First, I would like to remind you that the main justification for smoking bans, which are sometimes called clean-indoor acts, is supposedly to protect people from unhealthy air. I have 2 main criticisms ..First, if the intent is to protect nonsmokers from unhealthy air, then why don't these laws ever include radon limitations, regulate toxic mold, or prohibit people contagious with airborne illnesses from entering workplaces? And why don't they ban the use of fireplaces in hospitality establishments?

Radon kills approximately 21,000 and the flu approximately 35,000 each year; when other airborne infectious illnesses are included, this total easily rises to over 100,000 each year. Yet, the EPA has no regulations forcing owners of hospitality establishments to take steps if radon levels are high, and neither does Missouri, nor any locality in MO which currently has a smoking ban. Anyone who is ill, with an infectious disease, but can drag themselves out of bed, can lawfully enter any hospitality establishment, workplace, or public place anywhere in the USA; so while these antismoking groups attempt to justify their activism as in the interest of public health then why the narrow focus? if their motivations are pure, then they should be more concerned about radon, toxic mold, and airborne illness than secondhand smoke, because even by their own statistics, these are more dangerous.

Second, can any of you think of any other health laws which have emerged this way? Did we have activists push for real health laws at local levels and then try to force states to adopt them? Were real health laws ever put up on ballot referendums? did we get to vote on whether employees in hospitality establishments must wash hands after using restrooms, before returning to work? Of course not... if these smoking bans were real health laws, then they would be incorporated into our health codes just like real health laws are. The fact that smoking bans do not emerge as do real health laws leads me to conclude they really are not health laws to begin with....the real intent is to harass and denormalize smokers and those who wish to do business with smokers...nothing more.

Further proof that the real intent of smoking bans is purely to harass and denormalize smokers, can be garnered from recent efforts to ban outdoor smoking, where there are not even any questionable studies which claim it is harmful to nonsmokers. Examples include Bush Stadium, the Botanical Gardens, the St. Louis Zoo, and the Missouri Dept. of Health has banned outdoor smoking on all hospital campuses within the state. These policies are implemented to denormalize public smoking with the ultimate goal to get the public to accept smoking bans in indoor private property more readily. Right now, a smoker on a public sidewalk in Calabasas CA is required by law to ask anyone he gets within 20 feet of, if it's OK to continue smoking. The Belmont CA city council is considering a ban on all smoking inside attached housing, and all outdoor areas including private yards of single family homes, leaving the indoor areas of single family homes as the only place within the entire city where is legal to smoke. Most of you probably already know that outdoor smoking bans at parks, beaches, college campuses, etc., are being passed with increasing frequency throughout the USA. I have contacted Governor Blundt's office to determine if he has a position on the issue of outdoor smoking bans, and to urge him to prohibit outdoor smoking bans at any public facility supported by public funding. As of yet he has not responded. I ask all of you to support me in this effort to reverse outdoor smoking bans because the only possible intent is to harass and denormalize smokers.

Today, I wish to discuss the scientific evidence antismokers use to persuade lawmakers to pass smoking bans. You will see that all of this is really nothing more than selective research and then, selective reporting of research by the media. Local and state lawmakers simply are not aware that there is much evidence against the hypothesis that secondhand smoke harms nonsmokers. This is the main reason bans get passed. My webpage has a subsection where many studies, and public statements by professional epidemiologists debunk these secondhand smoke claims. It is also a rich resource of studies and reports that smoking bans do indeed cause economic harm to hospitality establishments. New information is constantly being added to my webpage, and I hope you will all use it as a resource of information to fight smoking bans. I have printed out cards with web addresses to my webpage, and other Internet resources where you can get more information to fight smoking bans. I will also be available throughout this conference and this evening to answer any questions you have, or to show you some of the resources I have brought with me.

