Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Her Various Symptoms Seemed Unrelated, Then One Doctor Put It All Together (nytimes.com)
131 points by CaliforniaKarl on Feb 18, 2018 | hide | past | favorite | 112 comments


> As for Huls, he is finishing up his fellowship at the hospital this summer. Wherever he ends up, he is certain he will find new cases to challenge his curiosity, cases he hopes to solve.

No, he won’t, because only early stage residents are given the freedom from economic realities to spend hours digging and thinking through records. Docs aren’t paid for their time, nor even for results - and their employers (read: hospitals) have no intention of going into the red.

Do note the key to this diagnosis wasn’t finding a brilliant or deeply experienced path: it was about putting it into the hands of a curious trainee, who had the time to chase that curiosity.

I’ve said it before and I’ll say it many times in the future: when it comes to complex patients, there’s enormous value in chucking insurance out the window and just paying the doc for their time. A moderately clever doc afforded more than 7 minutes to work on a case can solve a lot of problems.


Is there a way to actually request a specific doctor have some time dedicated to figuring your case out at your expense? i.e. "Please do more troubleshooting for me above what insurance pays."

I like your idea - but I fear the industry might have already ruled it out.

-edit- Granted, I've done this in my own life by going to specialists at their own offices/outside of a hospital or medical group.


Unfortunately, no. Your doctor is contractually bound to the insurer’s fee structure; they can’t take an extra penny off of you. You’d have to go to someone that doesn’t take your insurance at all (or just don’t tell them you have that insurance).

The side effect of all this, honestly, is that doctors stop learning how to turn off 7-minute-mode. When it comes to very serious problems, it pays to go to a practice that doesn’t accept insurance at all, because they never got into the bad habit to begin with.


> You’d have to go to someone that doesn’t take your insurance at all

There are some doctors at teaching hospitals who specialize in diagnosing rare diseases and don't take insurance for this reason.


You're right that it's difficult for physicians to "turn it off". The contractual requirements you reference, however, are basically unenforced. The only way a physician will get "caught" charging more than agreed is if the patient calls the insurer to complain. Presumably you wouldn't do that to her after she saved your life.


> Presumably you wouldn't do that to her after she saved your life.

Oh, you grossly underestimate what people are capable of. At least in China there are many reported cases of such horrible patients. Maybe US being a developed country have fewer these people, but definitely not zero.


You're right of course, and this isn't limited to the Chinese. It's best to have physicians in the family; marry one if there isn't one already! This isn't practical for everyone, so perhaps it's best to choose a capable young physician early in life, so that by the time such trust is required it has already formed.

Probably there are other professions in which one gets better service if one has an inside connection, but this situation really speaks ill of how medicine is practiced.


> Probably there are other professions in which one gets better service if one has an inside connection, but this situation really speaks ill of how medicine is practiced.

Having a doc in the family is a mixed blessing at best. You can get priority treatment, and the benefit of your family members full attention. On the flip side, all the docs I know that aren’t fools or morons don’t treat their own family for anything more serious than a scrape. You’re biased, you don’t want the guilt of an error, and you don’t want to be the bearer of bad news. I frequently ask my wife not to ask me about her mother’s health developments; I’d rather tell comforting lies than be a source of sour truth.

Being a VIP in a medical center due to your relatives gets you more ass kissing, not better care. People don’t want to step on toes, or give the impression that useful care is being withheld. And if your relative is a worse doc than the one you’re seeing? No one is going to overrule your family member to your face.


The value I would see in having a physician as a close relative is in learning how to communicate effectively with other physicians. Then one could in theory need fewer visits to their own doc to get to the bottom of a problem.


That’s actually a problem. When I myself see a doc, I make a point of being as colloquial about it as possible. Part of our training is about taking what people say and converting it into technical symptoms; when people start to “communicate effectively” it heavily biases us towards certain diagnoses. The more intelligent/educated the patient the more likely we are to take one of their word choices at face value.

Personally, all I ever want from a patient is a clear timeline of events, their best attempt to articulate what they felt, a list of scripts, and old medical records (especially imaging.) The patient that can give me that has made the task of helping them 1000x easier.


That’s really helpful to know.


> Presumably you wouldn't do that to her after she saved your life.

