Tobacco Documents Online

AdTI ISSUE REPORT November 30, 1994 PC Cancer Risks by Cesar V. Conda One reason the American people distrust Washington is that the facts often take a back seat to politics. Government regulators - and their special interest allies - often exaggerate potential health risks to scare the public into accepting more aovernment • control over private decisions. But when it comes to "politically correct" causes. it seems that some activities aren't as dangerous as others, even though science tells us they pose the exact same health risks. Witness the response to the controversial epidemiological study (human population studies) linking induced abortion to greater risk of developing breast cancer. The study, published in the November 2 issue of the Journal of the NationaI Cancer Institute, found that on average those women who had abortions had between a 50 and 150 percent greater risk of getting breast cancer. depending on their age. Put another way, the • relative cancer risk from induced abortion is between 1.5 to 2.5. Epidemiologists and science reporters correctly down-played the study's findings because the risk factor is simply too small. and therefore difficult to determine whether there is a cause-and-effect relationship: "Epidemiological studies in general are probably not able, realistically, to identify with any conf dence any relative risks lower than 1.3. In that conte.YtL the 1.5 is a modest elevation compared to some other risk factors a L E X I S de TOCQUEVILLE r N! T f T U T I 0 N that we know cause disease," said Dr. Eugene Calle of the American Cancer Society. "Even if this risk turns out to be real or a 50 percent increase...it's not a huge risk," said Dr. Tim Johnson, medical editor of ABC's "Good Morning America." "When you have a truly causai link between something, the risk is much higher. For example, the risk between smoking and lung cancer. if you smoke the risk for lung cancer goes up 3,000 percent So a 50 percent risk, even if real, is very smaIl." "In epidemiologic research, relative risks of less than 2.0 are considered small and are usually difficult to interpret," according to the National Cancer Institute. "Differences in risk of 50 percent (relative risks of 1.5 percent) are small in epidemiological terms and severely challenges our ability to distinguish whether it reflects cause and effect or whether it simply reflects bias," said Lynn Rosenberg of the Boston University School of Medicine. Most serious scholars of risk assessment techniques view any relative risk under 3.0 to be weak. This is because epidemiology is "too crude to identify risks this smaiI." says Steve Milloy, author of a new Department of Energy study entitled Choices in Risk Assessment. As the table nearby shows, common foods we eat = ranging from fried meat to white bread - and everyday activities such as electric razor use - have The Alexis de Tocqueville Institudon 2000 15th Street North. Suite 501 Arlington. VA 222D1 Tel.(7Q3)351.49b9 FA%(783)351-)090

2 relative risk associations that are too small and too weak to cause concern. Why then are the Environmental Protection Agency (EPA) and the Occupational Health and Safety Agency (OSHA) on a crusade to ban smoking in public places when their own epidemiological studies put the relative health risk from secondhand smoke at only 1.19? Why does the American Cancer Societ,v, on one hand, claim that a 1.5 relative risk found in the recent abortion-breast cancer • study is a "modest elevation," but on the other hand. testify before ConQress that a 1.19 relative risk factor from passive smoking justifies a public smoking ban? The only reason for this discrepancy is politics: smoking and the tobacco industry are "politically incorrect." If one were to apply the same standards uniformly to breast cancer and second hand smoke studies, there would be little basis for the EPA's conclusions regarding secondhand smoke. More e.xcamples: To date, the EPA has decided not to regulate or ban the use of chlorine in drinking water and electromametic power lines, both of which • pose higher relative health risks than second hand smoke. A 1992 study in the American J~urnai of Public Health found that the relative risk of developing rectal cancer from drinking chlorinated water at 1.38, and a 1994 study in the Journal of Cancer Causes and Control calculated the relative risk of developing cancer of the nervous system from living near electric power lines at 1.89. The EPA should not regulate these activities because the risks are too small and the cost to society of doing so would be enormous. For instance. a complete ban of all industrial uses of chlorine would cost the U.S. economy up to $90 billion a year. Further, the EPA and OSHA had to manipulate the science to exaggerate the health risk from secondhand smoke. For example, the confidence interval - the standard procedure for estimating the possibility of error - was lowered from 95 to 90 by the EPA so that it could assert its findings on secondhand smoke were "statistically significant." Yet even if one uncritically accepts the EPA's conclusions, the resulting increase in risk is about the same as the lifetime risk of being killed while riding a bicycle. The point of all this is not to debate abortion or defend the tobacco industry but to highlight the fact that we should put potential health risks in perspective. The health risks from secondhand smoke are infinitesimal. And the cost of OSHA's proposed smoking ban for U.S. businesses could run as high as $8 billion a year. If government regulators win the battle to ban smoking in public places, they will be emboldened to regulate more private activities and decisions, leaving individuals with less freedom and fewer choices. In summary, the public should view the claims made by the EPA, OSHA, and so- called public interest groups about health risks with a healthy dose of skepticism. Fortunateiy, the new Republican Congress plans to enact tough legislation requiring government bureaucrats to perform more scientifically rigorous. risk and cost-benefit analysis before imposing new regulatorv burdens on the American people. The AleAs de Tocqueville Institution 2000 15th Street North. Suite 501 Ariington, VA 22201 Tei.(703)351-4969 FAX(703)351-0090

3 • Eatposure/ Relative Risk Reference - Condition Beef 1.5 Esophagus Cancer Yu et. ai., 1988 Fried BacontHam 2.0 White Bread 2.4 Whole Milk 7-1 Lung Cancer Mettlin. 1989 Frankfurter ~ 12 Mortality Kahn. etaL, 1984 Salt and Pepper ~ 1.2 " " Tap Water 1.5 Spontaneous Abortion Her¢-Picciono. etaL, 1989 Fried/Grilled Meat I 1.7 I Pancrratic Cancer Noreil, et al., 1986 Male Baldness ( 1.4 Heart Disease Lesko. et. aL, 1993 Cooku (1 CookiuDay) 1.49 Heart Disease Willett. et. al., 1993 Eiectric Razor Use 2.15 Letzketnia Lovely, et. aL, Reported in WSJ Pork Meat Intake 1 2.4 Lung Cancer Mestiin. est ai..1989 Motor Exhaust 1.5 Hayes, et. aL. 1989 Drinking Water (Chiorinated) 1.38 Cancer Morris. et aL, 1992 Eiecaomagnetic Fields I 1-3 Cancer Various Studies Physical Inactivity I 1.6 Lung Cancer Albanes. esL aL. 1989 # # Cesar V. Conda is Executive Director of the :Uesis de Tocqueviile Institution in Arlington, Va. Note: This renort does not necessuiivi reflect the views of the Alexis de Tocqueviile Insdnuion and its direcmrs nor is it to be understood as an anempt to aid or hinder the passage of any iegislation befote ConBt+ess. TTte Aleads de Tocqueville InstitUtion 2000 15th Strnt North. Suite :fl1 Arlington. VA 2ZZ01 TeL(70Ci)35I-4969 FA%(703)351-8090 Table t_ Weak Relative Risk Associations