But first a little history:

Our peak year of active adult smoking was 1953, which was the year scientific evidence that smoking causes lung cancer began to be accepted by the public. In every year since, adult smoking rates have declined in this country, but that decline slowed in the mid 1980s. For this reason, public health specialists began to look for other ways to get this decline back on track. I brought with me a book containing proceedings of a tobacco symposium conducted by the International Agency for Research of Cancer 1986, in which it states that of the studies conducted to date, the results are consistent with either a small increased health risk from secondhand smoke, or no health risk at all. and in another chapter, that legislation and political activity will have to be implemented to eradicate smoking from our society. These include restrictions on smoking and establishing "nonsmoking as the norm." I also brought with me, a Missouri Dept. of Health booklet titled Towards a Tobacco-free Missouri which states in part, that to stop smoking they must get the media on their side to accomplish the mission: (to envoke) a tobacco free environment as the norm; tobacco control policies; and comprehensive tobacco control programs. In other words, "we are going to harass smokers until they give up their nasty little habit." And your tax dollars and mine are being used to accomplish this goal.

In 1986, the first surgeon general's report on secondhand smoke was released, but this report did allow for the possibility that it's findings could be explained by other factors. Six of the 13 studies completed by 1986 found a statistically positive relationship, but this is only half of the studies. It also reported that it took decades of exposure to harm nonsmokers and recommended that nonsmokers be given the option to avoid frequent exposure. (CDC) As a consequence, governments at the state and local levels began to pass laws requiring most workplaces to provide the majority of their space as nonsmoking. But these policies did enable smokers to continue smoking indoors, and failed to re-ignite the decline in smoking rates experienced from the 1950s to the 1980s.

Towards the end of the 1980s and since, secondhand smoke studies finding health effects have become more common that those which do not, by about a 2:1 margin. These studies resulted in the 1992 release of the report by the EPA claiming that secondhand smoke is a class A carcinogen in nonsmokers, and this started the chain of events which lead to some states and many localities banning indoor smoking or severely restricting it. (EPA)

In 1995, Library of Congress biostaticians reevaluated the EPA report and concluded that it was too scientifically flawed to justify declaring secondhand smoke a class A carcinogen. (University of Maryland)

In 1998 a federal judge ruled that the EPA report cherry-picked the science and vacated most of it's conclusions. (Heartland) The EPA appealed and in 2002, the appellate court ruled that the EPA, like any other government agency should be free to release reports, even if defective, that a ruling against the report would unleash a frenzy of litigation against many government agency reports, and that since the EPA did not use the report to attempt to regulate smoking, that the report did not cause any harm to the plaintiffs. (LP)

The Clinton administration had asked OSHA to look into banning smoking in all indoor workplaces, but they never took action, and in 2003, they announced that their own evaluation of the evidence did not justify banning indoor smoking because the levels were just too low, and that further, they had no general duty to do so because the spectrum of published studies were too inconsistent and statistically weak to act under the general duty clause. (OSHA) This prompted antismoking groups to again turn up the heat on state and local governments to ban smoking. They also began to develop the tactic of using ballot referendums to enact smoking bans.

Also, in 2003, the first report of a large decline in heart attacks after implementation of a smoking ban in Helena, MT ( 40%) caused a resurgence of smoking ban efforts throughout the USA. However, that report has been widely criticized because Helena is a small town, and similar fluctuations had occurred before, apparently without cause; the ban was widely violated because it was contrary to state law therefore public secondhand smoke exposure did not actually decline much; and lastly, this whole heart attack decline was actually limited to the first 3 months of the ban, when even the health dept. was not enforcing it, and the heart attack rate returned to normal in the second 3 months when they did attempt enforcement. We looked at states' heart attack rates before and after statewide smoking bans to see if they had heart attack declines like Helena did. We compared these trends to the USA and states without smoking bans, and found no difference in heart attack rates. We believe the Helena study is just one more example of selective research which is widely used by antismoking groups to mislead lawmakers and promote bans.