I wish that were so, but disease is not confined to the reasonable, the rational, the kind, the respectful, or the grateful. On the contrary, the stress of disease and it’s associated burdens (social, financial, psychological, etc) tends to diminish those qualities in even the best of us.

Harvard did a huge study on malpractice suits a while back. They found that most medical errors that caused the death of a loved one did not result in suits, and most suits did not originate from medical errors. That is to say, the good or ill you do a patient isn’t the reason they sue.


Almost makes the red envelope culture in Chinese hospitals seems appealing.


Not at all.

Having to slip the doc and the anaethetist a red envelope and hoping it's the enough that they'll take proper care of you is not a good system to live under.


https://en.wikipedia.org/wiki/Red_envelope

I’d never heard of this before.


Lol, that's exactly life in Eastern Europe/ex-USSR. There's socialized medicine, but if you want better service, you have to bring a gift (whisky, chocolate, money, the latter is always desirable). Not just hand cash over, that's disrespectful nowadays.


My grandmother was a pediatrician in Kiev in Soviet times. Her salary was about on par with anyone else, but she got tons of gifts.


On the situation in hospitals specifically:

https://www.seattletimes.com/opinion/the-red-envelope-capita...


Glad I live in a country with a free health service.


[flagged]


Did you really link 'free healthcare' with 'a country that does not pay doctors'?

Clearly healthcare can be free to the recipients while doctors still get a good salary.


Most countries with free healthcare do have woefully underpaid doctors. The NHS has been cutting back aggressively on doc salaries for years, letting them fall well behind inflation.

The only thing I can say in their defense is that at least docs in those countries don’t generally take on massive, private, un-dischargeable debt in order to do the job. In the US we treat the medical workforce as a public good, though they have to take on all the risk of entering that workforce through private financing.


I get to drive on most roads without paying, but I wouldn't say that the roads are "free"?


Compared to toll roads they are free...


We've uncovered a potential ambiguity in the original phrasing of "free health service". Could the now-dead response to that comment have been highlighting this very ambiguity?


Yes, I've done this. I paid an out-of-network specialist for a second opinion on a cancer treatment. My HMO ended up following that specialist's recommendation. While I likely would have survived either treatment, this one was shorter and had fewer long-term side effects. I paid for the consultation out of pocket. It was less than $1,000.

Not that outcome matters much to your question, but I have fewer scars, more original body parts, and better overall body function than I would have had with my HMO's original plan.


Some possible search terms that might help (but I don't actually know the answer): concierge medicine, direct primary care, fee-for-service, out-of-network


Isn't this what concierge doctors offer?


Some. Some push the same old practice model but with reasonably cheap prices, so they can keep their old patient base and make their income off the slightly higher billings and the reduced overhead of eliminating your billing (read: calling the insurer and arguing) staff.


Why doctor? It's the most expensive part of your healthcare. But it can be trainee/researcher/AI. Which could be absolutely out of expensive healthcare system. Service which you could buy for additional cost from company which has good guess/prediction ratio based on input parameters from doctor. Service which could be in any part of the world.


What? The doctor’s visit is the cheapest part of healthcare. Procedures cost far more, and pharm far more again. Perhaps you’re outside of the US, though I can’t think of a country where this isn’t true.


Don't forget diagnostic scans and tests. Those are quite expensive. However GP is proposing a service that doesn't actually exist now: paying a medical expert to spend a lot of time considering a challenging case. It would have obvious benefits, so we shouldn't dismiss it out of hand just because it would be too cheap.


There was a now-defunct attempt at this called MetaMed: https://en.m.wikipedia.org/wiki/MetaMed

I'm curious if any others have succeeded.


I have often thought that the Freestyle Chess model (a variation of Advanced Chess https://en.wikipedia.org/wiki/Advanced_Chess ) could be applied to medical diagnosis.

Create a team consisting of one or more human doctors with a medical diagnostic AI.

From the article it sounds to me that a lot of the research could have been done by an AI, with the human doctors making the final decision on the "best" diagnosis.


The technical term is "clinical decision support system".

https://en.wikipedia.org/wiki/Clinical_decision_support_syst...


There's an analogy here for software and engineering too. Getting to root cause and finding solid solutions - takes curiosity, energy... and time.


