Lung Cancer and the Power of Suggestion | Hacker News Reader

Lung Cancer and the Power of Suggestion

asserttrue.blogspot.com

19 pointstechdog51 comments

Were there millions of depressed people in hiding, in 1955, waiting to come out of the closet in the 1990s? Was 9.3% of the American adult population (the percentage now suffering depression) merely suffering in silence, back in 1955? Or did the drug companies and their media flacks convert a relatively rare psychological condition into a growth industry through massive advertising and media hype (in other words, through the power of suggestion)?

The author is trying to claim that many (most/majority?) cases of depression are made up and hyped. They are trying to claim that it's impossible for there to be that many actual depressed people now.

But turn it around. How many (openly) gay people/couples are there today? Millions! But there weren't that many in 1955! So what happened? Was it all media hype? <sarcasm>Or are we expected to believe that all those gay people were suffering in silence in the 1950s?!</sarcasm>.

Of course all the gay people were suffering in silence in the 50s. People don't doubt that. So why the doubt about mental illness?

I don't think that the author is trying to claim that depression is made up and hyped. What he seems to be claiming is that publicity surrounding depression could have caused an increase in the number of depressed people through the power of suggestion.

He's not saying that mental illness isn't valid, or that the people with depression aren't suffering. He's suggesting that the increase could partly be caused by the immense advertising campaigns surrounding anti-depressants; this is in no way de-legitimizing depression, just suggesting a possible cause.

Except America is the only industrialised nation that allows direct marketing of antidepressants to its populace.

Other industrialised countries - Australia, the UK, France and Germany, etc have also experienced a rise in patients under treatment for these conditions, in lockstep with the US experience, despite this difference you suggest is responsible

It's like an experimental study that removes a confounding factor!

The author never outright states one way or the other whether depression was just hidden or whether it was made up by hype (or some mixture of both, or .... ). However the implication is that the media hype has made more people being diagnosed with depression.

He also conflates bipolar with depression and anti-depression medication.

They are different conditions (albeit ones that have shaky scientific bases as 'diseases' due to the subjective nature of diagnosis).

Prozac/fluoxetine is not a prescription drug used for bipolar illness as the excess serotonin can in fact trigger manic episodes and it is thus very dangerous in bipolar patients

It appears you've made some sort of resolution to publish and promote a blog entry per day in 2013. 40 entries in 41 days this year vs. 46 in all of 2012. You should reconsider - whatever your reasons were, I doubt they included a desire to develop a reputation for presenting topics that were sensationalized and thinly researched [1] produced with a pace that ensures discredited theories dont get reviewed.

[1] http://asserttrue.blogspot.com/2013/02/drug-companies-stop-h...

Wow, nice spot and they have all been submitted to HN. I have never seen anyone's submission history be so hell bent on self promotion:

https://news.ycombinator.com/submitted?id=techdog

You'd be surprised, there are one or two people who are as blatant.

Just for fun, here are a couple of charts. (http://imgur.com/a/ywor0)

"Do Cigarette Warning Labels Help CAUSE Cancer?" No... Long answer: Hell no and stop doing this.

To the next person with a pseudoscientific gut-feeling-bear-with-me-just-a-bit-while-I-pull-a-half-baked-idea-out-of-a-number-set-that-does-nothing-to-corroborate, please don't. Corrolation doesn't equate to causation, no matter what amount of reality bending assertions you apply. Warning those at high risk for cancer know they're at high risk and subsequently confirm the hypothesis doesn't mean the warning was the cause.

There seems to be a trend of Very Smart People with no experience whatsoever in a given field, reading some article somewhere, pulling out R and some graphing software, and proclaiming "I've discovered something that nobody who actually works in the target field which I knew absolutely nothing about six hours ago ever found!". I wish those stories would stop getting posted to HN.

That graph did not look nice enough to come from R.

I've seen one of these (on a different topic) posted to HN repeatedly in comments, by a variety of participants here. Really, people need to learn how to evaluate sources better here on Hacker News. That's part of being an educated person.

http://news.ycombinator.com/item?id=5105664

I feel it is the hacker mentality. People here are very intelligent but (some often) make the mistake of feeling as though their experience and success in one domain is applicable to multiple, that they are, in fact, renaissance men and women when in fact they are largely unskilled in the Domains they are attempting to analyse in depth.