In 2003, epidemiologist James Enstrom published a huge cohort study of 35,000 nonsmoking Californians which found that those living with smokers had the same heart disease and lung cancer rates as those who did not. This is the largest and most complete study ever published. He was immediately attacked, both personally and ethically by antismoking activists who claimed he received funding from the tobacco industry. However, upon further analysis, it turns out that the funding used to gather the data was actually supplied by the American Cancer Society, funding to study the data was supplied by California tobacco tax revenue, but that funding was canceled when they found out his study would report no risk. It was only then, that he obtained tobacco industry funding to support himself while preparing the manuscript and awaiting publication. In 1997, Dr. Geof Givens, a mathematics professor at the University of Colorado, had performed a mathematical analysis of the all secondhand smoke studies reported to that date. He found from the distribution of these data, that it was highly likely many secondhand smoke studies reporting no risk had never been published. Considering the experience of Dr. Enstrom, it appears Givens is likely correct. Dr. Enstrom now has a website called scientificintegrityinstitute, which you can visit. There, he describes other persecution tactics antismokers used to stop publication of his secondhand smoke study and a complete record of personal attacks on him.

The last big development was last year's release of the new Surgeon General's report which restates the claims made in the EPA report, and is the first official govt. report to insist that secondhand smoke causes heart disease in nonsmokers. It should be mentioned that the surgeon general's press release claims that there is no safe level of smoke, and that the effects can be immediate, but there is insufficient supporting evidence in the actual report itself to back up these claims. Three weeks later, the surgeon general resigned and we do not really know why, at this time. However, this report, or more accurately, the press releases, which are not under govt. peer review or control, have precipitated a smoking ban frenzy which lasts to this day. I have a criticism of the surgeon general's report on my webpage and links to 2 other criticisms of the surgeon general's report, one by Dr. Enstrom, which points out that his own study was not included in the report, ( again selective research) and if it had, the most that could have been concluded is that the risks are small. The second criticism is to an article by Dr. Gio Gori, which was published in Regulation Magazine. This is the most extensive criticism I know of. He stated that the surgeon general's public statements are not supported by the contents of the report itself; that the studies he used did not actually measure secondhand smoke exposure of the test cases; that most of the studies he used did not show a statistical link to disease, and that the report did not control for other causes of these diseases.

Now to the scientific evidence:

I have with me an epidemiology textbook, published in 1967 which teaches students how to go about collecting epidemiological evidence and the proper interpretation of that evidence. In summary, there are many statistical and logistical flaws in all the arguments used to claim secondhand smoke causes disease in nonsmokers.

First, there are 2 kinds of epidemiological studies. Case-controlled, which are also called retrospective studies, and cohort studies, which are also called prospective studies. In case controlled studies of secondhand smoke, epidemiologists interview nonsmokers with diseases like lung cancer or heart disease, and ask them about their lifetime exposure to secondhand smoke, including if their spouse smokes, or if they were exposed a lot of smoke in other places. They might also test their bodily fluids for traces of secondhand smoke constituents. These studies are little more than surveys, and have all the fallacies typical of surveys. They are highly subject to recall bias by the nonsmokers, and interpretation bias by the epidemiologists. My book teaches students that the best use of these studies is to use them to decide if it is worth the time and expense to conduct cohort studies.

Now, what's a cohort study? In these, healthy enrollees are asked about their exposure to a risk, in this case secondhand smoke, and are followed up over many years to determine if more of the exposed enrollees developed the disease than the unexposed group. My book teaches that cohort studies are more expensive, and time consuming, but more likely to have fewer problems with bias. Now, and this is important, most cohort studies do not find a link between secondhand smoke and disease in nonsmokers. In fact the biggest, and most complete study ever done, which was the one by Dr. Enstrom, did not find any link between secondhand smoke exposure and lung cancer or heart disease in nonsmokers. Another study, which combined all the results of all previous cohort studies of heart disease, found a 5% elevated risk, which is within statistical probability of no risk at all. (Scientific Integrity Institute

But my textbook also teaches the following.... After you have completed your study, you do not have proof of a risk... you have a hypothesis; unless all of the following are true:
• It is very plausible that the risk mechanism is real, for example, since smokers develop more lung cancer and heart disease, it is plausible exposed nonsmokers will too.
• The percentage of exposed individuals who developed/or have the disease is more than 2-3 times the percentage of unexposed individuals who developed/or have the disease. ( This principle is based on overall experience that any kind of epidemiological study typically does not turn out to be true if the results are less than 2)
• all the other studies which have been reported find the same relative magnitude of risk you did.
• there is no other known cause of the disease.
• you have made sure that both the exposed and unexposed groups have the same exposure to other causes of the disease