You know what they call software organizations that don’t have time to do this? “Fast paced”


For some time my read on this situation has been that the "doctor shortage" that produces 7-minute diagnoses stems from a restricted number of residency slots, which is set (I believe?) by the federal government. I had thought the AMA was influential in lobbying to keep the number of such slots low to prop up doctors' salaries, but with some research it's unclear how true this is. Anyone know? How strong a lobbying force is the AMA, anyway?


The AMA has actually been lobbying quite strongly to expand the number of residency spots. Doctors salaries are generally pushed from the top down by Medicare and not by general supply-and-demand principles. So these shortages don’t lead to salary bumps, they lead to inferior substitutions (expansion of mid-level practice privileges), which actually puts a downward pressure on salary while also providing worse care for the public.

The problem with residency slots is that they’re funded as part of the larger Medicare legislation, and no one wants to push for the bump in funding/taxes for Medicare needed to expand it.

The reason visits are seven minutes long is because part of a trend that started with the HMOs in the 80s. The hypothesis on the part of insurers was that docs have a general target income - that they don’t work at maximum capacity (in general - obviously wonky things happen at the extremes.) If you reduce per-service billing, docs would maintain constant output to hit their personal target income goal. And... that’s exactly what happened. Since then, insurers have continued to hit against per-service fees, and docs and facilities have continued to shrink the time they alot to each service in order to keep afloat. In fairness, docs don’t have an awful lot of flex on this - they have to meet overhead costs which are fixed regardless of how small their reimbursement gets, and even with pitifully short visits doc real earning power has gone down quite a lot in the last 30 years. Docs work about 3x as hard as they used to, for about one third of the real income.

It’s actually a bit scary: there’s not a lot of leeway left in this, and medical education is more expensive than its ever been. Docs carry the financial risk and massive increase in burnout, patients bear the risk of the reduction in time and quality of care, and everyone else is squeezing the juice out of the grapes. Something’s gonna give.


Interesting. It seems like I was wrong about the AMA then. Is there a reason that a standalone bill cannot secure federal funding for addition residency slots, or is the stumbling block that increasing the number of residency slots will be expensive (even in federal money terms) regardless of how it's enacted?


Your comment wasn't wrong. In 1997, The AMA along with five other medical groups lobbied congress to limit the number of medicare-funded residency slots [1, 2]. This limit was enacted in the Balanced Budget Act of 1997 and hasn't changed since then [3]. Hospitals were paid hundreds of millions of dollars to voluntarily reduce the size of their resident training programs [4].

[1] http://articles.baltimoresun.com/1997-03-01/news/1997060012_...

[2] http://www.nytimes.com/1997/03/01/us/doctors-assert-there-ar...

[3] https://www.cnn.com/2017/03/13/health/train-more-doctors-res...

[4] http://www.cnn.com/HEALTH/9708/24/doctor.glut/


You posted this two hours after I replied with a list of google results showing you’re two decades out of date. Now it’s -clear- you’re being disingenuous, but for the life of me I cannot fathom why.


None of this is outdated. You're trying hard to to rewrite history with downvotes.

The 1997 cap on medicare-funded residency slots has remained in place since that time, unchanged despite population growth and an aging population. The current crisis and physician shortage is in large part a result of that two-decades-old legislation which was engineered by the AMA and other major medical groups.

Based on your comment history, you trained as a doctor, which is great and absolutely commendable [1].

However, do you feel compelled to troll and post obviously slanted information due to your personal association with the AMA?

For the record, I think doctors should get paid well and more than they currently do. Clinics should be run by physicians. But allowing guilds like the AMA to artificially restrict the availability of critical healthcare has resulted in millions of avoidable deaths and serious suffering across the entire population.

"The predicted physician shortages will result in decreased access to care for millions of individuals. [...] [A]dding one PCP per 10,000 people would reduce predicted all-cause mortality [...] by 5.31 percent. Translated nationally, this would avert 127,617 deaths." [2]

[1] https://news.ycombinator.com/item?id=16337410

[2] https://web.stanford.edu/group/sjph/cgi-bin/sjphsite/the-loo...


"In 1997, a consortium that included the AAMC, the AMA, and other major organizations declared that [...] 'the United States is on the verge of a serious oversupply of physicians'. The consortium recommended limiting the number of residency positions funded by Medicare, a goal that was partially achieved in the Balanced Budget Act of 1997" [1]

See also the 1997 senate finance committee hearings on graduate medical education [2].

