undefined | Better HN

0 pointsjklinger4101y ago0 comments

I don't mind point one, and it can be improved in many ways. AI being a huge one.

Point two, though, is just an example of how preventive medicine is forgone due to its cost. The system doctor's use to share data could also stand massive improvements.

0 comments

3 comments · 2 top-level

reaperman1y ago· 1 in thread

If you think doctors can be overly stubborn to change once they've made an initial diagnosis .... you're going to be very unhappy with an AI diagnosis model. As one of many tools, I can see some value. But I'd hate if there was no way to at least try to convince an actual doctor that my issue might be something less expected.

Auracle1y ago

One of the first things I did when GPT 3.5 came out was to give it some medical symptoms I had when I was a teenager that was wildly diagnosed by my primary care doctor that I finally (at least partially) figured out by doing my own research 10 years later.

It nailed what tests I should have done one the first try, one of which I still need to get done because my current primary care doctor doesn’t know how to run it and an endocrinologist straight up refused in the past. However, one of the other two tests suggested would have led to my quality of life improving to such a great extent I would be living a completely different life right now.

I had really, really bad fatigue and we had noticed my shoe size was shrinking as well. The doctor, who is now president of the clinic itself, diagnosed it as depression. The shoe size thing was explained as my arches maybe getting taller. As it turned out, I also lost 2 inches of height as well but I don’t know if we had noticed that at that point yet, but I’m willing to bet it was being charted.

nradov1y ago

Regardless of how good AI clinical decision support models are, they're still garbage in / garbage out. The hard part of diagnosis is gathering all of the signs and symptoms that might potentially be relevant and AI can't help with that.

Cost is only one of several factors for forgoing preventive medicine. Many preventive services haven't been proven effective, or carry a risk of iatrogenic harm. And the healthcare system overall only has a limited capacity which is difficult to increase for a variety of complex reasons, so delivering preventive care to one patient may reducing access for other patients who already have medical conditions.

There are a limited set of preventive care services which most US health plans must cover at no cost to the patient because they have been proven to be net beneficial.

https://www.healthcare.gov/coverage/preventive-care-benefits...

If you think that more preventive medicine should be covered then you can submit public comments to the task force responsible for writing the requirements.

https://www.uspreventiveservicestaskforce.org/

j / k navigate · click thread line to collapse