Tobacco Documents Online

SAVE 35% oo ON CO~c~h Complete this form today and save up to 35% off the cover price for a subscription to Consumers' Research. From economics to the environment, from health insurance to home appliances, Consumers' Research brings you the best in consumer reporting for only pennies a day! r-1 Two years (24 issues) for $39.00 F-1 One year (12 issues) for $21.00 (REGULARLY $60.00) (REGULARLY $30.00) D Renewal (please attach current label) ~ Payment enclosed Q Bill me later Name Address City State Zip Subscription rates for U.S. only. All other countries $8.00 per subscription year, payable in U.S. dollars drawn on U.S. accounts. PLEASE ALLOW 6 TO 8 WEEKS FOR PROCESSING

BUSINESS REPLY MAIL FIRST CLASS PERMIT NO. 7525, NASHVILLE, TN POSTAGE WILL BE PAID BY ADDRESSEE COf~SUSMeE~RS' NO POSTAGE NECESSARY IF MAILED IN THE UNITED STATES Subscription Department ~ Post Office Box 41533 ~ Nashville, TN 37204-9926

issue and left the statutory standard in place. While the agency rejected a variety of attacks on that standard, we are concerned with only one of the defeated arguments: the contention that the standard will force carmakers to produce sznaller, less safe cars, thus making it more difficult and expensive for consumers to buy larger, safer cars. We find that the agency has not coherently addressed this concern. When the automaking firms petitioned for a reduction in the model year 1990 standardd down to 26.5 mpg, and petitioners pressed the argument that failure to reduce the standard would cost lives, NHTSA had three basic choices. First, it might have concluded that the statute does not require it to con- sider safety effects when deciding whether to embark on a modification proceeding. It could then have dismissed petitioners' claims without further ado. While a court might have reversed, the statuto- ry framework is so loose and...that the agency would have had a fair shot at being upheld. Second, NHTSA might have seriously examined the record data. On its face this suggested (as we shall see) the overwhelming likelihood that a 27.5 mpg standard reduces the supply of safe cars avail- able to American consumers. Conceivably, of " ... consumers who do not want to be priced out of the market for larger, safer cars, deserve better from their government." course, a sophisticated analysis might have over- come the record's apparent implications, but even if it did not, all NHTSA would have had to do was face the trade-off. It could have said that while the 27.5 standard might cost, say 200-to-500 American lives a year for ten years, it would also reduce American oil imports by, say, 50,000-to-400,000 barrels a year, and that in its judgment the trade- off was worth it. And it could have expressed any such trade-off in less numerical terms. Finally, NHTSA could have fudged the analysis, held the standard at 27.5, and, with the help of sta- tistical legerdemain, made conclusory assertions that its decision had no safey cost at all. That is what it chose. The people petitioners represent, con- sumers who do not want to be priced out of the mar- ket for larger, safer cars, deserve better from their government. ...the agency insisted at oral argument that even if the 27.5 standard constrains the behavior of carmakers, it will not lead to smaller cars. Yet nowhere has the agency actually justified this claim or even purported to make such a finding. "Nothing in the record or in NHTSA's analysis appears to undermine the inference that the 27.5 mpg standard kills people..." It came closest in the following passage: "[T]here are still a number of fuel-effeciency enhancing methods that [GM and Ford have] not fully utilized throughout their fleets.... NHTSA believes that the domestic manufacturers should be able to improve their fuel economy in the future by these and/or other technological means, without outsourcing their larger cars, without further down- sizing or mix shifts toward smaller cars, and with- out sacrificing acceleration or performance." Why the agency expressed itself in the normative ("should be") is anybody's guess. At any rate, it has never claimed that domestic manufacturers will in fact meet the standard without downsizing their fleets, or even that there is a substantial probability that they will do so, or even that there is a substan- tial likelihood that they will use methods other than downsizing for the lion's share of the work. Presum- ably NHTSA does not assert such facts because it could not ground them in the record. Moreover, to the extent that carmakers choose technological innovation over downsizing (and fur- ther assuming that such innovation would not itself compromise aspects of auto safety), that choice would involve significant costs in implementation, even if we assume that research and development are complete. That cost would translate into higher prices for large cars (as well as small), thereby pres- suring consumers to retain their old cars and make the associated sacrifice in safety. The result would be effectively the same harm that concerns petition- ers and that the agency fails to negate or justify. The historical