailor's walk thru the solution space, than finding a leaf in a decision tree. I believe (but cannot prove) that'll be true of all rare multi-factor diseases.
The key point, advice is to never stop asking for help. Someone, somewhere out there has the answer.
I am appreciative when medical staff is required. But when it's not, ruthless automation must be out into place to contain costs.
Automated triage systems are already used at intake for urgent care clinics. They suck, partly because GIGO.
The challenge would be in creating software that can intelligently interact with a patient to surface the actual facts.
Humans are infinitely variable and read and interpret clues that we've barely identified let alone figured out how to automate reliably.
But that doesn't mean it's not possible. Machines will, one day, do this BETTER than humans can.
There isn't a direct decision tree to follow.
Most doctors are not trained researchers, like others have said. A researcher in a different field could be better equipped to solve a certain medical problem than a doctor, given that they have access to the right information.
Here, one should also consider that the doctor typically has 15 to 30 minutes to listen to a patient and come up with a solution, whereas a persevering patient can spend hours upon hours searching for solutions, correlations or clues on the internet and in the scientific literature.
(Background: My colleque (MD PhD) who works in a precision oncology project from the bioinforatics side just lost a friend where he and other specialists where involved in the treatment. And that was not an exception. The search space of what might be the issue is just to big, the human population is to small to solve some genetic twirks (yet, hopefully).)
Edit: Sorry, overread your point about the not allowed treatment. You are right, the colleque above is able to predict better treatment in 65% of cases that gets applied only in 25% of patients (of total). Still, I would not go so far and say that those alternative treatments should also be applied right away because we would not know what would happen and while it may be good, side effects are a serious issue that cost lives, the hospital potentially millions (we are in europe, thank god) and are of limited scientific value. Studies will be done in to these treatments at a point in the near future but that may not help the current patients. This might sound cruel to you but it is probably the best we can do. Untested treatments are a serious problem, no good doctor would want to have anything to do with that, sorry.
http://tim.blog/2016/10/20/my-life-extension-pilgrimage-to-e...
As a physician in training however, I can assure you that some of us indeed know about mTOR and VEGF inhibitors, the signaling pathways, and what we could potentially use them for. We learn about it in school. If anything, this story is more of a commentary on how mandatory genetic testing may be more helpful in the treatment of rare diseases. Castleman's is a broad definition for what is going on. Some patients benefit from Anti-IL6 therapy, but maybe some, like Dr. Fajgenbaum, benefit from mTOR inhibitors like sirolimus. That makes me feel like each Castleman's patient may have a different underlying cause for the same severe symptoms.
As far as I know, physicians use certain medications for off label use. VEGF is of interest to me and my research, so I know how we've used the VEGF inhibitor Avastin (bevacizumab), originally approved for colon cancer, for macular degeneration. More and more uses are becoming known for many drugs.
As a side note, this patient's use of sirolimus to treat Castleman's is still akin to using a blunt device on the signaling pathway that mTOR is a part of. It still hasn't truly gotten him any closer to figuring out the cause of his Castleman's disease.
It took me about thirty minutes with the test results and Google to make the diagnosis. I was lucky, in that I had a colleague who was acquainted with the head of Stanford endocrinology, who confirmed the diagnosis and found us a good surgeon. Wife is fine today.
It's anecdotal, but my experience with the medical system is that nearly all doctors see the same handful of things day in and day out, and are not trained to see beyond their expertise.
http://www.businessinsider.com/ibms-watson-may-soon-be-the-b...
I'd pay a small fee to get access!
I went to five different doctors for throat discomfort and chest pain. The first four didn't see anything in my mouth, so told me I can't be having any symptoms if they don't see anything after a two second glance at my mouth and refused to do anything further.
I was in enough discomfort I lost sleep, and after a few months of these symptoms I couldn't stop thinking the worst.
I had to practically threaten the last doctor to do an infection test which turned out a positive result for mono. Not my imagination after all.
As for my chest, I described my symptoms to my mother, who almost immediately suggested acid reflux. Sure enough, as later confirmed by a throat specialist, that's what it was, and the medication fixed my symptoms pretty much the next day.
