Wednesday, February 06, 2013

Blowing Second-Hand Smoke

My university is considering a campus-wide smoking ban, justified in part by the claim that second hand smoke kills more than fifty thousand people a year. I am generally suspicious of claims of that sort, so have been trying to explore that one. It turns out that it is a misstatement of a claim in a 2005 report from the California EPA. In that report 50,000 is the midpoint of a range of possible values. In the justification for the proposed ban, it has been converted to a lower bound.

More interesting is the question of where the number comes from. Reading the 2005 report I was unable to answer that question. The number also appears in a Surgeon General's Report, but reading that it is reasonably clear that it is simply repeating the CA EPA figure, not offering an independent estimate.

How could such an estimate be made, given the obvious problems in arranging controlled experiments on the effect of potentially lethal pollutants? One way is by using natural experiments. There have been a number of studies which looked at a city that imposed a smoking ban, compared heart attack death rates after the ban with death rates in comparable cities, and reported a surprisingly rapid and large effect.

There is a problem with that approach. Heart attack deaths in a single city vary randomly—with or without smoking bans, they sometimes go up and sometimes go down. If you want to argue that second hand smoke causes a lot of heart attacks, all you have to do is to find one city where a ban was followed by a decline and report that result. Given the pressure for anti-smoking measures, there are incentives for academics to do so. And even if the researchers are honest and pick their city at random, the study may be more likely to be completed and published if it gets a striking result, especially one that fits what many people want to believe.

One can always find possible problems with studies of controversial issues, especially ones that produce results one does not want to believe, but in this case there is at least some evidence. A 2009 NBER study analyzed all of the data and concluded that there was no effect from smoking bans—cities where the ban was followed by a decline in heart attack deaths were about as common as cities where it was followed by a rise. If that result is correct, it strongly suggests that the conventional view is the result of cherry picking the data.

Which fits my suspicion of scientific "facts" asserted in political controversies, especially ones supported mostly by the fact that  authorities such as the Surgeon General's Report and the CA EPA say they are true. It also fits my more general suspicion of the too popular idea of Official Scientific Truth, to be established by consulting the Official Scientific Authorities rather than by looking at arguments and evidence.

Of course, I too have my biases—not with regard to smoking, since I'm a non-smoker, but with regard to Official Scientific Truth. Can any reader help correct them by pointing me at convincing evidence not merely that second hand smoke has some negative effect, which strikes me as a priori likely, but for the size of the effect? Or at a convincing critique of the NBER paper?

---
"All previous published studies on the health effects of smoking bans share a common methodology: they compare the outcomes in a single community that has passed a smoking ban with outcomes in a small set of nearby communities that have not passed bans. A major contribution of this paper is that we simulate the results from all possible small‐scale studies using subsamples from the national data. We find that large short‐term increases in AMI incidence following a smoking ban are as common as the large decreases reported in the published literature."
(Shetty et. al., "Changes In U.S. Hospitalization And Mortality Rates Following 
Smoking Bans," NBER working paper 14790)



32 comments:

RP Long said...

My question is, how is it possible to disambiguate between the effects of second-hand smoke and other, similar environmental factors such as air pollution, candles, incense, barbecue grills, etc.?

David Friedman said...

Ryan:

In theory, you can do it using the natural experiments--places where smoking was banned over a wide enough area so that people who die of heart attacks in the area are probably people who lived in it and so were subject to the ban.

Strictly speaking, that's measuring the effect of the ban, not of second hand smoke. It might include the direct effect of a reduction in smoking due to people who otherwise would have smoked now finding it less convenient. And it doesn't reflect a complete elimination of second hand smoke because some people will smoke despite the ban.

But it may be the best one can do, and it does solve your problem.

EH said...

I do not have any data to bring to bear on the topic; nor do I expect you will find any that will withstand scrutiny.

My suspicion against the importance of second hand smoke is from the extremely improbable dose-response relationship implied by any significant effect. Scientifically establishing the direct effects of smoking, which I believe to be real, is no easy feat. Now dilute the dose of that smoke by a few orders of magnitude... should we expect to be able to measure anything but noise?

