/

THE TOBACCO INSTITUTEof NEW ZEALAND

THE TOBACCO INSTITUTE of NEW ZEALAND8th FLOOR, DILWORi1BUILD1NG, CNR QUEEN AND CUSTOMS STREET EAST.P.O. BOX 1582, AUCKLAND I, NEW ZEALAND. PHONE 797-393._

RESPONSE TO

A DISCUSSION PAPER BYTHE DEPARTMENT OF HEALTH

'CREATING SMOKEFREE INDOORENVIRONMENTS'

DECEMBER 1988

r.

CONTENTS

HeaderPage

BuffOVERVIEW

GoldINTRODUCTION andENVIRONMENTAL TOBACCO SMOKE ISSUES DISCUSSION

GreyHEALTH DEPARTMENT 'FEEDBACK' FORM

Lemon A CHAPTER-BY-CHAPTER DEBATE OFTHE HEALTH DEPARTMENT DISCUSSION PAPER

BuffCOMMENTS ON THE PUBLIC DISCUSSION PAPER -A PAPER BY DR W ALLAN CRAWFORD

Green STATEMENT BY GRAY ROBERTSON TOTHE HOUSE OF REPRESENTATIVES ONENVIRONMENTAL TOBACCO SMOKE

BlueENVIRONMENTAL TOBACCO SMOKE -A REVIEW OF THE LITERATURE

PinkCOMMENTS ON THE PUBLIC DISCUSSION PAPER -A PAPER BY DR LARRY C HOLCOMB

GreyTOBACCO OBSERVER - OCTOBER 1988"INDOOR AIR QUALITY"

Cream TOBACCO OBSERVER - NOVEMBER 1988"SMOKING AND THE STATE"

Green APPENDIX TO "CLEARING THE AIR" BYDR ROBERT D TOLLISONENTITLED 'THE RESEARCH EVIDENCE ON ENVIRONMENTALTOBACCO SMOKE

*** * * * *****

Page

1-6

1-12

1

1-19

1-29

1-16

1-87

1-10

1-4

1-4

OVERVIEW

If -the interests of informed and balanced public debate concerningenvironmental tobacco smoke (ET.S) are to be adequately served, thediscussion paper Smokefree Indoor Environments" published bythe New Zealand Department of Health should be considered alongside thissubmission.

It is our view that the discussion paper on its own does not meetacceptable criteria for a public discussion document. It presents one sideof the argument and pre-empts discussion by not only stating the authors'preferred legislative options but also by segmenting them from otheroptions by the use of creative printing techniques.

As this submission will demonstrate, the discussion paper is extremelyselective as to references, fails to provide all of the information essentialto a balanced discussion on the issue and concludes with a questionnairewhich does not meet recognised standards and which will inevitably gainthe responses the authors seek.

An objective review of all the information available on the subject clearlyestablishes that legislation is unwarranted, and if implemented, cannot bebased on scientific proof that the health of nonsmokers is damaged byETS.

This is not to say that there should be unfettered rights for smokers anymore than there should be for nonsmokers. This industry submission doesnot contend that only industry or smokers' views should apply. Itacknowledges that Government has an advisory role and that more researchis needed; it contends that market forces should be given the opportunityof achieving a solution; it seeks more balance in the media and does notpromote a case for smokers to do as they wish without any regard fornonsmokers.

In initially addressing broad issues, this submission establishes a soundcase for commonsense, good manners and corporate and workplaceresponsibility to deal with the issues generated by those opposed to ETS.

1

It establishes that smoking behaviour can be approached as a matter ofprivate not public choice, with various sectors adopting smoking policiestheir patrons prefer.

This submission opposes a bureaucratic choice of a single rule for all orthe predominance of personal preferences by one group over another bethey smokers or nonsmokers.

In a chapter by chapter debate this submission redresses the imbalanceinherent in the discussion paper. Perhaps most importantly this section ofthis submission highlights the occasions on which the discussion paperselectively quotes from authoritative reviews such as the National ResearchCouncil 1986 Report on ETS - a report which takes a much more balancedview than one would suppose from the way the discussion paper referencesare made.

This Report of the National Research Council which the discussion paperacknowledges as being authoritative,: -

does not suggest that the question of whether ETS has anysignificant health effects has been decided,

does not conclude that occasional exposure to ETS poses a healthrisk to nonsmokers,

does not suggest that ETS impairs adult respiratory function,

does not conclude that heart disease has been scientifically linked toETS

does not conclude that allergies to tobacco smoke have been firmlyestablished and,

does not conclude that valid evidence establishes ETS as a risk toasthmatics.

3

This submission also outlines why the public feedback form in thediscussion paper, which will no doubt be used to generate politicalinfluence, does not meet the standards set down by the Market ResearchSociety, nor those of the Survey Appraisals Committee of the New ZealandStatistical Association. As such, the questionaire cannot be interpretedas giving an accurate and unbiased indication of public opinion nor canresults compiled from the feedback form be used as a reliable indicator asto the direction in which social policy making on ETS should be steered.

An independent paper by a recognised expert in occupational andenvironmental health, Dr W Allan Crawford, highlights perhaps the majordeficiency in the discussion document. Dr Crawford maintains that thediscussion paper "would have benefitted much from adopting the NationalAcademy of Science and the International Agency for Research on Cancerapproaches by honestly telling the public what is known and what is notknown.

Dr Crawford's paper also gives weight to this industry's contention thatETS is becoming a scapegoat for non-ETS risks as a result of thediscussion paper giving no consideration "to other pollutants whichpervade ubiquitously our air environment at home, work or leisure andwhich are being shown to be environmental hazards to health."

A statement by Gray Robertson an indoor air expert, in a testimony beforethe United States House of Representatives, highlights the inherentdangers in applying tunnel vision to the issue of ETS in that "toconcentrate upon one element of this matrix, environmental tobacco smoke,would be inefficient and dangerous. It would be naive to assume that theremoval or control of ET S, the most visible indoor air pollutant, wouldsolve indoor air pollution problems. The step (smoking ban) ignores themany hidden sources of indoor contamination, and can lead - at least inthe short run - to a false sense of security."

Two issues of the US publication "Tobacco Observer" are included asappendices.The first highlights a problem also faced by the HealthDepartment in its New Zealand head office. The issue outlines theproblems faced by the US Environmental Protection Agency in its ownoffice and quotes a specialist from that department as saying that "(thebuilding) should be declared a Superfund toxic waste site."

4

That specialist is referring to the fact that the top US pollution police -the EPA - falling victim to sick building syndrome provides a goodindication of the severity of indoor air pollution problems. That theincident occurred despite severe smoking restrictions within the EPAoffice, demonstrates what many experts have been saying for years: ETSis a symptom not a cause of poor indoor air.

The Tobacco Industry in New Zealand is on record as having offered topay for a scientific air quality measurement and ventilation inspection ofthe head office of the Health Department of New Zealand in response to apublic statement that the office has air quality problems in spite of asmoking ban. To date, this offer has been declined.

Unlike the Department of Health in New Zealand the EPA did agree to callin specialists to examine its building and found that the building'sventilation system was inadequate and that air quality problems wererelated to the materials in the building, not to tobacco smoke. To itscredit the EPA officials have removed the materials which caused theimmediate crises but they are still determining how to clean up the air inthe clean air headquarters of that nation.

The second issue of the "Tobacco Observer", in dealing with claims of"social costs", interviews a prominent professor of economics who states "ifthe anti-smoking forces want to persist with their social cost calculations,they should be consistent and apply them to other commodities as well.In fact, an entire social cost "industry" could be set at work to calculatethe "social cost" of eating red meat, skydiving, skiing, leisure driving,cholesterol intake, auto racing, team sports, sedentary lifestyles and manyother everyday activities.

The "social costs" of these activities would run well over $200 billion peryear, but what does it prove? Should we control people's lives throughfederal regulations to squelch individual choice and voluntary risk-taking?And, what are the social costs of depriving individuals of their freedom ofchoice?"

An independent paper prepared by an American ventilation expert, DrLarry C Holcomb, B . A. , M. Sc. ,Ph.D., concludes that "if one is trulyinterested in achieving a healthy indoor environment one must look at allthe aspects of clean indoor air and determine the best solution for theoverall problem. Legislating smoking in public places and thus trying tocontrol exposure to ETS will not solve a problem with indoor air qualityand would provide negligible health benefits if any."

The final appendix to the submission is reproduced from the publication"Clearing the Air" by Dr Robert Tollison. It highlights the extent ofresearch evidence on ETS and in over ninety references, establishes thefact that the debate on ETS is not concluded and that the conclusions ofthe discussion paper and its preferred options are unwarranted andunnecessary based on the scientific evidence.

- Smoking may be annoying to some but the solution is not to ban it. Nosingle group's demands should prevail because the claim that ETS has beenshown to have significant health effects is not substantiated.

There are sufficient incentives in the private sector, without Governmentintervention for the right type of indoor environment to be provided andthere are sufficient incentives, again without Government intervention, forthe employer to provide the indoor environment preferred by his workers.It is not more Government legislative intervention that is needed, butrather administrative and educative support.

Indoor environmental air quality can be improved and at the same timeaccommodate both smokers and nonsmokers.

It should also be recognised that pressures for employers to improveindoor air quality may be reduced if Government legislates on ETS.Employers could consider that the responsibility to ensure an acceptablequality of indoor air has become the responsibility of central Governmentrather than their own.

This submission places in the public domain a wealth of information,evidence and opinions on the question of ETS which is sadly not includedin the Health Department discussion paper.

We invite the Department's views on the submission. We propose that theDepartment would be failing in its advisory function if it were not toacknowledge that the sheer weight of evidence presented brings its owndiscussion paper into question and renders any recommendation toGovernment for legislation on public smoking at this time as beingunsanctioned, unreasonable and unnecessary.

*************

INTRODUCTIONAND

ENVIRONMENTAL TOBACCO SMOKE ISSUES DISCUSSION

1

SUBMISSION TO THE EXECUTIVE OFFICERHEALTH PROMOTION PROGRAMME

HEALTH DEPARTMENT, P 0 BOX 5013, WELLINGTON

ENVIRONMENTAL TOBACCO SMOKE - ISSUES DISCUSSION

An unbiased non-selective review of the literature on environmental tobaccosmoke clearly establishes that a case has not been concluded and that theissue of environmental tobacco smoke raises many complex issues.

Some of the issues raised in a debate on environmental tobacco smoke(ETS) are:

Possible health effects of ETS on nonsmokers.

The measurement of ETS and other substances in indoorenvironments.

Economics and the appropriate public policy towards ETS.

Law and civil liberties.

Industrial relations.

The behaviour of interest groups and bureaucracies.

How smoking and ETS differ in a meaningful way from otherbehaviour to which some people may object.

The tone of the debate about ETS is often complex and confusing. It musttherefore be difficult for the average person to assess the issuessurrounding ETS. If anything, the public may have been influenced bythe anti smoking movement to believe that ETS is equivalent to a modernplague that kills and harms thousands of people yearly. This is clearlynot the case.

The basic facts about ETS are quite different from the stereotypical"facts" argued by the anti-smoking movement. The summary view of theliterature is that:

1. There is no compelling evidence that ETS in indoor environments is ahazard to health.

2. Sweeping public health legislation to control ETS is not justified bythe evidence on any perceived public health impacts of ETS.

3. ETS is a trivial factor in most indoor environments. In fact,building measurements indicate that there are indeed potentsubstances present in indoor environments, such as asbestos, butthat ETS is typically a very minor substance in indoor environments.For example, many buildings that have banned smoking have poorventilation systems based on recycled air. A smoking ban in such abuilding removes a minor substance in its indoor air while ignoringother major indoor air pollutants. Cleaner indoor air environmentsrequire better ventilation systems, not smoking bans.

4. There is a property rights perspective on ETS, with clear incentivesfor the owner to provide the type of indoor environment preferredby patrons. To do otherwise would be to risk losing money.Government intervention to ban smoking cannot improve upon theincentives that already exist in the private sector. Some restaurantswill ban smoking, and others will not. The point is that Governmentregulation is not necessary in deciding where and when people canor cannot smoke.

5. There is an industrial relations perspective. Public legislation aboutworkplace smoking gives to management unnecessarily arbitrarypower over workers. There is now sufficient incentive, withoutGovernment intervention, for the employer to provide the indoorenvironment preferred by his workers. Again, this does not meanthat all workplace environments will be smoke free. The point isthat the managers of companies will provide the environment that isin the best interests of their workers.

6. The press has sensationalised the issue of ETS, as has the anti-smoking lobby.

7. The value of personal autonomy ought to prevail in the debate aboutpublic smoking.Definable civil liberties outweigh alleged "socialcosts," which are no where proven or even satisfactorily advanced.

3

8.We all find some aspect of others' behaviour to which we object.However, if we try to regulate all the things to which some peoplemay object, we end up regulating everything, and we all end upworse off. A little tolerance goes a long way in a modern society.

There is a cogent case to make that ETS does NOT pose a huge socialissue calling for Government intervention into the affairs of privatecitizens. Smokers and nonsmokers will work their differences out,companies will provide indoor environments that their workers prefer,businesses will do the same for their customers, and the society andeconomy in general will go about its business in such a fashion as tomitigate any problems supposedly attributable to ETS. This does not meanthat smokers will have unfettered rights to smoke. To the contrary, theirbehaviour will be constrained by private action. The hand of Governmentregulation is not necessary.

THE ETS CAMPAIGN

Seen in the perspective of the anti smoking campaigners, the issue is quitesimple: not only is ETS an annoyance to nonsmokers (an "invasion of theirprivate space"), but it is an alleged matter of public health. ETS, albeitin a less intensive way, allegedly promulgates the same diseases thatsmoking supposedly does. Smokers thus constitute not only an annoyancebut are seen as a health hazard to nonsmokers. The scientific basis forthis belief is allegedly conclusive, having been "legitimated" by anauthority no less than that of the US Surgeon General. Anyone whoquestions the belief, therefore, is either motivated by wishful thinking(smokers, in the main) or by commercial interest (the tobacco industry).A "health hazard" having been established, Government action is obviouslycalled for. The campaign seeks progressively to limit and eventually toeliminate ETS in all places where smokers and nonsmokers mingle. Theopenly avowed end result is to proscribe smoking in any public area.

The reality is much more complicated. Very few people can claim expertisein physiology, yet people have convictions on ETS. These convictions arebased on faith in authorities. But since authorities conflict, the individualcampaigner chooses which authority to give credence to. People whobelieve very little of what Government tells them on other importantmatters (foreign policy, for instance) will regard an issue as closedbecause the US Surgeon General or the Health Department has spoken onit.

In the many years of controversy over smoking and health, the phrase'vested interests' has been used almost exclusively to refer to the tobaccoindustry, and it is again so used in discussions of ETS. The tobaccoindustry would be very pleased if there were evidence to show that ETShelps cure depression, the common cold, or some other condition that ailspeople. To concede this obvious point, however, in no way implies thatthere is no vested interest, no ideology, on the other side of this debate.

There is the vested interest of the anti-smoking movement itself. Themovement has been successful in enlisting allies in the bureaucracies. Thefreely proclaimed agenda of the anti smoking campaign is to stigmatize,segregate, and (at least partially) criminalise smoking. An oftenannounced goal of the campaign is to reduce smoking to a sociallyunacceptable activity except in private homes.

But the anti smoking agenda will run into problems wherever there is astrong tradition of individual rights. Smoking, after all, is a voluntaryactivity. The early period of the anti smoking movement was to alert("educate") smokers to the alleged health hazards. But as the movementincreasingly sought Government action to proscribe and regulate smoking,some way had to be found to argue that smokers constitute a danger notonly to themselves but to others as well. A doctrine of individual rightscould no longer stand in the way of coercive Government action.

A new figure was introduced into the debate, the innocent bystander.Thus, the ETS issue introduced a whole new category of "victims" into thedebate over smoking. Not surprisingly, this issue has , occupied anincreasingly important place in the anti smoking propaganda, and antismoking activists have been particularly angered by any public statementsof scientists or others to the effect that the evidence on the dangers ofETS is less than overwhelming.

HOW DO PEOPLE REALLY BEHAVE?

The ETS issue extends to punitive action against smokers. In the contextof this issue, the smoker is cast in the role of villain. A combination ofaggressive propaganda from the anti smoking lobby and the instinctivehostility of some toward corporate profitability has created a social climatein which smokers could become an oppressed minority.

4

5

There is an understanding that any attempt to impose one person'spreferences on the behaviour of another will set off attempts to have one'sown behaviour constrained in a like fashion. An attitude of "live and letlive," or mutual tolerance and mutual respect, may be better for all of us,despite the occasional deviation from ordinary standards of commondecency.

Such an attitude would seem to be held by anyone who claimed to hold todemocratic and individualistic values, in which each person's preferencescount equally with those of others. Those who think that his or her ownpreferences are "superior to," "better than" or "more correct" than thoseof others, will try to control the behaviour of everyone else, while holdingfast to his or her own liberty to do as he or she pleases.

We can resort to politics to impose our own preferences on the behaviourof others. But others can do the same. The political process is double-edged. If one group can use it to advantage in imposing personalpreferences over the behaviour of others, it can be used to theirdisadvantage by imposing the preferences of others on their ownbehaviour.

THE MEDIA AND ETS

On December 16, 1986, the US Surgeon General released a report onenvironmental tobacco smoke. The way its findings have been reported bythe press is disturbing. Upon publication of the report, the SurgeonGeneral added an obiter dictum that the report proves environmentaltobacco smoke is harmful to nonsmokers. That statement is untrue, yetthe press dug no further. The Surgeon General's report does NOTprovide irrefutable evidence of environmental tobacco smoke's mortal peril.In fact on no fewer than 147 occasions, the report demurs from makingthis claim and actually points out that it cannot conclusively establish sucha claim.

