Tuesday, May 26, 2009

New Study: Big Blow to Smoking-Ban Claims, More

A major argument in promoting smoking bans is that they can reduce the occurrence of the heart attacks or heart disease that are allegedly caused by secondhand smoke. The new study is the most extensive ever done on this issue and utterly demolishes this claim.

The study analyzes heart attack hospital admissions and mortality before and after implementation of all state and local workplace smoking bans in the nation. The study was conducted by researchers from the RAND Corporation, the Congressional Budget Office, University of Wisconsin, and Stanford University. It covered 217,023 hospital admissions for heart attacks and 2.0 million heart attack deaths in 468 counties in all 50 states over an eight-year period.

Michael Seigel, MD, who is both a medical doctor and public health official, has over 20 years experience in tobacco research and currently teaches at the Boston University School of Public Health. He is adamantly opposed to smoking, but of the new study he says, “The most important finding of this study is that there are just as many smoking ban communities in which heart attack admissions and mortality have increased...as there are smoking ban communities in which heart attacks have decreased ....The mean difference was found to be zero.”

Furthermore, he adds: “The study also simulates the results from the comparison of all possible combinations of regions in the U.S. by examining subsets of the data....The authors are therefore able to simulate what the results would be for each of the 19,406 possible comparisons of intervention and control regions in the country....Most interestingly, the simulation of all possible comparison studies of local regions in the U.S. finds that just as many studies would find an increase in heart attacks associated with smoking bans as would find a decrease in heart attacks. The mean difference in heart attack admissions among all studies was 0. The exact same result was found for heart attack mortality.” In his words, the “main result” of the study is “that the smoking bans had no effect on either heart attack admissions or heart attack mortality.” To read his full explanation, go to http://tobaccoanalysis.blogspot.com/2009/04/new-study-of-national-heart-attack.html

This study is opposite to what the public has been led to believe, because anti-smoking advocates have been so effective in convincing the public—and state and local governments—of the “dangers” of secondhand smoke. Dr. Seigel has cited over a hundred anti-smoking groups, including the American Cancer Society, the American Lung Association and the American Heart Association, for “campaigns of deception.” His website, www.tobaccoanalysis.blogspot.com, details an astonishing array of scientific misrepresentations, outright lies and hypocrisy by anti-smoking groups. These tactics have proven effective in obtaining passage of smoking bans, even as they have become ever more shrill and absurd. That monumental lies have been instrumental in the passage of smoking bans is a measure of the gullibility and scientific illiteracy of the general public and elected officials. Of course, it is also a demonstration of the dishonesty of the smoking ban activists and the absence of genuine evidence for their cause.

Last year Dr. Siegel ran a Most Ridiculous Secondhand Smoke Claim Tournament. The American Cancer Society managed to make the Final Four in this liars contest with this whopper: “Immediate effects of secondhand smoke include cardiovascular problems such as damage to cell walls in the circulatory system, thickening of the blood and arteries, and arteriosclerosis (hardening of the arteries) or heart disease, increasing the chance of heart attack or stroke.” The U.S. Surgeon General's Office also figured in the contest with this entry: "Even brief exposure to secondhand smoke has immediate adverse effects on the cardiovascular system and increases risk for heart disease and lung cancer."

That claim by the SG was made after issuance of his 700-page report on secondhand smoke. But as Dr. Seigel said at the time, “There is absolutely no evidence to support this claim. Certainly, no evidence is presented in the Surgeon General's report to support this claim. And certainly, the Surgeon General's report draws no such conclusion.”

Now the new massive study emphatically demonstrates the falsity of the claim. It is revealing, too, that this comprehensive study was ignored by the same news media that were so quick to report on the dangers of secondhand smoke based on the small and seriously flawed Helena, Montana study. That study covered a laughably small sample in a single small community. It touted a 40 percent reduction in heart attacks—from an average of about 7 per month to 4—during a six-month period a smoking ban was in effect. But an almost identical drop occurred four year earlier when there was no smoking ban. Only one-third of the heart attack victims in the study had never smoked. And three-fourths of the restaurants in the community were already nonsmoking before the ban. The study contained several other serious shortcomings. Yet the evening news on CBS, ABC, and NBC all reported this study as a noteworthy indication of the effectiveness of smoking bans in protecting nonsmokers from the dangers of secondhand smoke.