fact is, however, that carmakers respond to CAFE standards by reducing the size of their fleets. NHTSA itself has explicitly acknowl- edged as much in the past, and we ourselves have insisted that "the evidence shows that manufactur- ers are likely to respond to lower CAFE standards by continuing or expanding production of larger, heavier vehicles." Even in the decision below the agency acknowledged this link, explaining that "Chrysler's CAFE has been higher than that of GM or Ford in recent years primarily because it does not compete, or compete as heavily, in all the market segments in which GM and Ford sell cars, particu- larly the large car market." The agency now tries to obscure this reality by pointing out that "the average fuel economy of the new car fleet has improved steadily from 26.6 mpg in model year 1982 to 28.2 mpg in model year April 1992 11

"As NHTSA itself has amply docu- mented, however, minivans are considerably less safe than vans generally, with a fatality rate per registered vehicle about 25-33% higher than that of large cars." 1987, while the average weight of a new car increased two pounds during the same period." This argument misses the point. The appropriate comparison, which NHTSA must but did not address, is between the world with more stringent CAFE standards and the world with less stringent standards. The fact that weight has remained con- stant over time despite mileage improvements shows the effect of technological improvements, to be sure, but in no way undermines the natural inference that weight is lower than it would be absent CAFE regulation. Here we can be quite sure that it is lower, since, as NHTSA observed in this decision, economic recovery and declining gasoline prices sharply raised consumer demand for large cars over the relevant period ("consumer demand has shifted back toward larger vehicles") If consumers demanded substantially bigger cars, carmakers-absent regulation-would have pro- duced substantially bigger cars, not cars that remained, on average, within two pounds of the cars made when consumers favored smaller cars. More- over, NHTSA has given us no reason to think that whatever technological innovations permitted automakers to meet CAFE requirements while keeping weight constant did not also cost consumers more, again pricing some consumers out of the mar- ket for new large cars.' NHTSA also argues that even if the 27.5 mpg standard will deplete the supply of large GM or Ford cars, a consumer looking for a big car "will buy a large car from another manufacturer, or will buy a minivan, or will keep his or her older, large car.... [A]ny one of those alternative consumer out- comes is far more likely than the possibility that the consumer will buy a smaller car than he or she wanted to buy." Nothing in the record suggests that any of these will give consumers large-car safety at the prices that would have prevailed if NHTSA had made a less stringent choice. The reference to buying large cars from "anoth- er manufacturer" is somewhat in the spirit of Marie Antoinette's suggestion to "let them eat IIt is significant that even NHTSA makes no more than the lame claim that "[tlhis example illustrates the point that not all CAFE gains come by reducing weight." The issue is whether a material portion of the "CAFE gains" are likely to entail downsizing. NHTSA never even pur- ports to deny this. cake." By NHTSA's own hypothesis, the "other manufacturers" are Chrysler, which has essential- ly removed itself from the large car market, and foreign manufacturers, which are subject to CAFE standards on their U.S. sales. To the limited extent that foreign firms produce truly large cars at all, they are expensive ones. In suggesting minivans (which are exempt from the 27.5 standard), the agency disingenuously obscures their dangers by citing safety figures only for vans in general. As NHTSA itself has amply documented, however, minivans are considerably less safe than vans generally, with a fatality rate per registered vehicle about 25-33% higher than that of large cars. Finally, NHTSA's notion that the consumer should "keep his or her older, large car" ignores both its own finding that neww cars "appear to experience fewer accidents per mile traveled," and the plight of consumers seeking to buy a large car for the first time. By making it harder for consumers to buy large cars, the 27.5 mpg standard will increase traffic fatalities if, as a general matter, small cars are less safe than big ones. They are, as NHTSA itself acknowledges. The agency explains: "Occupants of the smaller cars generally are at greater risk because : (a) the occupant's survival space is generally less in small cars (survival space, in simple terms, means enough room for the occu- pant to be held by the vehicle's occupant restraint system without being smashed into injurious sur- faces, and enough room to prevent being crushed or hit by a collapsing surface); (b) smaller and lighter vehicles generally have less physical structure avail- able to absorb and manage crash energy and forces;. and (c) in most collisions between vehicles of differ- ent weight, the forces imposed on occupants of lighter cars are proportionately greater than the forces felt by occupants of heavier vehicles.