If you're a young male, doctors don't want anything to do with you. I usually Google things and make my own decisions now. I've started to view GPs as nothing but an unfortunate barrier on my way to a useful prescription or specialist.
Cough was gone. It was miraculous!
I'm sorry you had a bad experience, but as a young man who had recent health issues this wasn't my experience at all.
I expect computers to outclass MDs very soon. Likely an Android will interview you and a MD will just have a look on the diagnosis (XX% chance for a XX% chance for b, make test A,B and C to rule B out).
Unless you mean a tablet running android, in which case I agree.
How long do you need to build an MD?
I don't think IBM build Watson to win Jeopardy.
At this point she was admitted and being monitored 24/7. Her OB was in Hawaii and getting back in a few days. I told her straight out to ignore the Drs. And nurses at the hospital. She was only allowed to speak to me or the specialist, and disregard anything the regular staff said.
It amazed me that the staff had a diagnosis confirmed with a test yet they still gave her the wrong information because she did not fit the regular pregnant woman profile they were used to pushing through their assembly line.
Her Dr. got back and delivered the baby the next day as the monitor showed it going into distress. He said the hospital lost 2/3 babies every year from Choleostasis, yet Drs. seldom acted on symptoms and treated it correctly.
Scary.
Richard Feynman wrote about how when his wife got sick he figured out that she had TB, but because he was very young and a physicist her family and their doctors ignored him and his diagnosis. He acquiesed and did not push, which he later regretted.
He said that only after they had tried many other approaches and wasted time did the Drs come to the same conclusion, but by then it was too late for her.
At that point her realized he was much smarter than regular MDs and never took their word as gospel from that point on.
That story stuck in my head as my friends and family have had medical challenges.
I have a relatively rare spinal condition and I've only seen 2 out of about a dozen doctors who knew what it was before they googled it and one of those gave me figures I'd already seen in a research paper I'd read and the other had less knowledge about my condition than I did.
I don't blame them since there are literally thousands of medical conditions and a lot of them fall on the long tail in terms of what they see on a day to day basis.
I'm not afraid of disagreeing with my doctors anymore though since often I know as much about my condition as they do.
It would be better if they didn't affect an air of omniscience sometimes though.
The only reason they found the condition at all was because I had an A&E (ER) Doc tell me to go back to the GP and demand an MRI on my spine, he was a good doc, he knew something wasn't right.
In medical school we were tasked with identifying an unusual cause of a baby's fussiness after eating.
For the first time, everyone got the right answer. My professor wasn't sure whether to be proud of us for all doing deep reading into rare disease or to be angry that we had collaborated.
But in fact, everyone had just popped the unusual findings into Google, which spat back the diagnosis of ALCAPA (anomalous left coronary artery from the pulmonary artery).
Your doctor's brain is not the world's best device for information retrieval.
For example, My wife's car would sometimes fail to start. However, later it would start with no problems. After a while I noticed it had a much higher no start rate when it was raining outside. We took it to the shop and I explained the situation, so they did a "wet test" and it turned up nothing. They explained that they were powerless to fix it without being able to reproduce it.
I wasn't satisfied with that answer so I did my own research on google with the make/model of the car and concluded pretty quickly that the distributor cap was the problem. I took it back to the shop and specifically requested that they replace the distributor cap. They did, and the problem was resolved.
As a former doctor, agree 100%. Still the medical profession seems to look down upon "information aid"-like technology. Beyond accurate physical examination and correctly entering the patient's history into a well-trained pattern-matching AI engine, doctors really don't need to be doing much in a first-world country. They need to be re-inserted into the workflow only if the treatment recommended by the automated/AI approach does not help.
The brutal truth is that the vast majority of doctors, even specialists, will be incapable of correctly diagnosing and treating the very uncommon, rare and extremely rare medical conditions.
"The eyes see what the mind thinks" - is absolutely applicable to 99% of doctors, who, for various reasons, would be more concerned with helping the 95% of their patients with obvious/common/easily-identifiable conditions that respond to their first or at most second line of treatment, than bother spending an inordinate amount of time, effort and energy on the remaining 5% who's death/suffering may well be considered the "cost of doing business". Bitter truth, from experience.
Even though she had a diagnosis, she was on a regular maternity ward.