Nonetheless, the international press coverage of the Surgeon General'scomments reported the story as though the report had amassed unassailableproof of the mortal danger of environmental tobacco smoke to innocent, nonsmoking bystanders.

6

To understand the apparently unstoppable dissemination of exaggeratedclaims against environmental tobacco smoke, we must recognise the rightsof those who are militant on behalf of one perceived cause or another.However the rights of the reformists and those who report them must betempered by an obligation to be both accurate and fair.

The press is itself one of the leading institutions of reform. It is missionoriented. Its first mission is to discover and disseminate the news. Whatif the news is irrelevant to an audience? What if it is embarrassing ordisturbing? It is sent out anyway. And what if it is inaccurate? Thatis, what if a fact is wrong? Perhaps the facts are right but the emphasisis wrong. The press is not daunted by any of these questions. Theyknow that they serve a legitimate purpose: general enlightenment andimprovement.

Whatever is seen at the time as improving and enlightening humanity canbe considered news, and this news must be reported in such a way as toattract the interest of the audience. Reformist issues catch the fancy ofthe press and tend to get reported as reformers would have themreported.

ETS is now such an issue. The press can read the complete SurgeonGeneral's report contradicting his separate charge that environmentalsmoking causes cancer, but then go on to report the finding in the wayreformers would like that finding to be reported. The press will overlookall the report's qualifications and assert that the report argues thatenvironmental tobacco does cause mortal illness.

This is highly irresponsible. An important institution is discreditingitself. The press overstates the threat of environmental tobacco smoke aspresented in the Surgeon General's report, and once a line is establishedit is repeated endlessly. This "line" on environmental tobacco smoke isthat it kills thousands of people every year. Never does the press statethat most studies on the connection between environmental tobacco smokeand cancer have failed to establish even a statistical relationship, muchless a causal connection, or that a nonsmoker would have to sit behind adesk or in a restaurant for hundreds of hours non-stop to be exposed tothe vaporous equivalent of one cigarette.

The general attitude of the media in its reporting of the ETS issue iscapricious, intolerant and misinformed.

PRIVATE RIGHTS AND PUBLIC ACTION AND THEIR RESOLUTION

Whether an activity is alleged to be harmful to the individual is not therelevant criterion for law in a free society. Many human activities canresult in personal harm. Skiing, football, boxing, skydiving, andswimming are activities where a person puts safety at risk. Soft drinkconsumption by children may lead to hyperactivity and aggravate allergies.Medical experts agree there is no dietary reason for adding salt to foodand yet many people do so and subsequently increase their risk ofhypertension and other circulatory diseases. Sunbathing can lead to skincancer. A higher risk of coronary disease is associated with highcholesterol consumption. Diving, mining, and bridge construction oftenresult in injuries and fatalities. These are but a few ways we take riskswith the length and quality of our own lives. Yet no one would advocatelegal prohibitions against sunbathing, salt, red meat consumption, andother harmful and risky activities. Instead we leave it up to educationand persuasion but otherwise leave the individual free to choose.

There is the assertion that people who place themselves at risk by smokingmay become a financial burden to society. The conclusion reached by thisargument is that since society has the ultimate responsibility of paying thebill, it has a right to restrict people's right to take risks with their healthand safety.

Such a position raises an important question: if legislative intervention isdeemed legitimate in the case of cigarette smoking is it not a logicalextension of the same principle to regulate other activities, such assunbathing and salt consumption, that may apparently jeopardize healthand safety and cause people to become a financial burden on society?

7

HARM TO OTHERS

Harmful or beneficial side effects may arise in the course of consumingsome commodities and affect people not directly involved in the consumptionof the goods in question. Such external economies of consumption occurwhen one person's consumption indirectly benefits another. (When onehome owner makes increased expenditures to maintain property, he or sheunavoidably raises the attractiveness and hence the value of neighbours'property). External diseconomies of consumption occur when one person'sconsumption indirectly harms another, (an airplane flying over aproperty). In either case, there is no compensation for the harm orbenefit.

It is difficult to conceive of a world where all possible harm has beenoutlawed. In such a world, activities we normally take for granted couldnot occur.

Moreover, the nonsmoker, as well as the smoker, places subjective valuesand trade-offs on the harm or benefit he or she is willing to accept.Would a nonsmoker faced with a broken down car refuse the offer of apassing motorist who smoked? Would a nonsmoker turn down an interviewfor a job he has always wanted because the employer smoked?

PROPERTY RIGHTS

Resolution of conflict over smoking is straightforward and reasonablysimple where clearly defined property rights prevail - for example inprivate homes. However, the conflict of wishes between smokers andnonsmokers becomes a more contentious issue in publicly owned placesbecause property rights are ill defined in the sense of who owns what.Publicly owned places have been paid for by taxes and rates. Presumablysmokers as well as nonsmokers pay taxes, and both can claim partialownership of publicly owned places. There is nothing in principle thatsays whose rights should prevail. More often than not, whose wishesprevail is a result of which group has the dominant political power at themoment.

8

Smoking in factories, offices, restaurants, and stores, which cater to thepublic as workers or customers, are the current focus of the anti smokingmovement. Most of these establishments are privately owned. The factthat they do business with the general public does not make them publiclyowned. Private property rights are thus well defined and should resolvethe smoking controversy just as it does in the case of privately ownedhomes. The owner makes the decision whether smoking is to be permittedon the premises and accepts the economic and social consequences.

What are the options available to the nonsmoker or anti smoking workerunder these circumstances? The most obvious is for the worker towithhold his or her labour from employers who permit smoking on theirwork sites. Another option is to try to convince, cajole, or embarrasssmoking co-workers into abstention.

Employers have a common law duty, given the state of technology, toensure a safe workplace. There is a similar common law expectation thathome owners must take reasonable precautions to ensure safety for theirvisitors. At the same time as the common law holds employers and homeowners accountable to take reasonable measures to ensure the safety ofemployees and visitors, it also places a burden on the latter to takereasonable measures to protect themselves.

Unlike the cases of slippery stairs, radioactivity, and certain noxiousfumes, the fact that people are smoking is visible, apparent, and cheap todetect. The remedies available to a worker in a workplace where theemployer permits smoking are identical to those available to a nonsmokerwho visits a house where the owner permits smoking: avoid the situation.To rejoin to this remedy by saying the visitor has choices while theworker does not is the same as asserting the intersection of two improbableevents: there are no other jobs available, and all employers permitsmoking.

Obviously similar remedies are available to nonsmokers in their choice ofrestaurants, bars, and other places of entertainment. They are free towithhold patronage from establishments that permit smoking. To facilitatenonsmokers' options or, for that matter, smokers' as well, it is necessaryfor owners to make readily available information about whether smoking ispermitted at their establishments.

9

The tradition of common law respect for private property may explain whypeople who wish to abrogate the private property rights of others choosestatutory relief rather than appeal to civil courts.

Abrogation of property rights has become the most popular route in theanti smokers' drive for prohibition. They have realised the age-oldpractice of using legislation to accomplish what they deem more costlythrough voluntary market transactions. Their agenda is to use thelegislature, both Central and Local Government, to take away part of theowner's decision on how his or her property is used.

CONFLICT RESOLUTION

People exhibit many likes and dislikes. Political involvement in socialactivity results in the gain to one person becoming a loss to another.Market allocation reduces the potential for conflict.

Smoking may be an annoyance to those around the smoker. Such anobservation does not place smoking in a category by itself. Loud music,in discotheques and bars, may lead to hearing loss and headaches.Typically the person who is offended by the loud music either leaves theestablishment or does not enter in the first place. As long as louddiscotheque music is contained within the establishment, most peopleoffended by it would not lobby for enactment of a law specifying aparticular decibel range. There is no ugly confrontation. People who likeloud music go to discos, and those who do not go elsewhere. A similarresponse is available to nonsmokers.

Most human conduct is regulated not by law but by forces such as socialcustom, etiquette, and peer pressure. Behaviour regulated in this wayincludes (or included) men giving up their seat for women, boisterousbehaviour, child rearing, foul body odour, coughing around others, anddress codes. Violation of these customs is met with social sanctionsleading to general compliance without their being regulated by threat ofGovernment intervention.

10

Because cigarette smoking is a private act with which some peopledisagree, anti smokers call for Government regulation. However, it ispossible to show that every private act has external effects of one sort oranother. The list of private acts that can reduce the physical andpsychological well-being of other people is enormous and includes loudmusic, boisterous behaviour, poor table manners, dishevelled appearance,sexual promiscuity, obscene language, trespass, obesity, meat eating andflatulence. Therefore to call for Government regulation of private actssolely because they produce external costs and reduce the welfare of otherpeople is to call for Government regulation of everything.

11

CONCLUSIONS

1. An objective review of all the literature leads to the conclusion thatlegislation is unwarranted, unnecessary and an infringement onfundamental rights of New Zealanders.

2. Commonsense, accommodation, patience, scientific research, goodmanners, corporate and workplace responsibility - these can all dealwith the social issues posed by ETS.

3. Smoking behaviour can be approached as a matter of private choice,not public choice.

4. Entities in the private and public sectors can adopt the smokingpolicies that their workers and patrons prefer. This does not meanlaissez-faire for smokers. Some organisations will ban smoking, andsome will not. Some will manage their ventilation systems properly.Still others will segregate smokers and nonsmokers. And so on.The pattern that emerges will be the result of private alternativesand choices.

5. The public choice alternative involves having bureaucracy choose asingle rule for all.

6. The private choice alternative is consistent with the practices of afree society. The public choice alternative is the route to dominanceof personal preferences by one group over another. If that can beimplemented over smokers - who is next?

12

* * * * * * *

HEALTH DEPARTMENT DISCUSSION PAPER -THE 'FEEDBACK FORM'

THE 'FEEDBACK FORM'

Information collected by this feedback form should be interpretedwith extreme caution, for the following reasons:

(a) The construction of the questionnaire is not soundly based,and it is likely to induce biased responses. The questionnairedoes not meet the standards set down by the Market ResearchSociety, nor those of the Survey Appraisals Committee of theNew Zealand Statistical Association.

(b) The respondents cannot be said to be representative of thepopulation of New Zealand, nor can the number of responsesbe used to assess the level of approval or otherwise of theproposals contained therein.

(c) The format of the questionnaire invites a biased response.The clear encouragement in each question is to agree. Allthat is necessary is to place a tick in a box provided, whereasa respondent who may wish to disagree cannot do so withoutconsiderable additional effort, including making a counter-suggestion. In addition, the space provided for counter-suggestions is inadequate, having no additional lines on whichto write. This approach predisposes the respondent to takethe simpler option of agreement, and furthermore, discouragesparticipation by those who disagree with the proposals setforth.

(d) There is no provision made for respondents to indicate aneutral or indifferent stand with a "don't know/don't care"response.

(e) The use of the word "could" will distort results. Theproportion of respondents who may agree cannot necessarily betaken as approval/ support for any proposed legislation. Manyrespondents may simply agree to the fact that legislation couldbe enacted. Quite clearly, legislation could be enacted whichforbade a large number of lawful activities, and in fact thishas happened many times in the past.

(f) The invitation to respondents to use a photocopy of thequestionnaire will mean that the majority of those who send inphotocopy responses will not have read the background paper.

(g) It must be concluded that any data collected by means of thisquestionnaire cannot be interpreted as giving an accurate andunbiased indication of general public opinion, and should notbe used as an indicator as to the direction in which socialpolicy making should be steered.

1

THE HEALTH DEPARTMENT DISCUSSION PAPER -A CHAPTER-BY-CHAPTER DEBATE

The stated purpose of the Health Department publication is as a'discussion paper'. However, it can in no way be regarded as apolitically neutral scientific background paper.

The Hon David Caygill, Minister of Health, in the Foreword, sets thetone for what is to follow. It would seem that, contrary to hisassertion that "no decisions have yet been made", the issue isdecided and it remains only to decide on a course of action. Thediscussion paper (DP) is quite clearly a political document, designedto persuade, not to inform. Its selective use of information andmisinformation is statistically distressing. This is a global commenton DP, in that the language used throughout is of this nature, andto continually point out the emotive sentences and misinterpretationsof data would be both tedious and self-defeating.

This section addresses the issues raised in the Health Departmentdocument. The format will be to take DP chapter by chapter and toindividually state and refute the arguments put forward in thatdocument. Some chapters deal almost exclusively with matters whichare acknowledged to be matters of opinion; these will be largely leftalone. As these opinions rely upon flawed data or misinterpretationsof data, the opinions and options for action must also flawed. Theseopinions and options fall into several easily identifiable groups, andthey appear repeatedly throughout the document.

Irrespective of the flawed nature of the arguments upon which theyrest, these opinions and options are unreasonable, and will beafforded a brief review. Where opinions have been presented asfacts, this will be pointed out and counter-arguments put forwardwith supporting references. This paper may be regarded as a set ofcounter-propositions and arguments, but its presentation will not bein emotive or misleading terms.

Outline

This paper provides objective information which has been selectivelyomitted from DP, in order to provide more information so that abalanced assessment of any ETS (Environmental Tobacco Smoke)effects on health may be made. It will become progressively clearerthat legislative options for action are unreasonable and unnecessary.

The burning cigarette releases a large number of chemicals into theair (DP p4).The concentrations of these chemicals in the airhowever, are generally very small. It is worth noting that theburning of coal and wood likewise releases a great number ofchemicals into the air, and in fact burning coal also releases a notinsubstantial amount of radioactivity into the air. Furthermore,while it is true that some of the chemicals released by burningcigarettes can be detected in the urine, blood and saliva ofnonsmokers who are exposed to ETS, it is not true that these havebeen detected in concentrations that have ever been shown to beharmful.

It is clear that ETS may be a source of irritation to someindividuals; however, the evidence that it is associated with diseasesin children and other nonsmokers is inconclusive. In fact, theNational Research Council report on Environmental Tobacco Smoke[27] quite specifically does not conclude that heart disease has beenscientifically linked to ETS (Pp 265-266). There is further commenton the contents of this report later in this paper.

I

The Health Department conducted a poll in New Zealand in November1987, which suggested that about 80% of people believe that thehealth of nonsmokers can be damaged by other people's smoke. Thisis of course entirely irrelevant to the question of whether ETS isactually harmful. Similarly, comments on other countries havingsuch draconian legislation are also irrelevant.

The extent of exposure to ETS is considerably lower than is impliedin DP. Studies measuring exposure of flight attendants to nicotineand carbon monoxide have consistently concluded that exposure istoo low to result in any physiological effects [55], [56].

Throughout DP, there is an implicit (occasionally explicit) assumptionthat the case has been proven, the decision has been made, and thatthe purpose of the 'discussion document' is simply to plan forimplementation. This is certainly not the line taken by thedocuments which DP relies upon for its 'scientific' support.(Specific references are given later in this paper).

In the back of DP is a 'Feedback form'. It is to be hoped that thisform did not form the basis of the Health Department'squestionnaire; the form contravenes almost all standard statisticalpractice for questionnaire design and will be discussed in anappendix to this submission.

DP Introduction

Smoking touches the daily lives of most people. It is claimed in DPthat the 'medical and scientific evidence shows that tobacco harmssmokers and it kills over 4000 smokers each year'. This claim isbased on the paper by Gray et al [1] and the Health Departmentpublication 'The Big Kill' [2]. The claims made in these papers arenot universally accepted, and the Gray paper, for example, rests onthe work of King et al [ 3 ], who state quite clearly that theirestimates depend on a rather 'sweeping assumption' that the effect oftobacco smoking is causative in nature, and furthermore that all ofthe excess risk associated with smokers can be attributed tosmoking. These assumptions have been called into serious doubt bya number of authors, notably Burch [4],[5],[61; Feinstein and Wells[7], Seltzer [81,[91, and Heath [ 10 ] . The problems of estimationwhich can be induced by differential misclassification of risk factorshas been commented upon at some length by Kraemer [30].

DP states that the discussion paper has been written to provideresearch-based information on the nature of tobacco smoke andsmoking, to promote discussion on the issues, and to invite commentsfrom the public on the 'options for action'. As far as DP isconcerned, the 'issues' seem to be about how far it is possible to goin condemning the tobacco industry.

2

There is a statement to the effect that the 'creation of smoke freeenvironments is not intended to stop people from smoking but toprotect the health of nonsmokers', and that DP 'provides informationon tobacco smoke and its health effects'. The tone and contents ofthe document would seem to indicate that both of these claims areuntrue. This is clear from the Minister's Foreword, in which hestates that 'Creating smoke free environments is an idea whose timehas come'. It seems clear from this that the decision has beenmade; this is a marketing document, selling the 'idea whose time hascome'. On the one hand, DP is a political document designed topersuade the public to support a legislative change which wouldconstrain the rights of a significant proportion of the populace, andon the other hand, as shall become obvious, DP presents only acarefully selected portion of the evidence concerning the allegedhealth effects of cigarette smoking and of ETS.

Since the antismoking drive involves curtailment of smokers' rights,there should be strong scientific evidence against ETS before anyaction is contemplated. DP does not seem to acknowledge this point,but rather takes a biased selection of the available evidence. Thisselection is criticised in some detail later in this paper.

Themes in the Discussion Paper

There are a number of themes which are evident throughoutDP. These themes are repeatedly returned to, and are listedhere, with brief comments

(a)Nonsmokers should be able to breathe smokefree air at all times.

