Slashdot Mirror
88% Of Medical 'Second Opinions' Give A Different Diagnosis - And So Do Some AI (mayoclinic.org)
First, "A new study finds that nearly 9 in 10 people who go for a second opinion after seeing a doctor are likely to leave with a refined or new diagnosis from what they were first told," according to an article shared by Slashdot reader schwit1:
Researchers at the Mayo Clinic examined 286 patient records of individuals who had decided to consult a second opinion, hoping to determine whether being referred to a second specialist impacted one's likelihood of receiving an accurate diagnosis. The study, conducted using records of patients referred to the Mayo Clinic's General Internal Medicine Division over a two-year period, ultimately found that when consulting a second opinion, the physician only confirmed the original diagnosis 12 percent of the time. Among those with updated diagnoses, 66% received a refined or redefined diagnosis, while 21% were diagnosed with something completely different than what their first physician concluded.
But in a related story, Slashdot reader sciencehabit writes that four machine-learning algorithms all performed better than currently-used algorithm of the American College of Cardiology, according to newly-published research, which concludes that "machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others."
"I can't stress enough how important it is," one Stanford vascular surgeon told Science magazine, "and how much I really hope that doctors start to embrace the use of artificial intelligence to assist us in care of patients."
But in a related story, Slashdot reader sciencehabit writes that four machine-learning algorithms all performed better than currently-used algorithm of the American College of Cardiology, according to newly-published research, which concludes that "machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others."
"I can't stress enough how important it is," one Stanford vascular surgeon told Science magazine, "and how much I really hope that doctors start to embrace the use of artificial intelligence to assist us in care of patients."
So what the study finds is that doctors as well as technology disagree. There's nothing indicating that the second or third opinion is correct.
To Terminate, or not to Terminate, that's the question - SCSIROB
So, WebMD really is the better option... ;)
I would expect that many people who ask for a second opinion have a reason to ask for a second opinion: in fact, the article even mentions the situation where the first doctor recommended they do so. What would be more telling would be if they did a large study and gave EVERYONE second opinions, and then told us how many differed.
This smells like another case of "lies, damned lies, and statistics", which is designed to make the Mayo Clinic look good.
I'm not supposed that to doctors will give different diagnoses, but they also have to work with what they're given. A patient may give different details, especially if they've just spoken to a GP - they're likely to preempt questions that had to be asked previously.
This goes with every profession. A colleague of mine was being investigated by HMRC (tax) and was complaining that his accountant was still telling him it was ok to claim fuel, but the tax man wasn't. The problem was he didn't understand that by lying to the accountant (consciously or otherwise) she couldn't give him the correct info.
If only 12% of the time second opinions confirm the original diagnosis and 21% of the time they find something completely different, then we should obviously skip the initial diagnosis and just go for the second opinion.
I'm an American. I love this country and the freedoms that we used to have.
Another thing to keep in mind is that a chart is not some purely dispassionate recording of standardized metrics. Even before you get to the diagnosis, you are already looking at pre-processed information, and not raw data.
Although there are some parameters recorded for most patients regardless of the issues at hand (such as vital signs, or maybe listening to the patient's heart/lungs), other history and data is selected -- much of the history, review of systems, and physical exam is performed as a way of supporting or excluding items in a differential that the physician is already thinking about. The list of things that I could observe and write down about a human in a chart is practically unlimited, including a long and useless lists of things that patient doesn't have -- and if I choose to specifically mention what findings are absent, it is because I am making a case for one condition or another..
The other thing to consider is that part of the evidence that is used to support or refute a diagnosis is time. Often I ask a patient to follow-up over the next few days, weeks or months as a way to see if things change (or don't change) the way I expect it too. When a patient follows up with me, I may be making a "second opinion" each time, and the ability to follow someone over time is a valuable tool that gets discarded when someone bounces from doctor-to-doctor searching for answers, without ever looping back to previously visited doctors for follow-up. Sometimes there is a list of rare "zebra" diagnosis that I usually don't write down, because I would usually get laughed at by colleagues for jumping at shadows (but I may keep them in mind if things start not making sense in the future). In any case, a doctor who is asked for a second opinion gets the luxury of having both the first doctor's records (hopefully), as well as a data point occurring later in time.
Anyway, can the thought process for making a final analysis of the case a doctor has made from examining a patient be improved? Certainly, but keep in mind Garbage-In, Garbage-Out, if you feed the AI medical records from someone with excellent vs poor clinical diagnosis skills, the quality of your results will vary greatly. Presumably the best results would be obtained when applied to a doctor with good intuition and observation skills, but poorly organized decision making.
Point 1: Who is Al?
I've heard some call Paul Simon Al.
