Why Some Doctors Hesitate to Screen Smokers for Lung Cancer (2015) | Hacker News Reader

Why Some Doctors Hesitate to Screen Smokers for Lung Cancer (2015)

npr.org

91 pointsg3ph4z90 comments

I wonder if this is a "dont waste your time" issue coming from management?

As an anecdotal example, Im an engine mechanic who routinely sees older trucks from the 80s and 90s. mostly idler/pitman arm replacements, tires, diff fluid, etc... on these which is expected as many are pushing a million miles or more. I had a 1981 International S1700 limp into my garage one day with a misfire problem due to long, long overdue oil pressure problems. I fixed the oil pump and was getting ready to hook up the diagnostic computer when my boss stopped me and said "if you plug that thing in its going to light up like a christmas tree. You aint telling the driver anything they care about anyway."

I mean the overall incidence of lung cancer is too low for this to be the case, even in smokers. It's around 5-6 percent.

thank you for saying this ! The environmental "externalities" world is exactly like this! Saving generations of [insert-charismatic-species] from extinction in the wild is NOT PROFITABLE and will, in certain cases, be HIDDEN from investors and regulators.

Well, yes, paying for externalities is not profitable. That's the whole reason that people have to be forced to pay for them.

Do share more stories on working on old trucks! I have a '99 Ford with a 7.3, and everyone's saying it's not worth fixing, but I just like keeping it on the road...

What kind of MPG does that get?

22-24 highway, 15 city. It has zero emissions equipment though, so I'm sure it's killing us all

It would depend on what needs fixing? Are we talking engine internals of the stuff that hangs of the engine?

Nothing to my knowledge needs an overhaul yet, nearing 400k miles, and it's just been maintenance and consumables (except the turbo started leaking at 300k miles, which was replaced)

Typically, diesels are built with maintenance in mind (at least for trucks, no idea what you'd find in a diesel car). The cylinders are sleaved and meant to be replaced every couple 100k miles or so, so the cylinders themselves also last a long time (the sleaves take the brunt of the wear). I think theyre typically less complex than a standard gasoline engine. E.g. they have no spark plugs (and thus no distributor or ignition system), instead glow plugs to start the engine. Possibly more complex in other areas, such as emissions control.

I'm no expert on engines, so the the above may not be entirely accurate or current. Most of my knowledge comes from working for a semi truck/diesel engine manufacturer as an IT intern close to two decades ago and touring factories.

That's about right.

I thought the 7.3's were supposed to be super solid? How many miles does it have, what's wrong with it?

My extremely limited knowledge of diesel engines is something like "they go a ton of miles, then need a major overhaul, then go a ton more miles". If you're at some kind of major overhaul milestone, then I could see the response you're getting.

'99-ish Ford kicked off the Super Duty series, but it isn't sufficiently interestingly old to a lot of people, so I can understand their the why-bother attitude.

It's nearing 400k miles. It's had service - new turbo and engine control components, suspension and frontend, brakes, glow plugs, etc. Original everything else though, nothing fancy, just a work truck. But anything "newer", even 10 years old, is $25k minimum for any halfways decent duty truck.

Your 7.3 is gold, that engine will last till the heat death of the universe. And you’re maintaining it. For some reason people think that a well maintained vehicle is a grenade because it has service history, while one being run into the ground is great because they haven’t paid for anything recently.

Newer trucks aren’t more reliable. They’re just newer. You won’t find a truck cheaper to maintain than the one you’ve already got.

(The people who tell you your truck ain’t worth fixing sound like my mother in law who traded in her car ‘cause the windshield wipers broke.)

Please tell me the story about your mother-in-law is an exaggeration

Nope, literal truth. Lovely lady, but some decisions...

FWIW older generations' experience with cars is that they are unreliable money pits after 10+ years. That was true until relatively recently.

> they go a ton of miles, then need a major overhaul, then go a ton more miles

Yup. Not unusual for big rigs to go a million miles between overhauls.