As someone heavily involved in medicine j see this all the time when the subject matter is something medical or biological- people really so feel that they can talk about issues of medicine and science with just the same authority as they can when otherwise discussing programming/whatever their day job is, when in fact they dont know just how much they don't know

I found it to be thought provoking, even if it wasn't able to form any definite conclusions. Correlation is still interesting for many reasons.

Your post suggests that it has been researched before (and presumably proved to show no link), but searching Google did not yield any results for me. I, and I'm sure most everyone who has read this article, would love to look over the findings if you are able to share what you know.

The blog post in question states in the fourth paragraph he was inspired by the recent article about "placebos and nocebos", which is "the article" I'm referring to. That is, we're reading the product of someone's short-term random curiosity. Random curiosity is a great thing but IMHO the path from random curiosity to published findings needs to have some serious experience in the middle for it to be worth our time.

NO! I absolutely resent that idea that "science" should only be something for the experts, and everybody without a PhD should shut up about it.

As long as it is just random ranting without any data, it is more than welcoming that "regular" people try to understand expert opinion and in fact challenge it. If "facts" are so clear and indisputable, than those amateur research should easily be shown as wrong.

If not, amateur researcher make a highly valuable contribution. They highlight the fact, that social science, especially on social phenomenons, is (often) based on consensus. In this case here, the author makes an excellent contribution just by highlighting how easy correlation and causation can be (mis)interpreted.

For example the other day I read and article published in a well regarded management journal, by authors who claim having found out that the willingness of users clicking on ads is based on the hours of sunshine per day in that region. The correlation equals causation phenomenon is quite common among social scientists who try to show how scientific their contributions are.

You're right, science shouldn't just be reserved for the PhDs. But the PhD is likely to have insight and experience that will make a lot of assertions false, avenues of research redundant and generally be able to tell the different between science and not science.

There's no restriction on laymen studying, researching and trying to understand new frontiers of science, however just as the PhDs don't have a monopoly on science (and as you point out, they're far from perfect), neither do lay people have a monopoly on reinterpreting reality through the misinterpretation of facts.

tl;dr: You're entitled to examination of facts and to be examined for your own research. You're, however, not entitled to your own reality. Which is what happened in the article.

> I absolutely resent that idea that "science" should only be something for the experts, and everybody without a PhD should shut up about it.

making some off the cuff, armchair observations within a particular field of study (and blogging and serially promoting them to hacker news) without even making casual efforts to familiarize oneself with at least portions the vast amounts of knowledge publicly available on that field is, generously, pretty bad "science". And less generously, it's just hubris. This is not an "amateur researcher", this is someone who apparently based on posting history has a really high blogging-to-value ratio.

Of course, we all have the right to be clueless. I only requested that perhaps Hacker News doesn't want to encourage a culture of armchair cluelessness and particularly someone who appears to be a serial armchair blogger/HN promoter (see https://news.ycombinator.com/submitted?id=techdog).

If he had stopped at "Hmmm, there may be some cause or causes of lung cancer we're not paying enough attention to." that would have been fine.

Agreed. There's a big difference between "I don't know" and "I don't know, therefore...". People jump to therefore based on assumptions and that's not science.

The next step after "I don't know" should be "let's investigate objectively".

Corrolation doesn't equate to causation

Yes this is correct. And to go off on a general tangent, correlation tells you where to start looking. After all, if A & B are correlated, then one of these statements must be true: "A causes B", "B causes A" or "there is something else, C, which causes A and B".

Just because you can find a correlation, doesn't mean that finding is irrelevant. It is very interesting and tells you where to start looking.

> After all, if A & B are correlated, then one of these statements must be true: "A causes B", "B causes A" or "there is something else, C, which causes A and B".

You missed "it's purely coincidence", which is an entirely valid option.

You should do enough experiments to be able to state with statistical confidence that it's not a coincidence.