Ezekiel 23:20
concludes that "machine-learning significantly improves accuracy of cardiovascular risk prediction
Computer: Subject has 728% increase in cardiovascular heart disease risk due to obesity, sedentary lifestyle, nuggets, and left-handed masturbation to furry Dragon Ball drawings.
Impressed Doctor: God damn.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
One doctor saying you have cancer while another says it is the flu is definitely 'different'. If both say you have lung cancer, but the second says it is a slight variation of the cancer diagnosed by the first doctor, does that meet the definition of 'different' for this study?
Maybe the people who went for a 2nd diagnosis had reasons to do so.
It should be read as: of people who were UNSATISFIED with their first doctor to the point they got a second opinion, were correct in their belief.
Not that 88% of doctors visits are wrongly identified. 99.999% of people who get a cold, and get some antibiotics, and feel better in a couple of days, aren't seeking a second doctor.
Is it sad that a robot can diagnosis better then a real doctor? Kinda says these doctors are not very good, if the robot was correct.
Actually, it's closer to: real doctors with the assistance of richly informed AIs will be a lot better at diagnosis.
Google is already as good at diagnosis as our local doctors for most ordinary ailments. We only have to go to the doctor to get a piece of paper that allows us to buy the drugs we need.
Doctors are just another gatekeeper about to be made obsolete by technology.
You mean a computer can store much more data and crunch through it easier? There is nothing sad about that. That's pretty obvious really. Medical diagnosis is one of the key examples of AI from academia.
My condition is unknown by many oncologists. Forget about general practitioners. Individual conditions may be rare but there are enough of them that they impact a lot of people collectively.
Not everyone is House. Sometimes you have to know when to punt it to a specialist, or even a specialists specialist, or to Mayo.
A machine can serve as a handy specialist, or at least an automated way of having the patient diagnose themselves.
A Pirate and a Puritan look the same on a balance sheet.
Google is already as good at diagnosis as our local doctors for most ordinary ailments. We only have to go to the doctor to get a piece of paper that allows us to buy the drugs we need.
Doctors are just another gatekeeper about to be made obsolete by technology.
I would argue that as Dr. Google becomes more prevalent, the gatekeeper function of the physician (at least with regard to antibiotics) becomes increasingly important. Antimicrobial stewardship is incredibly important and one of the most important things a GP can do is refuse to write for antibiotics and/or prescribe narrowly whenever possible. It won't improve patient satisfaction, and it can take time to explain why antibacterials won't treat a URI, but giving you the antibacterial you 'know' you 'need' is actually neither good for you nor your neighbors.
I needed FIVE medical opinions to get to the correct diagnosis of my condition, and I had to go to Mexico to get it.
Thank God that the doctor in Mexico correctly diagnosed my lung cancer, because FOUR American doctors kept on putting me off, two of them accusing me of being psychosomatic and telling me "nothing is wrong with me," and the other two incorrectly diagnosing me of, get this, allergies and asthma, respectively... because coughing up blood and having a constant metallic taste in the mouth are definitely characterizing symptoms of allergies and asthma.
The good news is that I have civil suits against four well-insured American doctors and an insurance company now that will guarantee my family's financial well-being once I am gone, which will be in 3-9 months from now, especially given that, had my cancer been correctly diagnosed 3 years ago, it would have been operable and I likely would have had a good chance of survival.
be careful. just because two doctors conclude the same or similar doesn't make it right. doctors frequently just back each other up following your chart notes in terms of diagnosis and are afraid to change diagnosis. you may not have a condition at all and still receive multiple false diagnosis. it's a really big problem if say you need medical care and you're in a prison or state hospital and you need care but they deny diagnosis; and if you don't need care for mental illness but they frame you to look mentally ill each doctor you see merely follows the previous doctors notes very rarely deviating or checking any of the facts. in the community doctors are usually closely associated and cook the books for insurance companies, cohorts and themselves to avoid liability and lawsuits. it can be hard to even find a doctor willing to do consults over another doctor who you accuse of abuse.
https://www.obamasweapon.com/
"I used to work at a hospital and I think I can explain the basis for this. Doctors are clueless, refuse to use technology, they rush everything, and they're incredibly lazy. I don't mean some of them either. 100% of them."
Right....'Used to work at a hospital,' now more clueful than 100% of doctors, by his own estimate. With rippling abs I'll bet, too. Thank god we can all bask in this magnificent being's radiance! "My name is slashmydots, King of Kings; Look on my Works, ye Mighty, and despair!"
What an ass.
You're ugly too!
This sounds an awful lot like a Bayesian Trap, also called a Base Rate Bias. See here for a decent explanation.
https://www.youtube.com/watch?...
weylin
67.5% Slashdot Pure I guess I need to work on that....