This concept of a 'pure environment', is unrealistic, and that theblame for the lack of such purity should be placed at the feet of thecigarette smoker is equally unrealistic. The mere presence of ahuman body in a room raises the ambient temperature and humidity,as well as the carbon dioxide content of the room's atmosphere.Common office supplies and equipment release chemicals of variousnatures; clothing and furnishings contribute fibres; cleaning oftenincreases the airborne concentration of smaller particles. Cooking,coal and wood fibres generate both vast numbers of airborneparticles and various organic and inorganic gases. Adhesives are asource of organic pollutants, radon is produced from soil andburning coal, carbon monoxide and carbon dioxide are products ofcombustion, not only of tobacco products. Asbestos is of course awell documented carcinogen which for many years was used in a widevariety of building and convenience materials.

(b)Restrictive legislation is both practical and wellaccepted.

A six month trial ban of smoking on all carriages on inter-urbantrains in NSW came into effect on 5 April 1987. This particularpiece of restrictive legislation turned out to be very impractical, andwas certainly not 'well accepted'. The ban was subsequentlyrevoked in the third month of operation, on 13 July 1987. Thereasons given were that the majority of commuters favoured theprevious system of segregation. The Minister further stated thatthere had "been unfortunate incidents aboard trains with acts ofprovocation and retaliation between smokers and nonsmokers" whichhe found "distasteful".

3

A similar experience was Beverly Hills in 1987, when a smokingbylaw prohibiting smoking in restaurants and other public places wasoverthrown. Several court cases are pending on the grounds thatseveral local restaurateurs are suing the country council for loss ofearnings due to the imposition of the bylaw.

Laws in Texas, Florida, Illinois and Virginia have been amended orrevoked by various courts that have ruled that general prohibitionson public smoking are constitutionally broad, vague, indefinite andin violation of the constitutional guarantee of equal protection.

(c) The establishment of smoke free indoorenvironments as a norm in society and emphasison the importance of choice for nonsmokers.

Bolton [45] has stated that:"Business .. should be concerned when anything that is legal isbanned by decree. The role of Government should be to protectfreedoms and that includes not only the freedom to consume lawfulproducts (and cigarette and alcohol are both still lawful adultproducts) but also the freedom to advertise them."

Tollison [46] states:"If Government is going to be allowed to enact coercive measuresthat arbitrarily restrict the liberties and the rights of smokers,where will employment of this restrictive power end?

"If Government is not expressly prohibited from interfering with thefree expression of voluntary choice by peaceful adults in the case oftobacco, what is to restrain the Government from similarly imposingother restrictions on political expression or religious practice. Whatis to prevent a conservative Government from making liberalspeeches illegal, or vice versa?

"The liberty of all depends on drawing a line beyond whichGovernment may not cross. That line must represent the rejectionby the general public of the principle of unlimited Governmentintervention into the peaceful domain of social life.Tobaccoregulation is only one example. Every time Government powerextends beyond the realm of the simple protection of individualsagainst force and fraud the very existence of the free andcompetitive social order is increasingly - threatened.

(d) Legislation for 'public places' should be at thelocal authority level.

This has occurred already in Waitemata City and elsewhere. Thesebylaws were passed with few dissenters, and with even fewer effortsat consultation, either with the shopkeepers in the community whosefreedoms were at risk, or the members of the community itself.

(e) DP has been written to provide research-basedinformation on the nature of tobacco smoke andpassive smoking; to promote discussion of the issues,and to invite public comment.

4

The information within DP on the nature of tobacco smoke and theeffects of passive smoking has been carefully selected to persuadethe reader that it has been established beyond all doubt that ETScauses a number of diseases. This assertion is not true, and hasbeen refuted by many prominent scientists [5,8,9,10,11,23,53,54],The 'issues' upon which discussion is being promoted consist of the'options' as laid out in DP.t is clear that the 'issues' have beendecided, and what is needed is public comment on the variousrestrictive options proposed by the Health Department. Althoughthe Department's Health Promotion programme has stated that thedepartment 'did not lean in favour of any particular proposal at thisstage' [47], it has also stated that it is 'planning legislation to bansmoking in enclosed public places' [48]. Mr Caygill has gone onrecord as saying 'I would hope that in the end controls are placedon smoking in enclosed public spaces.' [49], and also that theevidence was 'conclusive that the effects of passive smoking onnonsmokers were harmful' [50]. This is of course not true. DP hasstated that 'One of the options would include new legislation calledthe Smoke Free Indoors Environment Bill' [51].

(f)Exposure to ETS is widespread and the health ofnonsmokers must be protected.Uberla [52] has conducted a large review of theliterature relating to the alleged relationship of ETSon lung cancer and other health effects. Hisconclusion was that there was insufficient evidence tostate that ETS had any deleterious effects on thehealth of nonsmokers. The principal study whichhas reported a significantly increased level of riskfor nonsmokers exposed to ETS is that of Hirayama[ 16 ] . Uberla raises a number of concerns about themethodology of the study, including the following:

*The study was a screening study, designed to generatehypotheses, not to test them.*The reported cohort was very different to the population aboutwhich inferences were to be made.*The estimate of exposure was non specific, and its reliabilityand validity were not assessed.*The assessment of the outcome was not adequately validated.*Confounding factors were not allowed for.*Classification bias was neither controlled nor excluded.*Sensitivity analyses on the likely effects of classification biaswere not carried out.*The information on the cases was very sparse, and notvalidated in any way.

* Temporal trends in lung cancer in Japan over the period ofthe study indicate that the increases cannot be accounted for byETS effects, and that there might be some other cause of lungcancer which was not studied.

* A large number of statistical tests were applied to the materialfrom Hirayama's study. Adjustments for multiple testing reduce the'significance' of Hirayama's reported increase in risk.

5

6

Mantel [23] states further that 'it is unlikely that any epidemiologicalstudy has been, or can be, conducted which could permitestablishing that the risk of lung cancer has been raised by passivesmoking'. Others to comment on this matter are Pershagen [53],and Adlokofer [54].

DP: Tobacco Smoke in the Environment

The question of ETS (Environmental Tobacco Smoke) and causation ishotly argued, with several studies giving inconclusive results [11,12, 13, 14, 15, 22, 23].

The question of environmental tobacco smoke and its health effects isvexatious, and as yet unanswered.

While some studies have claimed a positive effect, for exampleHirayama [16], Garfinkel et al [17], Knoth et al [18], others havefound no significant relationship, for example Buffler et al [19],Kabat & Wynder [20] and Garfinkel [21]. Relative risks of lungcancer for nonsmoking women married to smokers compared tononsmoking women married to nonsmokers range from a little over 2to around .75. Lee et al [22] found that among lifelong nonsmokers,passive smoking was not associated with any significant increase inrisk of lung cancer, chronic bronchitis, ischaemic heart disease orstroke.

Nathan Mantel, a prominent medical statistician, had this to say inthe BMJ 1987 [23]:

it is unlikely that any epidemiological study has been, or canbe, conducted which could permit establishing that the risk of lungcancer has been raised by passive smoking. Whether or not the riskis raised remains to be taken as a matter of faith according to one'schoice."

Auerbach et al in 1979, as a result of their study of changes inbronchial epithelium [43], stated:

"On the 57 nonsmoking cases, virtually all slides has "0%" atypicalcells. If passive smoking does indeed lead to lung cancer, it wouldbe expected that some of the nonsmokers who had been exposed tosmoke of others would have shown histological changes in theirbronchial epithelium similar to those found in the light (less than 10cigarettes a day) smokers, but no such cases were observed."

On the other hand, allegations that ETS is actively harmful rest onthe following pivotal studies:

(i) IARC study, [24](ii) Australian NHMRC report, [25]

Surgeon General's report, [26](iv) National Research Council, [27](v) Several studies of children, [ 28 ], [29]

It appears that these 'meta-studies' are not all independent, but relyheavily on quotations from each other. Since most are onlysummaries, or 'meta-studies', and involve no new experimentalevidence, this has the apparent effect of multiplying the weight ofwhat little experimental evidence does exist. Furthermore theestimates of the number of cancer deaths per year due to exposureto ETS (46 per year in NZ) are based entirely on overseasestimates, which may not be relevant to New Zealand, and may infact be entirely wrong. The problem of 'evidence multiplication'referred to above is also a problem here, with many of the relevantstudies quoting each other. Nathan Mantel [23] has also pointed outthe probable existence of a 'publication bias' which ensures that ofthose papers on this issue, those which are on the 'right' side ofthe fence have a greater probability of being published.

The problem of confounding variables has been completely ignored.In the indoor home environment there is passive inhalation of carbonmonoxide, of particles, of polycycic organic matter, of pesticides, offormaldehyde, of carcinogens, suspect carcinogens andcocarcinogens, including radon and its progeny, passive inhalation ofbacteria, viruses, fungi and allergens [24]. All these passiveexposures can and do take place in the absence of ETS andtherefore without their effects being confounded by ETS.

The well known genetic effect observed among smokers has also animportant part to play in the evaluation of risks to children ofsmokers, since these children, in general, are quite closely relatedto their parents!

Exposure

There are several claims made in DP which are unsubstantiated, orare not pursued in an appropriate manner. For example:

" .. the amount of chemicals inhaled can be calculated."Pg 7

A reference is given for this, but it is apparent from the referencesthat the calculated range is very broad, and no confidence intervalsare given. Since the distribution of particles in air is likely tofollow a right-skew distribution, the expected value is more likely tolie at the lower end of the calculated interval, i.e. much closer to0096 mg than to 0.5 mg for an eight hour period [31], P 125. The

calculated figure is neither quoted nor used in the following part ofDP.

"Cotinine levels in the blood take about 20 hours tohalve."Pg 8

Again a reference is given for this, but no indication of a likelyrange of values, nor any idea of how this estimate of the half life ofcotinine was arrived at, nor even of its significance. While notforwarding this as a 'good thing', we would nevertheless contendthat such statistics are meaningless without accompanying evidence ofsome sort of clinical significance.

"Comparisons between passive and active smoking,whether based on particles inhaled, nicotine absorbed orcotinine derived, may mislead."Pg 8

7

This would seem to suggest that the comments made earlier abouthalf life of cotinine and the 'particle burden', are not of anyimportance in themselves.

"It is mainly sidestream smoke which contains higherproportions of some cancer causing compounds."Pg 8

The reference quoted for the particle and chemical burden for thenonsmoker, [27] P 126, goes on to state that the "additionalcontribution to the total dose (from ETS) is negligible". This partof DP would seem to be simply a collection of short and somewhatmisleading or irrelevant statements taken out of context from thesources quoted. In fact the National Research Council document[31] states (P 153) that:

"Nicotine and cotinine, however, may not be directly related to thecarcinogenic potential of the smoke."

Furthermore, this report acknowledges (P 154) that:

"Knowledge of the effects of diet is needed when interpreting resultsfor mutagenicity of the urine of ETS exposed nonsmokers.."

"At least two thirds (nearly 500,000) of smokers are inpaid work. Most smoke at work and most work indoors."Pg 8

No authority is given for this, and furthermore, it is difficult toknow how these estimates could be obtained from census documents.

"At least 1,000,000 of New Zealand's approximately 2.5million nonsmokers are estimated, from informationavailable, to be regularly exposed to environmentaltobacco smoke at home or at work."Pg 8

The "information available" consists of the results of a telephonesurvey of 192 respondents. A sample of 192 is not generallyregarded as being a reasonable size for measuring attributeresponses. The following comment in DP about likely exposures ofnonsmoking children is based on census figures and the grossassumption that rates of exposure found in a survey of Form Onestudents can be applied across the board.

"Carbon monoxide levels are usually higher wheresmokers are present, but levels vary greatly."Pg 9

Presumably this means that the evidence is not sufficient todemonstrate that ETS is even associated with high levels of carbonmonoxide, and that significant differences have not been found.This type of statement is clearly meant to be suggestive, but in factis not at all convincing. We could as easily state that "Carbonmonoxide levels are often lower where smokers are present, butlevels vary greatly"! Clearly the evidence regarding carbonmonoxide levels is at best patchy.

8

9

"This does not mean that tobacco smoke will always harmnonsmokers or that the harm will always be very great.Tobacco smoke is, however, a risk to nonsmokers. Noone can predict who it will harm and who it will not,although the risk appears to be related to the amount ofexposure."Pg 10

No evidence for any of this is quoted, the only explainable reasonwould be because of its very dubious nature. We refer once more toNathan Mantel [23]:

"Whether or not the risk is raised remains to be taken as a matter offaith according to one's choice."

The dubious nature of the evidence for any true causative linkbetween exposure to ETS and health effects is reflected in DP (PTen), where the World Health Organisation's International Agency forResearch on Cancer is quoted as stating that "passive smoking givesrise to some risk of cancer".

Chemicals in Smoke

Of the many chemicals in tobacco smoke, DP singles out N-nitrosamines for special attention. The concentrations, however, arebased on laboratory measurements, and do not relate to levels in'room air'.

The main reason for using sidestream smoke (SS) in the laboratoryrather than ETS gathered under actual room conditions is that it ismuch easier to gather SS. Although we now know a great dealabout the composition of SS, there are serious limitations in theapplicability and interpretability of such data. When all is said anddone, it must be remembered that it is ETS and not undiluted SS towhich nonsmokers are exposed.

It is often said that exposure to ETS is equivalent to smoking somesmall number of cigarettes per day; a quarter of a cigarette is the'estimate' given in DP on page eight. There has never been ademonstrated elevation of statistical health risk associated with adose of this magnitude. Furthermore, this type of extrapolation isunwarranted and scientifically inaccurate for many reasons, amongthem the following: if nonsmokers inhale ETS, it is almostexclusively through the nose with a normal breathing pattern, whilesmokers typically inhale mainstream smoke and pause beforeexhaling.

The two major studies of the effects of ETS on health of nonsmokersin 1986 [26], [27] concluded that ETS has not been shownscientifically to increase the risk of cancers in tissues other than thelung or to increase cardiovascular disorders or deficits inrespiratory function in adults.

Action on ETS Pg 14)

This section starts with the bald statement that "Tobacco smoke isone of the leading environmental causes of cancer in New Zealand",an assertion with which many, if not most, scientists working in thisfield would disagree [19, 20, 21, 22, 23]. This statement and theassociated estimates on page fourteen of DP are based on an estimateof 34% excess risk for cancer of the lung for ETS exposed persons.This estimate has been put in some doubt elsewhere [44], but wewill comment briefly upon this here.

This estimate of the excess risk as 34% is based on a meta-studycombining the results of a number of studies including those byBuffler, Garfinkel, Hirayama, Kabat, Lee and Trichopoulos. If thestudies outside of the US are omitted from the combined estimate,the odds ratio is 1.14, with a confidence interval (0.92, 1.40), iethere is no statistically significant effect of ETS. In fact, the onlystudies to produce 'significant' relative risks were the Japanese andGreek studies. These results have been criticised on grounds ofboth methodology and sample size [32]. In this regard it is worthnoting that the problem of misclassification of smokers or ex smokersas nonsmokers is such that, if the extent of misclassification was 5%,then this would result in a relative risk of 1.4.

It is notable that, in a major study of cancer mortality in NewZealand [41], ETS is not mentioned. Extrapolating extremely smallrisks simply by multiplying by an appropriately large factor is aprocedure whose scientific name is 'guesswork', not estimation.

The risk that tobacco smoke is alleged to pose to the health ofnonsmokers is stated to be "needless and preventable". This raisesthe point of risk evaluation in general, and it has been well-demonstrated that those in the risk-evaluation business are notparticularly good at it, nor are they in total agreement as to how itshould be performed.

The report from the National Research Council [27], which is heavilyreferenced in DP, takes a much more balanced view than one wouldsuppose from the way the references are made:

* It does not suggest that the question of whether ETS hassignificant health effects has been decided; in fact most of thereport stresses the need for more and better data for furtherand more rigorous analyses.

*It does not conclude that occasional exposure to ETS poses arisk to nonsmokers.

*It does not suggest that ETS impairs adult respiratoryfunction.

*It does not conclude that heart disease has been scientificallylinked to ETS (P; 265-266).

*It does not conclude that allergies to tobacco smoke have beenfirmly established.

*It does not conclude that valid evidence establishes ETS as arisk to asthmatics.

10

Furthermore, the National Research Council document:*

stresses that ETS is physically and chemically distinct frommainstream smoke

* and acknowledges the difficulties of measuring exposure andaccounting for misc1asifications, bias, covariables andconfounding variables.

Important points to note from the document include:

"Sufficient data are not available to assess the relative genotoxicityof whole ETS." P59

Nicotine and cotinine "may not be directly related to the carcinogenicpotential of the smoke." Pp 152-153

"Three problems are especially important for studies of respiratorysymptoms in children, ie underreported active smoking on the partof the children, recall bias leading to overreporting of symptoms byparents, and the confounding variables of infections in parents. Allthree may lead to overestimation of symptom prevalences amongchildren of smokers." P 159

Another confounding variable, not considered in the report, is thediffering genetic makeup of smokers and their offspring to that ofnonsmokers [5, 8, 9, 10, 111.

"It is not possible to determine whether ETS is directly causing thedecreased lung function observed in children of smoking parents orif an increased infection rate in these children is responsible for thedecrease." P 197

(P 15)Comments on 'A clean environment' display a general moraldisapproval directed at the smoker. While the public do indeedexpect industrial pollution and the exhaust from cars to becontrolled, there is good reason for this, and such emissions arecontrolled. The levels of chemicals present in ETS are far below thestandards set by environmental agencies worldwide, and there is noevidence that there is any cause for further controls to be placedupon those who produce ETS, ie the smoking population.