°'2 The agency tries to skirt the obvious conclusion with two specious arguments. First, it essentially argues that the 27.5 mpg standard will have no effect on the availability of large cars (i.e., will accomplish nothing at all). This, we have seen, is simply untrue. Second, the agency observes that new cars now come with a variety of mandatory and optional safety fea- tures (airbags, anti-lock brakes, etc) that will pre- sumably compensate for a decline in size. There are two things wrong with this latter (See CAFE, page 35.) 2One might argue that the third factor indicated that if all cars were small, there would be fewer traffic fatalities. Any such inference appears quite doubtful. Cars can hit a variety of objects, including trucks, trees, and other cars; fatalities in car-to-car crashes do not account for even a majority of passenger-car occupants fatalities. Unless NHTSA outlaws trucks and trees, smaller cars will probably always mean higher fatality rates, as NHTSA recognizes., ("in single vehicle crashes, there is increased risk of serious injury or death"). 12 Consumers' Research

Special Report: Passive Smoking And Your Heart Gary L. Huber, MD, Robert E. Brockie, MD, and Vijay K. Mahajan, MD. n the July 1991 issue of CR, we defined the nature of environmental tobacco smoke (ETS), presented an overview of how the possible health risk of exposure to ETS is assessed, and reviewed the available literature on the alleged relationship between ETS expo- sure and the risk of nonsmokers developing lung cancer (see "Passive Smoking: How Great a Haz- ard?"). There are published now a total of 32 studies on ETS and lung cancer. Although some may cite these reports to mean otherwise, the majority of the published data do not support the conclusion that exposure to the residual con- stituents of ETS is associated with lung cancer in nonsmokers. That is, only 7 of the 32 pub- lished studies-or less than a fourth of the investigations that have examined this ques- tion-report a small, but statistically significant, increased risk. The reader is referred to our ear- lier publication for a more extensive analysis of these considerations. Exposure to environmental tobacco smoke has also been reported as associated with the develop- ment of cardiovascular disease. This is an impor- tant issue, in that the number of people in our society who develop cardiovascular disease exceeds by a substantial margin those that will develop cancer. A critical evaluation of this sub- ject requires placing the available information in some rational perspective within a broader under- standing of cardiovascular disease in general. The term "cardiovascular disease" is used to describe those illnesses of the heart, brain, and other organ systems that develop because of acquired abnormalities in the blood vessels that supply them. Cardiovascular diseases are by far the most common cause of disease and death in our society today. Over 60 million Americans suf- fer from these diseases and one million or more of them die each year, accounting for one death every 30 seconds. Cardiovascular diseases are responsible for almost one-half of all deaths in the United States. To place this in further perspec- tive, there are more than twice as many deaths Drs. Huber, Brockie, and Mahajan are with, respectively, the University of Texas Health Center, the Presbyterian Hospital of Dallas, and St. Vincent's Hospital-Medical College of Ohio. from cardiovascular disease as there are from all forms of cancer combined. Coronary artery disease, an illness that is due to a narrowing or blockage of the major vessels that supply blood to the heart muscle, is one of the most common forms of cardiovascular disease. If the coronary artery is partially blocked, the reduced blood supply to the heart muscle may cause reversible ischemic heart pain, or angina pectoris, to develop. If the blockage is more severe, myocardial infarction (irreversible dam- age to part of the heart muscle) can develop; "The fact that about half of all cardiovascular deaths can not be explained on the basis of spe- cific identifiable risks reflects how little we really know about these matters, and how extreme- ly difficult it is to study them with precision." worse yet, sudden death may occur. These are manifestations of coronary artery disease that we commonly call "heart attacks." Coronary artery disease and heart attacks cause about one death every minute in this country. The exact cause of coronary heart disease is not known. It is generally held that the primary problem is atherosclerosis, which is a gradual build-up of fatty deposits on the inside of the coronary vessels. The build-up of these deposits forms an atherosclerotic plaque, rendering the artery wall thicker, often with an irregular sur- face that may cause the blood within to clot. This is a slow process that begins in infancy or early childhood and progresses all through life. As the build-up of fats (primarily cholesterol) continues, a point is reached where the vessel opening becomes significantly narrowed and is more sus- ceptible to complete blockage. Many people who die of heart attacks, however, do not have an unusual amount of atherosclerosis in their coro- nary vessels, or even elevated cholesterol levels. Furthermore, the degree of development of April 1992 13