[1] http://annals.org/aim/fullarticle/717927

[2] https://www.finance.senate.gov/imo/media/doc/hrg105-901.pdf


I know the Mayo Clinic has a totally different pay structure for docs, maybe there's other places out there like that as well. I think this is why people go there for answers to hard problems. Docs get paid a salary, they then choose how many patients they see, and that's it. I assume that's also why it can be tough to get an appointment there.


Mayo is a bit of a special case. Guwande hyped their fee structure in comparison to private practice, but it’s largely the same as every hospital: the doctor is an employee with an annual salary, the hospital is reimbursed per unit service by payors. Guwande compared them to private practices, where the doctor acts as his own employer, and thus has to maintain their own volume.

The difference is that given their prestige and the number of out-of-state (read: mostly cash-paying) patients, they get much higher reimbursement than most places and can afford not to push the patient mill so hard. Normal hospitals have a patient base that is mostly insured or, if uninsured, the impoverished kind, not the globe-trotting wealthy kind. So their reimbursement is negotiated with insurers, so... they have to push the mill hard.


Well, I suppose he could end up in a country where health isn’t in the insurance straitjacket


> there’s enormous value in chucking insurance out the window and just paying the doc for their time

I'm sure there's enormous value in chucking economic realities out the window for everyone's job, but that doesn't make those realities go away.


You perhaps misread what I wrote. I don’t know how to reply, since I don’t know what it is you think I said. I don’t believe it “throws economic realities out the window” to say that sometimes you need an extended interaction with an expert, and that if you want more than 7 minutes of an expert’s time, you should engage with a reimbursement model that pays for more than 7 minutes. That doesn’t ignore the economic realities of healthcare at large; it’s saying sometimes you need a premium good, and those have premium prices.


I thought you were proposing just forgetting about insurance and having the doctor's time paid for by magic.


> Huls hadn’t heard of Schnitzler syndrome, either. He came upon it by using the database PubMed to look for a disease that matched the patient’s symptoms. He made a list of her symptoms and abnormalities. To get the full picture, he combed through her earlier electronic medical records, now archived in an old electronic warehouse, and found that her symptoms had started maybe a decade earlier. Then he looked for a disease that fit. It took hours before articles on this strange disorder began to appear. As he read, he suspected that she had it.

Is this database open? If so, anyone in the same situation as the daughter in this article could spend the hours, rather than hoping a professional will. Not to self-diagnose, but to narrow down the list to a number of candidates the professionals are willing to look through. Sure, what takes the M.D. 6 hours might take me 40 hours, but if the alternative is 10 years of suffering for someone I love, it's easily worth doing.

On a similar note, I have combed through a bunch of research on a particular type of medication recently, and I hope the fruits of my labour will pay off. Time will tell. Either way, I have learned a lot.

Once you are dealing with somethibg that is difficult to diagnose and/or resistant to treatment (I might argue the two are the same thing on a fundamental level), a surprisingly large portion of medical care turns DIY. They never tell you this but unless you are actually made of bucks and can visit the best specialists in the world, it won't take long until you know more than the doctors you meet about the specific problem you're dealing with.


Pubmed itself is open (just google "pubmed"). However, while many scientific publications are open access, many more are not.

I find that google scholar does a pretty good job at searching. I typically start with google before using pubmed.

Then you need a place to store your papers. I like to use F1000 workspace.

If your doctor is not helping you, get a new one, preferably in a big city with a university-affiliated hospital. These MD deal with highly educated levels of patients and colleagues, and they like it.


> many more are not.

True, but

https://en.wikipedia.org/wiki/Sci-Hub

https://sci-hub.tw/

sad that it needs to exist, but good that it does.


> I find that google scholar does a pretty good job at searching. I typically start with google before using pubmed.

It does, plus it has the great advantage that it will find a PDF of the paper posted by the author (or someone else) on a webfolder somewhere. Pubmed only links to the "official" paper publication, plus some open repositories in some cases.


> Is this database open? If so, anyone in the same situation as the daughter in this article could spend the hours, rather than hoping a professional will.

I've done this with various doctors, trying to figure out what the hell is wrong with me. In my experience, as soon as you start telling them they just think you're a crazy hypochondriac who sends too much time on the internet.


