Hacker News Reader

Comment by 0xWTF | Hacker News Reader

The problem with lab developed tests (LDTs) is they aren't very regulated, and in a malpractice situation, a physician could be liable if they made diagnosis or treatment decisions based on unverifiable LDTs.

It may well be that many LDTs are quite reproducible, but there's now oversight on that at the moment. You could set up a lab in your garage and start billing people out-of-pocket for the results of a Ouija board.

If you want some policy opinions from the College of American Pathologists, feel free to lodge them here: https://www.cap.org/advocacy/laboratory-oversight-and-regula...

View on HN

> in a malpractice situation, a physician could be liable if they made diagnosis or treatment decisions based on unverifiable LDTs.

That seems to be a senseless concern, since offered the evidence of a lab test, a physician should be prompted to investigate further. No physician (nor individual) should ever be making a diagnosis or treatment decision based on a single line of evidence.

More importantly, such results can help offer hard evidence to pressure one's physician to investigate a particular complaint, which can be difficult to convey or to convince the physician it's real. Physicians are often dismissive of things that they do not understand, or of certain classes of patients. Consequently many patients are forced to figure out the source of the problem for themselves.

With a lab based test result, you can go to a physician and say, "Hey, this isn't just my imagination nor some kind of over-reaction. It's tangible and externally verifiable."

And for those without a GP, waitlisted like myself (5+ years waiting here in Canada!), such tests can help one navigate our piecemeal medical system more effectively.

> No physician (nor individual) should ever be making a diagnosis or treatment decision based on a single line of evidence.

Talking in absolutes like this doesn't help the rest of your point. There are plenty of situations where a single result is sufficient to trigger a whirlwind of treatment decisions.

> More importantly, such results can help offer hard evidence to pressure one's physician to investigate a particular complaint, which can be difficult to convey or to convince the physician it's real. Physicians are often dismissive of things that they do not understand, or of certain classes of patients

If you've ever been sitting on the other side of a consult you would recognize that this statement isn't all that accurate of a characterization. As much as people like to believe they are outsmarting navel-gazing doctors, the reality is that with sufficient time and adequate resources most of these "complex issues" can be diagnosed and treated. The problem lies in that most providers have a 15 minute window to determine if the patient is part of the boring 80%, or if the patient's problem will require time you don't have. Until provider shortages are solved, that will be how it works in order to do the most good on the whole. Berating and pressuring them will just contribute to the burnout that in turn contributes to shortages. The source of the problem is above them.

> That seems to be a senseless concern, since offered the evidence of a lab test, a physician should be prompted to investigate further. No physician (nor individual) should ever be making a diagnosis or treatment decision based on a single line of evidence.

You talk like an engineer who only has one product/project/patient to work on for weeks/months/years. A doctor has to see 40 patients a day, and the next day there are a completely new set of 40 patients etc. To investigate further the quality of a lab test result isn't something a physician has much time to do unfortunately.

It sounds like we need to force the AMA to create more doctor openings and drastically increase the kind of care offered.

For me personally, there should be a substantial difference between a surgeon and a diagnostician. Someone who's only job it is to look at evidence and lab results and walk through what problems it could be. This is a wholly different job than actually doing any of the procedures and would attract new people.

Moreover, I know many doctors who are skilled surgeons but terrible at diagnosis.

Take my own wife, who, up until we figured out what worked for her and found a doctor willing to do it, had miscarriage after miscarriage (seven total), and the fix was fairly straightforward, but it took forever to get there, and most of the obstetricians and MFMs we saw (great surgeons, themselves), were not actually very knowledgeable on the latest research. Nor should they be. The intricacies of various blood disorderse, genetic recombinations, etc, has so little to do with their practice of gynecological surgery that it's crazy we confuse the two. At the end of the day it took my wife and I (both coming from scientific backgrounds and familiar with academia and statistics) took matters into our own hands, got familiar with the data, and suggested appropriate treatments (which have worked several times now in a row, so clearly something changed, unless probability is really trying to screw with us).

Yes, there is a big difference between surgeons versus doctors which are good at diagnosing. It's two completely different skills.

I think what is missing is also something in between nurses and doctors, who can take more formal responsibilities than a nurse and offload the doctors.

And there should be a track to directly become a surgeon instead of first become a specialist doctor and then do surgery education after that, those roles should be different. The education for doctors is overall just too long for little benefit.

They don't need to investigate the test quality, only to run whatever tests they trust based on the indications provided.

Exactly, the unofficial test is basically a check engine light that you need to get checked out. The doctor is like a mechanic and should then run additional diagnostics to confirm any issues.

Of course, right? And then the article quotes a doctor proud to "put “zero stock” in the tests because she cannot verify their accuracy". I guess she prefers to have her patients go through the massive hassle of finding another physician.

She probably would just repeat the tests if indicated with a laboratory that's trustworthy.

The problem isn't that there's isn't a lab with a trustworthy result. The LDT world can make any claim in the space of biologically feasible. No lab has the resources to set up every conceivable test in biology and validate it!

But isn't this an opportunity to have a "pre-test" and the doctor can say "ok, you had this result from an unverified lab, let's do a more thorough test and see what we find".

The public need to be educated that these tests are a lower cost "pre-test".

In Australia, we are given an FIT test for colon cancer when we turn 50. This test is 79% accurate. This is a simple at home test, no doctors involved, just mail in your sample, get a result, and then if the result is positive, go to a doctor for follow-up.

> But isn't this an opportunity to have a "pre-test

We call these screening tests.

> doctor can say "ok, you had this result from an unverified lab, let's do a more thorough test and see what we find"

The more likely scenario is that this is the 10th patient this week presenting for suspected cancer based on an at home administration kit with a low specificity (under ideal conditions), and it turns out it's not. Conducting such tests indiscriminately (voluntary and self administered) is going to waste a ton of doctor time that we don't have.

> The public need to be educated that these tests are a lower cost "pre-test"

Anything that requires education of the public about anything is unfortunately a losing battle.

> we are given an FIT test for colon cancer when we turn 50. This test is 79% accurate. This is a simple at home test, no doctors involved

This can work with a structured protocol. The problem is that at least in the US, we will have direct to consumer labs chasing next quarter's revenue numbers, so everyone will be shown commercials and be grabbing them off of shelves to avoid seeing expensive doctors that they don't trust. It's a recipe for a lot of unnecessary anxiety and time wasted.

These types of services are incentivised to over test and you may find many spurious results when you over test. Look up the term "incidentaloma". If you have to get a real doctor to interpret and re order these tests, you are contributing to the burden of the system rather than a "pre test". The specific example you mention is apples and oranges, since the FIT test is a specific validated screening test provided by government rather than a profit driven private lab. There is evidence it's not even that helpful as a screening test.

LDTs can only be carried out in a CLIA laboratory which you're unlikely to be able to operate out of your garage. I mean you could but there are a bunch of staffing and accreditation requirements that would make that very difficult and expensive.

Fyi the FDA play to regulate LDTs is being legally challenged, it's all pretty up-in-the-air right now.

Reply on news.ycombinator.com