Even for "consumer" diesels, the engine typically outlasts the vehicle around it.

I drove a '97 Dodge which had 300k on the odo, got tired of the poor interior and upgraded to an '06 with 200k. I even threw a mercedes diesel into a 67 chevy pickup, because diesels are great :) https://www.youtube.com/watch?v=yFd1ZL2hgMY

No, the argument is specifically that the harm from following up on false positives outweighs the benefit from finding true positives. While the 2011 study that the Medicare policy change is based on specifically concludes that this will not be the case, the concern is that the study found a lower false positive rate than would be seen in the wild, due to better skill and greater diligence from the radiologists participating in the study.

Harm from false positives is a real thing, especially for a test with a relatively high false positive rate and relatively invasive followup to false positives (they don't state it specifically, but the theater analogy implies 66X more false positives than true)

Furthermore, unlike an auto mechanic who might not be able to bill much for finding problems the owner doesn't want fixed, or following up on a diagnostic computer finding that turns out to not be a problem at all, the healthcare system at large stands to profit from both the screening and the follow up on those false positives, so the bean counting manager shouldn't be applying pressure to skip the screening.

I've watched my family go through smoking, and cancer hasn't been the issue. I don't know how representative it is, but my experience definitely reflects the premise of this article: cancer testing could have caused way more harm than good.

Both my mother and her father were still smoking frequently when they were on oxygen. In other words: When they should have known better. Her mother was looking really bad at 50 and when the Dr told her "it's killing you, and not in some abstract way", she cut it out cold turkey and pretty much instantly looked 10 years younger.

Grandfather was struggling in general, but it was twisted up intestines that did him in. My mom, one nurse told me, her lungs were "just shot". COPD. Funny thing was that she was getting enough oxygen (with O2), she just couldn't get rid of CO2. Basically took her outta the game 20 years early.

Hopefully, my kids who watched this, can use the experience to steer clear of some things. If you can't set a good example, show what a bad example leads to. :-)

This could be the most controversial thing comment I've posted on here, but...: Why are we even spending resources figuring out how to extend the lives of people who have smoked a pack a day, for 30 years, and are currently in old age? Even if we find cancer and save their lives now, will that prevent then from picking up smoking again? Will it prevent the heart disease they're going to encounter ten more years down the line? Dragging it out seems like a massive burden on the rest of society. I'm unsure how these studies are being funded but I just wonder if the money is better spent on helping people to stop smoking.

Those people (or their insurance) are willing to pay for treatment. If the market is there, why not fund some new research into treating lung cancer? Not every incidence of lung cancer is from smoking two packs a day, sometimes it just happens although its more rare. It's the idiots smoking two packs a day that are helping subsidize the research that can help save the person who just gets lung cancer randomly.

I think it's a hard choice between allowing for personal freedoms and improving society for the better by outlawing dangerous drugs like smoking and alcohol. I think there can be a balance by allowing everyone the same level of access to healthcare but increasing taxes on dangerous recreational drugs to subsidize the healthcare system. You can eat fast food and smoke all you want but it's going to cost you heavily financially.

Unlike some other forms of cancer, random lung cancer is exceedingly rare. The vast majority of lung cancer patients were smokers.

As a society we have limited resources for healthcare. So it's worth considering whether we might achieve better net results by shifting some resources from lung cancer treatment to smoking cessation programs.

I'm not sure what you mean by "exceedingly," but:

> Overall, 10 percent to 15 percent of lung cancers occur in non-smokers. (Another 50 percent occur in former smokers.)

> Two-thirds of the non-smokers who get lung cancer are women, and 20 percent of lung cancers in women occur in individuals who have never smoked. This percentage is significantly higher in Asian women.

https://www.verywellhealth.com/lung-cancer-in-non-smokers-22...

As for the smokers, they generally pay higher health insurance premiums so at least they're contributing their share of resources.