The correlation could also be due to a flaw in your experimental design (e.g., something in your lab is contaminating your samples, or the researcher has an unconscious bias that skews measurements in a particular direction), in which case repeating the same experiments will not solve the problem.

Sure, there's loads of ways an experiment can be flawed. But also, it might not be flawed. If you're got a good experiment that shows correlation, then it has to be one of the 3 results above.

No, it is statistically most likely to be one of the 3 results above. "Has to be" doesn't have much of a place in a statistics discussion. There is a non-zero probability of the correlation being due to chance.

In 1955, only 38,200 people in the U.S, were in mental hospitals as a result of disablement due to depression (C. Silverman, 1968, The Epidemiology of Depression, Johns Hopkins Press). Today, depression is the leading cause of disability in the U.S. for people aged 15-44 (according to NIMH stats). Depression now affects 20.5 million American adults: 14.8 million in the form of major depressive disorder and 5.7 million in the form of bipolar disorder. That's a pretty huge jump in numbers: 38,200 to 20 million.

He's comparing the number of of people insitutionalized in a mental hospital in 1955 to the total number of people who suffer from depression in the United States now. How is that comparison even relevant to anything? How many people are institutionalized for depression now? My guess is that those numbers either wouldn't support his narrative or can't be found in a few minutes of googling like the rest of his "research".

Also notice how he gets to 20 million (a very dramatic number indeed) by adding together 14.8 million suffering from a major depressive disorder and 5.7 million in the form of bipolar disorder. Are those thing mutually exclusive?

If there IS a nocebo effect here, it's mildly implausible that it would have a really narrow effect like "making people more susceptible to lung cancer". At a minimum, I'd expect them to be more susceptible to other cancers, and possibly other diseases and illnesses (the big fat warnings in your face, and having to bypass them to smoke, acting as some sort of immunosuppressant or something like that).

Perhaps, if this effect is real, further research could demonstrate that? Or perhaps there's still a confounding effect somewhere, maybe something related to how people self-select whether they're going to smoke and some factor which is correlated both with making that decision in the face of a big fat warning label and with increased susceptibility to lung cancer?

Anyway. Nice idea for a research paper. Not sure it belongs here until you've done more though. :P

+1. Interesting idea for a research article, but post when you've done the research, not when you've just had the interesting wild hypothesis.

I don't understand why it's so infeasible that improvements in cigarette technology could result in cigarettes that are worse for one's health because they've been more aggressively optimised for tar content as officially measured, or shelf life, or cost of manufacture.

As for the theory that modern smokers simply suck harder, I'm not sure that's so implausible. Light cigarettes are more common today, and most light cigarettes seem to get their lightness entirely from lengthier filters that have small holes in the paper. (If you use the tobacco for some other purpose, e.g., as filler for an unfiltered marijuana cigarette, then you'll find the smoke just as harsh and the roach paper no less tarry.) This requires a more forceful draw to avoid that unpleasant "sucking in air" sensation.

I've also noticed, on the occasions that I've checked recently, that cigarettes today tend to have lower advertised tar and nicotine content than the ones I smoked as a teenager. So perhaps that could result in a similar effect.

Light cigarettes are in fact Worse as a whole as previously heavy smokers switch brands and start smoking lighter cigarettes

The effect is real and had been studied.

In fact, all of the effect size attempted to be attributed to 'warning labels causing cancer' can be explained by this paper here

http://www.ncbi.nlm.nih.gov.ezproxy1.library.usyd.edu.au/m/p...

This URL requires a login from University of Sydney for access.

My bad

Smoking and lung cancer - Tuberk Toraks http://www.ncbi.nlm.nih.gov/pubmed/16100660/

Can we please get this website hell-banned?

Every single article posted by techdog is an article from this blog, and every single one of these articles is factually and demonstrably wrong, written by someone who has very little understanding of statistical analysis or any of the specific topics the article purport to address.

But in spite of all that, it has managed to provoke some interesting discussion here.

The first article graph of woman lung cancer rates raises some red flags - a huge increase in woman lung cancer caused by cigarettes would be expected a couple of decades after the huge increase in woman (as opposed to mostly-male) smoking in first half of 20th century.