As I said, this epidemiology textbook was published in 1967. Apparently today, epidemiologists are not being trained to adhere to these principles as much as in the past. As a result, we are constantly hearing about studies such as good-fat, bad-fat studies reversing previous conclusions, the same for vitamin E, and a whole host of other risk factors. About 10 years ago, researchers became concerned that increasing use hormone replacement therapy could be causing the increased prevalence of breast cancer. As a result, physicians began lowering the doses of these to their patients, and some discontinued treatment of their patents except in the most severe cases of need. Recent reports indicate, that after decades of increasing incidence of breast cancer, this incidence is beginning to decline. This helps support the hypothesis that HRT is a risk factor for breast cancer. This is how epidemiology is supposed to work. In the case of secondhand smoke exposure, however, despite decades of lowering exposure throughout the USA, nonsmokers' incidence of lung cancer had remained steady. But within the last few years, newer reports indicate that incidence of lung cancer in nonsmokers in Canada is beginning to increase (MSN), and similar but less extensive reports show the same thing in the USA (NIH and Post Gazette) This should be a wake-up call that strongly questions if these studies finding risk are true.

Incidentally, the US Centers for Disease Control states that 70% of all secondhand smoke exposure has been eliminated since 1988. This is based on nicotine measurements in nonsmokers. Yet antismoking activists still claim secondhand smoke kills 50,000 nonsmokers each year. I have a report with me from the Congressional Research Service which traced the claim secondhand smoke kills 50,000 Americans each year all the way back to an article published in Environmental International in 1988. The CRS reported that one of the three journal referees recommended the article be rejected because it was too speculative, the CRS also criticized the methods used in that report extensively. Yet that claim is still being made today. In fact, it is the source of claims made by state health depts in Missouri and Illinois... that secondhand smoke is killing 1500 Missourians and 2900 Illinoisans each year. All they did is take the Environmental International report, and prorate it's conclusions to the current populations of these states. Accordingly these state-specific estimates are just as speculative as the original claim blasted by the CRS.

Now, moving on the 1992 EPA secondhand smoke report. Remember I told you cohort studies are more reliable that case-controlled studies? Well, most of the 31 studies the EPA used to conclude secondhand smoke causes lung cancer in nonsmokers were case-controlled. Further, the EPA only used residential exposure studies, but it's findings are being used to justify workplace smoking bans. According to the 1995 Congressional Research Service report criticizing the EPA report and the proposed rules from OSHA, studies of workplace exposure and heart disease conclude that the overall risk is nonexistent, and I read directly from that CRS report:

Another example of selective research in the EPA report, is that they actually gave the studies, which did find risk, more statistical power than those which did not, when calculating the overall conclusion, from the 31 residential studies.

The EPA and the surgeon general's report try to do an "end run" around the criteria set forth in my text book which I told you about. For example, the EPA did consider the possibility that other causes were responsible for the results in some studies. The EPA negated the possibility that publication bias could explain the results, but now we know scientists like Dr. Enstrom were attacked and persecuted to discourage others who would publish null studies. So publication bias is real.

They did consider the possibility that some other causes of disease were actually responsible for the results. They considered personal history of lung disease, family history of lung disease, heat sources, cooking with oil, occupation, and diet. According to the EPA, without considering these, the conclusion was that secondhand smoke increased the lung cancer risk 30%. After factoring these other possible causes in, they recalculated that the increased risk was 19%. When I calculate the relative standard deviation of the aforementioned 31 residential studies the EPA used, I get RSD = 35%. In other words, the studies the EPA used, differ from each other by more than the EPA's final conclusion. This violates the consistency of association parameter my textbook teaches which must be met to determine if the hypothesis is real.

