In This Doctor’s Office, a Physical Exam Like No Other | Hacker News Reader

In This Doctor’s Office, a Physical Exam Like No Other

nytimes.com

26 pointsswampthing15 comments

I'm really conflicted. I don't think this approach will necessarily mesh with how we traditionally treat medicine. There are many cases where we consider something a disease, and then we find people who live with it with no issues for decades. What if, even though those people technically have diabetes, they're still fine? I guess you have to define what "fine" even is. How much disease can someone have before they're sick? How good is someone supposed to feel?

Thinking about this more... I don't think this was a well written article. They could have gone much more into the larger issues.

It would be nice of general practitioners had better tools though. It should be trivial for them to see which drug will work best based on someone's genes, for example. And the way we measure blood pressure these days is stupid. People's blood pressure goes up just because they're in the doctor's office. Giving people a glucose monitor, heart rate tracker, and blood pressure monitor to wear for a week should probably be standard.

From an outside perspective:

Imo as a society we should reframe the question from 'how to we cure people?' to 'how do we make people's quality/quantity of life better?' It feels somewhat as if there is a gradual creep of what qualifies as a 'disease' and is therefore worthy of treatment.

It also seems like there is an archaic adverseness to non diagnostic related measurement and data. The argument e.g. if we detect something that might be a tumor in a full body scan we have to instigate a series of painful and probably unnecessary biopsies. Imo (again) this represents an overly simplistic view of what the appropriate treatment should be. To say nothing of the dearth of information we have about the day-to-day and hour-to-hour variation in biomarkers in both the healthy and the sick and all the good that information might be used for.

A lot of the time people have some unexplained trouble. They are tired, have pains or repeating infections or what you can imagine. They can be "fine" mostly and never find explanations during their life. They might not even try. It is tedious and just costs so much for an uncertain outcome. But for some, maybe their life quality really could improve relatively easily, if only the problem was correctly identified.

> Giving people a [...] blood pressure monitor to wear for a week should probably be standard.

Wait, isn't that already standard? My English NHS GP gave me a blood pressure monitor for a week. The practice nurse taught me how to use it correctly, and told me why I was taking at least three readings a day. I paid a small deposit for the machine - £10, waivable if I couldn't afford it.

I'm surprised this isn't the case everywhere.

Personalized medicine is barely getting started. We only have data on a few drugs to determine safety, efficacy, and optimal dosage based on genotype. Much more research is needed. Practitioner tools aren't the major obstacle here.

Here's the main point from the abstract:

We discovered more than 67 clinically actionable health discoveries and identified multiple molecular pathways associated with metabolic, cardiovascular and oncologic pathophysiology. We developed prediction models for insulin resistance by using omics measurements, illustrating their potential to replace burdensome tests. Finally, study participation led the majority of participants to implement diet and exercise changes. Altogether, we conclude that deep longitudinal profiling can lead to actionable health discoveries and provide relevant information for precision health.

While this is an interesting study, I agree with the critical. "Clinically actionable health discoveries" does not mean that acting on those discoveries actually prolonged life or improved quality of life. Given the economic incentives in US medicine this has the huge potential for over-treatment.

That's true, but I think there is a lot of open space in the idea of evaluating "does discovering this health issue actually help people". Discovering prostate cancer often doesn't help because most prostate cancer grows so slowly you'll die of old age before it matters - but then some people have faster growing kind.

But I'm following my own health. I have a condition which can be treated well with simple cheap pills. If I hadn't known about it in my family, I wouldn't have pushed for blood tests early in my life. For most people that blood test wouldn't have mattered.

For several years now I've wondered why we don't have "Health Closets" (terrible name, I'm sorry) in our homes.

An HC would be a small room dedicated to measuring all the things that can serve as indicators of health. I'd love to have a room I could step into that would take high res photos of my skin to send to my dermatologist, analyze various samples of fluids, maybe take retinal photographs, etc.

With the advent of ML driven diagnosis in medicine there is the potential that if we collect a huge amount of data it can be processed automatically. In addition data could be aggregated for population wide analysis.

While this could be correlated with my full genetic profile, I'm much more interested in how regular measurement with lots of low cost sensors could impact healthcare.

Because over-testing leads to over-diagnosis, which leads to over-treatment, which doesn't lead to longer or better life and often leads to reduced quality of life.

I find this argument hilarious because it is literally admitting widespread iatrogenic harm, and the solution is not to do less harm, but to keep information away from the patient. How else would like to prove that the AMA is a regressive guild with state protection and financial support through the insurance industry? The only reason many doctors don’t like diagnostics is that it undermines their witchcraft, similar to how second-rate mechanics hate consumer OBD-2 port readers, while first-rate mechanics love them. The argument for less diagnostics is laughable. Diagnostics do not cause harm. The harm is being caused by doctors.

Because there aren’t any systematic perverse incentives baked into US healthcare or anything.

Like, say, many if not most preventive measures being unreimbursible while treatment (especially Stone Age chemo protocols) are reimbursed at profitable rates.

Nope, it’s those fucking physicians greedily trying to gut medicine. Like so:

https://www.nytimes.com/2018/10/26/health/private-equity-der...

Oh, whoops, my bad, it’s the usual malevolent actors and their stooges.

Yes there are shitty doctors (I’ve worked with plenty). However, I would propose that the majority of the rot in the US system is structural. Socialized or properly regulated private systems do not have these scale of problems.

Last but not least, I’m guessing you haven’t added up the cost of running a Snyderome per person. It’s not sustainable on a large scale at current prices. That is why cruder methods of risk stratification that cost less are the norm.

Larry Niven's known space series predicated this 50+ years ago. You'd get up every morning, put your sleeve in for a diagnosis first thing, it would treat you, tell you if there was a problem.

Data collections is only a fraction of what makes good medicine.

if it's a fraction, say 1/5, then what fraction would data collection + data presentation be?

1.1/5

Medicine is mostly about applying the right algorithm to an ill defined problem. The day you can read a problem text and have code autogenerated to select the right algorithm to solve it is the day you can start automating the valuable parts of medicine.

Reply on news.ycombinator.com