Lung cancer is still cancer, and while different types of cancers vary in many ways, they are also similar in many ways. Studies into the detection and treatment of lung cancer can (and have) contributed to the detection and treatment of other types of cancers.

From a practical perspective, setting aside issues like blame and responsibility and the dollar value of one life vs another, studying all types of disease increases our knowledge of human biology, and our knowledge of disease in general.

Keytruda is one example of a therapy that was developed to treat lung cancer (and melanoma), but now looks like it could be effective against other solid tumors that fit a genetic profile.

This is not to argue against smoking cessation (or more importantly, prevention) programs. Just that there are reasons to pursue lung cancer treatment research that go beyond the lungs entirely.

Just to give you a data point. I grew up in a family that smoked in the house. I've never smoked myself.

I hope you aren't suggesting that I should be condemned for the sins of my parents if I should one day get lung cancer.

Also keep in mind that pollution is going to be a large (if not the largest) contributor to lung cancer in the future.

Please read my comment again. I in no way suggested you be condemned. What I actually suggested is that we do an economic analysis to determine how to deploy our limited healthcare resources in order to maximize net benefits.

>>Why are we even spending resources figuring out how to extend the lives of people who have smoked a pack a day, for 30 years, and are currently in old age?

Slippery slope I guess (Plus it is immoral). Next somebody is gonna say why should certain kind of people get medical treatment instead of this other people who clearly deserve it more.

There's nothing inherently immoral about rationing healthcare. In fact healthcare is already rationed everywhere for most patients. In the US we mostly ration based on wealth, age, and employment status. But that doesn't necessarily mean other approaches are less moral.

But I think we should ration care based on effectiveness of the treatment not moral deservingness.

> There's nothing inherently immoral about rationing healthcare.

That a subjective statement, it is only true for you and those who agree with you.

>In fact healthcare is already rationed everywhere for most patients.

That is not an argument for the morality of rationed healthcare.

We could make some sort of objective argument based upon ethics though. Suppose the smoker became addicted to cigarettes as a by-protect of being drafted into the military where he served in a distinguished role perhaps saving lives and protecting your freedom. Being a draftee and not particularly well educated, this persons means in retirement were quite limited, often depending on government assistance in various forms.

A different person was ensured to be wealthy by inheritance then draft dodging then engaging in tax evasion, illegal migrant labor, various kinds of fraud and strong arm tactics. Yet this person didn't smoke and is quite wealthy.

Should we invest in healthcare for tobacco users?

Everything is rationed, because the Earth has finite resources. Healthcare is no different. Morality has nothing to do with it.

It is not possible to provide everyone with the best health care possible. For example, only a handful can get the services of the best heart surgeon.

Compromise is inevitable.

One of the truisms of medicine is that ultimately every disease and every treatment has a 100% mortality, so we can never truly save a life. What we can do is promote the quality and quantity of life. In the US, the rule of thumb is that it's worth spending $50,000 to keep someone alive for a year, with that being the approximate cost of dialysis therapy. So, to save 10 years of life would be expected to be worth spending a half a million bucks. Which is all to say that an additional 10 years of reasonable quality of life is worth a lot, both to society and to most individuals. The other strain in your comment suggests that people who are presumably in some way responsible for their disease are less worthy of care. This opens up a giant slippery slope. Start with drug use, smoking, alcohol use, being overweight, eating meat, drinking coffee, working too many hours a week... Ultimately only celibate teetotalling vegans are worthy of medical care.

>The other strain in your comment suggests that people who are presumably in some way responsible for their disease are less worthy of care

I am stating that people who are presumably in some way responsible for their disease are not MORE worthy of care. It is only these people who are receiving free screenings - as per the article.

Where do you draw that line? - I mean no person who understand the addictive power of nicotine would make this kind of statement.

Most diseases have some human choices involved. Living in a city with bad smoke, not running every day, poor diet, or drinking.