The paragraph on 'Development of a smokefree society' is misleadingin the extreme. The reference to children is designed to beemotive, and to persuade, not to inform.

Current Situation in New Zealand: Tradition and Legislation (P 19)The paragraph entitled 'Tradition' states that there are severalplaces already where smoking is not polite or appropriate. This isof course, quite true, but entirely irrelevant: these places also areinappropriate for other activities such as dancing, singing, sex,eating watermelon!

11

Legislation

There are a number of regulations already in force in New Zealand,restricting cigarette smoking in certain situations, it is true. DPadvances this fact as being a reasonable basis upon which to setmore regulations in place. However the existence of legislation doesnot confer any external validity on the argument that ETS is harmfulparticularly if the legislation is for reasons other than ETS (safety,hygiene etc).

Public Opinion

Even if 100% of the respondents believed that ETS was harmful,would this make it any more true? The 'widespread understandingthat the health of nonsmokers can be damaged by other people'stobacco smoke' bears testimony not to the harmful effects of ETS,but rather to the effect of the antismoking propaganda campaign.Similar comments apply to the statements about attitudes.

Proposals for Smokefree Indoor Environments

All of the proposals put forward are at the extreme 'antismoking' endof the opinion scale. The justification for taking the proposedmeasure to 'create smokefree indoor environments' is the need toprotect people from the consequences of passive smoking. In viewof the speculative nature of the evidence for such harm, this is nojustification at all. Implicit in all of this section of DP, of course,is the assumption that some action must be taken, and that there isno room for any different opinion about any putative effects of ETSon health. The question surely is whether there is any cause forany action at all.

Discussion of Appendix One

A Comparison of Selected Chemicals in Mainstream and SidestreamSmoke

The table quoted displays a number of chemicals present insidestream smoke, their concentrations and the ratio of theconcentration in sidestream smoke to that in mainstream smoke,presumably in the form of 95% confidence intervals. The accuracy ofthe figures quoted is not questioned, although some of the estimatesshow enormous confidence intervals, which makes assessment oflikely true levels very difficult. The relevance of these figures,however, is certainly questionable, since sidestream smoke isdefinitely not ETS. The dilution factors applied to sidestream smokebefore it becomes part of ETS are enormous, and of a magnitudesuch that the expected dose for a nonsmoker is less than 1% of thatfor a smoker. The relevance of analyses of Canadian cigarettes tothe New Zealand situation is also questionable.

12

Discussion of Appendix Two

The Action of Carcinogen Agents in Cigarette Smoking

The thrust of this appendix seems to be that tobacco smoke, andsidestream smoke in particular, contains a large number of chemicals,some of which have demonstrated carcinogenic properties in largeenough doses. There have also appeared some statements that theevidence for a threshold dose for carcinogenesis is in some doubt,but the experience with diethyl stilbestrol and other artificialestrogens seems to confirm that for some carcinogens at least, athreshold dose makes a great deal of sense. The appendix fails tomention the fact that all the major classes named are also all presentin any environment into which smoke or fumes from the complete orincomplete combustion of wood, vegetation, fossil fuels such as coal,natural gas, petrol, diesel, and even cooking, particularly by gasoccurs. The only exceptions are a few derived from nicotine, andas has been stated elsewhere, there is no conclusive evidence thatthese compounds are present in doses sufficient for carcinogenesis.It is also noteworthy that the greatest source of nitrosamines inhumans is from the ingestion of foodstuffs.

Discussion of Appendix Three

Deaths Attributable to Involuntary Smoking in New Zealand. APreliminary Estimate

One of the authors of this appendix, Pearce, has gone on record asbelieving that a health hazard having a relative risk of less than 1.5should be regarded as 'unreliable' [41]. The relative risk uponwhich the figures in this appendix are based is 1.34, from the meta-study of Repace and Lowry [42]. This figure arises from poolingthe results of a number of studies; however, when the Americanstudies alone are considered, the estimate becomes 1.14, with anassociated confidence interval of (0.92, 1.40); ie the relative riskfor these studies was not significantly different from 1.0. In fact,only the Japanese and the Greek studies found significantly elevatedrelative risks, and each of these has been subjected to severecriticism in the scientific literature.

In studying the lung cancer deaths allegedly 'attributable toinvoluntary smoking in New Zealand', Kawachi and Pearce usemethods similar to those used in the US, Canada and Australia; they.then proceed to offer the similarity of their estimates to those of theoverseas studies as evidence of validity and consistency. Given thefact that they are using a similar model, it is scarcely surprisingthat when a model predicts about 50% 'attributable deaths' amongnonsmokers in the US, the same model predicts about 48%-51% forNew Zealand! Repeated views of the same piece of 'evidence' do notconstitute confirmation by independent study.

Other causes are mentioned, indeed, mentioned at length, but thereis nothing of any substance here. When all is said and done, theessence of this section is: "Consensus on the causal nature of thisassociation awaits the reports of further detailed studies." In otherwords, this is not proven. The studies of deaths from cancersother than the lung, in particular, have been criticised for severemethodological flaws.

13

Kawachi and Pearce attempt a summary of the potential biases in theestimations, and mention two, each of which would tend to make theestimates under-estimates. They ignore the converse possibleerrors:

(i)misclassification of neversmokers as eversmokers, and(II)misclassification of nonexposed nonsmokers as exposed

nonsmokers.

14

REFERENCES

1. Gray AJ, Reinken JA, Laugesen MThe cost of cigarette smoking in New ZealandNZMedJ 1988, 101: 204-7.

2. Department of Health (1988)The Big Kill. The Human Cost of Smoking in New Zealand.

3. King DR, Smith AR, Salter DMMortality attributable to smoking in New ZealandNZMedJ 1983; 96: 195-9.

4. Burch PRJPeriod and cohort trends for mortality and cigaretteconsumption in England and Wales, 1946 to 1980, with emphasison sex ratios.J Clin Epidemiol Vol 41 No. 4, 373-384 1988.

5. Burch PRJSmoking and Lung Cancer: the Problem of Inferring Cause.JR Statist Soc A (1978), 141,4,437-477.(with discussion)

6. Burch PRJ (1986)Can Epidemiology become a rigorous science? How big is TheBig Kill?IRCS Jour of Med Sci 14: 956-961.

7. Feinstein AR, Wells CK (1974)Cigarette Smoking and Lung Cancer: the problems of"detection bias" in epidemiologic rates of disease.Trans Ass Amer Phys, 87, 180-185.

8. Seltzer CC (1963)Morphologic constitution and smoking.J Amer Med Ass, 183, 639-645.

9. Seltzer CC (1967)Constitution and heredity in relation to tobacco smoking.Ann NY Acad Sci, 142, 322-330.

10. Heath CW (1958)Differences between smokers and nonsmokers.Arch Tnt Med., 101, 377-388.

11. Fisher RA (1958)Lung cancer and cigarettes?Nature, Lond, 182, 108.

12. Fisher RA (1958)Cancer and cigarettes?Nature, Lond, 182, 596.

13. Friberg L, Kau L, Dencher SJ and Johnsson E (1959)Smoking habits of monozygotic and dizygotic twins.Brit Med J, 1, 1090-1092.

15

14. Todd GF and Mason JI (1959)Concordance of smoking habits in monozygotic and dizygotictwins.Heredity 13, 417-444.

15. Mantel NDoes passive smoking stunt the growth of children?Letter to the editor,Tnt Jour of Epidemiology, 15, 3, 427-428

16. Hirayama T (1981)Nonsmoking wives of heavy smokers have a higher risk of lungcancer: a study from Japan.Br Med J, 282,183.

17. Garfinkel L, Auerback 0 & Joubert L (1985)Involuntary smoking and lung cancer: A case-control study.J Nati Cancer Inst, 75, 463.

18. Knoth A, Bohn H & Schmidt F (1983)Passive smoking as cause of lung cancer in female nonsmokers.Med Kiln, 78, 54.

19. Buffler PA, Pickle LW, Mason TJ & Contant C (1984)The causes of lung cancer in Texas.In 'Lung Cancer. Causes and Prevention'. Mizell M & CorreaP. (eds).Verlag Chemie International Inc.

20. Kabat GC & Wynder EL (1984)Lung cancer in nonsmokers.Cancer, 53, 1214.

21. Garfinkel L (1981)Time trends in lung cancer mortality among nonsmokers and anote on passive smoking.J Natl Cancer Inst, 66, 1061.

22. Lee PN, Chamberlain J & Alderson MR (1986)Relationship of passive smoking to risk of lung cancer andother smoking-associated diseases.Br J Cancer, 54, 97-105.

23. Mantel NLung cancer and passive smoking.letter, Brit Med Jour, 294: 440-441 1987.

24. World Health Organisation, International Agency for Researchon Cancer.IARC Monographs, Evaluation of the Carcinogenic Risk ofChemicals to Humans: Tobacco Smoking V 38.IARC, Lyon, France: 1986.

25. National Health and Medical Research CouncilEffects of Passive Smoking on HealthCanberra: Australian Government Publishing Service, June1986.

16

26. The United States Surgeon GeneralHealth Consequences of Involuntary SmokingWashington DC: Department of Health and Human Services,1986.

27. National Research CouncilEnvironmental Tobacco Smoke: Measuring Exposures andAssessing Health Effects.Washington DC: Academy Press, 1986.

28. McClellan VAA National Survey of the Smoking Habits of Form One Studentsin New Zealand SchoolsWellington, Department of Health 1987.

29. Fergusson DM et alParental Smoking and Respiratory Illness in InfancyArchives Diseases of Childhood, May 1980, 55:5, 358-361.

30. Kraemer HCThe Robustness of Common Measures of 2X2 Association to BiasDue to Misclassifications.Amer Statist, Nov 1985, 39, 4, 286-290.

31. The United States Surgeon General (1986)Smoking and HealthWashington DC: Department of Health and Human Services.

32. Lee PN, Chamberlain J & Alderson MRRelationship of passive smoking risk of lung cancer and othersmoking-associated diseases.Brit Jour Cancer 1986 54, 97-105

33. Trichopoulos D, Kalandidi A, Sparros L and MacMahon B(1981)mt J Cancer, 27, 1-4.

34. Thomas CB, Fargo R and Enslein K (1970)Personality characteristics of medical students as reflected bythe strong vocational test with special reference to smokinghabits.Johns Hopkins Med J, 127, 323-335.

35. Freire-Maia A (1960)Smoking and PTC sensitivity.Ann Hum Genet, 24, 333-341.

36. Thomas CB and Cohen BH (1960)Comparisons of smokers and nonsmokers. 1 A preliminaryreport on the ability to taste phenylthiourea (PT C).Bull Johns Hopkins Hospital, 106, 205-214.

37. Cohen BH and Thomas CB (1962)Comparisons of smokers and nonsmokers. 11 The distributionof ABO and Rh(D) blood groups.Bull Johns Hopkins Hospital, 110, 1-7.

17

38. Doll R and Pike MC (1972)Trends in mortality among British doctors in relation to theirsmoking habits.J Roy Coil Physcns Lond, 6, 216-222.

39. Multiple Risk Factor Intervention Trial Research Group (1982)Multiple Risk Factor Intervention Trial. Risk Factor Changesand Mortality Results.JAMA Sept 24 1982 248, 12, 1465-1477.

40. Stallones RA. (1983)Mortality and the multiple risk factor intervention trial.Amer J Epid, 117,6,647-650.

41. Pearce NE and Howard JE. (1986)Tnt J Epidemiol 15, 456-462.

42. Repace TL and Lowrey AH. (1985)An estimate of Non-smokers' Lung Cancer Risk from PassiveSmoking.Environ Tnt 11:3-22.

43. Auerback 0, Garfinkel L, Hammond EC (1979)Change in bronchial epithelium in relation to cigarette smoking1955-1960 vs 1970-1977.New Eng J Med 300:381-386

44. Burch PRJ (1986)Health Risks of Passive Smoking: Problems of InterpretationEnivorn Tnt, 12: 23-28

45. Bolton R (1988)Tobacco and Liquor AdvertisingMarketing Magazine, July 1988

46. Tollison R (1986)Smoking and the StateLexington Books, Toronto

47. The Chronicle (Levin)Views sought on bans on smoking15 Jan 1988

48. Auckland StarPlan for all-out ban on smoking15 Jan 1988

49. TV OneTV News29 Mar 1988, 6.44 pm

50. NZ HeraldSmoke Signals from Caygill28 Aug 1988

51. The DominionNon-smokers at risk27 Sep 1988

52. Uberla K (1987)Lung cancer from passive smoking: hypothesis orconvincing evidence?Tnt Arch of 0cc and Env Health 59: 421-437

53. Pershagen G (1986)Review of epidemiology in relation to passive smoking.Arch Toxicol, Suppi, 9: 63-73

54. Adlkofer, F (1987)The problems of passive smokingDer Kassenarzt, 27: (51/52) 29-39

55. Duncan DB, Geaney PP (1984)Passive smoking and uptake of carbon monoxide in flightattendantsJAMA 251 (20)

56. Foliart D, Benowitz NI, Becker CE (1963)Passive absorption of nicotine in airline flight attendantsNew Eng J Med 308 (18): 1105

A PAPER BY DR W ALLAN CRAWFORD, OBEM.B., Ch.B., Ph.D., M(FOM) R.C.P., D.P.H., D.I.H., F.R.A.C.M.A.,

F.A.C.O.M., F.R.S.H.

CONSULTANT IN OCCUPATIONAL AND ENVIRONMENTAL HEALTH

Dr. W. ALLAN CRAWFORD, O.B.E.MB.. Ch.B.. Ph.D.. M.(FOM) RC.P.. D.P.H.. D.IH., F.R.A.C.M.A.. F.A.C.O.M.. F.R.S.H.Consultant in Occupational 29 BATTLE BOULEVARDE

and Ent'ironmenzol Health SEAFORTH. N.S.W. 2092.

TELEPHONE: 94-4747

COMMENTS ON CREATING SMOKE FREE INDOOR ENVIRONMENTSDEPARTMENT OF HEALTH, NEW ZEALAND 1988

Introduction

The Minister of Health, the Honourable David Caygill, in his forewordto "Creating Smoke Free Indoor Environments", a discussion paperproduced by officers of the Health Promotion Programme of theDepartment of Health, Wellington, New Zealand, asks the question,"What can be done to ensure that everyone's health is protected byensuring clean indoor environments in our schools, offices, hospitalsand workplaces?"

The answer is not merely making such places tobacco smoke free, butin a resolute program of environmental and occupational health andhygiene.Environmental tobacco smoke (ETS) is the visible andolfactory evidence that the air in such places is inadequatelycontrolled and that the many noxious gases, fumes and particulatesfrom other sources are also not controlled.

My conclusion on page 21 of this document is "I offer the opinion thatthe discussion document itself fails to provide the information neededto promote discussion. It falls to provide information on what is notknown, what is in contention, the biological meaning of some scientificdata, and the difficulties inherent in interpreting confounded andconflicting epidemiological data related to low level risks to humanhealth arising from exposure to environmental tobacco smoke. Noconsideration has been given to other pollutants which pervadeubiquitously our air environment at home, work or leisure and whichare being shown to be environmental hazards to health".

The overall scenario in 1988 is that the effects of ETS on the healthof workers in industry has received virtually no investigation. Inthose industries recognised as posing threats to the respiratory tractfrom fumes, dusts, fibres and gasses no specific data has beenlocated on active research relating to exposure to ETS.

There has been an upsurge in research into the environmentalconditions in offices and buildings and health effects of a minornature are frequently observed and are clearly multifactorial andreferred to as the Sick Building Syndrome and in which ETS may playa minor role.

We are at the stage in 1988 when the only scientific data are:1. There is inadequate data on the epidemiology of ETS in the

workplace.2. There is no evidence of increase of lung cancer in nonsmokers

during the decades of increasing prevalence of smoking.3. There are competitors as potential (possible) causes of lung

disease and lung cancer in air pollution in industrialisedcountries.

4. There are competitors in the workplace as potential causes forthe induction of lung disease and lung cancer.

5. That no atypical cellular changes have been found in the lungsof nonsmokers.

6. That nonsmokers do absorb one chemical peculiar to cigarettesmoke - nicotine and this is measured as the metabolite cotinine.

MERSEY HOLDINGS PTY LTD.(Inc. in N.5.W.)

Epidemiology of ETS in the Workplace

There are, to my knowledge, only three papers on the subject.Kentner 1984 (1), Kentner 1988 (2) and White and Froeb 1980 (3).The Kentner studies found no adverse effects and the White andFroeb study acquired suspect results.

No Evidence of Increase in Lung Cancer in Nonsmokers

Dr Wynder (4) raised a question which has not yet been answered:

"If passive inhalation in fact increases our risk of lung cancer, thenthere should have been a steady increase in the incidence of lungcancer among nonsmokers. This should have been observed in thefemale population for the last 40 years. As more and more mensmoke, more women passively inhale the smoke of their husbands.Garfinkel and I agree that there has been no significant increase inthe incidence of lung cancer in male or female nonsmokers. In thiscase we have a

Doll and Peto (5) in their assessment of avoidable risks of cancer inthe US, commissioned by the Office of Technology Assessment of theUS Congress also state independently of Garfinkel:

"Pending further data, we remain unconvinced that any materialtrends in true lung cancer death rates among American nonsmokershave occurred in recent decades [although some such increases shouldbe expected if the effects of "passive" smoking reported by Hirayama(1981) and Trichopoulos et al (1981) are confirmed]. A fortiori, weare even less convinced that any such trends not attributable topassive smoking among nonsmokers have occurred."