None of the nurses / Drs. Seemed to really want to learn about her condition or make sure she was being treated correctly for her disease. She "didn't fit". Her death or the baby's death would have been "the cost of doing business".
Her doctor said 2 or 3 babies died of Choleostasis every YEAR, yet the nurses and doctors on the maternity ward were quite clueless.
You really need to take charge of your treatment. It can mean the difference between life or death, or the difference between being a victim or living a normal life.
In my own personal experience, I've unraveled medical mysteries affecting my health that have stumped some pretty good docs, and in one case, a super specialist. On the other hand, I've been greatly helped by docs.
All I can say is don't give up just because docs say they don't know or don't seem to be making you better.
Sadly, a lot of doctors just make shit up out of thin air to help gloss over what they can't be expected to read, which is probably a good time to consider switching doctors.
Can you describe how this process differs between people with an illness like cancer and people with an illness like Morgellons?
We rely too much on them and delegate everything in our society to them.
But first, there are varying degree of quality among experts. The delta is huge.
Secondly, they are only human, and make mistake all the time.
Thirdly, they have an agenda too. It may not be to help you the best they can, but to make a living with the least problem possible.
Now another problem is that a lot of people reject expertise in a such a way that any honest attempt will look terrible after them. The typical anti-vaxer speech has no credibility at all, preventing any sane debate on vaccines. The pro consider it "not even a question". But everybody should always raise questions about any medical act before deciding to got for it or not. This include vaccines, band aids, whatever.
The notion or "expert" also tends to remove responsibility from the people. But your health is YOUR concern, you choose take great responsibility in it.
In the end, we should all learn to not ask experts to solve our problems, but to use them so we can solve our problems.
He's a licensed MD and doesn't accept insurance; cash only at $300 an hour. Totally worth every penny.
He actually listens, and is willing to get creative with things.
If you're not satisfied with your PC provider, find a cash only doctor. I think you'll find they care more.
A buddy of mine is a world-renowned specialist, but in a niche too small to allow him to go all-cash. He gets a patient a couple of years back who came with literally boxes of medical records, years of failed doctor visits, and in excess of 100k$ in worth of tests and interventions. He spent two days poring over her records, and got her the diagnosis no one else did - got her on the right tx, and gave her her life back.
Insurance reimbursed him precisely the same (<$300 gross) that it would have for an ordinary visit of ordinary length, more or less. (Fine, he billed a level 5 eval and management instead of a level 3; difference of nearly peanuts). How often do you think he sets aside the time to pause and really think deeply on a problem?
If he was charging cash and had the chance to charge people for his time and expertise, knowing him, he'd be happy to do nothing but spend days at a time tackling the hardest problems available.
Well, quite frankly, this is part of the problem and it's a social adjustment we need to make as well. We see this with virtually any position where you need to diagnose; 99% of the time it's a simple issue from column A, but occasionally you get a symptom that looks like the issue is from column A but it's actually in column B. In technology, we fly through column A solutions because they're cheap, fast to test, and often done without cost, and column B solutions are more drastic, more expensive, and sometimes more dangerous.
The difficulty in translating this to medicine is that a lot of times column B solutions are dangerous to the point of no return. Combine this with the fact that there really aren't rollbacks when it comes to health, you have actual life or death decisions to be made.
Taking the time to review the whole history can really help with this and to get a full timeline of events, but even then it's hard to get past the column A/B thinking. I've been dealing with a very annoying sinus issue for 3 years across about 8 doctors from across the planet (due to moving because of work). Every time I've started anew with a doctor, they go down the exact same treatment list as the previous, and when they're told that I've done that already, they just want a repeat. I understand this thought process (what if I'm lying, what if they bungled the test, what if something changed, etc), but I've done this 8 times now. I've been told each time it's an allergy, but every time we see an allergist, they can't get me to react to any allergen they have available to test for and are left scratching their heads as to why antihistamines don't have any effect. They also can't find a common cause that fits with allergies, but in their minds, it just looks so much like allergies they're not sure what else to test. I secretly suspect that if I could just get a doctor to spend more than 20 minutes in a consult to review the case and what I've done so far we could try something else, but regardless of where I've been, I just can't get a doctor to look too much longer. They either don't believe the symptoms as I report them, don't believe the steps I've already tried to resolve it, or they just don't have the time to care past the initial look at the symptoms.