However, there are many more causes of lung cancer in nonsmokers which were not factored into the EPA report. They did not study how differences in socioeconomic status, alcohol consumption, or radon exposure could affect their conclusion. I have studied how urban residency would affect the EPA results. This full report is on my webpage. If the EPA had factored in urban residency, their recalculated risk would have been reduced further, to only a 5% increased risk of lung cancer in nonsmokers. These examples tell us that it is really impossible for epidemiologists to measure small risks when much larger risks are also present. Statistically, secondhand smoke researchers are trying to find a "needle in a haystack"

In the EPA report, and many other studies which I have looked at, most of the ones which do find secondhand smoke causes disease cherry-pick to factor in only those other factors which could potentially alter the conclusions, which are evenly distributed among exposed and unexposed individuals, and they ignore those other factors which really could offer alternate explanations of the findings. This is yat another example of using selective research in secondhand smoke studies.

The EPA also stated that their conclusion is unlikely due to chance. Yet, epidemiology textbooks teach that it takes a risk 2- 3 times higher than the one the EPA reported to conclude the risk is real. In fact, the 2006 surgeon general's report, actually states within the report, that it is possible misclassifying former smokers and nonsmokers as never smokers, could actually alternately explain all his conclusions. Before getting away for the surgeon general's report, it actually states that many diseases like stroke in adults, and ear infections in children are suggestive, but insufficient to prove that secondhand smoke causes them.

So, what all this boils down to, is that they do have some case-controlled studies where all they did was survey nonsmoking lung cancer and heart disease patients about their past exposure to smoke in residences. They claim that this means that workplace exposure must be causing disease in nonsmokers too. They don't really know if other factors are really responsible for the observed health differences they blame on secondhand smoke. They do not have even ONE study which claims that patrons can get disease from secondhand smoke in hospitality establishments.

Now, antismokers do have a series of studies claiming nonsmoking hospitality workers get 50% more lung cancer and heart disease, than most other kinds of workers. But, other studies also find non-hospitality workers who work near busy roads also carry this excess risk of 50% of these diseases. () and (McMaster) Most studies also find nonsmokers who live close to busy roads also have 50% more lung cancer and heart disease than would be otherwise expected. (EHP Online) and Oxford Journals) So, in the studies that the antismokers use to convince lawmakers that hospitality workers are at risk from smoke, it is likely the real reason for these effects is due to the fact that most hospitality establishments are located near busy roads. Alternately, even if these studies of hospitality workers are true, they were conducted before modern filtration/ventilation equipment came into use, (EHP Online) and it is likely that even if it were ever true that hospitality workers were at increased risk, that it is no longer true.

To help insure that lawmakers will not reject bans by rationalizing that proper ventilation will work, antismokers have now started claiming that there is no safe level of smoke, and that it takes tornado like ventilation to protect nonsmokers. One Internet resource I have provided to you is run by a ventilation engineer who posts many studies showing it really does work. (Clean Air Quality) However, the idea ventilation cannot protect nonsmokers is preposterous. Ventilation is used in a wide variety of workplaces under OSHA regulations and protects workers from real airborne hazards all the time. We have a fireplace in our home, which can produce more carbon monoxide in one evening than a smoker produces in one year. We also have a carbon monoxide detector located on a wall about 20 feet from our fireplace. It is sensitive to one-part-per-million. The OSHA 8-hour limit for carbon monoxide in the workplace is 50-PPM. When we use our fireplace, that detector never registers even 1-PPM, unless we close the damper too soon after we think the fire is out. And our fireplace does not have a tornado-like draft. When I go outside, it looks like the smoke is coming out about 5 mph. It is so obvious the antismokers are making false claims about ventilation, and pointing this out when fighting bans, becomes an easy way to convince lawmakers that antismokers' claims are not scienced-based.