The Pollutants and Public Health

The public are exposed passively to multiple pollutants largelyproduced by the public's personal activities. The public are largelyurbanised and most urban developments are industrialised andtherefore further polluted. The public are passive inhalers andingestors of numerous substances known to have toxic properties.What the public do not know is that such toxic properties may be onlyof academic interest until the amount or concentration and durationsof absorption produce some adverse physiologic response, manifestillness, or unaesthetic conditions. The threat of some ill-effect is apowerful motivator for action to be taken by "Government" on behalfof the public. Such action, however, to be acceptable, must bebased on solid non-contentious research findings of cause-effectrelationships and should not inconvenience the public nor individuals,and affect their unalienable rights to freedom to pursue happiness andwell-being at work, leisure and at home unless the hazard is real andimportant as it is in the case of AIDS. Most relevant activities inregard to happiness and well-being occur indoors in the presence ofpollutants.In the indoor home environment there is passive inhalation of carbonmonoxide, passive inhalation of particles, passive inhalation ofpolycyclic organic matter, passive inhalation of pesticides, passiveinhalation of formaldehyde, passive inhalation of carcinogens andsuspect carcinogens, including radon and its progeny, passiveinhalation of bacteria, viruses, fungi and allergens. All these passiveexposures can and do take place in the absence of environmentaltobacco smoke and therefore without their effects being confounded byenvironmental tobacco smoke.

2

Lance Wallace (6) of the US EPA Total Exposure AssessmentMethodology (TEAM) study concentrated on the volatile organics groupof chemicals and selected 20-26 of the 37 hazardous air pollutantslisted as possibly requiring regulation. Of these 20, about 12 areknown or suspected human carcinogens which were always found, ornearly always found, inside homes. Risk assessments of 6 compoundsfor cancer yielded an estimated range of 1000 - 5000 excess cancercases per year in the USA. These six volatile organics are carbontetrachloride,chloroform,benzene,trichlorethylene,tetrachiorethylen, P-dichlorobenzene.The estimate for otherchemicals with mutagenic properties were too highly speculative forinclusion in the report. The assays included breath, personalbreathing zones and adjacent outdoor air. A great range of exposurewas found from <lug - >100 ug/m 3 . Personal air concentrations weregreater than outdoor exposures and in many cases the concentrationsin breath exceeded outdoor concentrations.

Doll (7), referring to the urban-rural lung cancer rate gradient,expresses the view that the differences were not generally large,being commonly about two-fold but that it was notable that mortalityincreased progressively with the size of the town. He refers also tospecific carcinogenic agents in the urban air environment such asPOMs, asbestos, arsenic and radon. He also notes the substantialdifferences in sex incidence between males and females in all countriesstudied and the wide range in rates from country to country. Asexample, Scotland had age standardised lung cancer rates per 100,000of 78.14 in males to 11.71 in females while Portugal had 10.91 and2.74 respectively. A similar wide range of rates in towns is reportedfrom the polluted to not so polluted areas.

That lung cancer is more common in urban areas does not by itselfgive evidence sufficient to infer that urban air pollution is theattributable factor for lung cancer. There are confounding factors inurban living related to smoking, drinking, eating habits, work,lifestyle, indoor pollutants from heating and cooking and familialinheritance.

Carnow (8) also notes the evidence from studies of migrants whichadd weight to the hypothesis that air pollution in industrial areas isrelated to lung cancer, eg, migrants from Britain going to relativelyless polluted countries such as South Africa, Australia and NewZealand.

The Britain - South Africa migrant lung cancer changes are describedby Dean (9) 1959, the Britain - Australia also by Dean (10) 1964 andBritain - New Zealand by Eastcott (11) 1956.

Public concern about air pollution, although certainly understandable,is frequently in inverse proportion to the concentration of thepollutants present in the environment. In some quarters there areconcerns about the presence of minute amounts of various substancesin the air. At times such concerns have been almost irrational,failing to consider that the improvements in health and longevity haveoccurred in the presence of these low levels of pollution. If clean airis to be defined as air free of all pollutants, all industry, allcommunity activity and, indeed, breathing itself will have to behalted.

3

The current lack of a sense of perspective with regard to ETS andother pollutants is to be deplored because it could lead, by politicalpressure, to the adoption of draconian measures against a largesection of the public, despite the absence of incriminating evidence.In this regard Professor Patrick Lawther (12) of the United Kingdomwrites:

"The lack of sense of history and perspective often accompanies themost laudable zeal, but the distortion which follows careless, albeitpassionate, thought and clamour for action will impede progress in thesearch for the real effects of pollution and the implementation ofmeasures to abate it. We must be on our guard."

Competitors for ETS in Workplaces

The concept of carcinogenic risk of inhaling environmental tobaccosmoke is biologically plausible as it, like any other smoke from theincomplete combustion of vegetation, fossil fuels and animal products,will contain some carcinogenic polycycic aromatic hydrocarbons(PAll), polycycic organic material (POM) and nitrosamines. ETS isunique as it specifically contains nicotine and nicotine derivatives.Nicotine per se is not a carcinogen.

Doll and Peto (5) 1981, report the established occupational causes ofcancer of the lung as arsenic, asbestos, bishloromethyl ether,chromium, ionising radiations, mustard gas, nickel and polycydicaromatic hydrocarbons.Occupations with raised StandardisedMortality Ratios for lung cancer are also tabled: Miners andquarrymen, gas, coke and chemicals makers, glass and ceramicmakers, furnace, forge, foundry, rolling mill workers, electrical andelectronic workers, engineering and allied trades workers n • e. c.,woodworkers, leather workers, clothing workers, food, drink andtobacco workers, construction workers, painters and decorators,drivers of stationary engines, cranes, etc., labourers n.e • ctransport and communication workers, warehousemen, storekeepers,packers, bottlers, and service, sport and recreation workers.

Kvale, Bjelke and Heuch (13) 1986 concluded, in their study of lungcancer and occupational risks in Norway of 11,995 men, that some 13-27% of the 125 lung cancer cases could be attributed to harmful workexposure. They stratified for age, place of residence and smokinghabits. They note that exposures to carcinogens most commonlyoccurs to the lower socio-economic groups of workers.

The key documents which list permissible levels of pollutants in theworkplace are the TLVs, Threshold Limit Values for ChemicalSubstances and Physical Agents in the Work Environment. They areprepared by the Chemical Substances TLV Committee of the AmericanConference of Government Industrial Hygienists (ACGIH) andregularly updated.

The following definition of a TLV is given:

"The Threshold Limit Value-Time Weighted Average (TLV-TWA) - thetime-weighted average concentration for a normal 8-hour workday anda 40-hour workweek, to which nearly all workers may be repeatedlyexposed, day after day, without adverse effect." Appendix A to the1983-84 edition is dedicated to carcinogens in workplaces. Ofparticular note is that for some carcinogens and suspect carcinogensthere are TLVs listed.

4

This is contrary to the publicly held view that no exposure tocarcinogens should be permitted and has already been referred to inrelation to environmental considerations. The following carcinogenicsubstanceshavebeenallottedTLVs.Asbestos,bis (chloromethyl) ether, chromite ore processing, some chromium (VI)compounds coal tar pitch volatiles, nickel sulphide roasting fume anddust and vinyl chloride.There are also TLVs for 25 chemicalssuspected of having carcinogenic potential for man.

Union Response (AFL-CIO) on Workplace Smoking Cessation

The US Surgeon General's 1985 report on smoking and health in theworkplace emphasises the opportunities to conduct anti-smoking andsmoking cessation campaigns in these relatively closed communitieswith their relatively captive employees.

There was an immediate hostile response from the AFL-CIO (14), theUS Trade Unions organisation which included:

"The Surgeon General's report on The Health Consequences ofSmoking - Cancer and Chronic Lung Disease in the Workplace willseriously set back efforts to protect the health of American workers.

"The report minimises and misrepresents the risks posed byoccupational hazards, and suggests that if workers stop smoking theywill eliminate the risk of occupational disease. In fact, the way toeliminate occupational disease is by reducing workplace hazards.

"Unfortunately, this latest report adds no new scientific information toour knowledge of the risks posed to workers on the job.

"This report can only hurt workers by undermining efforts to controlworkplace hazards. It will lead. to misdiagnosis of occupationaldisease. It will make it even more difficult for workers sufferingoccupationally-related disease to secure compensation to which theyare entitled. And it will be used by those who are responsible forpoisoning workers to avoid legal liability."

Professor Weiss (15) from Harvard University in his review of theUSSG report on smoking in the workplace notes:

"The Surgeon General, clearly recognising that the government cannotregulate what goes on in private homes, is taking the only actionavailable to him in suggesting that workplace exposure be banned.The health impact of exposure at work, except for its irritation, islargely unknown. The report is on its weakest ground scientificallyhere. However, if the effect of restricting smoking at work is toreduce the total number of smokers then the action would seemjustified despite the paucity of scientific data supporting it.Restricting exposure is only part of the solution to this problem.Further research and education are still needed."

Hirayama (16) contributed to the problem by saying:

"To me, "direct passive smoking" means the subject is within adistance of 1 or 1.5m from the source of exposure. That kind ofexperimental setup must be possible. The two issues of distance anddose should be differentiated when considering the polluting of roomair by tobacco smoke. In actual experience, these two factors acttogether.

5

This is true in the example of an office worker whose dose willdepend on the distance he sits from a heavy smoker. In my opinion,room air contamination may not play a major role in a carcinogeniceffect of passive smoking - at least in Japan - because my own datasuggest that direct passive smoking is far more important."

Basically, from the studies on ETS reported, there is inadequate dataon respiratory effects in workers because of the massive and diverseconfounding factors in the workplace and general industrial areas.There is even less data on cardiovascular effects, if any at all, inhealthy adults. This is the conclusion of the National Academy ofScience (17) advisors.

The most potent tool in the hands of those urging exposure to ETS tobolster anti-smoking campaigns aimed at the public, the workforce andpoliticians is the threat of lung cancer induction. The bases for thistool are the US Department of Health (18) 1986, the National Academyof Science/ National Research Council (17) 1986, and the InternationalAgency for Research on Cancer (19) 1986. There have also beenconflicting personal views expressed, Aviado (20) 1986, Wald (21)1987, Lebowitz (22) 1986 and Uberla (23) 1987. Various symposiahave been published on the issues, Geneva (24) 1983, Vienna (25)1984 and Essen (26) 1986.The conclusions reached by theseagencies, people and workshops are in conflict. (See Appendix 1).

There are over six million organic and inorganic chemicals listed bythe American Chemical Society and while of industrial origin mainlythey are not confined to the factory environment but may polluteambient air and the indoor air of homes. Some untold number will befound in homes. Some 60,000 - 70,000 chemicals are in common use.Of these 7,000 have been adequately tested in animals. About 30have been linked to cancer in humans and 1,500 linked to cancer inanimals.

Non-smoking policies for workplaces are based on three studies, onextrapolations mainly from studies of spouses, and extrapolations fromthe epidemiology of smokers, faith in the no threshold theory foreffects of toxic substance, an absence of an established no effectlevel, and possible plausibility. Such policies must therefore beconsidered as social ones and not health policies. The use of passivesmoking as a tactic in the smoking cessation strategy is based onweak grounds apart from irritations.

Competitors in the Lung Cancer Causation Stakes

Tobacco smoke is frequently stated to contain 3,800 or more of thesechemicals or about 0.06% of the chemicals listed. A few chemicals arespecifically associated with the combustion of tobacco of which a dozenor so of these chemicals originate from the chemical nicotine. Dareone suggest that if the same analytical effort was applied to somekitchens when cooking is in progress, particularly with gas stoves orother fossil fuels, some 3790 could be found and 3800 if the cookpermits smoking in the kitchen.

6

Doll and Peto (5) 1981, in their table 19 give a numerical descriptionof the lung cancer deaths in the US in 1978 noting that misdiagnosisof secondary cancer and some mesotheliomas may be included. Anextract from their table is as follows and relates to deaths from lungcancers that can be ascribed to occupation:

7

No. of deathsrecorded in 1978

Type ofCancer

Male Female

Lung 71,006 24,080

Cancer deaths ascribed tooccupational hazards in 1978(United States)

MaleFemale

No%No%ascribed ascribed ascribed ascribed

10,651151,2045

Variables which require consideration [listed by the NRC (17) 1986]are occupation and industry of employment, exposure to specificrespiratory carcinogens, such as asbestos, arsenic, radon, etc, inoccupational or non-occupational settings, dietary factors, familyhistory of cancer, residential history, housing characteristics, yearsof education, and socio-economic status. The International Agencyfor Research on Cancer (19) 1982 adds further data.

Cellular Changes in the Lungs of Non-Smokers

Hammond and his colleagues Auerbach and Garfinkel (27) in a studyof the histological changes in the airways of the lung in 1979 andreferred to at a conference in Vienna stated:

"On the 57 non-smoking cases, virtually all slides had 11 0%" atypicalcells. If passive smoking does indeed lead to lung cancer, it wouldbe expected that some of the non-smokers who had been exposed tosmoke of others would have shown histological changes in theirbronchial epithelium similar to those found in the light (less than 10cigarettes a day) smokers, but no such cases were observed. Thisdoes not exclude the possibility that the mechanism by which passivesmoking could lead to cancer might be totally different from that insmokers."

Nicotine Absorption

Some nicotine is absorbed from inhaling ETS and excreted as thematabolite cotinine in nanogram amounts (1000 millionth of a gram).Jarvis in the 1987 IARC (28) report, P.56 offers the opinion:

"It seems unlikely that non-smokers could absorb more than one ortwo mg of nicotine a day, even if they spend the majority of theirwaking hours in heavily smoke-polluted atmospheres. Comparison ofcotinine concentrations in urban nonsmokers and active smokerssuggests that average nonsmokers may receive a dose of about 0-. 2mgof nicotine per day. This is a preliminary estimate which depends ona number of assumptions and will be subject to revision as dataaccumulate on more representative samples.

SPECIFIC COMMENTS ON THE NEW ZEALAND DISCUSSION PAPER

Foreword

This document is limited to a single hypothetical hazard issue withinan environment of multiple hypothetical hazards to health now beingassessed internationally and particularly by the United States NationalAcademy of Science (17) with international participants. The WorldHealth Organisation through its International Agency for Research onCancer (IARC) (19) is also involved. These organisations areobjective and scientific and their reports do not emerge fromorganisations which also have active anti-tobacco smoking programsare thus not confounded by in-house conflicts of interest or biases inpolicies as this document produced by members of a health promotionprogramme.

The air pollutants to which people are exposed and which havereasonably similar hypothetical adverse morbidity and mortality risksare numerous and are to be found in homes, in workplaces and theenvironment of towns and cities. The "harm" as the document fordiscussion in the foreword by the Minister states, "can range fromdiscomfort to danger of disease". "That the risk is needless andpreventable, and one which no one should have to accept" is alsostated. That there are multiple risks is not apparently recognised.

The statement follows that "passive smoking is a major environmentalcause of cancer in New Zealand" and that "it can cause lung cancereven in healthy nonsmokers" has no basis from science orepidemiology in New Zealand (or Australia) but only extrapolationsfrom confounded epidemiological studies which are dominated bycontentious contributions from environmentally disadvantaged areas inAsia.

A cursory perusal of the national Academy of Sciences publications onIndoor Air Pollution (29), Polycyclic Aromatic Hydrocarbons (30) andPolycydic Organic Matter (31), and the US Environmental ProtectionAgency Total Exposure Assessment Methodology (TEAM) Study (32)will make it very clear that Environmental Tobacco Smoke is but oneof many complex materials which may pose a risk to human health andone of many that need to be assessed.

Dominating the list of sources of materials which may induce ill-healthby inhalation are the incomplete combustions of the fossil fuels: coal,petroleum, diesel oil, kerosine, natural and coal gas and wood.These would be followed by man-made chemicals which pervade ourhomes, factories and streets.

This health promotion discussion document would have benefitted muchfrom adopting the National Academy of Science and the InternationalAgency for Research on Cancer approaches by honestly telling thepublic what is known and what is not known. The views of ProfessorWeiss (15) of the Harvard School of Public Health are consideredworthy of reflection. (ibid page 6)

8

Overview (P.4 & 51 The research programs on the health effects ofETS continue. The results are conflicting and all studies areconfounded. (See Appendix 3)

While ETS is generally the only visible pollution there are many othersources of invisible pollutions with carcinogenic potential for humansand other toxicological properties found to be emitted to indoor airfrom heaters, cookers, cooking, furnishings, building materials andwork processes.

In view of the presence of other important indoor pollutants, eg,oxides of nitrogen, formaldehyde, benzene, organic compounds,ionising radiation, benzo-a-pyrene, biologicals, and nitrosamines theBill should encompass all and be named A Pollution Free IndoorEnvironments Bill.

Of the 3800 chemicals said to be released, only nicotine derivedchemicals are tobacco specific and they are few in number - the vastmajority are also produced by incomplete combustion of fossil fuels,plant and other organic materials.

In the most recent IARC report the following statement is made, IARC(28), 1987 p.43: "Reported effects have ranged from increasedrespiratory illness in early life in children of smoking parents toimpaired lung function and lung cancer in adults. These findings,however, remain controversial, and the reality of the risks frompassive smoking is not yet generally accepted. One reason for this isthe paucity of data quantifying the exposure of those at risk. Thedegree of exposure has been ascertained largely by unvalidatedquestionnaire methods (for example, inference of exposure when thespouse smokes). Our understanding of risks from passive smokingwould benefit greatly from the availability of reliable quantitativemarkers to supplement questionnaire data and provide estimates of themagnitude of the dose received. However, the problem is a complexone because of the multitude of potentially harmful substances foundin differing concentrations in mainstream and sidestream smoke andbecause of the difficulties involved in measuring the lowconcentrations likely to be found in nonsmokers."

