Hacker News Reader

Comment by mrcwinn | Hacker News Reader

Happy to see this. I've struggled with an injury for the past five years. I've been to multiple sports-focused physicians, had various scans. Responses from doctors have ranged from "everything seems fine, can't really figure this out" to [completely wrong hypothesis]. Tried acupuncture. Tried a chiropractor. I remember one doctor, though, had an interesting thought that seemed to make sense - but I've been so discouraged from so many false starts or misplaced hope, I didn't bother following up.

Finally I typed in my entire history into o3-deep-research and let it rip for a while. It came back with a theory for the injury that matched that one doctor, diagrams of muscle groups and even illustrations of proposed exercises. I'm not out of the woods yet, but I am cautiously optimistic for the first time in a long time.

View on HN

Why would you trust a LLM over a battery of human experts? I find it hard to believe that the doctors never proposed exercises or some kind of physical therapy for you, at least in the US.

I can't speak to the OP's condition, but having seen plenty of doctors and physical therapists in the US for over a decade:

Yes, they propose exercises.

No, they don't work.

For certain (common) conditions, PT seems to have it nailed - the exercises really help. For the others, it's just snake oil. Not backed by much research. The current state of the art is just not good when it comes to chronic pain.

So while I don't know if an LLM can be better than a battery of human experts, I do know that those human experts do not perform well. I'm guessing with the OP's case, that battery of human experts does not lead to a consensus - you just end up with 10 different treatments/diagnoses (and occasionally, one is a lot more common than the other, but it's still wrong).

I've never used LLMs for this, but as someone who's been through a lot of sports-related injuries, I find doctors more or less useless (except for prescribing painkillers and performing surgeries.)

No doctor or physio has ever been able to fix my chronic issues, and I've always had to figure them out myself through lots of self-study and experimentation.

Much of modern medicine's expertise is focused on areas that reimburse well (in some part because those industries subsidize the research).

I would not expect most physicians to have a deep fund of literature-backed knowledge to draw from regarding exercise. Telling someone to do an exercise probably doesn't compensate well.

That said, I'm also pretty negative about the availability of rigorous literature regarding much of nutrition, dentistry, podiatry, physical therapy, etc... you know, the things that affect the health of most human beings that have ever lived.

Sports medicine specialists do this all day every day. They don't hesitate to prescribe specific exercises if they think it will help, and as for compensation they get paid the same for that service as any other regular office visit (typically CPT code 99212 or similar).

Because there is so much variability in individual injuries and physiology it's extremely difficult to do rigorous studies comparing different treatments. Like even something common like a rotator cuff tear isn't one single thing that can always be treated the same way. Patients and practitioners will often have to follow a trial-and-error process until they figure out what works in a particular case. Experienced providers who see a lot of cases eventually develop a lot of tacit knowledge about this in a way that's difficult to codify or explain.

Do you therefore agree that most physicians do not "deep fund of literature-backed knowledge to draw from regarding exercise"?

> Because there is so much variability in individual injuries and physiology it's extremely difficult to do rigorous studies comparing different treatments.

This is generally true of research in all areas of medicine; I don't see it as a fundamental reason why we wouldn't hope for equally powered and rigorous studies focusing on the issues that affect most of the population, even if there are not pharmaceutical or device companies that will profit from the recommended interventions.

I'm only a decade in, but my clinical experiences in my area of expertise vary widely from those of my colleagues, which vary from those of other colleagues, in spite of years of shared experience in a fairly homogenous demographic. Expert opinion is not a level of evidence that carries much weight in my mind.

It may be the best we have in many cases, I just wish we had something better.

It's a lot harder to do clinical research in some areas than others just due to the availability of subjects. Like if you're developing a new drug to treat type-2 diabetes then it's a lot easier and cheaper to identify patients who meet the inclusion criteria than to find patients with some particular type of shoulder injury or whatever. And the study endpoints are also problematic since there are no clear objective signs; usually they have to rely on patient reported symptoms.

Some specialist physicians have a deep knowledge of exercise for injury prevention and rehab. But it also helps to get input from a good physical therapist and/or personal trainer. As a patient it can be tough to discriminate the good ones from the quacks.

I don't think "easier" and "cheaper" are valid barriers to research regarding the most prevalent maladies affecting most people of the species. If there is a low morbidity gold standard for in-vivo diagnosis, it seems like a valid inclusion criteria. If there is no such gold standard available, it's kind of hard to validate a claim of someone having deep expertise.

The endpoints for such research don't have to be problematic; regarding musculoskeletal complaints it's fairly straightforward to measure self-reported pain, self-reported level of function, or more objective measures like working days lost, strength, stamina, range of motion. The more difficult issue is determining an appropriate control group / placebo intervention.

(Also, I'll point out that my initial comment also included nutrition, dentistry, and several other fields under the same umbrella of "not enough good research" -- or to further specify, my complaint is that the burden of disease is disproportionate to the availability of rigorous literature.)

> Some specialist physicians have a deep knowledge of exercise for injury prevention and rehab.

By what measure? As in "they have a lot of clinical experience" -- (the "plural of anecdote")? Or as in they have read a lot of books on the subject? In which case, you seem to be agreeing that the evidence base is weak, so I'm not sure how authoritative this body of literature could be.

> As a patient it can be tough to discriminate the good ones from the quacks.

I imagine they would be virtually indistinguishable as a patient. Based on the context of this conversation, how would you recommend that I, as a physician, tell them apart? (It's hard enough to do with other physicians.)

Moved

Sounds like you needed a good PT not a physician.

Good DPT with a team of PTs and PTAs, and possibly also see a DO who can do osteopathic manipulation therapy (OMT).

Start with a dedicated orthopaedic practice who will make referrals to the above and then order CATs/MRIs if the above isn’t helping and evaluate if surgery is necessary.

One challenge is that it is hard to even know how to navigate all this. LLMs could actually help with that.

You have to be realistic and assume that treating most major musculoskeletal injuries will be a lengthy trial-and-error process. Every injury is unique and imaging studies or other tests can only give limited clues about what's going on. There are patients in terrible pain whose scans look normal, and there are other patients whose scans indicate serious pathology yet they are pain-free and completely functional. Instead of getting discouraged, keep trying different modalities (except for the pseudo-science BS) until something works.

In general a lot of those injuries will eventually heal on their own. So it's easy to fool yourself into believing that a particular treatment was effective even when the real cure was time.

i mean it sounds like that doctor had it right then? i dont think this reads as a win for "online self-diagnosis", ai-empowered or otherwise.

From their story, it sounds like the one doctor had it right because the AI-empowered self diagnosis seems to agree. Or do I misunderstand?

You trusted known grifters acupuncturists and chiropractors, but not doctors. Then you trusted a computer known for telling lies.

I think you should take a step back and re-assess your internal heuristics.

They literally said they saw sport focused physicians.

I hope recent cuts to government science have managed to hit enough of the safetyists and industry captures who keep us from just trying out new healthcare approaches like this and learning. They'd like nothing better than to replace the help you got with "As a large language model, I am unable to offer medical advice."

Reply on news.ycombinator.com