I think a lot of it depends on how you present it and who you see. I have had the best results when leading the doctor on just to the point of tricking them into thinking they came up with the idea themselves. Play to their vanity!


I'm guessing they're just pattern matching as most of the time, they'd probably be correct.


Yep, that makes sense... problem is when something doesn't match the patterns, they just shrug and give up.


My first reaction to these things is that there should be a large database of diseases and symptoms along with some statistics about how strong the correlation is between the disease and symptoms. I don't mean "database" like pubmed. I mean something more machine readable with consistent coding so a symptom might be "elevated IgM" in english. The trick is coding something like that with consistent names for the tested item and consistent definitions of "elevated". Perhaps the definition of elevated would depend on the disease?

One should be able to then filter the disease database based on symptoms. Doctors seem to do this in their head, but they are not comprehensive stores of information. I suppose the problem is in the encoding of symptoms, it's not that simple.


The only appropriate definition of elevated is higher than normal for that patient. Unfortunately, doctors rarely have a baseline to work from, so they use statistics instead. A relative of mine had a PSA test that read statistically normal. Luckily he looked at the actual number and saw that, while it was within statistical normal, it was in fact higher than any other prior reading for him. Further testing came up positive for prostate cancer. If he hasn't been paying attention, that would have been another year of cancer until the next check.


It’s hard to convince institutions to release data locked up in EHRs (eg EPIC) due to the labor involved (EPIC is not well structured for this), risk of accidentally leaking PHI, and general motivation (getting hold of useful data is the bread and butter for career researchers).


The main problem here is human body works in mysterious ways.

We are by definition reacting different, and unfortunately statistics are not helping on this case.

Basically, we have:

- too many illnesses (I guess 10% of top common health problems, covering 90% of all problems)

- for each illness, we have N number of symptoms (which unfortunately all or only a subset of this can occur)

- then we have fake symptoms (statistically out of threshold values but we are living fine with them)

this is generating a big number of combinations.

With this much combinations, it is requiring extensive amount of tests.

As everything health is also optimized for the greater good. Which requires optimum cost management.

So basically medicine today, with statistical approach, is happy to skip 1%, if it can fix 99% with 90% less cost.


> - too many illnesses (I guess 10% of top common health problems, covering 90% of all problems)

I think this is probably the easy thing to fix when it comes to searching, as you could pretty much leave out the common diagnoses as they'd likely have been tried - Imma call this the "Dr House, is it Lupus?" filter.

From there, you could have "try this" filter, and see what the results are. I wonder what the overlap is between tests and symptoms, but surely with 10 tests and 10 symptoms, most illnesses could be mostly identified?


This was one of the early areas of AI research, which seems to have gone nowhere: https://en.wikipedia.org/wiki/Mycin


It does make one wonder how broken the system is. Consider how often we hope this is the norm, and yet it is held up as something exceptional:

———————

“When I see people suffering and I know that if I took the time and effort, I could figure it out,” he told me, “then I have to do something.” He looks for unexplained pathological findings — in this case, the high level of IgM.


Schnitzler syndrome, whilst rare, is well known to dermatologists and is in the differential diagnosis of chronic urticaria (hives). Not to take away from the resident that made the diagnosis but if the patient had been seen by a dermatologist I would be surprised if this would have been missed.


Article said a dermatologist was consulted early on and who failed to make the connection. Doesn’t support your ascertion.


Of course it is possible to miss the diagnosis. My point was that this is not an obscure disease that a dermatologist would not know about.


Anecdata, but... I personally know so many missed cases (from “mundane” cases like mono, to the more serious cases like neuroblastomas) that were diagnosed by googling or a guess from a Med student after being dismissed by a specialist, that leads me to suspect the problem is inherent, rather than the occasional miss.


I agree that there are many missed and incorrect diagnoses made not least by specialists.


But what you actually said is this:

if the patient had been seen by a dermatologist I would be surprised if this would have been missed.

I was diagnosed in my mid thirties with a genetic disorder. It is a more recently recognized milder manifestation of a well known condition. Prior to that, I was basically treated like a hypochondriac and routinely accused of being lazy.

I have read a bit about the difficulty in getting a proper diagnosis. It is routine for people to get treated like they are nuts because the doctors do not immediately have the answer. This winds up being incredibly threatening. It puts you in a position of fearing for both your freedom and your life if you keep talking about your symptoms and keep trying to find an answer, while knowing that not getting an answer may well kill you.

