Michael MacGuinness on Environmental Tobacco Smoke

This is a consolidation of six e-mails that Michael Mac Guinness sent me during November (whilst I was overseas), in response to my post in support of Ireland's proposed ban on smoking in public. This included reference to a study in Helena, Montana purportedly showing a close correlation between a smoking ban and a reduction in heart attacks.

Michael starts starts by chiding me for swallowing this study too readily (I accept this !), and goes on to explain why he is skeptical about the scientific relationship between ETS and harm to health.

Readers may make their own judgement.

ENVIRONMENTAL TOBACCO SMOKE

From: Michael Mac Guinness
Sent: 1st, 4th, 5th, 5th, 22nd, 28th November 2003
Subject: ETS: Correlation, Causation and Helena, Montana

Reference your article on ETS (Environmental Tobacco Smoke) and the Helena, Montana study, Steven Milloy has made a dissection of the Helena study and includes a link to the slide presentation (pdf, 273 kb) of the study. It appears that in this instance the authors may have been a little too quick to jump to a conclusion, forgetting that “correlation does not imply causation!” See also here.

I have never smoked but have always been, and remain, extremely skeptical regarding the degree of risk attributed to ETS. Unfortunately, in this area as in others, there is a lot of selective study and reporting by activists seeking to advance their agenda. See a review of such studies here.

My understanding (I’m an engineer, not an epidemiologist) is that the levels of increased risk of lung cancer the studies attribute to ETS (about 20%, adjusted risk ratio 1.2) is within a range where it could equally well be caused by other factors not controlled for. I believe epidemiologists require an adjusted risk ratio of 2 before concluding that causation is demonstrated. Since even the studies quoted by proponents of the smoking ban fall well below this threshold, the scientific basis for a ban is insufficiently solid to justify legislative action.

I have been dismayed, but not surprised, that all of the debate (in the media and for example the exchange between Frank at Internet Commentator and Gavin Gavin's Blog) on the passive smoking issue has uncritically or implicitly accepted the contention by proponents of the ban that the science is settled.

In fact, so far as I can determine, the science is far from settled. Studies funded by tobacco companies reporting no significant effect from ETS are routinely dismissed as biased. This may be true, however it is unproven - but see later discussion of bias in studies.

On the other hand, an important component of the smoking ban proponent’s case, the US EPA’s (Environmental Protection Agency) 1992 525-page study, “ Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders”, EPA/600/6-90/006F (December 1992) was challenged in court (yes, by a tobacco company). The challenge was successful in 1998.

The judge's findings are worth reading in full. Key observations :

“Rather than reach a conclusion after collecting information, researching, and making findings, EPA categorized ETS as a “known cause of cancer” in 1989. EPA, Indoor Air Facts No. 5 Environmental Tobacco Smoke, ANR-445 (June 1989) (JA 9,409-11).”

“In conducting the Assessment, EPA deemed it biologically plausible that ETS was a carcinogen. EPA’s theory was premised on the similarities between MS, SS, and ETS. In other chapters, the Agency used MS and ETS dissimilarities to justify methodology. Recognizing problems, EPA attempted to confirm the theory with epidemiologic studies. After choosing a portion of the studies, EPA did not find a statistically significant association. EPA then claimed the bioplausibility theory, renominated the a priori hypothesis, justified a more lenient methodology. With a new methodology, EPA demonstrated from the selected studies a very low relative risk for lung cancer based on ETS exposure. Based on its original theory and the weak evidence of association, EPA concluded the evidence showed a causal relationship between cancer and ETS. The administrative record contains glaring deficiencies”

“In this case, EPA publicly committed to a conclusion before research had begun; excluded industry by violating the Act’s procedural requirements; adjusted established procedure and scientific norms to validate the Agency’s public conclusion, ...”

An editorial in Investor’s Business Daily on 29 July 1998 summarises some of the findings.

Having read the whole judgement (but not the original report), it is clear that whatever adjectives may be used to accurately describe the EPA’s report “scientific” is most certainly not one of them. For some reason, the adjective “biased” comes to mind, but surely not. After all, it was not funded by a tobacco company.

As noted above, I have always been skeptical of the alleged hazard of passive smoking. In my reading, which extended to more than one website, as well as other issues I wanted to check out, I have come across no evidence that I found convincing. The evidence that is available demonstrates conclusively that many issue-driven (ie not big tobacco or big oil) activists and researchers have few scruples about manipulating and selecting data to ensure their papers will support their pre-determined conclusions in order to advance their agenda, irrespective of truth. On controversial issues a healthy skepticism is in order, for bias is not solely the preserve of the black hats.

If the science is unproven and/or if, as in this case, one of the pillars supporting the case for taking legislative action is a fraud, then the policy and the legislation are indefensible.

I don’t like tobacco smoke but so long as it’s not blowing directly in my face I’m not bothered. If they really want to ban something, my vote is ban vinegar. The smell of it is revolting, I cannot abide it, it really does turn my stomach.

Here are a couple of other links that may be of some interest; plenty more if you do a site search for EPA and ETS:

http://www.junkscience.com/news/sws/sws-chapter6.html http://www.junkscience.com/news/euwsjets.htm

The 1998 ruling was, however, overturned on appeal in 2002. But, as the author of Numberwatch writes:

“The appeals court did NOT overturn Judge Osteen’s findings that EPA had used deceptive science to support a policy on environmental tobacco smoke that it had adopted before it even began its study. It did not even address those issues. Instead the appeals court ruling focused only on whether the lower federal court had jurisdiction to review EPA actions in this matter and found that it did not.”

I’ve read the ruling in full. Numberwatch’s statement is perfectly accurate.

Meanwhile, to link to a page linking to a variety of studies, including the EPA’s and a rebuttal of the findings in the 1998 ruling, click on this here.

This last link is to the website of the Center for Tobacco Control Research and Education referred to by Steven Milloy in his rebuttal of the Montana report mentioned at the beginning of this message.

It also links to the following article (pdf, 127 kb) “Why Review Articles on the Health Effects of Passive Smoking Reach Different Conclusions” by Deborah E. Barnes, MPH; Lisa A. Bero, PhD JAMA, May 20, 1998—Vol 279, No. 19

The principal conclusion of this article is that, after controlling for various factors, it finds that review articles written by authors with tobacco industry affiliations are 88 times more likely to conclude that passive smoking is not harmful. I’ve seen several reports about this article and reporting its conclusion.

However, reading the article reveals a few interesting points:

	Mean article quality was 0.36, meaning that the average review article in the study satisfied only one third of the criteria used for quality assessment. Low quality was expected to correlate with a non-harmful conclusion, however it did not. Mean quality scores *overall* were very low.
	77% of articles failed to disclose sources of funding. Affiliation to the tobacco industry was determined indirectly by investigating the prior relationships of authors with tobacco-affiliated organisations. There was no investigation of other sources of funding, eg, anti-smoking advocacy organisations.
	To support its conclusion regarding the influence of tobacco affiliation, the article quotes other studies that have found an association between conclusions of review articles and the affiliations of the authors.

It appears to me that the authors may have a blind spot. While they do note that it may have more general implications, it does not occur to them that author affiliation may also affect articles that conclude that passive smoking is harmful, and they do not examine this possibility. In this context, the article itself was funded by the Robert Wood Johnson Foundation, whose mission statement includes the following : “To reduce the personal, social and economic harm caused by substance abuse — tobacco, alcohol, and illicit drugs.” A laudable objective, but perhaps it is not a totally neutral, impartial, disinterested entity in this area of research. I find it notable that the authors found that the articles were generally of low quality regardless of the conclusion.