On a related note: I am no fan of smoke, second hand or otherwise, but what I find particularly irksome is that private property right gets trampled on, while there seems to be neither legal nor moral rules against people blowing smoke in your face at the busstop.

Dave said...

I don't think the "dose response" argument EH makes is very strong. Individuals' reactions to smoke vary dramatically, and people with very bad reactions are not smokers. For example, my wife is very allergic to smoke, and just walking by a smoker can give her difficulty breathing for a while afterwards. One might expect that any impact on health from secondhand smoke would be concentrated in individuals that were particularly vulnerable in some way.

Consider by analogy the effects of "secondhand peanut" exposure, which can kill those with severe food allergies. You can't argue that it must be harmless on the basis that the exposure to non-peanut-eaters is orders of magnitude less than the exposure to peanut eaters.

DR said...

Why only look at heart attacks? Much recent research has revealed a plethora of evidence indicating that nicotine has beneficial health effects on the brain.

Most adults should be consuming small regular quantities of nicotine. Very preferably in smokeless form. Particularly those with a family history of neurological conditions.

Smoking tobacco is probably the worse route of nicotine administration health wise. But those exposed to second-hand smoke, particularly the very moderate kind of passing someone on the street, should have only very minor exposure.

Since nicotine has decreasing marginal returns, and the negative components of tobacco smoke have increasing marginal cost, that probably indicates that there is some very small, but non-zero, optimal level of tobacco smoke exposure. (Assuming that the choice is between complete nicotine absence and minor tobacco smoke exposure). If anyone is close to that non-zero optimal it's probably those who are exposed from passing smokers on the street.

The increased heart attack levels should also be counter-balanced by looking at rates of Alzheimer's and Parkinson's. This is harder to do in a "natural-experiment" fashion since these diseases develop slowly and you wouldn't see a big discontinuity as with heart attacks.

EH said...

Dave: I am not saying there are no negative effects of second hand smoke: what it does to the odor of my clothes and hair is all too real; and I too quickly develop a cough in very smokey indoor areas.

But that is a rather different question than that of serious long term health consequences. Some people may die from peanut residue; yet they are all over my desk and no collegue has ever raised an eyebrow.

In any case, if I were in charge of a private institution like the one David discusses, I would find the obvious consequences of second hand smoke annoying enough to warrant a ban of smoking from all areas not explicitly designated for that purpose. Maybe id even try to sell the ban to the public on grounds of health concerns; after all, that seems like a much less contentious battle ground than picking a side between annoying non-smokers or smokers. 'sorry guys, the surgeon general hath spoken'.

Tim Worstall said...

This is the guy to talk to/read his blog.

velvetgloveironfist.blogspot.pt

Very much his field. He's a researcher at the IEA in London and taking apart the various studies about second hand smoke is very much one of his sports.

Unknown said...

Expecting science and reason to prevail is a lost cause. No experiment has established the existence of God or of the efficacy of prayer, yet 80% of Amerikans assume them as fact and make the rest of us put up with the nonsense from public figures, on currency and in anthems and pledges.

If you can believe those things, it's not much of a stretch to believe in global warming, insurance, ghosts and other superstitious nonsense.

Unknown said...

I really dislike the number of people killed as a statistic. Let's assume my expected lifespan is 80 years. If I engage in risky behaviour or exposed to "bad" environment, I will die sooner. But "sooner" could mean 39 and 79 which are very different outcomes, but both are reported as "death caused by". Can't we get a better statistic

Milhouse said...

In the '90s (if I remember correctly) the EPA published a claim that they had finally proved that "second hand smoke" caused cancer, and it turned out that they'd done this only by relaxing the threshold for statistical significance, and then reexamining surveys that hadn't found anything statistically significant under the usual standards, and lo and behold, under the special this-time-only standard their findings were significant! Someone sued the EPA and it was ordered by a federal court to withdraw the claim and the publication that contained it.

This tells me not only that anything that relies on that claim is also suspect, but also that there must not be any real evidence, or they wouldn't have had to fake it. And the fact that after so many decades of looking hard for such evidence they still haven't found it tells me that it's extremely unlikely to be there.

Tibor said...
This comment has been removed by the author.
Tibor said...