plaque formation does not consistently correlate with the site of an eventual occlusion or with death from this disease. Atherosclerosis appears to be responsible for the largest share of heart attacks and related deaths in this country. How or why atherosclerotic plaques occur and develop, however, is not known.' In the absence of a proven mechanism for the development of coronary heart disease, emphasis has been placed on the identification, through epidemiological studies, of "risk factors" that are associated with an accelerated rate of for- mation of atherosclerotic plaques. Most often, however, results of such epidemiological studies are expressed as death that is attributable to heart disease, not as a quantification of atherosclerosis. Unfortunately, death certificates, from which mortality rates are often derived, are notoriously inaccurate for diagnosis of heart disease. Risk Factors A risk factor is the term that describes a char- acteristic of behavior or of lifestyle, or an envi- ronmental exposure, or an inherited characteristic, that on the basis of epidemiologi- cal data is reported to be associated with the development of disease. The risk is usually expressed as an "odds- ratio," or a "risk ratio," which measures "relative risk" in comparison to some control group or pop- ulation which has not been exposed to the factor in question. If there is no difference in the disease rates associated with the factor, compared to the disease rate for the non-exposed or control group, the relative risk will be calculated as "unity," or 1.0. If there are differences in the disease rate that are associated with the factor studied, these differences will be expressed as a relative risk that is some variation of unity. If the relative risk is less than 1.0, the average exposed individuals would have less chance than the nonexposed con- trol individuals for the development of the dis- ease. If the relative risk is greater than 1.0, the exposed individuals would have an increased chance for the development of the disease. How- ever, the degree of increased or decreased risk must be "statistically significant" by acceptable biostatistical criteria before a relative risk can have any meaningful importance. Relative risk relationships are only mathemati- cal associations. When they are consistent and 1There currently are two widely investigated, speculative theories: (1) Atherosclerotic plaques develop in response to an initial injury to the blood vessel wall, or (2) the plaques are an uncontrolled growth of sorts, with replication within the vessel wall that results in a build-up of cholesterol-laden cells that eventually will cause a blockage. strong, there is an implication of a potential causal relationship. Even when very strong, how- ever, risk factors by themselves do not represent anything other than a statistical association. They must always be considered in the context of other scientific information. They must also be evaluated in the context of whether or not the reported association makes any biological sense. The strength of a statistical association does not necessarily determine its importance. For instance, a weak association, if statistically signif- icant, that affects very large numbers of people may be important because of the magnitude of its effect on the population at risk. A strong statisti- cal association that has no biological relevance may be unimportant or meaningless. Even strong relative risk associations must be viewed cau- tiously when there exist numerous potential caus- es of a disease. The greater the number of potential "causes" of a disease (usually identified as risk factors), the more difficult it becomes to analyze (or implicate) any one of these "causes" (or risks) to the exclusion of another. When the relative risk is less than 2.5 or so, the association of the identified risk factor with the development of the disease is, by convention, considered to be weak. That is to say, the associa- tion of the risk factor with the development of disease may have only limited or no real meaning. The weaker the relative risk, the greater must be the care and responsibility exercised in its inter- pretation. When the relative risk is less than 2.0, there is a strong possibility (or probability) that the association is artifactual-that is, the relative risk may actually be due to confounding factors where two or more potential associations cannot be separated or distinguished. (A confounding factor, in this context, can be defined in the most simple of terms as "an alternative explanation. ") When the relative risk is less than 2.0, there are enormous problems of controlling the biases of the investigator as well as biases that are inher- ently present within the experimental design of every epidemiological study. Bias, in this context, means the introduction of error by failing to con- trol for or to consider other important influences. Risk Factors for Cardiovascular Disease. To compli- ca.te matters further, it is extremely unlikely that cardiovascular disease is ever caused by one factor. The development and progression of this disease are associated with many factors. Over the past 25 years, in fact, more than 300 identifiable risk fac- tors have been reported as potentially important to the development of cardiovascular diseases. Even with this large number of risk factors, leading authorities in cardiology emphasize that it is remarkable that most people who develop 14 Consumers' Research

atherosclerosis and most people who die from car- diovascular disease do not have a readily identifi- able specific risk factor to explain their disease. For example, only slightly less than 50% of all cardiovascular disease and death has been associ- ated with specific risk factors. The fact that about half of all cardiovascular deaths can not be explained on the basis of specific identifiable risks reflects how little we really know about these matters, and how extremely difficult it is to study them with precision. Cardiovascular risk factors are usually classi- fied as unmodifiable or modifiable. Unmodifiable risk factors are ones that represent an association that cannot be changed, such as age, gender, race, genetic determinants, family history, and so on. Modifiable risk factors in many ways are poten- tially more important, because once identified they hopefully can be reduced or controlled. Modifiable risk factors number literally in the hundreds, but the most important ones are thought to be high blood pressure, diabetes melli- tus, and elevated blood levels of cholesterol and triglycerides. Excessive life stress, excessive alco- hol intake, lack of regular exercise, cigarette smoking, obesity in a certain body distribution, and other life style factors may be almost equally important. Most physicians try to reduce modifi- able risk factors in the hope of reducing the mor- bidity and mortality due to cardiovascular diseases and, especially, coronary heart disease. Active Smoking Active tobacco smoking is reported as a major and an important risk factor for the development of cardiovascular diseases and for coronary heart disease. Active smoking is called a "major" risk factor because of the large numbers of people who smoke. Smoking rates may be under-reported now because of the associated "social taboos" of smoking. In spite of this consideration, it is esti- mated that at least 50 million Americans contin- ue to smoke on a regular basis. The Office of the Surgeon General has emphasized that reducing the magnitude of this active smoking population would have a major national health impact in reducing cardiovascular disease mortality. Although classified as a "major" risk factor for heart disease on the basis of the sheer number of active smokers, it may come as a surprise to many readers to learn that active cigarette smoking is not a strong risk factor. For instance, in 1983 the Surgeon General's report focused exclusively on tobacco smoking and cardiovascular diseases and estimated the relative risk for coronary heart dis- ease in smokers at 1.7, compared to nonsmokers. Other estimates, now based on over 20 million person-years of epidemiological assessment, have set the relative risk in active smokers for coro- nary heart disease at a level of from as low as 1.3 to as high as 2.0 or slightly greater, but with a four-fold greater risk for sudden death, compared to nonsmokers. Indeed, when critically analyzed, most epidemi- ological studies report that active tobacco smok- ing alone is, in the absence of other potential risk factors (such as high blood pressure and high serum cholesterol levels), an extremely weak risk factor for the development of cardiovascular dis- eases. Given the presence of additional risk fac- tors, however, such as high blood cholesterol levels or hypertension, tobacco cigarette smoking has been reported to influence the net overall risk for death from coronary heart disease. That is, adding active tobacco smoking to another under- lying risk may result in a net effect that 'a greater than simply the sum of the two individual risks combined. There are several possible explanations for this, which will be addressed in the discussion that follows. It is not clear, for instance, whether tobacco smoke itself is actually important in the development of atherosclerosis or whether simply that tobacco smoking is an epidemiological "marker" for a life style characterized by multiple high risk behaviors. It must be remembered, how- ever, that increasing the strength of the risk asso- ciation does not allow the inference that the risk factor in itself is causal, for active tobacco smok- ing or for cholesterol or for any other factor. 'Passive' Smoking Nine epidemiological studies (see table, page 17) have reported the relative risk for develop- ment of cardiovascular disease in nonsmokers exposed to ETS. Since the residual constituents of ETS are so dilute, it is extremely difficult to mea- sure them directly. In that context, then, it is important to emphasize that none of the nine epi- demiological studies actually measured exposure to ETS, but rather projected or estimated an exposure to ETS on the basis of a surrogate. The surrogate was usually the historical identification (by answer to a questionnaire) of a smoker living in the household of a nonsmoker. The nine stud- ies contain 12 sets of epidemiological data, seven sets of which are data for nonsmoking females who were married to or living with active male smokers, and in some instances who reported ETS exposure in their workplace. Limited data are available on four sets of nonsmoking males who reported a surrogate equivalent of exposure to ETS. April 1992 15 ~ I