Epidemiologists use “relative risk” (RR or Risk Ratio and informally including the similar Odds Ratio computation) as a means for measuring the severity of risk. The U.S. Surgeon General stated the relative risk for secondhand smoke is between 1.20 to 1.30. This is far below the minimum level at which any meaningful risk might be indicated. Both the World Health Organization and the National Cancer Institute have clearly stated that RRs below 2.0 are too low to be relied upon. The same is true of the federal Reference Manual on Scientific Evidence and textbooks such as Breslow and Day’s Statistical Methods in Cancer Research. A report by the independent health consulting firm Littlewood & Fennell characterizes RRs below 2.0 as “dancing on the tiny pinhead of statistical insignificance.” An article by Drs. Huber (pulmonary specialist), Brockie (cardiologist), and Mahajan (a hospital director of internal medicine and professor of medicine) states: “A strong relative risk would be reflected by a risk ratio of 5 to 20 or greater. Weak relative risks, by conventional definition, have risk ratios in the range of 1 to 3 or so....No matter how the data from all of the epidemiological studies are manipulated, recalculated, 'cooked,' or 'massaged,' the risk from exposure to spousal smoking and lung cancer remains weak. It may be 1.08 or it may be 1.34 or it may be 1.42, but all of those still represent a weak relative risk.”

RR of 2.0 is the bare minimum for a credible risk. Geoffrey Kabat, Senior Epidemiologist, Albert Einstein College of Medicine, says, “An association is generally considered weak if the odds ratio [relative risk] is under 3.0 and particularly when it is under 2.0.” David Hitt, author of Statistics 101, states: “As a rule of thumb, a RR of at least 2.0 is necessary to indicate a cause and effect relationship, and a RR of 3.0 is preferred.” John K. Sutherland, a health physicist with 40 years experience, has written: “Any scientist worth his qualifications knows that a RR of less than two or even three is unreliable and too shaky to place much credible reliance upon.” Dr. Marcia Angell, editor of the New England Journal of Medicine, one of the world’s leading medical journals, said, “As a general rule, we are looking for a relative risk of 3.0 or more before accepting a paper for publication.” Dr. Robt. Temple, director of drug evaluation for the FDA, said, “My basic rule is if the relative risk isn’t at least 3 or 4, forget it.” And EPA declined to regulate high-voltage power lines because it said the RRs “seldom exceeded 3.0” and anything less than 3.0 was just “statistical static.” Keep in mind, too, that the municipal tap water that tens of millions of Americans drink every day in thousands of cities across the U.S. has RRs 2.0 to 4.0. Yet we hear no reports of deaths from drinking these waters.

The 1992 report Revised Comments on the 1986 Surgeon General's Report...EPA...and NIOSH states: “Risk estimates below 2.0 or 3.0 are described as 'weak' and thus any conclusions drawn from them are unreliable.” The summary of this 47-page document concludes: “...these reports [Surgeon General's, EPA, and NIOSH] do not provide a defensible basis for regulation of smoking in the workplace.” Comments in the report are supported by 113 references in the scientific literature.

People unfamiliar with the term “relative risk” may easily fail to appreciate the magnitude of difference between RRs of 1.2 and 2.0. Just looking at those numbers, they may conclude that the latter is less than twice the risk of the former. In fact, however, a relative risk of 2.0 is five times greater than one of 1.2. This is because for relative risk, zero risk is set at 1.0, not 0.0. (RR below 1.0 indicates a negative risk, that is, a beneficial preventive effect rather than an increased risk.) A claim that a RR of 1.2 causes heart disease or lung cancer is based on mere one-fifth of the minimum required, while even at 2.0 and beyond, other factors in the research must be carefully considered. Instead, when the issue is ETS, alarmism over utterly meaningless results has been the chosen policy of anti-tobacco activists, including those in public health agencies.

The 2006 SG's report claims 46,000 deaths annually due to heart disease from secondhand smoke. But the American Heart Association website lists the following RRs for ETS: 1.25 for Cardiovascular disease, 1.18 for ischemic heart disease, and 1.13 for arrhythmic heart failure or coronary arrest mortality. None of these suggests credible risk. Death estimates are derived from relative risk. If a RR is meaningless, so are the estimates of deaths based upon it. So the big scary estimate of 46,000 deaths has no validity. It is simply a phony number put out to scare people and panic them into political action. If such death estimates were valid, the new study would not have found that smoking bans have zero effect on heart attack mortality.