What I find most puzzling about the second-hand-smoke-kills argument is that its advocates often say that second hand smoking is actually more dangerous than the first hand one. That seems extremely improbable to me, since if you smoke first hand, you also smoke second hand (unless you are blowing the smoke out of a window or something...but in that case there can be no second hand smokers at all).

For the record I am a non-smoker myself (with a very very rare exception of an occassional cigar) and I think for a private university to ban smoking on its grounds it is perfectly ok (as it is for a pub) whether or not there are significant health problems associated. The problem arises when the state tries to impose general bans.

But I guess the point of an article is a much more general than just smoking. People say that statistics doesn't work. But it does - as long as it is used properly. Which raises a problem - the field of probability and statistics is most notoriously known for easy mistakes. Some things work differently than our first intuition tells us, sometimes there are more factors in play than someone counts with, and the cause and the consequence can be often all too easily swapped.

So for that reason a person who arranges the experiment should be someone who knows it really well - that is a mathematical statistician. But most of the research I have seen especially in medical or social areas is done by medical doctors or (perhaps worse yet) sociologists. These people often have poor knowledge of mathematics and statistics (doctors maybe a little better, but not all that great either) and easily make those mistakes (especially in social questions where everyone is initially biased and tries to keep that view as a posterior one as well, wether consciously or not).

I just had an exam yesterday and then I was talking with the professor for a while and somehow we got to this topic too - he told me he constantly gets phonecalls from sociologists asking things like "what does it mean that on the level alpha the test is negative?". If this is the average researchers knowledge of statistics, then combined with the necessary bias it really generates a lot of noise in the results (whatever they are). And another layer of noise then comes fro journalists (from who most people get the information) who more often than not make their own "interpretations" of the result - for example when the scientific report says "hypothesis A was found inconclusive by the testing", they write "hypothesis B was confirmed to be true".

So the best thing to do if one is really interested in something is to take the data and "do the math" yourself. But most people cannot really do that and also the data (with a complete methodology of data collecting) is often not available. So often you are left with believing someone who you believe is an unbiased researcher - but that belief of yours is of course biased.

A good example of poeple believing what they want to hear is this:
http://gregmankiw.blogspot.cz/2013/01/he-didnt-know-secret-handshake.html

Fred Mangels said...

I think for a private university to ban smoking on its grounds it is perfectly ok ....The problem arises when the state tries to impose general bans..

But it's the same old slippery slope thing. Those of us who opposed the first smoking bans were considered alarmists when we suggested eventually they'd try to ban smoking in our own homes. Now they're doing exactly that.

A number of cities in the Bay Area have passed, or are considering ordinances that ban smoking in apartments and even homes. Some years ago, Eureka banned outdoor smoking on its bayside boardwalk, probably the most windswept area of the city.

As far as second hand smoke goes, it's ludicrous to suggest someone smoking a cigarette while walking down the street is affecting the health of someone 30 feet away. It's just the age old need for us to find a boogey man to attack. That's why all the hysteria over second hand smoke has prospered as it has.

Anonymous said...

In America,CDC and Surgeon General data,say that never-smokers exposed to SHS have 5 lung cancer deaths per 20,000 exposed never-smokers per year.
Of those 5, one is claimed to be caused by the SHS exposure.

However, since there is no difference in the lung cancers, there is no way of knowing which one is the one that can be said to be caused by SHS exposure.

Each of the lung cancers has a 1 in 5 probability of being the one caused by SHS. That is a 1/5 chance.

Each of those lung cancers has a 4/5(80%) probability of not having been caused by SHS.

As for people being allergic to SHS. Allergies are a reaction to a protein in something. SHS has no proteins in it.

An allergic reation to SHS is probably a mental problem rather than a physical problem.

Gary K.

Tibor said...

Fred: Well, the reason for someone to ban smoking in his own business (a pub, a university...whatever) does not have to do anything with health issues. Some people (me included) prefer to go someplace where there is no smoke because then I don't come back home smelling like an ashtray (that is most of the time...I value the time spent with some of my friends wo are smokers and unwilling to go to a nonsmoking pub more than not smelling afterwards or there can be other reasons for me to go to a smoking pub, but given two otherwise equal circumstances, I choose the non smoking pub). All I'm saying is that there is no business for the government to interfere here and while anyone should be free to argue for or against any private ban, ultimately it is the owner of the place who should decide. And if the owner decides the place is going to be non-smoking (e.g. to attract people who don't want to smell to a non-smoking pub), noone (the government included) has the right to force him otherwise - and the same goes for the situation where the owner decides to let people smoke there.

I suppose you were not really arguing for having the state say (and enforce) "people have right to smoke wherever they want to", but if you don't acknowledge the owner's right to ban smoking, then I don't really see any other way you could achieve that.

Anonymous said...

Is SHS exposure more deadly than other things that kill us?


Doctors claim that there is no safe level of exposure to SHS and that SHS kills about 50,000 never-smokers per year.

There are about 68 million never-smokers exposed to SHS on a regular basis and that gives a death rate of 1 per 1,360.

Surgical errors will kill about 32,000 people out of the 15 million operated upon.
That is a death rate of 1 per 469 people.

Exposure to doctors with sharp knives is about 3 times as deadly as exposure to SHS.

Gary K.

Anonymous said...

Just a bit more.

The 1/20,000 never-smokers exposed to SHS dying from lung cancer per year tells us two things.

1.In any given year, a never-smoker exposed to SHS has a 0.00005 probability of dying from lung cancer because of it.

2.In any given year, a never-smoker exposed to SHS has a 99.995% probability of not dying from lung cancer because of it.

A sane person would not worry much about such extremely small odds.

Gary K.

Parmenion said...

Robert Nilsson,
Professor of Molecular Toxicology, Stockholm University, Department of Genetics, Microbiology and Toxicology. Senior Toxicologist, Swedish National Chemicals Inspectorate.

"The one-sided preoccupation with ETS as a causative factor of lung cancer in nonsmokers may seriously hinder the elucidation of the multifactorial etiology of these tumors."

In the book What Risk? Professor Nilsson puts children's risk of passive smoking in this perspective:

"Looked at another way, a child’s intake of benzo[a]pyrene during 10 hours from ETS is estimated to be about 250 times less than the amount ingested from eating one grilled sausage."

David Friedman said...

Gary K:

Do you know how that 1/5 number was calculated? If not, why do you believe it? The CDC and Surgeon General, like anyone else involved in such disputes, have their own biases, their own preferences about what results they get.

Unless you actually know where the numbers come from, not from whom by how calculated, you should treat them not much more seriously than the claims in advertisements about how good the advertised product is.

Tibor said...

David Friedman: Of course the ideal way to judge the results is to look at how the results were obtained from it. But not only you often have neither the data nor the methodology available, I think for most people (considering the average understanding of mathematics) even that would not be very helpful.

Someone skilled can do real wonders with the data to achieve the desired results and make it look that the methodology is ok. Or (worse yet? I don't know what is worse really) someone unskilled can make a mistake which leads him to false results. But not many people are actually skilled enough to find the mistake (intended or not). Of course, if the data and methodology are available, someone will point it out, but perhaps he is also biased and/or incompetent and if you don't understand how the results were obtained, you can't tell who's right. Then you will most probably believe the guy "on your side" - that is the one that doesn't require you to change your current opinion, since pretty much everyone is biased against what doesn't fit into their present worldview.

Fred Mangels said...

Tibor wrote, "I suppose you were not really arguing for having the state say (and enforce) "people have right to smoke wherever they want to", but if you don't acknowledge the owner's right to ban smoking, then I don't really see any other way you could achieve that.".

Probably poor writing on my end. I agree that owners of private property should be the ones to decide whether smoking is allowed on private property. That includes grocery stores, bars or whatever.

Truth be told, though, most owners of private property (perhaps with the exception of bars) would opt to not allow smoking. So be it.

I don't have a problem with smoking bans in government buildings which should accommodate everyone and where many people might have no choice but to have to be inside.

I misread your comment and didn't notice you specified "private" university. Sorry.

Ann W. said...

Here's one to think about David. And written by one of the authors of both the SG report and the CA EPA report.
Is Glantz really saying that smoking makes you more tolerant to impurities?
" Of note, the cardiovascular effects of ETS appear to be different in nonsmokers and smokers. Nonsmokers appear to be more sensitive to ETS than do smokers, perhaps because some of the affected physiological systems are sensitive to low doses of the compounds in ETS, then saturate, and also perhaps because of physiological adaptions smokers undergo as a result of long-term exposure to the toxins in cigarette smoke. In any event, these findings indicate that, for cardiovascular disease, it is incorrect to compute "cigarette equivalents" for passive exposure to ETS and then to extrapolate the effects of this exposure on nonsmokers from the effects of direct smoking on smokers. These results suggest that heart disease is an important consequence of exposure to ETS. The combination of epidemiological studies with demonstration of physiological changes with exposure to ETS, together with biochemical evidence that elements of ETS have significant adverse effects on the cardiovascular system, leads to the conclusion that ETS causes heart disease. This increase in risk translates into about 10 times as many deaths from ETS-induced heart disease as lung cancer; these deaths contribute greatly to the estimated 53,000 deaths annually from passive smoking. This toll makes passive smoking the third leading preventable cause of death in the United States today, behind active smoking and alcohol. "
Passive smoking and heart disease. Epidemiology, physiology, and biochemistry. Glantz SA, Parmley WW.

Art said...

In what ways are non-smokers on your campus exposed to second-hand smoke?

Joey said...

This conversation got me interested in this debate. I just read "The Impacts of Passive Smoking at Home on Respiratory Disease: Results from the Indonesia 2001 National Survey Data" (PDF available for free from World Bank).

I'm not sure how widely-cited this paper is; it was the first one I clicked. Anyway, it looks for a relationship between passive smoking, measured by proxy by the number of cigarettes smoked by household members, and respiratory illnesses of kids between the ages of 1 and 10. It divides household smoker status into four groups:
No smokers
Smokes 1-14 cigs a day
15-29 cigs a day
30+ cigs a day

They use logistic regression to account for about 8 other independent variables. The data was collected by survey, the people surveyed live in Indonesia.

The only (P<.15) differences were between the 1st and the 4th groups. The % with disease isn't monotonically increasing through the groups, which doesn't make it look dose-dependent. Although this may be due to the logistic curve they fitted the data to or the fact that P was high. Stats isn't my specialty.

Anyway, what I thought was funny:
If you're going to analyze data about second-hand smoking, the first thing you should control for is whether the subjects of the study, the kids aged 10 and under, are nonsmokers. Fifty-four percent of males older than 10 smoke. They know that because they only asked people who were older than 10 about their smoking status. What an omission!

If the authors really wanted to be honest, don't you think that they would have just looked at the health status of people they knew to be nonsmokers?

My rule of thumb is to be suspicious about the findings of any paper with "Policy Recommendations".

Anonymous said...

http://vitals.nbcnews.com/_news/2013/01/28/16741714-lungs-from-pack-a-day-smokers-safe-for-transplant-study-finds?lite

Lungs from pack-a-day smokers safe for transplant, study finds.

By JoNel Aleccia, Staff Writer, NBC News.

Using lung transplants from heavy smokers may sound like a cruel joke, but a new study finds that organs taken from people who puffed a pack a day for more than 20 years are likely safe.

What’s more, the analysis of lung transplant data from the U.S. between 2005 and 2011 confirms what transplant experts say they already know: For some patients on a crowded organ waiting list, lungs from smokers are better than none.

“I think people are grateful just to have a shot at getting lungs,” said Dr. Sharven Taghavi, a cardiovascular surgical resident at Temple University Hospital in Philadelphia, who led the new study...........................

Ive done the math here and this is how it works out with second ahnd smoke and people inhaling it!

The 16 cities study conducted by the U.S. DEPT OF ENERGY and later by Oakridge National laboratories discovered:

Cigarette smoke, bartenders annual exposure to smoke rises, at most, to the equivalent of 6 cigarettes/year.

146,000 CIGARETTES SMOKED IN 20 YEARS AT 1 PACK A DAY.

A bartender would have to work in second hand smoke for 2433 years to get an equivalent dose.

Then the average non-smoker in a ventilated restaurant for an hour would have to go back and forth each day for 119,000 years to get an equivalent 20 years of smoking a pack a day! Pretty well impossible ehh! OSHA ON SECOND HAND SMOKE.................

Michael J. McFadden said...

A couple of quick comments to start:

1) The Glantz/Parmley thing theorizing about nonsmokers being "more sensitive" to ETS is simply something pulled out of a very dark place in order to try to justify their claims of instant heart attack reductions after smoking bans. Of course if you actually *READ* Glantz's famous Helena study, you'll find that he carefully avoids mentioning any effect of the ban on nonsmokers as a separate group (saying the sample was "too small" -- although it was obviously at least 2/3rds or so of the total that they so solidly proclaimed about.)

The original 50,000 figure traces back to Glantz and a few of his friends as well: Parmley, Wells, Judson, and Repace. That little clique started pushing the 50,000 deaths per year claim back in 1988. By the end of 1992 when the EPA Report came out, the claim was *still* seen as so ludicrous that the EPA refused to include it -- even though it was willing to lower its standards from 95% to 90%!

In terms of outdoor campus exposures, and (for argument's sake) accepting the EPA Reports claim of 3,000 deaths, a 19% increase over the base rate of 1 lifetime lung cancer per thousand people, once reasonable adjustments are made for intensities and durations of exposure and assuming that students have to walk through at least one fairly dense cloud of smoke a day by doorways, it would take on the order of 250,000,000 years (i.e. two and a half million centuries) of schooling before the average student would develop lung cancer from their campus exposure.

To see more of how the Antismokers juggle their figures to push their arguments, Google "V.Gen5H" and read my "Lies Behind The Smoking Bans" there. It's superficial, with big bombastic print designed for quick reading in the dim light of a bar or smoke-pit under attack, but its facts are accurate and their presentation is honest. Feel free to offer any specific, substantive criticisms: I promise I won't mind and I'll try to stop back to respond.

Michael J. McFadden
Author of "Dissecting Antismokers' Brains"

Michael J. McFadden said...

Joseph Miller, good catch on the research negligence (If the authors really wanted to be honest, don't you think that they would have just looked at the health status of people they knew to be nonsmokers?)

And the answer there is that the authors of these things generally do NOT "want to be honest." Instead they want to produce results that will either bring them more millions of dollars in grant money OR produce more pressure on smokers to quit (because the researchers usually believe that people would be better off healthwise if they quit). Usually there's a mix of the two motivations at work, and there are, very occasionally, some few researchers who simply look for the truth rather than worrying about grants or trying to produce results to enhance a social engineering goal.

Unfortunately, at least from what I've seen in the area of antismoking research, those sorts of honest researchers are few and far between. *Do* check out the Lies booklet I referenced. And remember, the studies I examine there are the Biggest and the Best of the bunch: the ones that get brought up all the time at the hearings on bans.

- MJM

Joey said...
This comment has been removed by the author.
Joey said...

Michael,

"Lies Behind the Smoking Bans" was good. Lies (exaggerations, at least) are definitely what they look like. Scientists are skeptical of research conducted by researchers with ties to the tobacco industry, but are uncritical when citing EPA or other gov't-funded research figures. You can see this naivety in "Why Review Articles on the Health Effects of Passive Smoking Reach Different Conclusions", in which the authors compared review articles conducted by tobacco-affiliated authors to that of researchers who weren't, as if biases only run in one direction.

Frank Koza said...

An interesting report from the National Association of Scholars: "A Crisis of Competence: The Corrupting Effect of Political Activism in the University of California."

http://www.nas.org/images/documents/a_crisis_of_competence.pdf

Fred Mangels said...

This article in my local paper shows more political correctness from the medical "researchers". The article is about the decline in life expectancy in females. The article twice mentions smoking, or increased smoking rates as one possible contributor.

This, despite smoking declining over the years to now they're saying only 10% of Californians smoke? Somehow that suddenly contributes to a decline in life span among women?

Seems to me it's just another attempt by the politically correct to manipulate statistics throw their boogeyman, smoking, into the mix.

http://www.times-standard.com/localnews/ci_22763381/study-shows-declining-life-span-some-u-s

johnhallal31 said...

Great post, Mr. Friedman. Sometimes, public data does seem like conjecture rather than something that has been found using scientific methods. Incidentally, heart attacks are surely not the only side-effects of smoking. What about respiratory disorders?