When I had been on antibiotics for several weeks straight with no diagnosis, I expressed my concern that I could die if they couldn't find the answer. The doctor asked if I would like a psych referral. My legitimate concern that I could die in the face of being deathly ill and doctors failing to readily diagnose it was treated like I was merely being neurotic and needed therapy.

Doctors frequently do a terrible job of identifying anything that is at all obscure and they treat patients in a way that makes it both very challenging and genuinely threatening to even try to get a proper diagnosis. I did all in my power to not wind up hospitalized because I was relying heavily on home remedies, like hot baths, to keep my symptoms down to a dull roar. I was convinced that ending up hospitalized would deny me the ability to do that and would be the death of me because doctors can be such assholes at times.

After finally getting a diagnosis, I began growing stronger and healthier. I had always been sickly and did a lot of self care. I now had an explanation as to why I was always sick and it was very empowering.

People with my condition typically get regular checkups with a specialist four times a year. As I grew stronger, my doctor began scheduling me fewer checkups and expressed zero curiosity as to how on earth this was happening when it is supposed to be impossible. When I moved because I was getting divorced and could not afford to remain in California, I did not bother to try to find a new specialist. I can manage my condition better than a doctor can and I didn't want the hassle when doctors don't really want to see me anyway because, I guess, I'm too competent to be a desirable patient.

One outgrowth of that: people on the internet conclude I am some nutcase rabidly rejecting conventional treatment. Doctors did not really want to see me. I no longer need a physician's care. But good luck explaining that to legions of internet strangers who are quick to judge and have no desire to be confused with the facts.

If you have an obscure condition, it is not unusual to know more about it than most general practitioners will. This fact is not at all respected by the general public. Claiming you know more gets pissed all over in internet discussion.

The system does an incredibly poor job of serving most people with obscure conditions. Given the wealth of information and tools available today, I find it appalling that this is so.


I've had a similar experience. I had symptoms for which I saw several doctors and eventually went to ER and was repeatedly just told by them and everybody else "it's just stress." Finally, I went to a specialist and they diagnosed my condition accurate and immediately. I don't even have any side-effects any more because now that I actually know what my condition is, I changed my diet accordingly. No drugs or anything, all I needed was someone to tell me what my condition is and that has improved my life immensely.


How did you finally get a correct diagnosis? I am in the same boat, and in my 20s. I have been to some reputed doctors/institutions but none of them seem to be in the least interested in my records or medical history or symptoms beyond 5 minutes, more in veiled accusations of being either a hypochondriac or an uncooperative patient.


How did you finally get a correct diagnosis?

Kismet? Fate?

My ex was career military and I had the brilliant foresight to be a supportive wife and homemaker, allowing him to shine at work. This turned out to be critical to saving my life as his last 3 duty stations before we divorced were all special duty assignments, including his assignment to Travis Air Force Base as part of a tiny contingent of Army people stationed there.

Travis has a regional hospital serving military members from 8 states. It's longest hall is like a quarter mile long and it has its own dedicated gate on the base. It also is a teaching hospital that works closely with UC Davis Medical Center in Sacramento 45 minutes away and I think has connections to stuff in San Francisco, where some of our (family) testing occurred.

I had already been very sick for some weeks before we moved to Travis. Just before Christmas, we moved from temporary quarters into an apartment. January 4th, 10 weeks of chronic sinusitis turned to pneumonia and I ended up bedridden until some time in April.

I had about 2 doctors visits a week during that time and they ran all kinds of tests. All but two tests came back negative. I was told I have a ragweed allergy and a genetic disorder.

Because I was in the SF Bay Area, where by all rights an infantryman would not have been stationed normally, there were like 3 hospitals in the area that dealt with this disorder. My official diagnosis was atypical cystic fibrosis based on sweat chloride results in a historic grey area. Had I been anywhere else in the world, I probably would have died, having never received a proper diagnosis.

Most doctors have heard of CF. They have not heard of atypical CF, which presents more mildly and is frequently misdiagnosed as asthma, a diagnosis I had at one time. I felt it was wrong, but it got me useful treatment, so I didn't argue it at the time.

I also did some of the following things which I list in hopes of answering what I assume is your real question of "Do you have any insights into what I can do to try to successfully get a proper diagnosis?":

I persisted in the face of a lot of nonsense from medical people.