In workplaces, particularly blue collar industries, environmentaltobacco smoke is the least important of the air pollutants to whichemployees are exposed.

The Sick Building Syndrome has emerged and with it the finding ofmany indoor air pollutions as competitors to ETS in the health effectsmarket. These competitors include airborne phenolics, asbestos,carbon monoxide, carbon dioxide, plastics, paints, ozone, solvents,disinfectants, formaldehyde, trichlorethylene and other gasses andparticulates. Some others penetrate the building from the generallypolluted air environment of the city streets and inevitably containpolycydlic aromatic hydrocarbons from motor vehicle exhausts. Thelatter, like ETS, is known to contain low concentrations ofcarcinogens.

The syndrome has been described variously but basically consists ofcomplaints of eye irritation, dry mouth and throat, nasal irritation,lethargy, headache, stuffiness and vague general feelings of ill-health.

9

In England reports by Robertson (33) et al and Finnegan (34) et alindicate that it is a problem of air conditioned buildings more so thannaturally ventilated buildings. In the USA and Canada there aregovernment reports from the National Institute of Occupational Safetyand Health and Welfare Medical Services on the possible origin ororigins of the complaints. Interestingly, complaints of ETScontributed only 2% to the suspected problems. Inadequate ventilationwas the dominant complaint having a 50% problem rating. FromHonolulu a report by Kodama and McGee (35) specifically reports onbiological pollutants and the adverse findings in air conditionedpremises.

Here again are data which do not seriously incriminate ETS as ahazard commonly encountered and, when it is, control by ventilationis the answer. If that ventilation cannot control ETS it cannotcontrol the other pollutants as indicated by the AFL/CIO (14) attitudenoted above. In addition the presence of radon and radon daughtersin buildings which have carcinogenic effects on the lung is nowactively being researched in the US, Europe, UK and Australia.

Suspected Causes of Problems From health and Welfare Canada MedicalServices Branch Indoor Air Quality Investigations

No. %ofTotal

Inadequate ventilation 64 68Poor air circulationInadequate outdoor air (CO 2 .800 ppm)Poor temperature/ humidity control

Outdoor contaminant 9 10Re-entry of building exhaustMotor vehicle exhaust

Indoor contaminant 5 5Copy machinesTobacco smoke

Building fabric 2 2Glues and adhesivesFormaldehyde and organics

Biological contaminants 0 0No problem identified 14 15

Total 94 100

10

Suspected Causes of Problems for Health Hazard Evaluation Branch,National Institute for Occupational Safety and Health

No.% of Total

11

Inadequate ventilationContamination (inside)Contamination (outside)HumidityContamination (building fabric)Hypersensitivity pneumonitisCigarette smokingNoise /illuminationScabiesUnknown

Total

98 48.3

36 17.7

21 10.3

9 4.4

7 3.4

6 3.0

4 2.0

2 1.0

1 0.5

19 9.4

203 100.0

Adapted from Mellus et al, 1984 by Sterling (36) 1987

Airliners

In modern aircraft separation into smoking and nonsmoking zones hasbeen shown to be very effective and significant amounts of smoke donot spread throughout the whole environment.

The pioneering studies of Muramatsu et al (31) in 1984 recorded anarithmetic mean level of nicotine of 15.2 ug/m 3 in commercial aircraftin Japan prior to the introduction of smoking and nonsmoking areas.These authors further reported in 1987 levels of 5.3 ug/m in thenonsmoking areas and 13.5 ug/m 3 in the smoking sections. Also in1987, Oldaker and Conrad (38) reported on a detailed study instandard domestic airline flights.

Oldaker and Conrad summarise their results as follows, (NS -Nonsmoking, S - Smoking, ND - Not Detected):Table M. Summary of Results from Sampling Nicotine in Aircraft

aircraft seatingnicotine concn. ugJmtypesectionNrangemean

727-200NSS

737-200NSS

737-300NSS

TotalNSS

10ND(0.03)-24.22.6

80.4-42.26.8

29ND(0.04)-40.27.7

11ND(0.08)-112.46.5

10ND(0.4)-17.24.2

70.7(2)-76.721.5

49ND(0.03)-40.25.5

26ND(0.08)-112.49.2

They concluded that segregation is effective as are the ventilationsystems and that nonsmoking passengers are exposed to the meannicotine equivalent of smoking 0.0041 to 0.0082 of a cigarette inflights of about 1 hour duration and that such levels of exposure arevery low. The range of equivalent values were 0.0004-0.0369.

A detailed report entitled Estimation of Effect of EnvironmentalTobacco Smoke on Air Quality within Passenger Cabins of CommercialAircraft II has been published by Oldaker et al (39). A review,"The Impact of Environmental Tobacco Smoke on Airline Cabin Quality"by Holcomb (40) 1988, covers the recent national and internationalcontroversy on possible health effects of ETS and other substancesand conditions in airliners. He concludes:

"The available scientific evidence does not support the prohibition ofsmoking on commercial aircraft. The data that are available reveallow concentrations of substances that can be traced to ETS in smokingsections, and even lower concentrations in nonsmoking sections, thusconfirming the efficacy of current in-flight smoking policies. Theavailable data also suggest that factors or substances other than ETSmay be major contributors to subjective complaints or discomfort bypassengers and flight crew. Finally, given the limited andintermittent occasions for exposure, even in the case of compromisedindividuals and cabin attendants, adverse health effects from exposureto ETS aboard aircraft are highly unlikely."

The only study on humans in aircraft is that of Foliart et al (41). itreads as follows:

"There is concern that nonsmokers may suffer adverse health effectsfrom exposure to side-stream cigarette smoke (March 27, 1980, issue).Airline flight attendants are regularly exposed to cabin air that iscontaminated with cigarette smoke. We conducted a study todetermine how much carbon monoxide and nicotine are absorbed bynonsmoking flight attendants during transoceanic commercial flights.

"Participants were sought among nonsmoking flight attendants on theSan Francisco-Tokyo-San Francisco route. Before the departure ofthe flight, blood samples were obtained from participants, and theycompleted questionnaires and were given containers for a urinecollection during the return leg of the flight (Tokyo to San Francisco)Within one hour after their return to San Francisco, a second bloodsample was obtained.

"Six nonsmoking women between 30 and 40 years of age participatedin the study. Only one participant lived with someone who smoked.All were full-time flight attendants who worked 68 to 73 hours permonth. Five of the six attendants served in smoking sections on theTokyo-to-San Francisco leg of the flight. The bloodcarboxyhemoglobin concentration (mean + 0.2 percent before take off,and there was no marked difference in the post-flight sample (0.7 +0.2 percent). Blood nicotine concentrations, measured by gaschromatography, increased in five of six attendants, from a mean of1.6 + 0.8 ng per milliliter (range, 0.8 to 2.7) to 3.2 + 1.0 ng permilliliter (range, 1.6 to 4.5; P< 0.05 by Student's paired t-test).These concentrations are extremely low compared with concentrations(15 to 45 ng per milliliter) found in typical cigarette smokers.Urinary excretion of nicotine during the eight-hour flight averaged12.9 + 6.5 ug (range, 6.8 to 21.7) and was lowest in the flightattendant who worked in the nonsmoking section. On the basis ofurinary-excretion data and known pharmacokinetic data for nicotine,we estimated that the flight attendants, on the average, were exposedto 0.12 to 0.25 mg of nicotine, and that the flight attendant exposedto the largest amount of smoke received 0.22 to 0.43 mg during theflight. We conclude that there is passive absorption of nicotine fromtobacco smoke by flight attendants during a transoceanic flight butthat the quantity consumed (equivalent to one cigarette) is relativelysmall compared with that consumed by cigarette smokers, and theconcentrations achieved are unlikely to have physiologic effects."

12

A detailed report entitled Estimation of Effect of EnvironmentalTobacco Smoke on Air Quality within Passenger Cabins of CommercialAircraft II has been published by Oldaker et al (39). A review,"The Impact of Environmental Tobacco Smoke on Airline Cabin Quality"by Holcomb (40) 1988, covers the recent national and internationalcontroversy on possible health effects of ETS and other substancesand conditions in airliners. He concludes:

"The available scientific evidence does not support the prohibition ofsmoking on commercial aircraft. The data that are available reveallow concentrations of substances that can be traced to ETS in smokingsections, and even lower concentrations in nonsmoking sections, thusconfirming the efficacy of current in-flight smoking policies. Theavailable data also suggest that factors or substances other than ETSmay be major contributors to subjective complaints or discomfort bypassengers and flight crew. Finally, given the limited andintermittent occasions for exposure, even in the case of compromisedindividuals and cabin attendants, adverse health effects from exposureto ETS aboard aircraft are highly unlikely."

The only study on humans in aircraft is that of Foliart et al (41). itreads as follows:

"There is concern that nonsmokers may suffer adverse health effectsfrom exposure to side-stream cigarette smoke (March 27, 1980, issue).Airline flight attendants are regularly exposed to cabin air that iscontaminated with cigarette smoke. We conducted a study todetermine how much carbon monoxide and nicotine are absorbed bynonsmoking flight attendants during transoceanic commercial flights.

"Participants were sought among nonsmoking flight attendants on theSan Francisco-Tokyo-San Francisco route. Before the departure ofthe flight, blood samples were obtained from participants, and theycompleted questionnaires and were given containers for a urinecollection during the return leg of the flight (Tokyo to San Francisco)Within one hour after their return to San Francisco, a second bloodsample was obtained.

"Six nonsmoking women between 30 and 40 years of age participatedin the study. Only one participant lived with someone who smoked.All were full-time flight attendants who worked 68 to 73 hours permonth. Five of the six attendants served in smoking sections on theTokyo-to-San Francisco leg of the flight. The bloodcarboxyhemoglobin concentration (mean + 0.2 percent before take off,and there was no marked difference in the post-flight sample (0.7 +0.2 percent). Blood nicotine concentrations, measured by gaschromatography, increased in five of six attendants, from a mean of1.6 + 0.8 ng per milliliter (range, 0.8 to 2.7) to 3.2 + 1.0 ng permilliliter (range, 1.6 to 4.5; P< 0.05 by Student's paired t-test).These concentrations are extremely low compared with concentrations(15 to 45 ng per milliliter) found in typical cigarette smokers.Urinary excretion of nicotine during the eight-hour flight averaged12.9 + 6.5 ug (range, 6.8 to 21.7) and was lowest in the flightattendant who worked in the nonsmoking section. On the basis ofurinary-excretion data and known pharmacokinetic data for nicotine,we estimated that the flight attendants, on the average, were exposedto 0.12 to 0.25 mg of nicotine, and that the flight attendant exposedto the largest amount of smoke received 0.22 to 0.43 mg during theflight. We conclude that there is passive absorption of nicotine fromtobacco smoke by flight attendants during a transoceanic flight butthat the quantity consumed (equivalent to one cigarette) is relativelysmall compared with that consumed by cigarette smokers, and theconcentrations achieved are unlikely to have physiologic effects."

12

NZ Paper Page 6. The 4000 smokers is at best a guesstimate withoutscientific foundations. All the diseases from which the 4000 are saidto have died have multiple causative factors including genetic, socio-economic, work and lifestyle factors. The proposition that "tobaccocan cause disease in nonsmokers" is the subject of much contentionand active international research.

NZ Paper Page 7. The NRC (17) data on dose is as follows: 0.0086mg to an extreme of 0.5 mg compared to 196-280 mg of a smoker.

NZ Paper Page 8. Duration of Exposure. What is of importance isnot the duration of exposure to infinitely variable concentrations butthe amount absorbed. For nonsmokers not exposed to ETS,nonsmokers exposed to ETS and smokers have the following ratios forabsorption of nicotine as measured by cotinine excretion 1, 3, 300.

NZ Paper Page 9. Chemicals in Tobacco Smoke. With the exceptionof nicotine derived organic chemicals, a few only, these chemicalshave confounding origins from other combustions.

NZ Paper Page 9. Some Other Harmful Chemicals:

Carbon Monoxide: Environmental assays of carbon monoxide andmeasurements of carboxyhaemoglobin in many studies have_ , indicatedthat with the exception of badly ventilated highly polluted indoor roomconditions carboxyhaemoglobin levels are in the range of <1 to 2%.This is below the level at whièh physiological effects can be found indedicated studies. The studies by Aranow which were often quotedhave been discredited and have no longer a mention since 1983 by theUSSG.

HCN."Hydrogen cyanide is formed primarily from protein attemperatures above 7000 (Johnson and Karg, 1971). Thus,smoldering of tobacco at 600 0C does not favour the pyrosynthesis ofhydrogen cyanide to the extent that it occurs during MS generation."NRC p 31 (17).

Nicotine. The USSG is the only and political user of addiction asdescriptive of what others describe as habituation.

NZ Paper Page 10. What little evidence there has been since 1986continues the contentious nature of the issue. The conclusions ofmajor authorities are as follows.

International Agency for Research on Cancer (28) 1986:

"Several epidemiological studies have reported an increased risk oflung cancer in nonsmoking spouses of smokers, although some othershave not. In some studies, the risk of lung cancer in nonsmokersincreased in relation to the extent of spouses' smoking. Each of thestudies had to contend with substantial difficulties in determination ofpassive exposure to tobacco smoke and to other possible risk factorsfor the various cancers studied.

13

The resulting errors could arguably have (artificially) depressed orraised estimated risks, and, as a consequence, each is compatibleeither with an increase or with an absence of risk. As the estimatedrelative risks are low, the acquisition of further evidence bearing onthe issue may require large scale observational studies involvingreliable measures of exposures both in childhood and in adult life."The 1987 IARC scientific publication suggests methodologies to ensurea more cohesive international approach to research on ETS and health.

National Academy of Science/ National Research Council (17):

"Two alternative explanations can be given for the finding of anincreased risk in the epidemiologic studies. The finding mayrepresent a direct and causal effect of ETS exposure on lung cancerin nonsmokers; or it could be due in whole or in part to bias, eitherin the form of systematic errors in the reporting of information or aconfounding factor that is associated both with lung cancer and thefact of living with a spouse who smokes" and "considering theevidence as a whole, exposure to ETS increases the incidence of lungcancer in nonsmokers", which appears in the executive summary.

The conclusions of the two agencies which have smokingcessation/anti-smoking activities and are thus likely to be biased insupport of their programs are:

US Surgeon General (18):

"Previous reports of the Surgeon General have reviewed the dataestablishing active cigarette smoking as the major cause of lungcancer. The absence of a threshold for respiratory carcinogenesis inactive smoking, the presence of the same carcinogens in mainstreamsmoke and sidestream smoke, the demonstrated uptake of tobaccosmoke constituents by involuntary smokers, the demonstration of anincreased lung cancer risk in some populations with exposures to ETSleads to the conclusion that involuntary smoking is a cause of lungcancer."

The Australian NH & MRC's (42) limited and brief study:

"Council recommends that, in view of the evidence that passivesmoking increases the risks of at least some diseases of childhood andadulthood, particularly disorders and diseases of the respiratorysystem, the following policies should be considered by healthauthorities in Australia as means of protecting the health ofnonsmokers:

1.

Development of procedures, regulations, or laws facilitating orrequiring the restriction or prohibition of smoking:a. Within the work environment. The available evidence

indicates that exposure to passive smoking is, onaverage, substantially higher at work than at home. Anatmosphere free of tobacco smoke should be regarded asthe workplace norm. Employers should be reminded ofobligations to provide a safe working environment.

b. In enclosed public places, hospitals, restaurants andtransport.

In recommending this policy, Council notes the apparentlysuccessful approach in some US States of declaring smoking tobe prohibited in all public places and workplaces unlessspecifically permitted.

14

15

2. Explicit inclusion in smoking-related health education material ofinformation about the known or suspected effects of passivesmoking upon health. This material should assist decisions inthe home (including decisions about ventilation) as well as inthe workplace and in public places.

"In making these recommendations, Council acknowledges that furtherresearch is necessary to confirm and elaborate these effects of passivesmoking upon health. However, mindful of the delays involved,historically, in both accepting and acting upon the evidence of theserious health consequences of active smoking, Council considers thatthere is now sufficient evidence pertaining to the effects of passivesmoking to require prudent public health action of the kindrecommended above. t,

NZ Paper Page 11. Para 2. The US Surgeon General's report wascompleted prior to the NRC 1986 report and is not based on thatreport. There are major differences of interpretation of the evidencereviewed by the USSG and NRC, the latter being more objective.

There has only been one new study in the workplace, by Kentner(2). The results did not indicate essential impairment in lungfunction from exposure to ETS.

No new studies on adults with regard to respiratory andcardiovascular disease have been located.

One new report, on nonsmoking spouses of smokers, an update byLam (43) in Hong Kong on lung cancer has been published.

NZ Paper. Para 3. Children. The reports of adverse effects onchildren up to 1 year of age have some consistency. Babies shouldnot be exposed to ETS particularly by their mothers.

Eye Irritation. This occurs in conditions of inadequate ventilationand is a sign that the ventilation and control of all air pollution isineffective.

NZ Paper Page 12. Asthma. Only one study is cited. The state ofthe art is that the Royal College of Physicians found that allergy totobacco is rare, that asthmatics seem more likely than non-asthmaticsto find tobacco smoke irritating and that the results of studies areconflicting. The studies of Dahms (44) and Shepherd (45) conflictwith regard to positive and negative responses as do the more recentstudies by Knight and Breslin (46) and Weidemann (47). Studies inchildren do not appear to have been undertaken.