I appreciate the difficulty in having to do this: when column A fails, it looks great if you pick the right item from column B, but it also gets really expensive while you shop around to find the right item from column B. They make it look so easy on House, but no place is willing to really do that sort of patient history and testing, and if the condition is not life threatening, then it's just a "deal with it" situation, we're gonna keep throwing the easy stuff at the symptoms until the person stops complaining.
I don't know what the problem is or how to embolden doctors to take a little more time, but some change would be nice. Though, hearing stories from my mother who has been a nurse for 20+ years, sometimes emboldened doctors look for problems that aren't there so they can look like heroes.
It's a crazy thing with health and unfortunately the value of such emboldened decisions tends to be a very ad-hoc assessment. If it worked well, it was a gutsy but necessary maneuver. If it did nothing, it was reckless. If it caused harm, it was malpractice.
https://en.wikipedia.org/wiki/Concierge_medicine
Annual fees can run in the 10s of thousands of dollars but quality and access to care is top notch. An appropriate comparison would be flying Spirit vs private. It's such a different experience (and in the case of medicine - outcome) that it can hardly be called the same thing.
Access tends to be a lot better, but only in those practices that are 100% cash. Those practices that try to "transition" or do a mixed customer base tend to be the same as all-insurance practices, since they fail to transition operations/processes towards the cash-based context.
- Top-tier education. Medical school and residency at top 5 programs in the United States.
- Availability. They have total patient rosters in the low-hundreds, and many of these patients are seasonal. Response was always same day and more often than not immediate (phone, text, e-mail). I got the sense I maintained this consistent level of ready access because I am a relatively "simple", "low maintenance" patient.
- Time. Full labs and blood draws were done quarterly. Results were tracked and compared over time. Each draw was accompanied by a follow-up that could last as long as an hour, depending on how long you wanted it to (for whatever reasons). I never saw anyone look at a watch or a clock. Making informed decisions regarding diet, cholesterol, and lifestyle (for example) is much easier when your doctor can spend 30 minutes discussing your lipoprotein trends over the past five years against 30 years of varying data, trends in medicine, and studies of actual impact on mortality and quality of life.
- Network and peer credibility. Their "med school buddies" were often in nationally recognized programs and specialties. Made an incredible difference in access at least once for me personally. Any referrals were met with similar levels of access, even in all-insurance or mixed practices.
I've often said the single most important benefit of concierge medicine (for whatever underlying reasons) was that your physician actually LISTENS to you and treats you like a real person, a whole person. Not only does that make substantial differences in your perception of care (and overall experience) it made definite and substantial differences in outcomes in my experience.
He took and analysed blood samples over many months -> logging.
When he identified something in the logs that looked odd, he took a drug that should affect that behaviour -> debugging.
/s
https://campus.aynrand.org/works/1985/01/01/medicine-the-dea...
Outside of hospitals (hospitals and physicians are not equivalent; think Walmart and your local mom-and-pop shop, if you want a better idea of the relationship between the two), physicians are reimbursed directly by volume. Not quality, though - just volume.
I think we also have that same problem in economics. People consume more when times are good, so let's get them to consume more and times will be good.
http://www.econtalk.org/archives/2016/02/adam_cifu_on_en.htm...
It's amazing how many doctors operate in "tunnel vision." A few years back after a routine procedure that put him under, my dad awoke to non stop hiccups. At first the anesthesiologist said it was a temporary side effect of the anesthesia and brushed it off. Well the hiccups continued for 2 freaking weeks non stop.
As you can imagine this is a) a major disruption to a normal life (he couldn't sleep, couldn't eat, couldn't do anything for two weeks) and b) annoying as hell to everyone around you.
We went his GP multiples times, consulted the same idiot anesthesiologist and went to the emergency room 4 times over those two weeks and the reactions varied. Everything from the doctor literally laughing out loud seeing a middle age man in agony over non stop hiccups to another doctor prescribing pineapple juice. Yup, pineapple juice. Apparently some people use that to stop hiccups.