Maybe you're perfect and run every day, living in the country side, only eat fruits and vegetable and have never had a cigarette or beer your entire life.

But most of us aren't perfect. And honestly if you are the perfect specimen you are going to cost society far more because everyone dies of something. And slowly dying of alzheimer's disease in your late 90's is way more expensive because you'll live in in a nursing home with full time care for your last ten year and you've been using medicare and social security for 30 years.

Dying in your 50's from lung cancer is it's own punishment.

I think you're oversimplifying the situation. Alzheimers could be caused by human choice and 40 years of human life has more than monetary value.

As does the life of a smoker.

And cardiovascular disease and exercise definitely seem to influence Alzheimers. Nigerians for instances have high rates of APOE4 but very low rates of Alzheimers and I think the reason is they also have very low levels of cholesterol.

Consider passive smokers - not guilty of anything, they just couldn’t avoid it. Like kids in the smoking families...

For the same reason we attempt to rehabilitate other types of drug addicts; they're still people with families and friends who want them alive for as long as possible. There are many societal constructs which are intended to help people who made poor choices.

I think it's a legitimate question though. I started smoking when I was 9, pack a day by 16.

> Why are we even spending resources figuring out how to extend the lives of people who have smoked a pack a day, for 30 years, and are currently in old age?

Are those people really that old?

Imagine if you started smoking at 13 (which is pretty common). If you smoked for 30 years, now you're 43. If you potentially live to ~80, you've barely passed the half way mark.

Not only that but someone at 43 is likely still working and paying taxes, so they are putting money into the system and could potentially be doing that for another 20ish years.

The benefit is only for those from the ages of 55 through 77 per the second sentence of the article.

Ah ok. I'm more of a "comment first, read later" type of guy, so I missed that!

As this was where you made you question, I'll stick with you here in the ruthlessly utilitarian minefield (i.e. people and $$$'s are fungible).

First, a clarification, it's 30 pack-years. There are lots of ways to get to that number. 30 years @ 1 pack/day and 15 years @ 2 packs a day are both 30 pack-years. That number sticks with you even if you quit 20 years ago.

So, especially for the younger end of that range (55yo), people between 55-65 are often in their peak earning years (finally gotten seniority or management) and this is when they are most often finally paying off their debts (mortgage, etc). It's also when they are likely doing peak elder care and also sending their kids to college. So this is the time where they are probably doing their peak economic and social good. If you can catch something that keeps them productive for an additional 10ish years, you've probably come out ahead for it.

False positives will be much more rare in practice with the new guidelines used (Lung RADS) which cut false positives by 50% or more while mostly preserving sensitivity.

That doctor should really update his diagram to reflect the current standard of care for lung cancer screening. Using the false positive rate from the 2011 study is intellectually dishonest!

In addition the amount of lung cancer deaths prevented was doubled (!!) in a recent European study by tracking patients for 5 years instead of 3 years.

Honestly curious: are most radiologists actually following guidelines like Lung RADS in practice, especially just after publication? Is it part of any standardized CME?

It's my impression that there's often a big gap between ACR best practices and what's happening on the ground in most places, which was a key point the doctor in the article was making.

Adoption of Lung-Rads is essentially required to be designated as an ACR screening center. To be reimbursed for Lung Cancer Screening, you must provide registry data to CMS (the registry is run by the ACR).

It's not codified into Federal Law, but if reimbursement requires a Lung-Rads code, then it's more or less mandated.

I'd argue out of all the fields of medicine, Radiology is pretty quick to adopt new technology or reporting schemes "on the ground". See the proliferation of Lung-Rads, TI-Rads, Cad-Rads, LI-Rads, PI-Rads, etc.

That's good to know, thank you for the response!

yeah, the cost benefit ratio of scans probably would improve significantly w/ less misinterpretation of scans...radiologists tend to notoriously over-call findings for liability reasons, and most MD's don't look at the images, they just read what the radiologist wrote.