Lung Cancer. The reference in the New Zealand Health Departmentdiscussion document to the NRC is grossly incomplete. It shouldread:

1. "A summary estimate from epidemiologic studies places theincreased risk of lung cancer in nonsmokers married tosmokers compared with nonsmokers married to nonsmokers atabout 34%. Assuming linearity at low-to-average doses and aconstant proportionality of nicotine and carcinogens inmainstream smoke and ETS, extrapolation from studies of activesmokers using relative urinary cotinine places the risk atabout 10%.

2. "To some extent, misclassification (bias) may have contributedto the results reported in the epidemiologic literature.However, bias is not likely to account for all of the increasedrisk. The best estimate, allowing for reasonablemisclassification, is that the adjusted risk of lung cancer isincreased about 25% (ie, RR = 1.25) in nonsmokers married tosmokers compared with nonsmokers married to nonsmokers.When one allows for exposure to nonsmokers who reportthemselves as unexposed, the adjusted increased risk is atleast 24%. The adjusted increased risk to a group ofnonsmokers married to nonsmokers is at least 8% (ie, RR =1.08) compared with truly unexposed subjects. This excessrisk may come about from exposures in the workplace or otherpublic places."

Lung cancer in nonsmokers who are not exposed to industrialcarcinogens is a rare disease. The IARC (19) best "Estimate" oflung cancer mortality rates per 100,000 population per year by agegroups in nonsmokers (based on Garfinkel 1981) is given in theIARC Vol.38 p.232 (10).

Age group (years)35-4445-5455-6465-7475+

Men 2.486.2313.2025.1043.50Women 1.173.588.9219.3237.71

The 34% additional risk is based on all studies published by 1986 andis dominated by two flawed studies in Tokyo and Athens and theother studies in Asia. When US studies are subjected to a meta-analysis the estimated increase in risk from ETS is only 14%. Thatis the risk relative to a non-exposed nonsmoker is 14%, ie, RR 1.14.This increase in RR not only includes the no effect (null) hypothesisbut is considered too small to be identified in any epidemiologicalstudy. Pearce and Howard (48) 1986, in a study of cancer in NewZealand draw a line of no significance epidemiologically at RR of 1.5(50% extra risk) stating that in their analysis of occupational groups(table 4) was "restricted to relative risks greater than 1.5 to limitits size and because relative risks less than 1.5 are particularlylikely to be due to bias or confounding".

16

Further the NZ paper (reference 32) page 12 para 3 is that ofRobbins in an appendix, not reviewed by the NRC (17) scientificpanel. The reference is in error and should cite P296. Reference33 refers to an aged 1965 estimate by Pike and Doll. Robbinscautions readers as follows:

"Exercises on quantitative risk assessment serve several usefulpurposes. First, public health decisions must often be made withoutcertainty as to the magnitude of the likely health benefits that wouldresult from implementing the various policy options. Quantitativerisk assessment can aid in the decision-making process byquantifying this uncertainty. Second, difficulties encountered inproviding precise estimates in quantitative risk assessment highlightareas where scientific knowledge is inadequate. Thus, exercises inrisk assessment can serve to help focus future research. (myemphasis)

"All quantitative assessments of risk rely on assumptions. Intervalestimates of quantitative risk are reliable only insofar as:

(1) the assumptions under which they were derived are valid and

(2) the range of parameter values used in the estimation processincludes the true value.

It follows that no quantitative risk estimates can be guaranteed to bereliable. Nevertheless, some risk estimates are more (or less)reliable than others." (my emphasis)

Cancers Other than Lung Cancer. While Wells NZ reference 34,expresses concern about ETS exposure and other cancers theauthoritative IARC (19), p.305 states:

"Positive associations have been reported between passive exposureto tobacco smoke and cancers at all sites (Hirayama 1984, Sandler etal 1985), for nasal sinus cancer and brain tumours (Hirayama 1984),and for cancers at many individual sites other than the lung(Sandler et al 1985). The Working Group noted that these findingswere at present difficult to interpret, as many related to sites thathave not been strongly associated with active smoking."

The NRC p.255 states:

"These recent observations on a possible connection between ETSand various forms of cancer have created much discussion and someconfusion. The lack of consistency with other data on tumors amongchildren of smoking mothers and the appearance of tumors that arenot clearly smoking-related call for further epidemiologic research.Any new studies in this area will, hopefully, have a very careful,rigorous design, so that more definitive evaluation of this possiblehealth hazard from ETS exposure is possible.

"There is no consistent evidence at this time of any increased riskof ETS exposure for cancers other than lung cancer.

"Smoking related cancers other than lung cancer need to be studiedwith adequate numbers and good exposure data and withconsideration of the potential confounding effects from other knownrisk factors for these cancers."

17

Further the NZ paper (reference 32) page 12 para 3 is that ofRobbins in an appendix, not reviewed by the NRC (17) scientificpanel. The reference is in error and should cite P296. Reference33 refers to an aged 1965 estimate by Pike and Doll. Robbinscautions readers as follows:

"Exercises on quantitative risk assessment serve several usefulpurposes. First, public health 'decisions must often be made withoutcertainty as to the magnitude of the likely health benefits that wouldresult from implementing the various policy options. Quantitativerisk assessment can aid in the decision-making process byquantifying this uncertainty. Second, difficulties encountered inproviding precise estimates in quantitative risk assessment highlightareas where scientific knowledge is inadequate. Thus, exercises inrisk assessment can serve to help focus future research. (myemphasis)

"All quantitative assessments of risk rely on assumptions. Intervalestimates of quantitative risk are reliable only insofar as:

(1) the assumptions under which they were derived are valid and

(2) the range of parameter values used in the estimation processincludes the true value.

It follows that no ctitative risk estimates can be guaranteed to bereliable.Nevertheless, some risk estimates are more (or less)reliable than others." (my emDhasis '

Cancers Other than Lung Cancer. While Wells NZ reference 34,expresses concern about ETS exposure and other cancers theauthoritative IARC (19), p.305 states:

"Positive associations have been reported between passive exposureto tobacco smoke and cancers at all sites (Hirayama 1984, Sandier etal 1985), for nasal sinus cancer and brain tumours (Hirayama 1984),and for cancers at many individual sites other than the lung(Sandier et al 1985). The Working Group noted that these findingswere at present difficult to interpret, as many related to sites thathave not been strongly associated with active smoking."

The NRC p.255 states:

"These recent observations on a possible connection between ETSand various forms of cancer have created much discussion and someconfusion. The lack of consistency with other data on tumors amongchildren of smoking mothers and the appearance of tumors that arenot clearly smoking-related call for further epidemiologic research.Any new studies in this area will, hopefully, have a very careful,rigorous design, so that more definitive evaluation of this possiblehealth hazard from ETS exposure is possible.

"There is no consistent evidence at this time of any increased riskof ETS exposure for cancers other than lung cancer.

"Smoking related cancers other than lung cancer need to be studiedwith adequate numbers and good exposure data and withconsideration of the potential confounding effects from other knownrisk factors for these cancers."

17

The NZ reference 35 is highly misleading, and the actual viewsshould be noted, Page 262 of NRC (19):

"Experimental studies of subcutaneous or intravenous administrationof nicotine in rabbits (Schievelbein et al, 1970; Schievelbein andRichter, 1984) and monkeys (Liu et al, 1979) have demonstrated thatlong term exposure leads to arteriosclerotic lesions. Exposure tocarbon monoxide also leads to atherosclerosis in rabbits, pigeons andother animals (Astrup and Kjeldsen, 1979). Studies of whole tobaccosmoke indicate that total serum cholesterol concentrations areincreased and the ratios of the various lipoprotein fractions arechanged (McGill, 1979). The contribution of whole tobacco smoke tomodifying the lipoprotein fractions is not conclusive. However,there have not been experimental studies of the effects of ETSexposure or administration of ETS extracts."

The reference 36 again by Wells is better stated by the NRC (19) asfollows (P.265):

1. "No statistically significant effects of ETS exposure on heartrate or blood pressure were found in healthy men, women, andschool aged children during resting conditions. Duringexercise there is no difference in the cardiovascular changesfor men and women between conditions of exposure to ETS andcontrol conditions.

2. "With respect to chronic cardiovascular morbidity andmortality, although biologically plausible, there is no evidenceof statistically significant effects due to ETS exposure, apartfrom the study by Hirayama in Japan."

NZ Paper Page 14. "Tobacco smoke is one of the leadingenvironmental causes of cancer in New Zealand". Has there beensuch a study? I think not. In the major New Zealand study (48) ofcancer mortality environmental tobacco smoke does not rate amention. The Doll and Peto view (Reference 38 in the New Zealandpaper) is but a wild guess that 100 times an immeasurably small riskbecomes a measurable risk. This is merely another anti-smokingcampaign use of a throw-away line.

NZ Paper Page 15. A Clean Environment. See Pages 2, 3, 4 and 5of this document.

NZ Paper Page 32. Aircraft - ETS in Airliners. Refer to Pages 12-14.

NZ Paper Page 39. Appendix 1. While no doubt scientificallycorrect the concentrations of chemicals in sidestream smoke (SS)were measured within 3 millimetres from the burning end of thecigarette. That is not ETS. The contents of ETS largely originatesin the SS and is then greatly diluted. It is reiterated that only theabsorbed dose, less than 1% of that of a smoker, is of any note.

18

NZ Paper Page 40-43. This appendix II is a brief summary of theIARC (19) Vol. 38, Evaluation of the Carcinogenic Risk of Chemicalsto Humans. It fails to reasonably inform the non-scientific public ofthe significance of the chemicals with regard to a potential orhypothetical threat to health. Such chemicals are generally innanogram amounts (1000 millionths of a gram) per cigarette.

The reader is not informed that with a few exceptions, mainlyderived from nicotine, all the- major classes named, N-nitrosamines,volatile carcinogens, polycycic aromatic hydrocarbons, polycydicasa-arenes, are all present in any environment into which smoke orfumes from the incomplete combustion of wood, vegetation, the fossilfuels, coal, natural gas, petrol, diesel fuel and even cooking,particularly by gas, occurs. For example, Robbins (NRC p.304)(19) indicates that 98% of Benzo-a-pyrene, an animal carcinogen andtherefore a suspect human carcinogen, found in such combustion,comes from cooking, coal burning and other sources and that thecontribution from environmental tobacco smoke may be only 2% orless.It is also of note that human intake and absorption ofnitrosamines is dominated by the ingestion of foodstuffs. Thisappendix will induce undue concerns in people not well educated inthe sciences of chemistry, carcinogenesis and toxicology.

NZ Paper Page 44-46 Appendix III. Deaths Attributable toInvoluntary Smoking in New Zealand. It was noted above that oneof the authors, Pearce (48), considers that a hazard to health whichis a relative risk of 1.5 or less (50% increase or less) are unreliable.However, it is not noted in this appendix that the studies on ETS asassessed by the NRC are 1.34 overall and 1.14 for United Statesstudies, ie, less than 1.5.

Doll and Peto 1981 (5), have indicated that Relative Risks of lessthan 2.0 are difficult to interpret and Rose (49), in the UK, notesthat in multi-factorial diseases a factor that increases the risk byless than a half (<50%) will almost certainly be undetectable. All thediseases rated as related to the smoking of tobacco are multifactorial.

The reader is not informed that all the studies of ETS and lungcancer have been on the nonsmoking spouses of smokers and aremainly on Asians and Asian countries. All the studies arerecognised as confounded and thus not reliable. Thus they cannotmeet the epidemiological criteria of causation cited such as thestrength of the association (less than RR 1.5), consistency, andspecificity (4 studies with statistically significant results, 15 withstatistically non significant results). Only the biological plausibilityhas been met as it also has been for many urban indoor and outdoorpollutions.

Lung Cancer Deaths Attributable to Involuntary Smoking in NewZealand. These are based on the estimates by Repace and Lowry(50) in the USA and are private views not accepted by the USEnvironmental Protection Agency (EPA) which employs Repace. Theestimates have also been rejected by other researchers. The upperlevel of Repace of 5000 lung cancer deaths per year in the US hasbecome the most used estimate by those making emotive calculations.The EPA prefers the 500 level and Arundel and Sterling (50)calculate 12 per year in the US - a number which could not be foundin epidemiological studies. The EPA TEAM Study (32) (see page107) is of note in this attributable risk numbers estimates.

19

20

Deaths from Ischaemic Heart Disease.The National Academy ofScience (17) 1986 p.265 has reviewed this subject and concludes:

1. "No statistically significant effects of ETS exposure on heartrate or blood pressure were found in healthy men, women andschool aged children during resting conditions. Duringexercise there is no difference in the cardiovascular changesfor men and women between conditions of exposure to ETS andcontrol conditions.

2. "With respect to chronic cardiovascular morbidity andmortality, although biologically plausible, there is no evidenceof statistically significant effects due to ETS exposure, apartfrom the study by Hirayama in Japan."

Deaths from Cancers other than the Lung.Insufficient dataavailability is agreed.

The Feedback Form. This is either in my view unbelievably badlydesigned or deliberately designed to acquire the results that theauthors wish to obtain. An expert in the design of questionnairesshould re-design the form.

To conclude, I offer the opinion that the discussion document itselffails to provide the information needed to promote discussion. Itfails to provide information on what is not known, what is incontention, the biological meaning of some scientific data, and thedifficulties inherent in interpreting confounded and conflictingepidemiological data related to low level risks to human health arisingfrom exposure to environmental tobacco smoke. No consideration hasbeen given to other pollutants which pervade ubiquitously our airenvironment at home, work or leisure and which are being shown tobe environmental hazards to health.

REFERENCES

1. Kentner M., Triebig G. and Weitle D. (1984) Prev. Med., 13,656-659.

2. Kentner M. and Welte D. in Indoor and Ambient Air Quality.Ed Perry, Selper, London.

3. White J.R. and Froeb H.F. (1980) Small-Airways dysfunctionin nonsmokers chronically exposed to tobacco smoke. N. Eng.3. 'Med. 302, 702-723.

4. Wynder in Vienna (1984) Symposium: Medical Perspectives onPassive Smoking, Preventive Medicine 13 (6).

5. Doll R. and Peto R. (1981) J. Natl. Cancer Inst., 66, 1193-1308.

6. Wallace L. A. (1987) The Total Exposure AssessmentMethodology (TEAM) Study: Summary and Analysis: Volume 1US EPA Washington DC.

7. Doll R. (1978) Environ. Health Perspect 22, 23-31.

8. Carnow B.W. (1978) Environ. Health Perspect 22, 17-21.

9. Dean G. (1959) Br. J. Med 2, 852-857.

10. Dean G. (1964) Proc. Roy. Soc. Med. 57, 984-987.

11. Eastcott D.F. (1956) Lancet 1, 37-39.

12. Lawther P.J. (1985) The chemical industry and the health ofthe community, Ed. F.J.C. Roe. pp.25-27.

13. Kvale G., Bjelke E., Heuch I. (1986) Int. J. Cancer 37:185-193.

14. AFL-CIO (1985) Executive Council Statement on Smoking andthe Workplace, Bal Harbour, F1.21-437.

15. Weiss S.T. and Speizer F.E. (1987) (Editorial) Am. Rev.Resp. Diseases, 135, 1225-1226.

16. Hirayama T. (1981) Br. Med. 3., 282, 183-185.

17. National Research Council (1986) Environmental Tobacco Smoke,Measuring Exposures and Assessing Health Effects,Washington, D. C.

18. U. S. Surgeon General (1985) The health consequences ofsmoking: cancer and chronic lung disease in the workplace.,U. S. Department of Health and Human Services, Rockville, Md.

19. IARC Monographs on the evaluation of the carcinogenic risk ofchemicals to humans, Vol 38, Tobacco Smoking (1986) Lyon,France.

21

22

20. Aviado D. (1986) Smoking and Society, Ed. Tollison. p.137-165.

21. Wald N.J., Nanchahal K., Thompson S . G. and Cuckle H. S.(1987) Brit. J. Med., 293, 1217-1222.

22. Lebowitz M.D. (1986) Environ. Int., 12, 3-9.

23. Uberla K. (1987) Int. Arch. Environ. Health 59, 421-437.

24. Geneva (1983) ETS - Environmental Tobacco Smoke. Eds. R.Rylander, Y. Peterson and M. -C. Snella.

25. Vienna (1984) Symposium: Medical Perspectives on PassiveSmoking, Preventive Medicine 13 (6).

26. Essen (1986) Proceedings of the International ExperimentalToxicology Symposium on Passive Smoking, Ed.E.Mohtashamipur, Toxicology Letters, Amsterdam.

27. Auerbach 0., Garfinkel L., Hammond E. C. (1979) Change inbronchial epithelim in relation to cigarette smoking 1955-1960vs. 1970-1977. New Engl. J. Med. 300:381-386.

28. TARC (1987) Environmental Carcinogens Methods of Analysisand Exposure Measurement Volume 9 Passive Smoking Lyon.

29. NRC (1981) Indoor Air Pollution.

30. NRC (1983) Polycydic Aromatic Hydrocarbons.

31. NRC (1972) Polycycic Organic Matter.

32. EPA (1987) The Total Exposure Assessment Methodology.(TEAM) Study: Summary and Analysis: Volume 1 US EPAWashington DC.

33. Robertson A.S., Burge P.S., Hedge A. (1985) Br. Med. J.,291, 373-376.

34. Finnegan M.J., Pickering C.A.C. and Burge P.S. (1984) Br.Med J., 289, 1573-1575.

35. Kodama A.M. and McGee R.T. (1986) Arch. Environ. Health.,41, 306-311.

36. Sterling T.D., Collett C.W. and Sterling E.M. (1987) J.Occup. Med., 29, 57-62.

37. Muramatsu M., Umemura S., Okada T., Tomita H. (1984)Estimation of personal exposure to tobacco smoke with a newlydeveloped personal nicotine monitor. Environ. Res. 35:218-227.