The claim that there is no safe level of smoke violates the common rule of toxicology that "the dose makes the poison." Remember, most of the common studies of active smokers conducted in the 1950s, through the 1970s, found that those light smokers of fewer than five cigarettes/day, had the same health effects as nonsmokers. Many of these were reviewed in the 1964 surgeon general's report. The EPA report said that the average nonsmoker was exposed to the equivalent of 1/2 of a cigarette a day in the late 1980s. The new surgeon general's report says that 70% of that has been removed since then. If so, then nowdays, the average nonsmoker is being exposed to 0.15 cigarettes/day. Oak Ridge National Labs has reported that nonsmoking bartenders wearing air monitors were exposed to less than 1/10th of a cigarette/shift. This is in good agreement with my estimate from above. Common sense should tell all of us that it is highly implausible that smoke actually harms nonsmoking hospitality workers in today's situations.

I have gathered many economic studies and reports which show bans do indeed harm the hospitality industry. These are also posted at my webpage. These include harm from local bans in Arnold and Ballwin Missouri. I also have an example where the Missouri Department of Health studied the Maryville Missouri smoking ban. That ban took effect just when Northwest Missouri State University initiated an expansion program which increased the student population. Further, the ban exempted the seven establishments most likely to be hurt by the ban. Suffice it to say, here, that small college towns' economies do not behave like most local economies in Missouri, and if the Dept. of Health really wanted us to know how bans affect business in Missouri, then they could have studied Springfield, or Arnold and now Ballwin, which have more typical economies and where those local bans definitely have harmed business. The fact that the Missouri Dept. of Health did not select a locality for an economic study with more conventional economics, again suggests selective research was used by our own health department to promote the claim bans do not harm business in Missouri.

Our main problem is that state and local lawmakers don't know of all of this information and the media will not tell them this. These lawmakers have become saturated with false information by the media over the last 20 years. Dissent is effectively silenced by the media, in partnership with health departments. I really do not think health departments consider all the evidence when they tell lawmakers secondhand smoke is dangerous too. So, this current trend towards smoking bans will continue unless the public becomes aware of these secondhand smoke facts which I have just presented to you, and the public starts loudly speaking out to lawmakers about these facts.

Legally, Missouri law professor Thomas Lambert has prepared a legal brief against smoking bans which is on my webpage This is recommended reading for anyone opposed to bans. Additionally, I have two personal opinions concerning the legality of bans.

First, no court has ever tested if bans interfere with the First Amendment right of association between owners and smoking patrons. A New York court did rule that bans do not interfere with the right of smokers to associate with each other because they can assemble outdoors, or in private residences. To the extent that bans reduce patronage in establishments in areas covered by bans, then owners, who cannot go outside to associate with patrons while they are smoking or associate with them in their homes, are being denied this most important of Constitutional rights.

Second, when bans grant exemptions, they may be interfering with federal antitrust laws. These kinds of bans effectively give smoking-allowed monopolies to the exempted establishments. A case in Colorado recently struck down it's ban because it violated equal protection clauses.

Regarding Missouri, I have already suggested to the Missouri Restaurant Association, that they lobby for an amendment to existing state smoking statutes. Current statute allows smoking in all of bar, tavern, and restaurant space, if the restaurant has fewer than 50 seats and posts a we-do-not-have-a-nonsmoking-section sign. If the restaurant has over 50 seats, then only 30% of seats can be designated smoking. This needs to change. If Missouri allowed a restaurant of any size to be totally smoking, provided they post appropriate signage, then restaurants like these would drain smoking patrons out of restaurants with limited smoking space. This could ultimately reduce the number of mixed-smoking-choice restaurants. If that happened, it would ultimately satisfy the desires of more patrons than the current situation. Fewer patrons who object to smoke exposure would be sharing space with smokers. Virginia recently considered exactly this change to it's smoking law, but it was defeated after the governor attached an amendment to it banning smoking in all restaurants.

However, now we have too many localities with bans for my idea to work fairly. Currently, Texas is considering a version of a ban which would allow owners to opt out, even if the local government already has a ban (KHOU) What we need now, is to ask the state to create a new class of hospitality business which could be called a "smoking establishment." These would be preempted from local regulation. They could be required to maintain proper ventilation and be subject to inspection by health departments similar to the ban which is currently in effect in Chicago. Any number of these licenses would be available, and could be obtained by owners in localities which currently have bans, but are suffering downturns in business.

David W. Kuneman