In order for that graph to facilitate truth instead of misleading, it needs to include either (a) woman smoking rates by year, or (b) male lung cancer rates, not woman.

Good point. If at some point the median man smoked 20 cigarettes a day, and the median woman smoked 0, then lots of men will get lung cancer and women won't. If things change to 10 and 10, then the average number of cigarettes will stay the same, but the number of people who smoke at least 10 will have doubled. This is how the lung cancer rate can go up with the number of cigarettes going down.

Another way to measure it would be to look at "how much lung cancer is there per smoker".

Seems more like an variation of what psych and behavioral economics calls the "what the hell" effect than "power of suggestion". "What the hell" effect says once your willpower fails and you give into temptation to begin doing negative behaviors, you are much more likely to go all in (think eating a single donut on a diet then polishing off the box). By making the cancer threat real post purchase and reminding of the ultimate consequence every time you light up, they may be pushing people to throw their hands up and just keep smoking.

Of course, like this article, this is just a untested theory I pulled out of my ass.

I'm not sure how the article goes from interesting hypothesis (which it is) to near certainty.

>If we rule out changes in cigaret design and environmental factors as explanations for the enormous increase in lung cancer, what are we left with?

Seems like a false dichotomy to me. It can't be X, so it has to be Y.

>All of these would affect smokers and non-smokers equally. And yet lung cancer rates have not gone up for non-smokers.

This isn't necessarily true. It could be a synergistic effect, where some external factor increases cancer rates when smoking is also involved.

Some data to help think up such effects: http://www.kidon.com/smoke/percentages3.htm

Differences in diet was an obvious idea, but it's not obvious to me what Japan and Israel might have in common.

Or, very simply, life expectancy has increased significantly during this time period. Since, "Of course, it takes years of smoking to get lung cancer", more people live to the age where they get lung cancer.

What about the use of additives in cigarettes that started around the 1960s? From wikipedia on 'cigarette': "According to data from the World Health Organization,[40] the amount of tobacco per 1000 cigarettes fell from 2.28 pounds in 1960 to 0.91 pounds in 1999, largely as a result of reconstituting tobacco, fluffing and additives."

Maybe as smoking went from an "everyone" activity to a more marginalized activity, you get worse health outcomes in that group for other reasons.

How does the fitness of the median smoker today compare to the fitness of the median smoker in the 1950s? What about wealth, access to treatment, etc?

One other explanation I can think of for higher lung cancer rates is that people are living longer, so more likely to succumb to cancer in the end.

But like someone else said, the idea that these trends have not been examined and explained to death by better-informed people is risible.

The only problem I see here is his argument about placebos - while the placebo effect is a strong one (and I believe in it), there is a current lack of control over placebos (I e sometimes drug companies try to use placebos that cause side affects) that are currently used in the industry. Otherwise. This is an interesting idea.

If you're testing your drug against a placebo, and the drug has a negative side effect, then you should try to make sure the placebo has the same/similar side effect. The whole point of double blind testing against a placebo is to test against something that appears similar.

Remember doctors aren't supposed to know what patients in a trial are on drugs vs placebo (that's the 'double' in double blind, the patient and the doctor is blind). If the real drug has a horrible side effect and the placebo doesn't, then the doctor will be able to figure out what patient is getting what, and you no longer have a double blind test.

I am curious about "cigaret," wiktionary says it is a dated form of cigarette.

I have trouble looking at the graph without being distracted by "cigaret" but firefox's dictionary is not marking it as a misspelling so it must be accepted somewhere. Where is this spelling still commonplace?

Or it could be all the coal burning pollution catching up to us?

Or any of the thousands of other chemicals, such as flame retardants, that we breathe in all our lives.

Or you know, people who have smoked is more likely to have kids with cancer or who where second-hand smokers at an age they can't remember now.

Also, campaigns against brest cancer had an exponential grow too but somehow the rate of breast cancer is about the same, or do your hypothesis includes lungs being more affected by perceptions than any other organ?

Reply on news.ycombinator.com