I tried to put my emotional baggage down before appointments and keep focused on getting answers rather than venting my spleen at random doctors.

At some point, I wrote everything down. Printing this out and taking it to a doctor was a mistake because he did not have time to read it. That appointment did not go well. But getting it all straight in my own head and able to tick off some of the most important details in chronological order and in a nutshell proved very valuable.

I eventually did some of my own research online as well.


Two different dermatologists, actually.


Are you "ascerting" that dermatologists never make mistakes?


Hi HN

Clinical research types may find this still imperfect tool to be of use none the less.

https://monarchinitiative.org/disease/MONDO:0018304#phenotyp...

Full disclosure: guilty guilty guilty


Is it just me or is this site exactly what several other people in the comments here are proposing as a major part of the solution?

I’m not sure, since others are basically saying it’s too hard to do.

Cross referencing conditions and diseases along with the scientific and common names for things could indeed be very useful!


That is the goal but it is a very wide and deep problem you won't catch me trivializing but we are chipping away at and seeing success. Our group working on laypersons synonyms because most folks with rare diseases do not know the terminology.

https://human-phenotype-ontology.github.io/2016/03/24/layper...

I should probably be more effusive about all we do as we always seem to be further along than most attempts(and we give it away), but as one of the sausage makers I always worry someone will run into something that is not fixed yet.


Surprisingly, if I google ”hives elevated levels of IgM” or ”hives elevated IgM”, ”hives fever elevated IgM” or several variations, the top hit is https://rarediseases.org/rare-diseases/schnitzler-syndrome/.

For ”hives elevated level of IgM” and ”hives IgM”, that’s the second hit.

Does Google react that fast to nytimes articles?


It's possible and even likely that Google did react -- it is enough for people to google "schnitzler's syndrome" and click the link to rarediseases.org -- as google does track which result people choose, and rank it higher for the future (for different searches as well).

It's also likely that "hives elevated IgM" is the right search in retrospect; Before diagnosis, the search would likely have been more along the lines of "hives weight loss elevated leukocyte elevated IgM fever rashes at night" which likely didn't point as easily to schnitzler's syndrome (or maybe it did, I don't know).

The info was distilled and purified for this article. It likely wasn't that well presented to start with - in my experience, that's the case for many rare diseases: once you know what it is, it is clear that in retrospect you should have seen it all along. But that's just 20/20 hindsight.


I have zero medical expertise, but it sounds like this situation could be improved with software. Given a database that contains an exhaustive list of every illness and symptoms:

    illness      symptom     diagnosis_symptom_set
    -------      -------     ----------------------
    id           id          id
    name         name        symptom_id (FK->symptom.id)
                             illness_id (FK->illness.id)
             

    (Pseudo SQL)

    SELECT diagnosis.name as name 
    FROM diagnosis_symptom_set diagnosis_sympt
    JOIN illness diagnosis ON diagnosis_sympt.illness_id = diagnosis.id 
    WHERE symptom.name in (hives, chills, weight-loss, .....)

This will at least give a patient awareness of what has been tried/tested and what remains, possibly saving them months or years.


Much easier said than done. There are several startups out there attempting to achieve something like that. The issue isn't in the technical difficulty, but the bureaucracy and red tape involved in getting the patient data, getting enough illness/symptom data to make diagnoses possible (which is a lot), proving to the hospital that it's worth the effort to adopt the platform, and making a solution agnostic enough to extrapolate to enough hospitals to turn a profit.


The issue is also the technical difficulty. There is huge variance and the number of variables to consider is also great.

A computer/database is a better tool for the job, of course, but the balancing act between false negative and false positives is extremely hard.

See, e.g. the (very) simplified case of base rate fallacy[0] ; There is an inherent information problem in doing diagnosis, that a large database and fast search do not address.

[0] https://en.wikipedia.org/wiki/Base_rate_fallacy#Example_1:_D...


Agree 100%. I never claimed it would be easy, just that software could help.


https://www.gideononline.com/about/gideon/ is like that for infectious diseases.



[flagged]


Haha


My guess is the doctors who don't spend extra time trying to solve the case probably shouldn't be paid for the time separately as is being suggested here.

It would appear to me that they have already lost the innate curiosity (if they ever had it) that made them want to become doctors in the first place.