The 2006 SG's report also claimed 3,000 deaths annually from lung cancer due to secondhand smoke. That figure came from the 1992 EPA study, which claimed secondhand smoke had a RR of 1.19 and that, the agency said, meant 3,000 lung cancer deaths annually. But the Congressional Research Service, at the request of the U.S. Congress, looked at the same data as EPA and concluded: “It is possible that very few or even no deaths can be attributed to ETS [environmental tobacco smoke].” Further, it stated that nonsmokers exposed to pack-a-day ETS every day for 40 years have “little or no risk of developing lung cancer”—much less dying from it (emphasis added). The CRS is part of the Library of Congress and has all the resources of that esteemed institution at its disposal. It is highly respected, nonpartisan, accepted by both Republicans and Democrats as fair and impartial, has no ties to tobacco companies, no regulatory or other agenda, and accepts no outside funding.

There's more. The U.S. House of Representatives then held a Congressional Investigation of EPA's findings. It concluded: “EPA could reach that conclusion [3,000 lung cancer deaths] only by ignoring or discounting major studies, and deviating from generally accepted scientific standards.” Further, it found EPA guilty of “conscious misuse of science and the scientific process to achieve a political agenda that could not otherwise be justified.” It also stated: “The agency [EPA] has deliberately abused and manipulated scientific data in order to reach a predetermined, politically motivated result.” (emphasis added.) Over the next seven years, five similar studies (meta-analyses) of secondhand tobacco smoke were performed by other researchers who, unlike EPA, followed correct scientific standards. The RRs of these studies showed a range of 0.98 to 1.03 and an average RR of 1.01, compared to EPA's RR of 1.19. Levois and Layard performed a meta-analysis of all the original studies utilized by EPA and came up with a RR of 1.00. Furthermore, these studies all had the standard 95% confidence level. The EPA study did not qualify for that. Instead, EPA used a degraded confidence level of only 90 percent, thus doubling the likelihood that its results were mere chance.

It should be pointed out, too, that EPA left out 19 earlier studies, including all studies with RRs less than 1.0, because they were counter to what EPA was trying to prove. This list included the Wu-Williams/Blot study which reported an important negative effect, rather than a positive one that EPA desired, for nonsmoking women married to smokers. The Littlewood & Fennell report to the National Toxicology Program Board of Scientific Counselors on Carcinogens states: “Inclusion of the Wu-Williams/Blot study in EPA's analysis would have forced EPA to reverse its conclusions about ETS and lung cancer.” Not surprisingly, EPA staff decided to exclude that study even though Dr. Blot was on the agency's Science Advisory Board.

When a study showed a RR of 1.5 for abortion and breast cancer, Dr. Eugenia Calle, director of analytic epidemiology for the American Cancer Society, was quick to state that a RR of 1.5 was too low to regard abortion as a risk factor for breast cancer. So why should we regard ETS, with a RR below 1.5 as a risk factor for cancer? Or anything else? Dr. Calle went on to say that RRs below 1.3 are too low even to be realistically identified.

In its own study, EPA violated not only many scientific standards but even basic honesty. It claimed various research studies by others were positive for ETS and lung cancer when the researchers themselves said they were not! Here are some examples:

“no evidence of an adverse effect of environmental tobacco smoke in the workplace.”—Janerich, et al, 1990

“the association with exposure to passive smoking at work was small and not statistically significant.” —Kalandidi, et al, 1990

“No association was observed [in women] between the risk of lung cancer and smoking of husband or passive exposure at work.” —Shimizu, et al, 1988

“no statistically significant increase in risk associated with exposure to environmental tobacco smoke at work or during social activities.”—Stockwell, et al, 1992.

The Littlewood & Fennell report noted: “These researchers carefully adhered to the sound principles of interpreting relative risks. The fact that their reports were deliberately misrepresented as 'positive' findings is an appalling corruption of accepted scientific and statistical standards.”

The brief quotes from research studies above give lie to the SG's claim “there is no safe level of exposure to secondhand smoke.” Of course, the SG's claim is refuted, too, by a significant number of studies reporting secondhand smoke having a preventive effect (RRs less than 1.0). These include a World Health Organization study that found children were 22 percent less likely to get lung cancer if both parents smoked than if neither did.

Two internal documents from EPA's Environmental Criteria Assessment Office, available under the Freedom of Information Act, state that the agency exaggerated the effects of secondhand smoke and that its conclusions were unwarranted. EPA was found guilty of violating six federal statutes for using harassment and intimidation to force employee support for its position on ETS—the same position that smoking-ban activists ever since have enthusiastically and uncritically supported. And the SG report continues to utilize that figure of 3,000 lung cancer deaths and treat it as though it were valid. What does that tell you about the value of the SG's report—and the integrity of the SG? What does it tell you about the integrity of the activist groups that use such phony numbers to promote smoking bans? —and about the politicians who vote for such bans even after they have seen the facade of lies demolished and the truth laid bare before them?