This sounds familiar. I have watched now quite a few people go through serious diseases. One pattern I saw was tests, tests and more tests. A lot of them very expensive, time consuming and painful. But when you ask what they will do with the test results there often is silence. To me it looks like they are doing something because “we need to do something”.

Reminds me a little of data collection practices at companies. Sucking up more data feels like you are doing something. But using the data is much more difficult and often doesn’t happen.

Cancer runs pretty deeply in my family, and I had read the same about breast cancer screenings years ago, re: false positives causing more harm than good, so it'd make sense to see the same trends for other cancers as well.

As far as breast cancer screening goes, the recommendations are for the general population; woman with a family history or other specific risk factors may benefit much more from screening than the general population and/or have different intervals than the general population. This is why communication with physicians is so important.

As a patient (or a patient advocate for a loved-one), anytime a test is proposed it is important to ask what actionable information can come out of it. Basically look 2 steps ahead.

Some tests will help confirm a course of treatment, or propose a new one. But what if that treatment is something you've already decided you won't do? For example in an older patient too old for a surgery or a baby that you've already decided you'll keep.

But you are right, some tests are ordered out of a desire to show that "something is being done". It's important to talk with the doctor about the cost (not just money but pain & recovery) and benefit. Everything is a tradeoff.

"Some tests will help confirm a course of treatment, or propose a new one. But what if that treatment is something you've already decided you won't do? For example in an older patient too old for a surgery or a baby that you've already decided you'll keep.

I had the impression that a lot of doctors go through the same test protocols independent of patient situation. They are more deliberate with the actual treatment but there seems less thought about the necessity of testing.

It also varies by doctor. Some like to order more tests than their peers. That's why it's important to have a doctor with whom you can openly discuss these tradeoffs. The vast majority are open to that, but some will give you the stink eye if they interpret that as you questionning their algorithm.

> But when you ask what they will do with the test results there often is silence.

With chronic and serious conditions, like in oncology, some tests are pretty much the only way to see if there's any progress with the treatment and how the general health levels of the patient are (bloodtests).

> That's because some cancers grow slowly and never become dangerous...These false-positive tests led to more follow-up testing, including risky procedures like a biopsy, which inserts a needle into the lung.

> "Not surprisingly," Welch says, "sometimes that creates problems like causing someone's lung to collapse."

These outcomes should inform whether or not the biopsy needs to take place or whether they should instead follow up with further screens to monitor whether the cancer progresses. Of course, the major downside of my suggestion is that patients are extremely reluctant to hear "cancer" without a plan to rapidly classify it as malignant or benign.

Classic base rate problem, the rate of lung cancer rate is low compared to the false positive rate. It seems that one option is to follow up with a second scan 3-6 months latter and see if the tumor has increased in size. However there probably serious liability issues with this approach even if it means you're still catching the cancer earlier on average compared to waiting for physical symptoms. I'm curious how counties like France, where smoking rates are high yet have high life expectancies, deal with screening for lung cancer currently.

Also, detection only really matters if the treatments are effective.

If you can detect Stage 4 lung cancer one week earlier, it probably does very little to improve the quality of life.

Anecdotally, as an ex-smoker who quit after my mother died from lung cancer, I spoke to my doctor about screening. He did a lot of dismissing and hand waving about it. It upset me greatly at the time. This article explains it all well enough. Doctor/patient communication is very important.

If you don't feel like you doctor is addressing your concerns, seek a second opinion. You are your own best patient advocate.

How long did you smoke?

My own anecdote:

I was athletic when I was younger, but I was a regular smoker through most of my twenties.

I quit a few years back. My doctor would check my breathing, etc, with everything I'd told him about my habit. He gave me the surprising news that my lungs would be virtually back to virgin health in 4-5 years (after nearly 10 years of smoking). And to get back to exercising more.

I smoked for 7 years, the quit, then picked it back up 4 years later for 5 more years. I didn't really meet the criteria outlined in this article of x "pack years and over 55". I have read that quitting before 40 reduces your chances of lung cancer to that of the general population. Honestly within two years it was stark how much a difference there was.

VOMIT, Victim Of Modern Imaging Technology, is a term among doctors for a reason.

Generally the more you know the better, but sometimes things can be stumbled upon that might otherwise be irrelevant yet the intervention can then introduce new risk or some other unique problem or undesired side effect. There aren't always easy answers.

> "but [can also] find cancers that were never going to matter."

I'm not sure, but isn't better to know that you have it and then investigate further, do more tests, perhaps do some period follow-up with a shorter interval?

Otherwise it's just betting: I bet that I don't have a lung cancer that will be agressive....

"investigate further" is the sticking point.

You're swimming in the ocean and notice something beneath you. Is it a friendly dolphin or a dangerous shark? Do you hold your breath and do a deep dive to figure out which it is? Keep in mind the deeper you go the greater the chance that you'll drown.

It's a strained analogy but I think it gets at the problem. What if you dive a little bit deep - some minor test - and still aren't sure? Go deeper - a more invasive dangerous test - or go back to the surface?

And what if you dive and are pretty sure it's a shark? Swimming all the way to shore (i.e. cancer treatment) is very risky but the shark will almost certainly kill you. How sure are you, really?

“You're swimming in the ocean and notice something beneath you. Is it a friendly dolphin or a dangerous shark? Do you hold your breath and do a deep dive to figure out which it is? Keep in mind the deeper you go the greater the chance that you'll drown.”

Does it really make a difference what it is? Will you do anything differently?

I ask this question about metrics constantly. "If you had this information a year ago, what decisions would have been different?"

It usually ends in dropping the request for another metric

Absolutely. So called "friendly" dolphins are often very sexually aggressive. With knowledge of what aquatic beast lurked in the depths below, I could tailor my preparedness to either be on guard against unwelcome mammalian advances, or remain blithely unconcerned over some incurious, nimrod fish.

I understand your point, but let's dispel the myth of dangerous sharks. The vast majority of shark species never attack humans at all and it's totally safe to swim near them.

* Choice A: X% chance of aggressive cancer, which worsens as it is left untreated.

* Choice B: X% chance of finding and treating aggressive cancer. Y% chance of false positive leading to significant side effects.

If Y is zero, then the choice to do further testing is obvious. If Y is large, then it depends on how large the side effects are, and what the chance of running into them is. You have risk in either case, and you will never reduce risk to zero. The question is how much risk you have, and what cost there is to mitigate that risk.

Nope. When it comes to tumors, doctors always play 'better safe than sorry', but the treatments available for cancer are so detrimental that it's going to do way more harm than good to receive treatment for something that was benign.

A little bit of information can be a dangerous thing.

The issue is that testing often comes with its own health costs and many tests return a fairly high rate of false positives, which in turn leads to surgeries that can take a high toll on the body.

Related to this, I recall reading about a study of elderly men with slow growing prostate cancer. Having surgery to remove the cancer did decrease the chance of dying from prostate cancer, but actually reduced their overall life expectancy. Prostate cancer surgery is hard on the body and an older person with a slow moving cancer has a pretty good chance of dying from some other cause before the cancer gets them.

I believe this is indicative of a frustrating phenomenon in the U.S. healthcare/legal system where many doctors are prone to being overly cautious due to the threat of malpractice litigation. You rarely hear of doctors being sued for unnecessary testing. They are much more likely to be sued for missing a cancer. This happens even when it was reasonable to miss a cancer. A doctor who is overly concerned about litigation can thus cause harm through the patient stress and physical effects of unnecessary followup testing when it is not justified. Rather than sticking our heads in the sand by not screening because we are worried that some doctors will have a harmful false positive rate, we should be addressing the root causes of the high false positive rate.