Or they're not confident enough in their ability. Doctors are already paid substantially compared to most other professions, so for them to not put in extra time on the exceptional cases makes me feel they probably feel entitled. Too much so to have any impact on the rare cases.

The doctors who could realistically find the solution are the ones who will likely want to take the case on for free.


The idea that we should count on people to spend additional time on their job for free because they enjoy it seems overly idealistic. Sure, such people exist, but the easiest way to incentivize more time on a job is to just pay people accordingly.


Related startup: http://www.pacmed.ai/en/


What makes macrophages the "most primitive part of the immune system"? How about antimicrobial peptides, which are simpler and evolved earlier?

https://en.wikipedia.org/wiki/Antimicrobial_peptides


I had almost exactly these symptoms twice in my life, as a kid. It stumped doctors then, too. It was not Schnitzler syndrome: it was not chronic and in the longest case only lasted a few weeks to maybe a couple of months. However, while I never did get a "name" for the illness doctors came to the conclusion that my immune system greatly overreacted to something and went into a crazy state of attack - hives, persistent fever, painful skin, pain all the way through my bones, etc.

I wish I could put a name to what it was - I do not think my case was rare and a name does exist, but I was too young to have been told at the time.


Am surprised that a Grand rounds was not convened for this. There's an online service of the same name which has the same function.


I will not be reading this article because of its title.


This was like a real life episode of Dr. House MD


TLDR: It was Schnitzler Syndrome, a rare autoimmune disease.

Clickbait titles and suspense-novel narrative structures in articles like this really annoy me for some reason. Just get to the point.


>Clickbait titles and suspense-novel narrative structures in articles like this really annoy me for some reason. Just get to the point.

You are right, the suspense approach, is terrible, I had the impression that the protagonist of the story was the Dr. May, and then it comes out as being only another doctor that remained clueless making every kind of unneeded tests for seven months until - by sheer luck - a curious trainee managed to find the right clue, reportedly by digging into early medical records.

I may add that it looks to me somehow "disturbing" the Senseless Capitalized Initials of the title and paragraphs.


Thanks, I didn't find the format interesting, and yours was the first comment to mention or discuss the finding.

Info on Schnitzler Syndrome: https://en.m.wikipedia.org/wiki/Schnitzler_syndrome


Thanks. I'd love it if the first comment on every post was a link to a wikipedia article or other reasonably dry/authoritative source describing the substantive subject of the story. Often (but not always) I'm just not interested in reading the "human element" aspects that often dominate general/popular news stories. A title like this pricks my curiousity but I know I'll have to wade through substantial back-story to get to the meat of the article.


Right; it's like someone has watched too many episodes of House and thinks that's how doctors are/should be.


In this case it is the other way around. AFAIK Dr. Sanders was the inspiration for House.


Or like internet publications are killing dead tree ones and then having trouble getting monetized because of the ongoing ad blocker wars, so publications are desperately trying tricks to be more interesting.


This is from New York Times Magazine.

Think something like GQ, Vanity Fair, Rolling Stone.

It's not meant to be a news article, it's a story.


Further, it’s a popular recurring story arc that appears in many popular magazines (Discover, which for some reason I always buy when flying has a regular series) and in many books (eg, ``Awakenings’’ by Oliver Sacks)

Whether this particular submission meets the criteria for a HN discussion could be debated, although it does seem to have generated at least some useful comments.


I call this the Ira Glass effect. At some point in the 2000s everything became This American Life.


"Symptom Number Two will shock you!"


I'm not sure why you were downvoted. That title is pretty much the definition of clickbait.


Maybe because sarcastic quips are a poor way to get a point across. Simply pointing out the title is clickbait and perhaps suggesting an alternative would be useful.


Who am I to say what a good title is? I'm certainly not the editor or the NYT (no one is, they fired their editorial staff).

But I am the person to say "this title reminds me of tropy drivel which I've seen over the past few years, and I'll supply the appropriate subheading."


Something like that would have worked better, indeed. To be clear, the HN guidelines for commenting [1] say "don't be snarky".

https://news.ycombinator.com/newsguidelines.html


A bunch of hardasses who don't like jokes or people who actually think the BuzzNow content on their Facebook is enjoyable.


Or people who understand the reality of getting stories read.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: