Where's the Consensus on Secondhand Smoke?

Health Care News ^ | November 1, 2007 | Joseph Bast

[Eric Blair 2084]

More than a year has passed since U.S. Surgeon General Richard Carmona said, "The debate is over. The science is clear: Secondhand smoke is not a mere annoyance, but a serious health hazard."

At the time, Carmona released a seemingly impressive 727-page report on secondhand smoke, the introduction of which claims secondhand smoke killed approximately 50,000 nonsmoking adults and children in 2005.

Carmona's report stated the new orthodoxy in the anti-smoking establishment: There is a "consensus" on the dangers of secondhand smoke. But did his report actually make the case?

Junk Science and Courtrooms

Understanding Carmona's report requires familiarity with a different report--the Federal Judicial Center's 2000 "Reference Manual on Scientific Evidence, Second Edition," the official guide for judges to understand and rule on science introduced in courtrooms.

According to the manual, nearly all the studies cited in Carmona's report wouldn't pass muster in a court of law because they are observational studies, the sample sizes are too small, or the effects they show are too negligible to be reliable.

For example, the Reference Manual states, "the threshold for concluding that an agent was more likely than not the cause of an individual's disease is a relative risk greater than 2.0." Few of the studies Carmona cites found relative risks this large, and most found risks in a range that included 1.0, which means exposure to secondhand smoke had no effect on the incidence of disease. In the world of real science, that's a knockout blow.

Most of the research Carmona cites was rejected by a federal judge in 1993, when the Environmental Protection Agency (EPA) first tried to classify secondhand smoke as a human carcinogen. The judge said EPA cherry-picked studies to support its position, misrepresented the most important findings, and failed to honor scientific standards. Carmona's report relies on the same studies and makes the same claims EPA did a decade ago.

Missing Study

Did Carmona and coauthors cherry-pick the data? Absolutely. They ignore the largest and most credible study ever conducted on spouses of smokers, by Enstrom and Kabat, published in the May 12, 2003 issue of the British Medical Journal. The authors found:

"The results do not support a causal relationship between environmental tobacco smoke and tobacco-related mortality. The association between tobacco smoke and coronary heart disease and lung cancer may be considerably weaker than generally believed."

Carmona mentions the Enstrom study just once, in an appendix listing studies too recent to include in the report. But Enstrom's study was published four years ago, and Carmona cites more recent studies. In fact, Carmona's principal "findings" were taken from a 2005 report--not a scientific study, merely another report--from California's Clean Air Resources Board, mostly citing the very studies the federal judge rejected in 1993.

Additional Confirmation

The Enstrom study isn't the odd exception among all the available studies on secondhand smoke. A 2002 analysis of 48 studies, also published in the British Medical Journal, found only seven showed a relationship between secondhand smoke exposure and lung cancer, while 41 did not.

A 1998 World Health Organization (WHO) study covering seven countries over seven years actually showed a statistically significant reduced risk for children of smokers and no increase for spouses and coworkers of smokers.

False Findings

No one is saying being around smokers is good for kids' health. The WHO study simply shows the largest and longest studies on secondhand smoke are most likely to find no effects.

There is a reason for this. In an August 2005 essay in PloS Medicine, Tufts University epidemiologist John Ioannidis explains:

"There is increasing concern that in modern research, false findings may be the majority or even the vast majority of published research claims. However, this should not be surprising. It can be proven that most claimed research findings are false."

Ioannidis writes that when tens of thousands of researchers are conducting thousands of small and short-term epidemiological studies, all of them seeking to find evidence of a small or nonexistent effect, and when academic journals are predisposed to publish studies claiming positive correlations (no matter how small) that support the conventional wisdom, the result is that "most published research findings are false."

Who's Claiming Consensus?

Far from being the last word on the health effects of secondhand smoke, Carmona's report and its uncritical acceptance by frequent commentators on smoking raise questions about bias, error, and the deliberate orchestration of public opinion. The commentators who echo the Surgeon General's claim fall into one or more of five groups:

Liberal advocacy groups such as the Center for Tobacco Free Kids, American Cancer Society, and American Legacy Foundation, which clearly profit from increased public attention to secondhand smoke.

Government agencies, including the Office of the Surgeon General, the Department of Health and Human Services, and EPA, which exist largely for the purpose of discovering and publicizing health risks, even if they are backed by dubious research.

Some corporations--notably Johnson & Johnson, which makes smoking-cessation aids--which give liberal advocacy groups hundreds of millions of dollars to demonize smoking and compel more consumers to use their products.

The news media, which simply publish the news releases from the first three groups.

Politicians, who read the newspaper stories and hear from the advocacy groups and rationally calculate their odds of being reelected improve if they proclaim deep concern over secondhand smoke and propose solutions that will cost taxpayers and consumers billions of dollars annually.

Heavy-Handed Government

The idea that smokers and nonsmokers might solve this problem voluntarily is dismissed out of hand by those who claim secondhand-smoke exposure is a public health crisis. The "solutions" they want all require bigger government: higher taxes on cigarettes, bans on smoking in public, restrictions on advertising and health claims, etc.

Oddly, these solutions all work to advance the self-interest and agendas of the five groups that repeat Carmona's claim of "consensus." What are the odds this correlation is coincidental?

[Eric Blair 2084]

the Reference Manual states, "the threshold for concluding that an agent was more likely than not the cause of an individual's disease is a relative risk greater than 2.0."

[Eric Blair 2084]

Big Brother, Nanny State, Grant Junky, Bogus Study, Junk Science, Social Engineering, Sheeple Control, Health Nazi

This is an extremely low volume ping list. 6 per week max. To be added to or deleted from this ping list, please click one of the follwing:

Eric, you are a genius add me to your ping list or Eric, you are a jackass, take me off this ping list

To request that I stop using these gay html colors and fonts, please send me a private message below.

____________________________________________________________________

Constitution of the Left Wing Nanny State of Liberals:

Picture courtesy of unixfox. All rights reserved. Copyright MMVII. Any use of the pictures descriptions or accounts of this ping without the express written consent of unixfox, Eric Blair, or Major League Baseball is strictly prohibited. Some restrictions apply. Ping not available in all states. For erections lasting longer than four hours, call the Guiness Book of World Records. Use only as directed.

Preamble:

We the People Sheeple of the United States Nanny State, in Order to form a more perfect Union Socialist Utopia, establish Justice Socially engineer a country of non smoking, physically fit, tea totallers, insure domestic Tranquility Smoking bans in bars, limits on unhealthy food and social drinking, provide for the common defense Universal Healthcare, promote the general Welfare health of the population whether they like it or not, in order to save above mentioned Universal Healthcare entitlement program from bankruptcy, and secure the Blessings of Liberty Dependency to ourselves progressive liberals and our Posterity Hitler Youth who we brainwash through public school education, do ordain decree and establish this Constitution for the United States of America. Nanny State of Liberals

[johnthebaptistmoore]

I certainly do hope that FR assiduously attacks Mike Huckabee often for his “pro-nanny politics”! There are still too many conservatives who don’t even know about his wanting a national ban on smoking. It’s bad enough that Mike Huckabee increased taxes several times while Governor of Arkansas and is weak on truly dealing with the illegal immigration issue. Huckabee also practicing nanny politics makes things even worse for him as a possible GOP Presidential candidate.

[TigersEye]

I remember when scientific consensus said oatmeal lowered cholesterol.

[Eric Blair 2084]

You’re right. Huckster is a nanny state liberal Democrat. He has only two positions that are conservative: 2nd A, pro life.

I get sick when I see MSM opining about how this liberal theocrat is the ideal GOP candidate for the “religious right” party.

[Eric Blair 2084]

As Rush is fond of saying re Gorebull Warning “Consensus is NOT Science”.

[TigersEye]

Huckabee might be a concern if he had a snowball’s chance in hade. But it doesn’t hurt to inform people of his big gov tendencies.

[TigersEye]

I remember when scientific consensus said that salt caused high blood pressure.

[mysterio]

The smoking war is lost. Americans have decided that they hate freedom and property rights.

[Eric Blair 2084]

We are trying to get a Huckster ping list together. FReepmail ob if if you want on.

That’s 3 of us! Anybody else want to be on the Huckabee Nanny State list?

[Eric Blair 2084]

I know how you feel. I feel like giving up sometimes too.

Viva la Resistance!

[wardaddy]

I despise smoke nazis with my every fiber.

They epitomize liberal do-goodery

..and no I don’t smoke.

[Eric Blair 2084]

I’ll pretend I didn’t hear from you. It will ruin my enjoyment and Schadenfreude when I see other unsuspecting citizens who pissed on smokers herded into box cars.

[Eric Blair 2084]

Check out how many SHS studies have a RR < 2.0 much less 3.0

Reference Manual on Scientific Evidence, Second Edition :

"the threshold for concluding that an agent was more likely than not the cause of an individual's disease is a relative risk greater than 2.0."

TABLE I

EPIDEMIOLOGICAL STUDIES RELATING TO LUNG CANCER AMONG NONSMOKERS MARRIED TO SMOKERS

Author	Year	Location	Sex of the subject	Number of lung cancers	Average Relative Risk	Relative Risk fluctuation (min/max) (95% confidence interval)
Garfinkel 1	1981	USA	F	153	1.18	(0.90 - 1.54)
Chan	1982	Hong Kong	F	84	0.75	(0.43 - 1.30)
Correa	1983	USA	F M	22 8	2.07 1.97	(0.81 - 5.25) (0.38-10.32)
Trichopoulos	1983	Greece	F	77	2.08	(1.20-3.59)
Buffler	1984	USA	F M	41 11	0.80 0.51	(0.34-1.90) (0.14-1.79)
Hiramaya	1984	Japan	F M	200 64	1.45 2.24	(1.02-2.08) (1.19-4.22)
Kabat 1	1984	USA	F M	24 12	0.79 1.00	(0.25-2.45) (0.20-5.07)
Garfinkel 2	1985	USA	F	134	1.23	(0.81-1.87)
Lam W	1985	Hong Kong	F	60	2.01	(1.09-3.72)
Wu	1985	USA	F	29	1.20	(0.50-3.30)
Akiba	1986	Japan	F M	94 19	1.50 1.80	(0.90-2.80) (0.40-7.00)
Lee	1986	UK	F M	32 15	1.00 1.30	(0.37-2.71) (0.38-4.39)
Brownson 1	1987	USA	F	19	1.68	(0.39-6.90)
Gao	1987	China	F	246	1.19	(0.82-1.73)
Humble	1987	USA	F M	20 8	2.20 4.82	(0.80-6.60) (0.63-36.56)
Koo	1987	Hong Kong	F	86	1.64	(0.87-3.09)
Lam T	1987	Hong Kong	F	199	1.65	(1.16-2.35)
Pershagen	1987	Sweden	F	70	1.20	(0.70-2.10)
Butler	1988	USA	F	8	2.02	(0.48-8.56)
Geng	1988	China	F	54	2.16	(1.08-4.29)
Inoue	1988	Japan	F	22	2.25	(0.80-8.80)
Shimizu	1988	Japan	F	90	1.08	(0.64-1.82)
Choi	1989	Korea	F M	75 13	1.63 2.73	(0.92-2.87) (0.49-15.21)
Hole	1989	Scotland	F M	6 3	1.89 3.52	(0.22-16.12) (0.32-38.65)
Svensson	1989	Sweden	F	34	1.26	(0.57-2.81)
Janeric	1990	USA	F M	144 44	0.75 0.75	(0.47-1.20) (0.31-1.78)
Kalandidi	1990	Greece	F	90	2.11	(1.09-4.08)
Sobue	1990	Japan	F	144	1.13	(0.78-1.63)
Wu-Williams	1990	China	F	417	0.70	(0.60-0.90)
Liu Z	1991	China	F	54	0.77	(0.30-1.96)
Brownson 2	1992	USA	F	431	1.00	(0.80-1.20)
Stockwell	1992	USA	F	62	1.60	(0.80-3.00)
Liu Q	1993	China	F	38	1.66	(0.73-3.78)
Du	1993	China	F	75	1.09	(0.64-1.85)
Fontham	1994	USA	F	651	1.29	(1.04-1.60)
Layard	1994	USA	F M	39 21	0.58 1.47	(0.30-1.13) (0.55-3.94)
Zaridze	1994	Russia	F	162	1.66	(1.12-2.46)
Kabat 2	1995	USA	F M	67 39	1.08 1.60	(0.60-1.94) (0.67-3.82)
Schwartz	1996	USA	F M	185 72	1.10 1.10	(0.72-1.68) (0.60-2.03)
Sun	1996	China	F	230	1.16	(0.80-1.69)
Wang S-Y	1996	China	F	82	2.53	(1.26-5.10)
Wang T-J	1996	China	F	135	1.11	(0.67-1.84)
Cardenas	1997	USA	F M	150 97	1.20 1.10	(0.80-1.60) (0.60-1.80)
Jöckel-BIPS	1997	Germany	F M	53 18	1.58 1.58	(0.74-3.38) (0.52-4.81)
Jöckel-GSF	1997	Germany	F M	242 62	0.93 0.93	(0.66-1.31) (0.52-1.67)
Ko	1997	Taiwan	F	105	1.30	(0.70-2.50)
Nyberg	1997	Sweden	F M	89 35	1.20 1.20	(0.74-1.94) (0.57-2.55)

The data in this table were obtained from the studies listed. In the Swartz (1996), Jöckel-BIPS (1997) and Nyberg (1997) studies, relative risk and confidence interval data were reported for the sexes combined. These data were separated based on the respective number of cases by sex, assuming the same relative risk for each sex.

TABLE II

EPIDEMIOLOGICAL STUDIES RELATING TO LUNG CANCER AMONG NONSMOKERS REPORTEDLY EXPOSED TO ETS IN THE WORKPLACE

Author	Year	Location	Sex of the subject	Average Relative Risk	Relative Risk fluctuation (min/max) (95% confidence interval)
Kabat 1	1984	USA	F M	0.68 3.27	(0.32-1.47) (1.01-10.62)
Garfinkel 2	1985	USA	F	0.93	(0.55-1.55)
Wu	1985	USA	F	1.30	(0.50-3.30)
Lee	1986	UK	F M	0.63 1.61	(0.17-2.33) (0.39-6.60)
Koo	1987	Hong Kong	F	1.19	(0.48-2.95)
Shimizu	1988	Japan	F	1.18	(0.70-2.01)
Janerich	1990	USA	F & M	0.91	(0.80-1.04)
Kalandidi	1990	Greece	F	1.70	(0.69-4.18)
Wu-Williams	1990	China	F	1.10	(0.90-1.60)
Brownson 2	1992	USA	F	0.79	(0.61-1.03)
Stockwell	1992	USA	F	no statistically	significant association
Fontham	1994	USA	F	1.39	(1.11-1.74)
Zaridze	1994	Russia	F	1.23	(0.74-2.06)
Kabat 2	1995	USA	F M	1.15 1.02	(0.62-2.13) (0.50-2.09)
Schwartz	1996	USA	F & M	1.50	(1.00-2.20)
Sun	1996	China	F	1.38	(0.94-2.04)
Wang T-J	1996	China	F	0.89	(0.46-1.73)
Jöckel-BIPS	1997	Germany	F & M	2.37	(1.02-5.48)
Jöckel-GSF	1997	Germany	F & M	1.51	(0.95-2.40)
Ko	1997	Taiwan	F	1.10	(0.40-3.00)
Nyberg	1997	Sweden	F & M	1.60	(0.90-2.90)

The data in this table were obtained from the studies listed.

TABLE III

EPIDEMIOLOGICAL STUDIES RELATING TO LUNG CANCER AMONG NON-SMOKERS REPORTEDLY EXPOSED TO ETS IN CHILDHOOD

Author	Year	Location	Sex of the subject	Average Relative Risk	Relative Risk fluctuation (min/max) (95% confidence interval)
Correa	1983	USA	F	no statistically	significant association
Garfinkel 2	1985	USA	F	0.91	(0.74-1.12)
Wu	1985	USA	F	0.60	(0.20-1.12)
Akiba	1986	Japan	F & M	no statistically	significant association
Gao	1987	China	F	1.10	(0.70-1.70)
Koo	1987	Hong Kong	F	0.55	(0.17-1.77)
Pershagen	1987	Sweden	F	1.00	(0.40-2.30)
Svenson	1989	Sweden	F	3.30	(0.50-18.80)
Janarich	1990	USA	F & M	1.30	(0.85-2.00)
Sobue	1990	Japan	F	1.28	(0.71-2.31)
Wu-Williams	1990	China	F	0.85	(0.65-1.12)
Brownson 2	1992	USA	F	0.80	(0.60-1.10)
Stockwell	1992	USA	F	1.70	(1.00-2.90)
Fontham	1994	USA	F	0.89	(0.72-1.10)
Zaridze	1994	Russia	F	0.98	(0.66-1.45)
Kabat 2	1995	USA	F	1.63	(0.91-2.92)
Sun	1996	China	F	2.29	(1.56-3.37)
Wang T-J	1996	China	F	0.91	(0.56-1.48)
Jöckel-BIPS	1997	Germany	F & M	1.05	(0.50-2.22)
Jöckel-GSF	1997	Germany	F & M	0.95	(0.64-1.40)
Ko	1997	Taiwan	F	0.80	(0.40-1.60)

The data in this table were obtained from the studies listed.

TABLE IV

EPIDEMIOLOGICAL STUDIES RELATING TO LUNG CANCER AMONG NON-SMOKERS REPORTEDLY EXPOSED TO ETS IN NON-HOME/NON-WORKPLACE SETTINGS

Author	Year	Location	Sex of the subject	Average Relative Risk	Relative Risk fluctuation (min/max) (95% confidence interval)
Garfinkel 2	1985	USA	F	1.42	(0.75-2.70)
Lee	1986	UK	F M	0.61 1.55	(0.29-1.28) (0.40-6.02)
Janerich	1990	USA	F & M	0.59	(0.43-0.81)
Stockwell	1992	USA	F	no statistically	significant association
Fontham	1994	USA	F	1.50	(1.19-1.89)
Kabat 2	1995	USA	F M	1.22 1.39	(0.69-2.15) (0.67-2.86)

The data in this table were obtained from the studies listed.

[xjcsa]

Ping for later.

[Tunehead54]

"A 1998 World Health Organization (WHO) study covering seven countries over seven years actually showed a statistically significant reduced risk for children of smokers and no increase for spouses and coworkers of smokers."

Rush regularly mentions this report but the times that I've tried I have not found a link to this report at Rush's site or otherwise. Anyone?

[metesky]

Nice find, Eric. Simple enough for even the most obtuse to understand.

[libertarian27]

Thanks for the Ping, Great article.

You’ll like this:
http://www.epi-perspectives.com/content/4/1/12/abstract

[wardaddy]

I’ll pretend I didn’t hear from you

Why don't you pretend to read my post again son, I'm on your side doofuss.

[Madame Dufarge]

Very, very good.

But doesn't "stinky" trump fact?

I get so confused.