After the third trip, with no seeming progress and my dad basically asking us to kill him so that he can get some normal rest, I decided to start googling, which I should have done when this all started. I found a interesting research paper from Cal Berkeley which described a similar case after surgery, which described the mechanism that actually causes hiccups (its a spasm of the diaphragm cause by the vagal nerve) and how they were able to treat it.
We decided to go to a more than decent hospital in the big city armed with this paper in hand. In to the emergency room, after the standard tests they did at the other places and the scopes and all that, the doctor goes yeah we have no idea what's going on let me look up some stuff and I'll be right back.
He went to some computer terminal and started clacking away while I was watching over his shoulder unbeknownst to him. After about 3 minutes of searching some medical portal the hospital uses and seeing his head shake repeatedly I walked over, tapped him on the shoulder and said, hey I think I might have found something helpful on my ipad and handed it to him.
He read the paper in 2 minutes flat said "oh my god," called his supervisor (he was the attending physician, not a resident) they both agreed to call this doctor at Cal and after a 5 minute phone call wrote a script for a drug called Baclofen which treats muscle spams. 2 of the worst weeks of my Dad and our lives' came to an end within 15 minutes.
I was glad this place being a teaching hospital, was open minded enough to consider what they were presented
The hospital thought I was some kind of Doogie Howser and the head of the ER goes to me "You should really consider becoming a doctor." I replied with "Why would I ever do something silly like that. I'm an engineer." The doctor was both amused and not amused at the same time.
Just goes to show you, how narrow their field of vision is. Most doctors get trained a certain way with certain tools and for the rest of their career that is all they know.
Applies to most of us whatever the value of x.
As engineers and scientists I would like to hope that most of us on HN are constantly looking for new tools and novel approaches. There will always be new tools and techniques for development, new frameworks, more optimized algorithms, better debugging, etc. We should all always be looking to integrate, update and enhance our toolset and skills and inspiration comes from many places, not just technology. It's a part of being the best at what you do--which for us happens to be critical thinking and problem solving.
At least I try to do this constantly myself.
Maybe that's part of the problem. Most people are trained to be whatever they are without a real focus on true problem solving. They're given one framework to solve the issues of their field and told to run with it.
That's a big part of this doctors story too, that he was willing to enhance the state of his art by thinking outside the box, pulling together an uncoordinated, distributed team, sharing information and research and opening the channels of communication.
Hmmm, sounds familiar...
Same thing happened with recent tremors related non-invasive surgery at Imperial college London. Data played extremely crucial part to localize it. (http://www.bbc.com/news/health-38157770)
Really excited to envision a purely data-driven healthcare system. It will have a significant impact on our understanding of different diseases.
Anyone knows where to find such anonymous health/disease related data for further research?
I don't mean any disrespect by this comment. Clearly he is an extraordinary individual.
Personally I don't see how the title misrepresents the content in the article, or tricks anyone into reading something they'd be better off not to have read.
https://news.ycombinator.com/item?id=10077754
https://news.ycombinator.com/item?id=11979596
https://news.ycombinator.com/item?id=10161193
https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme...
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
The bottom line is that occasionally having to stop and work a little acts as a speed bump that slows down the usual internet reflexes, making reflective responses a bit more likely. Reflective responses are what we're hoping for here.
If you introspect a bit you can catch yourself getting irritably impatient at having to take 5 seconds rather than half a second to process something on the internet. That's the reflex I'm talking about. The problem is that in that mode our responses are automatic, therefore predictable, therefore uninteresting.
Why not move the buttons on HN to random parts of the page? Or randomize the article titles every time you reload HN, so you're never sure whether you've already read an article? Or swap one comment between threads to keep people on their toes? Wouldn't those all be equally plausible ways to get people to read things more carefully by sacrificing usability?
Your other comments suggest that you would be against tl;dr's, which outside the internet are known as "abstracts", and (like descriptive titles) are known to be highly useful.
Note that we're currently engaged in an unhelpful meta-discussion about whether or not the title is clickbait, which IMO is exactly what the guidelines are designed to prevent.
While I didn't say so earlier, I largely object to the needlessly inflammatory word "stumped" which perpetuates the meme that patients frequently know better than doctors.
This case is an interesting exception, not the rule. There's a debate to be had about the fact people do indeed know their own case history than their often time-starved doctors, but whether or not the article has merit, the title is a distraction.