I periodically smoke, yet I don't call myself a smoker. Yet when I buy insurance I buy the smokers plan, the idea being that if I do need to use the insurance I expect the Insurance company to use any information necessary to deny my claim. Should I be worried about this?

Have fun getting any physical activity injury claim denied. Smokers aren't expected to do sports.

What does '/s' indicate?

A great and seemingly minimally biased resource for this type of preventative care(mentioned in the article):

https://www.uspreventiveservicestaskforce.org/BrowseRec/Inde...

It presents an evidence based pros v cons listing on screening and tries to present a informed approach, with less financial bias. The most toxic component of this entire issues also has to do with a form of defensive medicine that physicians engage in. Basically if you don't want problems don't look for them, not to say all physicians engage in this type of behavior. This mentality is very common amongst the older generations of physicians who fear the repercussion associated with minor mistakes.. Ie. doctor sends patient for CT suspicious finding are noted, doctor forgets to order future followup or some other variety of mistake... Patient ends up having cancer and the physician is slapped with a malpractice suit.

>Lazris says he shows the theater diagrams to many of his patients and gets a wide range of responses. Some patients, he says, point at one of the three blackened seats and say, " 'That's probably me. I'm not taking any chances, I'm getting this test.' "

>"Other people," he says, "will see [the same diagram] and will say, 'Are you kidding me? I'm not going for that; that's not worth it.'"

>But in either scenario, Lazris says, he has done his job — he has helped his patients understand the odds and then let them make the choice.

This is wonderful!! Good on him!!

This is how medicine should always be practiced. Much too often a doctor/PA just orders a bunch of tests with ZERO discussion (other than maybe "we'll screen you for XYZ"). No discussion of risks. No discussion of the upsides of testing. No discussion of individual risk factors. No discussion about how the results will be actionable.

Sometimes they order tests before even meeting a patient for the first time or even telling you what they are testing you for. My last primary care doctor did just that, I made an initial appointment with him for a routine physical (I need referrals for my health insurance) and someone from his office calls before the appointment and says "Dr So and So has ordered some blood tests for you, please go to the lab and get then done before the appointment." Didn't think telling me any other information was important, like, uh, what I'm being tested for and why.

Of course, at least some of the tests he ordered were not evidence based.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5638475/

On top of that, I have untreatable anemia due to a genetic condition. His first tests found the anemia and he ordered follow up tests. Mind you, this is before I ever met him or even spoke to him. If he had a conversation with me, the followup tests would have been completely and totally unnecessary.

A different doctor did that to my husband recently as well ("we've ordered some tests for you" before even meeting him nor telling him which tests were ordered.) I looked at the lab work afterwards and turns out PSA was one of the tests they ordered, which has shown to cause more harm than good. ZERO discussion about risks vs benefits. No discussion of benefits of testing at your age vs an advanced age. No theater diagrams.

https://www.medicalnewstoday.com/articles/260087.php

https://www.uspreventiveservicestaskforce.org/Page/Document/...

I read a story about an elderly woman in good health with no symptoms who went to the doctor for a routine physical. The doctor ordered a bunch of blood tests and a urine test, turns out she had an asymptomatic urinary track infection. She was given ciprofloxacin and had a bad reaction to it.

Turns out:

A quarter of elderly women have asymptomatic urinary track infections. Detecting and treating them have not shown to have any benefit.

Ciprofloxacin is an inappropriate treatment for uncomplicated urinary track infections in the elderly due to the high risks of serious side effects.

https://www.nytimes.com/2019/03/15/health/antibiotics-elderl...

The story doesn't say if the doctor discussed benefits and risks of doing those blood and urine tests, but I'm doubting they did (considering both the testing and the treatment was against recommendations).

I'm not, as a rule, against screening tests. I'm against doctors not explaining the risks vs benefits of screening tests. I'm also against doctors ordering screening tests that have shown to have no benefit.

Start calling them Physicians.

It's important to break up the medical field by service.

Physicians are drug dealers and the literal key to getting insurance paid for.

Doctors of Physical Therapy do not perscribe medication, but are the experts in muscle and skeletal matters.

Primary Care Physicians are this weird beaurcratic step in our healthcare process.

Didn't we see the exact same thing a decade ago with mammograms and breast cancer?

I remember there being two issues, one being computer aided detection in mammograms at the time was terrible for false positives compared to an experienced doctor (the the computer-aided was a selling/marketing point so people tended to opt for it).

The second was mammograms no longer being recommended for certain age groups, as they were shown in studies to just be ineffective. There was a backlash against such recommendations, under a kind of 'better safe than sorry' argument.

What this and many other similar situations argue for IMO are better blood screens for cancer biomarkers.

For it turns out even with early detection, metastasis has already happened. It's just that those metastasized cancer cells take a long time to grow.

I mean the doctor is right, but it sure sucks to be one of the people who really has cancer and you skipped the screening, no?

https://www.sciencedaily.com/releases/2016/12/161214145615.h...

tl; dr: the false positive rate is too high.

What does a false positive look like to a doctor, and how will it not harm the organism? I can't imagine a cancer that can be ignored safely, nor an early stage treatment that is counterproductive.

Lots of things look like cancer and can be ignored, and there are benign cancers.

Every treatment has risks.

That's the cruddy math.

So the old thought about cancer was "every cancer will grow and kill the patient so we always want to find the cancer early and treat it before it spreads"

Turns out that's not accurate, and its complicated, and we are still learning.

There's three types of cancer, birds, turtles, and rabbits. (I didn't make up this terminology, see links)

The birds have already flown away and left the barnyard, early detection isn't going to help.

The rabbits need to be contained before they hop away, if we shut the barnyard gate early enough we can prevent their escape. These are the cancers where early detection is useful and saves lives.

The turtles just chill out, moving so slow that they never escape. These cancers cause no symptoms and will not go on to ever harm the patient and may even resolve themselves over time. Finding turtles is bad, treating turtles causes very serious side effects (cancer treatment isn't exactly benign), wastes time and money, causes anxiety, and doesn't reduce mortality one bit.

We've discovered there's a lot of turtles out there.

Of course, once we find early stage cancer we usually treat it like its a rabbit, even if its destined to be a turtle, because a lot of the time we can't tell the difference.

South Korea is finding this out the hard way, aggressive screening for thyroid cancer has produced a TON of thyroid cancer patients (15-fold increase (!) over the past 20 years) but ZERO reduction in thyroid cancer mortality.

https://sciencebasedmedicine.org/a-skeptical-look-at-screeni...

https://fivethirtyeight.com/features/the-case-against-early-...

https://www.skepdoc.info/a-skeptical-look-at-screening-tests...

    Welch says, "but [can also] find cancers that were never 
    going to matter."  That's because some cancers grow 
    slowly and never become dangerous, he says.

This exact same phenomenon is already occurring with breast cancer. Everybody is a "breast cancer survivor" these days, because they keep finding these turtle tumors that were never going to be a problem. If you look at mortality rates between those who screen early and those who don't, they are the same.

https://link.springer.com/article/10.1007/s10549-018-4691-4

While it is true that there is plenty of uncertainty regarding the value of finding DCIS, it is well established that there is significant benefit to finding small invasive breast cancers such as invasive lobular carcinoma and invasive ductal carcinoma. The value is particularly significant if the lesions are found before exceeding 1 cm in size, which they often are nowadays. The paper sited refers to one study focused on DCIS and another Canadian study on breast cancer screening during the 1980's and 1990's before there were good mammography screening standards. The Canadian study essentially just showed that poor screening is of limited value.

a.k.a. every layman's writeup of bayesian reasoning ever.

Reply on news.ycombinator.com