What Is ETS? Environmental tobacco smoke (ETS) consists of "secondhand" residual smoke constituents emitted by active smokers into their sur- roundings. The nonsmoker may be exposed to these residual constituents in very dilute con- centrations. ETS residual constituents are remnants of exhaled mainstream and of sidestream tobac- co smoke that are so dispersed in environ- mental air that it is somewhat of a misnomer, or misconception, even to refer to them as "smoke," per se. Under real-life conditions, only about 100 or so of these environmental tobacco smoke remnants have been identified to date-and then only at extremely low con- centrations-in the environment of smokers. This is in contrast to the several thousand constituents that have been reported for the mainstream smoke that is inhaled by active smokers. Five of the twelve data sets report very small increases in relative risk that reach statistical sig- nificance, and seven of the data sets report changes that are not statistically significant at conventional levels of biostatistical acceptance. A relative risk ratio is an estimated average change in the disease rate associated with the studied variable; in all of these studies, then, the relative risk is the projected estimate of risk for developing cardiovascular diseases for nonsmok- ing individuals married to or living with smokers. Confidence intervals also are included for all data presented in the table.2 Seven of the 12 data sets have as a lower limit of their confidence intervals a relative risk of less than unity (1.0), indicating, by universally accepted epidemiological stan- dards, that spousal smoking may not be associat- ed with the development of cardiovascular diseases in nonsmokers; in other words, confi- dence intervals that reach less than unity for rel- ative risks indicate that there is insufficient evidence that the experimental group is different from the control group at the specified level of confidence. How Valid? None of the ETS epidemiological studies is a "high validity" randomized prospective interven- tion study designed to evaluate whether or not a reduction in the level of exposure to ETS is asso- ciated with a reduction in risk for cardiovascular disease. Three of the ETS studies are low validity case-control studies and six of the ETS studies are cohort "quasi-prospective" assessments. That is, they are drawn from what were prospective studies designed for another purpose and, as such, are valid only for generating hypotheses, not for confirming them. None of the six cohort studies were initially designed to evaluate the effect of ETS as a risk for cardiovascular disease. They all represent "data dredging" by "retrospec- tive" assessments of "nested case-controls" fol- lowed prospectively for another purpose. The studies are of diverse design and draw con- flicting conclusions. The six cohort studies, for example, do not report comparable data. Some report disease rates and others report death rates, some report prevalence statistics and oth- ers incidence data, and some assess the broad cat- egory of cardiovascular diseases in general and others assess certain manifestations of only the more specific coronary heart disease. For these reasons, the data from the various studies cannot be legitimately combined in so-called meta-analy- sis to see if stronger conclusions can be drawn.3 Even though combining such diverse data as are available from these studies is not generally considered scientifically acceptable (at least not by currently justified procedures), two publica- tions nevertheless have attempted to do so.` A third publication (Steenland, 1992), did not pool results of epidemiological studies, but developed and employed an elaborate model based on an extensive number of untested assumptions. Steenland projected a 2.2 percent greater chance for nonsmoking males and 1.2 percent greater chance for nonsmoking females of dying from coronary heart disease by age 74, when living with a smoker, in comparison to those living with a nonsmoker over a lifetime. Dosimetry and Trends. None of the studies on ETS and cardiovascular diseases measured or in any way directly quantified actual exposure to environmental tobacco smoke.s In the absence of direct measurements of exposure, these studies, like all epidemiologic studies on ETS, have used 2 A 95% confldence interval is a statistical expression of a range of val- ues that have, as listed here, a 95% probability of including the true value for the effect of nonsmokers living with smoking spouses, com- p3ared to nonsmokers living with nonsmoking spouses. Meta-analysis is a way of pooling or combining several studies, by sta- tistical analysis and integration of the results of low-power or weak reports, in order to draw conclusions that may be stronger than those demonstrable in any of the weak studies alone. 4The first publication (Wells, 1988) derived an estimated "pooled relative risk" for ETS surrogate exposure and heart disease of 1.2 for females and 1.3 for males. The second publication (Glantz and Parmley, 1991) failed to provide the reader with the methodology employed, and project- ed an overall "pooled risk" of 1.3 for both males and females. 5 Five of the nine studies on ETS and cardiovascular disease attempted to assess via a questionnaire on household smokers whether or not there was a "dose-related" association between the number of smokers 16 Consumers' Research

Studies of ETS and Cardiovascular Disease in Nonsmokers Study Sex Type of Study# Number of Cases+ Relative Risk** 95% Confidence Interval Variables Controlled Hirayama, 1984++ F P 494 1.16 0.9-1.4 Husband's age Garland, 1985 F P 19 2.7 0.9-13.6 Age, blood pressure (BP), weigt, choles- t i t l Lee, 1986 F C 77 0.9 0.5-1.6 erol mart a sta us. Age, marital status. M C 41 1.2 0.5-2.6 Svendsen, 1987 M P 13 2.1 0.7-6.5 Age, BP, plasma lipids, weight, income, He/sing, 1988 M P 370 1.3* 1.1-1.6 education, alcohol. Age, education, marital status, income. F P 988 1.2* 1.1-1.4 He, 1989 F C 34 1.5* 1.3-1.8 Age, BP, cholesterol, race, residence, Humble, 1990 F P 76 1.6 1.0-2.6 alcohol, other factors (but data not available). Age, BP, weight, cholesterol. Hole, 1990 M/F P 84 2.0* 1.2-3.4 Age, BP, weight, cholesterol, social class. Dobson, 1991 M C 22 1.0 0.5-1.9 Age, sex, prior coronary heart disease. F C 43 2.5` 1.5-4.1 ~ P=prospective cohort study; C=retrospective case-control study ' Statistically significant at the conventionally accepted level (5%). '` Weak relative risks have risk ratios between 1.0 and 3.0, or so. Any risk below 1.0 represents a negative relationship. Note that none of the studies report a strong average relative risk.. Data reported are from the author's papers or from review articles. + Cases contains coronary heart disease deaths and/or cardiovascular disease, with or without death. ++ Some of the data from Hirayama were reported as "statistically significant" with unconventional 90"/% confidence intervals (relative risk of 1.3 with 1.1-1.6 confi- dence intervals); recompilation of all of his data available reveals a nonsignificant relative risk. surrogates of exposure. For nonsmokers, the sur- rogate of ETS exposure has been an estimation of the number of active smokers living in the same household (usually an actively smoking husband with a nonsmoking female or an actively smoking wife with a nonsmoking male) or an estimate of smokers present in the work-place of the non- smokers. These surrogate "exposure" estimates were derived exclusively through various ques- tionnaires. No study employed actual direct quan- tification of ETS or ETS constituents in the environment of the nonsmoker. We reviewed in some detail the serious shortcomings of this approach in our previous publication in CR, noted above, and the reader is referred to that contribu- tion for a more extensive discussion. As with other studies on ETS and potential dis- ease risks, some of the reports on ETS and cardio- vascular disease contain data on some of the population subsets that, in the absence of other- wise significant differences, suggested to the authors or to other reviewers a "trend" of the sta- tistically insignificant data toward a meaningful association. Although in most of science "trends" in these kinds of data do not count, there are in the household and the amount of cardiovascular disease. The results were inconsistent, with some reporting a dose-response relationship, most reporting no significant effect, and some data suggesting a reverse dose-response relationship-less disease reported with high lev- els of exposure to ETS. "Six of the nine studies report a relative risk for cardiovascular or coronary heart disease...that is approximately equal to or in excess of that reported for active smokers. Intuitively, that makes no biological sense whatsoever." legitimate ways to assess whether or not such "trends" might have some "statistical signifi- cance." Seven of the nine original reports claim and discuss "trends" in their results, even when their own published statistical analyses of these data demonstrate that the proposed "trends" had no statistical significance. In other words, these contributors seemed to ignore their own biostatis- tical analyses and to adapt new rules to fit hypotheses otherwise not provable by their own reported data or by conventionally accepted bio- statistical principles. Other Scientific Evidence. Nonsmokers in the environments of active smokers typically are exposed to only extremely small amounts of a very limited number of residual remnants of ETS. Potential cardiovascular effects would not be expected from exposure to such small concentra- tions of these smoke constituents. April 1992 17

"All of the more than 300 cardio- vascular risk factors that have been identified are confounding variables and many have the same approximate relative risk or risk ratio as that reported for spousal smoking." Under real-life conditions,e it has been estimat- ed that nonsmokers are exposed to approximately as little as 1/10,000 to at most only about 1/100 to 2/100 per hour or so of certain constituents of cigarette smoke to which the active smoker is exposed in the same period. Exposure of nonsmokers to the highly diluted residual constituents of ETS is at a concentration well below that level which would be expected to produce any long-term pathological effects or dis- ease. Remarkably, then, in the face of this extremely low level of exposure, six of the nine studies report a relative risk for cardiovascular or coronary heart disease associated with ETS expo- sure that is approximately equal to or in excess of that reported for active smokers. Intuitively, that makes no biological sense whatsoever. Something clearly is wrong with either the design or with the gathering and calculation of the epidemiological data in these studies. Cigarette smoking remains a high frequency event in our society, with an estimated 50 million or more active smokers today. Approximately 500 to 600 billion cigarettes are consumed each year in this country alone. Death from cardiovascular disease is also a high frequency event in our soci- ety, with over one million cardiovascular deaths each year and one death every 30 seconds. It is not surprising, therefore, that considerable inves- tigative effort would be spent on studying the potential association between these two high fre- quency events. When such potential associations are studied, great care must be exercised to control for the influence of confounding factors on the reported results. This is particularly true for cigarette smoking and cardiovascular diseases because the reported association between the two is quite weak, the number of additional risk factors is extraordinarily large, and less than half of all car- diovascular disease mortality is reported to be associated with specific identifiable risk factors. 6 By "real-life" conditions it is meant conditions encountered in day-to- day living conditions in the world, as opposed to the artificial con- straints of the experimental laboratory or a sealed environment chamber. Confounding Variables. A confounding variable is one that can cause or prevent the outcome of interest (in this case, death from cardiovascular disease) and is not associated with the factor under investigation (in this case, reported expo- sure to ETS). All of the more than 300 cardiovas- cular risk factors that have been identified are confounding variables and many have the same approximate relative risk or risk ratio as that reported for spousal smoking. If these confound- ing variables are not evaluated and controlled for in an epidemiological study on ETS, how then can ETS be implicated to the exclusion of the other factors? It cannot be, of course, but that is exactly what has happened in the nine studies on ETS and car- diovascular diseases. For instance, the two largest studies (Hirayama with 494 cases and Helsing with 1358 cases-together representing well over 80% of all reported cases in these studies) do not even control for blood cholesterol levels, do not control for high blood pressure, and do not con- trol for diabetes mellitus, the three strongest risk factors associated with cardiovascular disease. Indeed, none of the nine studies controls for more than a limited handful of the potential 300 or more reported identifiable risk factors. In the absence of controlling these variables, the reported outcomes for implied ETS exposure are impossible to interpret with any confidence or meaning. In fact, it is now scientifically unaccept- able to undertake an investigation (or, for that matter, unacceptable to accept the contention of a published study) on cardiovascular disease with- out properly controlling for the three best known and widely accepted risk factors-high plasma cholesterol levels, high blood pressure, and dia- betes mellitus. Lifestyle Factors Active smokers are different from nonsmokers in a remarkable number of ways. In general, smokers as a group appear to have a lifestyle that results in a clustering of several adverse health risk factors. Smokers tend to drink more alcohol than nonsmokers, drink more coffee, live a more stressful life, behave more aggressively, have a lower socioeconomic status, exercise less, sleep less, and spend less time on enjoyable hobbies. Smokers tend to be less safety conscious, not to wear seatbelts, to have accidents more frequently, and to behave in ways that increase their risks for injury. Smokers, on average, are less educated than nonsmokers. Smokers are less "health con- scious" than nonsmokers, and they have a more negative attitude about modifying behavior to 18 Consumers' Research