As an engineer, I associate low quality with unreliability. I draw the conclusion from this article that research on passive smoking is generally of low quality, is potentially riddled with bias on both sides and is therefore utterly unreliable regardless of its conclusion. I cannot see how low quality research can lead to accurate, reliable results.

I am astounded that the authors do not appear to consider the quality issue significant. I haven’t had much time to do any more research on the issue but came across ActivistCash.com. Serious money available for pro-activist research. Accusing tobacco-affiliated researchers of bias looks more and more like the pot calling the kettle black.

I tried to find some hard numbers quantifying lung cancer rates in non smokers. Googling turned up a huge variation with lung cancer rates among non-smoker anywhere between 4 and 27 times lower than smokers but I had no way of evaluating either the reliability of the data or its applicability to Ireland.

I did find the National Cancer Registry Ireland report (pdf, 165 kb) “Cancer in Ireland 1994-2002 Incidence, Mortality, Treatment and Survival ”, but it doesn’t provide a breakdown of smoker and non-smoker lung cancer cases. I could not find such a breakdown anywhere (in the limited time I’ve had).

Bottom line, the more I read, the more skeptical I become regarding the alleged dangers of passive smoking. Case not proven.

Unfortunately, it’s become a crusade.

Regards,

Michael Mac Guinness

Addendum

Below are links are to

	the Office of Tobacco Control(OTC)/HSA report (pdf, 253 kb) on passive smoking dated December 2002 and
	P J Carroll's submission (pdf, 206 kb) dated 21 May 2003 to the HSA.

The OTC/HSA report is basically a review and summarisation of the results of studies on passive smoking/ETS. The referenced studies appear to have been taken at face value, there does not appear to have been any questioning of data quality or study methodologies or procedures. And usually that's fair enough. Normally there should be no reason to for this type of report. However, although the report makes quite a few references to the EPA's 1992 report, there is no reference to the 1996 ruling vacating the report, nor to the scientific/technical criticisms of the report raised in the case and summarised in the judgement. As a bare minimum, given that the OTC/HSA report is being used as a basis for legislative action and the EPA report's status as a keystone document, it would surely be a reasonable expectation that the OTC/HSA report would address the ruling and criticisms even if only in the form of acknowledgement and rebuttal. It is a curious omission.

In terms of timing, the OTC/HSA report was prepared prior to conclusion of the EPA's appeal against to 1996 ruling. Both were issued in December 2002.

The PJ Carroll submission is strongly critical of the OTC report. People should read both, but I expect many will simply disregard a tobacco company's position as being irredeemably biased.

I've had some thoughts on the issue of bias in studies.

Bias may be suspected, however, it can only be demonstrated by a thorough analysis of the conduct of the study including the data collection/selection process and the procedures and methodologies used in it. Bias may exist, but the results may still stand up to critical scrutiny, or they may not. Equally, there might not be bias, but critical scrutiny could expose fundamental errors in the conduct of the study not picked up during peer review. Suspicion of bias alone is insufficient grounds to dismiss a study. Innocent till proven guilty.

The following comment is issue-independent and may (or may not) be applicable to many issues where the conventional wisdom may not be quite the “consensus” claimed.

It occurred to me that:

One the one hand, researchers who have a high expectation of having their work reviewed with hostility or dismissed as being biased due to affiliations with “vested interests” actually have a powerful external incentive to adhere to high standards to ensure that their work will withstand intense scrutiny.

On the other hand, researchers working within a dominant paradigm and who expect their results (or whose results are expected) to conform to the conventional wisdom, and who are likely to have a high expectation of having their work favourably received, are likely to be primarily dependent on internal motivation to adhere to high standards and may be vulnerable to “don't rock the boat” type influence/pressure.

Finally, one man's “vested interest” is another man's “concerned sponsor”. It just depends which side of the issue you're on.

Return to the current post to place your own comments,
and to view the response of an ETS expert in Ireland.

You can also write to blog-at-tallrite-dot-com
(Clumsy form of my address to thwart spamming software that scans for e-mail addresses)

Return to top of page

Now, for a little [Light Relief]

Hit Counter

Gift Idea
Cuddly Teddy Bears
looking for a home

Click for details

BLOGROLL

Blog-Irish (defunct)

Rolfe's Random Review

Jihad

Terrorism
Awareness Project

Religion

Iona Institute
Skeptical Bible

Leisure

~~Graham's Sporting Wk~~

Blog Directory

Eatonweb

Discover the World

Irish Times

My Columns

What I'm
currently reading

N E W !
Stiff: The Curious Lives of Human Cadavers,
by Mary Roach, 2003

It takes a strange person to write a book about cadavers, and even stranger to make it informative, interesting and - funny. Ms Roach is such a person.

The variety of what she discusses is astonishing -

The process of decay; dissection through the ages; severed heads for plastic surgeons to practice on; head transplant research; corpses for crash dummies, for crash evidence, for weapons practice; cannibalism; composting as an eco-friendly alternative to cremation.

To get a flavour, and to see how she makes this stuff both educational and amusing, have a look at her online article, reproduced in her book, about a dead human brain being
“a terrible thing to waste”.

Or her archive.

+++++

This is the definitive account of the most foul human being ever to have walked the earth. No other monster comes close - not Stalin, not Lenin, not Hitler, not Pol Pot, not Genghis Khan, not Ivan the Terrible.

The book is meticulously researched, magnificently structured, beautifully written - and drips innocent Chinese blood from almost every one of its 971 riveting pages.

Moa Tse Tung was obsessed with simply killing as many of his countrymen as he could by whatever means in order to maintain the remainder in such a permanent state of terror that the idea of turning on him would never even cross their wretched minds.

He also starved peasants in their hundreds of millions in order to confiscate the food they grew to pay the Soviets for a gargantuan armaments infrastructure.

Most terribly, Mao was absolutely right. He proved that terror is the most effective way of retaining power. Too many despots have tried to emulate him, but none with the same single-minded ferocity.

Disgustingly, people name restaurants in his honour

+++++

English historian
Charles Foley's
fascinating account
of an honourable man who introduced the concept of Special Forces to the German military during World War 2.

In that role, as Hitler's trusted operative, he recounts much derring-do, such as rescuing Mussolini from mountain top captivity, bluffing the then Hungarian strongman into surrendering, wreaking covert havoc on the Allied invasion of France.

Particularly moving is his account, from the German viewpoint,
of the invasion of the Soviet Union and
the stoic, stolid, suicidal resistance of the Russians.

This page-turner of a book concludes with a forecast of the role of Special Forces in future conflicts, which has turned out to be surprisingly prescient.

It was written in 1954.

+++++

The purpose of this
500-page novel is to present in graphic detail the horrors of living, fighting and - above all - dying in (and under)
the trenches during
the First World War.

It does so,
both commendably
and shockingly.
You certainly cannot come away with other than feelings of
deep admiration and sympathy for what those young men endured,
not to mention the distraught families at home, in their tens of thousands, when the dreaded news of their sons' demise arrived.

But the book is spoilt by the introduction of a storyline which is sentimental and distracting. Much of it is frankly boring. You might enjoy the sex which is detailed and graphic, but it's unnecessary.

Also, the interminable, repetitive description, going on for over 40 pages, of being
buried alive in a collapsed tunnel,
just ends up
being irritating.

About 200 pages should have been edited out.

+++++

Other books here

Click on the logo
to get a table with
the Rugby World Cup
scores, points and rankings.

After 48 crackling, compelling, captivating games, the new World Champions are, deservedly,
SOUTH AFRICA

England get the Silver,
Argentina the Bronze. Fourth is host nation France.

No-one can argue with
the justice of the outcomes

Over the competition,
the average
points per game = 52,
tries per game = 6.2,
minutes per try = 13

Click on the logo
to get a table with
the final World Cup
scores, points, rankings and goal-statistics