38. Oldaker G.B., Conrad F.C. (1987) Estimation of effect ofenvironmental tobacco smoke on air quality within passengercabins of commercial aircraft. Envorn. Sci. Technol. 21:994-999.

23

39. Oldaker G.B., Stancill M.W., Conrad F.W. (1988) Estimation ofeffect of environmental tobacco smoke on air quality withinpassenger cabins of commercial aircraft II. In press.

40. Holcomb L. C. (1988) Impact of environmental tobacco smoke onairline cabin air quality. Environ. Tech. Letters 11 June1988.

41. Foliart D., Benowitz N.L., Becker C.E. (1983) Passiveabsorption of nicotine in airline flight attendants (Letter) N.Engl. J. Med. 308.

42. National Health and Medical Research Council (1987) Effects ofPassive Smoking on Health Australian Government PublishingService, Canberra.

43. Lam T.H., Kung I.T.M., Wong C.M., Lam W.K., KleevensJ.W.L., Saw S., Hsu C., Seneviratne S., Lam S.Y., Lo K.K.,Can W.C. (1987) Br. J. Cancer 56:673-678.

44. Dahms T.E., Bolen J.F., Slavin R.G. (1981) Passive smoking,effects on bronchial asthma. Chest 80:530-534.

45. Shephard R . J., Collins R., Silverman F. (1979) Passiveexposure of asthmatic subjects to cigarette smoke. Environ.Res. 20:392-402.

46. Knight A., Breslin A.B.X. (1985) Passive cigarette smokingand patients with asthma. Med. J. Aust. 142:194-195.

47. Weidmann H., Mahler D.A., Loke J., Vergulto J.A., SnyderP., Matlay R.A. (1986) Acute effects of passive smoking onlung function and airways reactivity in asthma subjects.Chest 89:180-185.

48. Pearce N.E. and Howard, J.E. (1986) Int. J. Epidemiol., 15,456-462.

49. Rose G. (1987) Br. Med. J., 294, 963-965.

50. Repace J.L., Lowrey A.H. (1985) An Estimate of Nonsmokers'Lung Cancer Risk from Passive Smoking. Environ. mt . 11:3-22.

51. Arundel A., Sterling T. and Weinkam J. (1988) Indoor andAmbient Air Quality. Ed. Perry, Selper, London.

APPENDIX 1

To illustrate some basis for the controversial findings and the lackof general acceptance the following quotations have been extractedfrom government and WHO sources. They are from specifically thereports of the United States Surgeon General (USSG) Report 1986,the US National academy of Science/ National Research Council (NRC)1986, the Australian National Health and Medical Research Council(NH & MRC) and the International Agency for Research on Cancer(IARC) (1986).

These statements demonstrate some of the flaws in the Jewel of theCrown of those agencies with anti-smoking programs which relyheavily or entirely on the hypothetical effects of exposure toenvironmental tobacco smoke produced by other people - the so-called passive smoking.

Clearly in the area of ETS, knowledge is limited, and furtherresearch is needed. This must be kept in mind in consideringpublic health measures that could affect the life-styles of millions ofpeople. This is obviously not the time to introduce legislation.Rather it is the time to inform the public in a dispassionate mannerof what is known and what is not known and to review and takeaccount of new data as they are produced.

Respiratory Function

"The physiologic and clinical significance of the small changesin pulmonary function found in some studies of adults remainsto be determined. The small magnitude of effect implies that apreviously healthy individual would not develop chronic lungdisease solely on the basis of involuntary tobacco smokeexposure in adult life."

From USSG, p.62.

"The effects of ETS on the lungs of adults are likely to beconfounded by many other factors, making it difficult toattribute any portion of the effect solely to ETS."

From NRC, p.10.

"In any future studies, great care should be taken, as it wasin the two cohort studies, to account for potential confoundingvariables in the analyses, such as socio-economic status andgas cooking."

From NRC, p.202.

"In view of the large number of factors that affect lungfunction, it is not surprising that it is difficult to documentthe extent to which a single type of exposure affects lungfunction. The lungs of adults have been subjected to manyenvironmental exposures and potential insults over a lifetime,making it unlikely that a specific effect could be isolated."

From NRC, p.200.

24

"A review of these and other population studies by a NationalInstitute of Health workshop concluded that it in adults therewas no clear evidence of a clinically significant effect onpulmonary function of prolonged passive smoking, and thereappeared to be only slight and inconsistent effects on lowerrespiratory symptoms.

From NH & MRC, p.23.

"Smoking and nonsmoking families differ in various respectsand, although such differences have been recorded andadjusted for in many studies, it remains possible that some ofthe apparent effect of passive smoking arises from otherenvironmental or behavioural confounding factors that havebeen inadequately controlled."

From NH & MR C, p.18.

Cardiovascular

"In summary, for normal young adult males and females, nosignificant acute effects of ETS exposure on heart rate orblood pressure have ben reported, either under resting oraerobic conditions."

From NRC, p.260.

"With respect to chronic cardiovascular morbidity andmortality, although biologically plausible, there is no evidenceof statistically significant effects due to ETS exposure, apartfrom the study by Hirayama in Japan."

From NRC, p.265.

Misclassification

"Misclassification of the extent of ETS exposure can alsooccur, and may reduce the observed risk if a nonsmokingspouse of a smoker is not exposed to smoke at home.Friedman and colleagues (1983), reporting on a survey of38,000 subjects, noted that 47 percent of nonsmoking womenmarried to smokers reported that they were not exposed totobacco smoke at home."

From USSG p.35.

"The Garfinkel study (1985) is the only study which verifiedcase classification: "This verification showed that 13 percentof the cases classified as lung cancer were not primary cancersof the lung. This study showed that 40 percent of the womenwith lung cancer who had been classified as nonsmokers (orsmoking not stated) on hospital records had actually smoked,compared with 9 percent of the controls."

From USSG p.85.

25

Risk Evaluation

"Two alternative explanations can be given for the finding ofan increased risk in the epidemiologic studies. The findingmay represent a direct and causal effect of ETS exposure onlung cancer in nonsmokers; or it could be due in whole or inpart to bias, either in the form of systematic errors in thereporting of information or a confounding factor that isassociated both with lung cancer and the fact of living with aspouse who smokes."

From NRC, p.231.

"The resulting errors could arguably have artifactuallydepressed or raised estimated risks, and, as a consequence,each is compatible either with an increase or with an absenceof risk. As the estimated relative risks are low, theacquisition of further evidence bearing on the issue mayrequire large-scale observational studies involving reliablemeasures of exposure both in childhood and in adult life."

From IARC, p.308.

"In both the Lebowitz and Ferris studies, adjustment was madefor the effect of parental symptoms or respiratory symptoms,but did not eliminate it."

From NRC, p.190.

Asthma

"The few studies of the effect of short-term ETS exposure ofasthmatic patients and of nonasthmatics are not consistent.This may be because they have not been conducted underadequate control and have examined persons with considerablevariability in the severity of asthmatic disease and airwayresponsiveness. Future studies should carefully define thepopulations when addressing issues of frequency of reaction toETS and should be done separately on hyperresponsive andnonhyperresponsive patients when addressing issues ofseverity of reaction to ETS."

From NRC, p.217.

Environment

"We lack data on the presence and concentrations of toxic andcarcinogenic components in tobacco-smoke-polluted enclosedenvironments."

From NRC, p.47.

26

"Confounding factors that should be considered in the design,collection, and use of questionnaire data are other risk factorsassociated with the disease that may or may not be correlatedwith exposure to ETS. In the case of lung cancer, such riskfactors include, but are not limited to:

*occupation and industry of employment,* exposure to specific respiratory carcinogens, such

as asbestos, arsenic, radon, etc., in occupationalor nonoccupational settings,

*dietary factors,*family history of cancer (Ooi et al., 1986),*residential history,*housing characteristics,*years of education, and*socio-economic status."

From NRC, p.115.

Dose - Absorbed nicotine and excretion of cotinine

"Generally, concentrations of nicotine and cotinine in plasma,saliva, and urine of nonsmokers exposed to ETS amount to lessthan 1% of the mean values observed in physiological fluids ofactive smokers, even though some nicotine measurements inplasma give a higher reading (Jarvis et al., 1984)."

From NRC, p.138,141.

"Much of these data, collected in the United Kingdom (Wald etal., 1984), showed that nonsmokers had, on average, about0.4 percent of the concentration of urinary cotinine found inactive smokers."

From NRC, p.224.

Cyanide

"Hydrogen cyanide is metabolized by the liver to thiocyanate.In addition to tobacco smoke, certain foods, particularly leafyvegetables and some nuts, are sources of cyanide. Cyanide isalso present in beer."

From USSG, p.200.

"Hydrogen cyanide is formed primarily from protein attemperatures above 7000C (Johnson and Karg, 1971). Thus,smoldering of tobacco at 6000C does not favour thepyrosynthesis of hydrogen cyanide to the extent that it occursduring MS generation."

From NRC, p.31.

27

APPENDIX 2

The extent of these confounding factors is to be found in the NRCreport and is as follows:

Table 11-1 Potentially Confounding and EffectModifying Factors in Epidemiologic Studies ofExposure to Environmental Tobacco Smoke

Unreported active smokingTobacco productsMarijuanaClove cigarettes

Developmental factorsMaternal smoking during pregnancy

Factors related to outdoor environmentOutdoor temperature, humidityRespirable and nonrespirable particulates, eg., fugitive dustPollens and other allergens

Factors related to indoor environmentCrowdingNumber and age of siblingsTotal number of people/animals in dwelling unitTotal number of smokers in dwelling unitHousehold conditionsFrequency of air exchangesTemperature and humidityUse and condition of air conditioning unitsConditions of child care facilitiesUnvented combustion products from heating! cooking stovesRespirable and nonrespirable particulates, e. g, wood smokesPollens, molds, mitesAllergens and infectious organismsFormaldehyde

Factors related to work/hobbiesWork/hobby-related exposure to gases, fumes, particulates

Miscellaneous factorsAnnoyance response to tobacco smokingReporting biases

To take account of these factors is a formidable task for anyresearch group. In addition, there is the onerous task of acquiringa sufficient number of subjects and controls. Without such accountand adequate case numbers the interpretations of the findings in anyresearch is difficult. When efforts are made to combine results ofseparate studies the results are at best speculative if not invalid,ie., Wald, (1986).

28

No. of cases

FMFMFMFFMFF

FMFMFF&MFMFFFFMFF

733

335

52503414

29288

64

1467

202813

55188

2008

153338

APPENDIX 3

The results of the studies so far are as follows:

29

CLSignificanceof Result

Authors

*Akjba

*Buffler

B rownson

*Chan*Correa

*Garfinkel 1981*Garfinkel 1985pS)*Gillis

* Hirayama

Humble

*Kabat

* KooKooLam

*Lee

*pershagen*Trjchopoulos

R. R.

1.482.450.800.501.681.010.752.032.291.12

OR1.18

1.003.251.632.25

OR1.700R2.20

0.791.001.541.651.651.031.301.282.13

0.88- 2.500.46-13.060.32- 1.990.14- 1.790.39- 2.970.42- 2.410.44- 1.300.83- 5.030.31-16.500.74- 1.690.90- 1.54

0.20- 4.910.60-17.651.25- 2.111.04- 4.850.60- 4.31.00- 4.900.26- 2.430.20- 4.900.90- 2.641.16- 2.351.16- 2.350.41- 2.470.38- 4.420.75- 2.161.18- 3.78

NS

IMS

N S (adenocarNS cinoma)NSNSNSNSNS(lsubgrou

NSNS

SS

NS?SNSNSNS?S

SNSNSNS

S

The NAS analysis of all the studies available to them reported (asmarked*) an overall odds ratio (OR) of 1.34, 95% CL 1.18-1.53. Forthe case control studies an OR of 1.24 CL 1.04-1.50 and for Cohortstudies OR% 1.44 CL 1.20-1.72. However when the American onlystudies were analysed overall (thus excluding the Japanese andGreek and Swedish, Hong Kong and British studies) the OR was1.14 CL 0.92-1.4. This is not significant statistically. The Swedishand British and Hong Kong studies were also not statisticallysignificant.

There is a major problem to be faced of confounding bymisclassifying smokers or ex-smokers as nonsmokers as discussed byGarfinkel and Lee at the round table discussion at the ViennaConference 1984; they have indicated that a 5% misclassification ofthat nature alone would result in an OR or RR of 1.4. Theproblems are under study by the IARC 1987 and addressed in theirscientific series report.

STATEMENT OF GRAY ROBERTSON, PRESIDENT, ACVA ATLANTIC INC.BEFORE THE UNITED STATES HOUSE OF REPRESENTATIVES,

COMMITTEE ON ENERGY AND COMMERCE.CONCERNING THE CONTRIBUTION OF ENVIRONMENTAL TOBACCO SMOKE

TO INDOOR POLLUTION

STATEMENT OF GRAY ROBERTSON, PRESIDENT, ACVA ATLANTICINC.,

BEFORE THE UNITED STATES HOUSE OF REPRESENTATIVES,COMMITTEE ON ENERGY AND COMMERCE, CONCERNING THE

CONTRIBUTION OF ENVIRONMENTAL TOBACCO SMOKE TO INDOORAIR POLLUTION

INTRODUCTION

I am President of ACVA Atlantic, Inc., a company that specialises inthe study and assessment of indoor air pollution.I received aBachelor of Science Degree with honors in Chemistry and Botanyfrom London University (Queen Elizabeth College) in 1964. Prior toentering London University, I was employed as a bacteriologist withEvans Medical Company Limited, (part of the Glaxo Group ofcompanies) in Liverpool England, and studied for a HigherCertificate of Education in Botany and Zoology at Liverpool TechnicalCollege.

After extensive experience with the Johnson Matthey Company inEngland, from 1964 through 1973 as a chemist primarily involved inthe use of precious metal catalysts in the oil, gas and chemicalindustries, I became head of that company's chemical division in theUnited States for the next five years. I then became GeneralManager of Phosyn International Limited, a company specialising inthe development of trace element technology and the manufacture ofspeciality chemicals. In 1980, I became Director of WintonLaboratories Limited, an English company specialising in the study ofindoor air pollution throughout buildings in Europe, Asia andScandanavia. While at Winton Laboratories, I personally wasinvolved in the study of a number of buildings and in designing anddeveloping methodology to identify the sources of indoor airpollution. I have been licensed to market in this country thetechnology developed by Winton Laboratories to assess and resolveindoor air pollution problems. Pursuant to that license, Peter Binnie(Technical Director of Winton Laboratories) and I incorporated ACVAin the United States in 1981.

1

ACVA is often called upon to investigate what has become known as"sick building syndrome" in large commercial office buildings. ACVAseeks to identify the causes of indoor air quality problems, and torecommend remedial steps. Over the past five years, ACVA hasstudied the air quality in 125 major office buildings with over 27million square feet of space occupied by over 135 thousand people.Clients include major banks, insurance companies, propertydevelopers, hospitals, colleges and government agencies. ACVA hastested the indoor air quality in the Department of Health and HumanServices, the Social Security Administration, the LongworthCongressional Building, the United States Supreme Court, GSARegional Headquarters, the Customs and Excise Building and theCoast Guard Building.

H.R. 4488 and H.R. 4546 suffer from the fundamental misconceptionsthat "tobacco smoke is a major contributor to indoor air pollution,"and that "tobacco smoke is the greatest source of ... harmfulpollution" violating particulate matter air quality standards.Mybuilding studies, and research conducted by NIOSH, 1 do not supportthese assumptions. We frequently investigate buildings as a resultof complaints from occupants with symptoms such as eye and noseirritation, fatigue, coughing, rhinitis, nausea, headaches, sorethroats and general respiratory problems. Our clients often assumethat these symptoms are due to environmental tobacco smoke (ETS).However, identical symptoms are found in individuals exposed toformaldehyde, carbon monoxide, oxides of nitrogen, and ozone.Moreover, similar symptoms are reported by individuals with allergiesto Aspergillus, Cladosporium, and Penicillium species of fungi, amongothers, as well as other miscellaneous bacterial aerosols. Exposureto household dusts, cotton fibres, and fiberglass fragments resultsin many identical symptoms.

This statement will describe some of ACVA's experience with indoorair pollution. ACVA's research bears directly upon the nature andextent of exposure to environmental tobacco smoke experienced byoffice workers in commercial buildings. ACVA's primary finding -that environmental tobacco smoke rarely is the cause of the indoorair pollution problems found in these buildings - it is important tothis Committee's work and contrary to one of the primary findingssupporting each of these proposed bills.

2

ACVA'S EXPERIENCE

Over the past five years, ACVA has been called upon to investigatebuildings due to air quality complaints from employees, for hygienicreasons, such as to investigate visible dirt in or around heating,ventilating and air conditioning (HVAC) grilles; for preventivemaintenance purposes and sometimes as a matter of general interest.On many occasions, initial complaints are thought to be related tothe presence of ETS, when symptoms in office workers such as soreand watery eyes, headaches, nausea, coughing and generalrespiratory problems are reported. While sensitive individuals mayexhibit many of these symptoms upon being exposed to highconcentrations of tobacco smoke, ACVA seeks to determine the actualnature and origin of such complaints. We have found that it can bemisleading to focus only upon the symptoms described by buildingoccupants in attempting to determine the actual nature and origin ofsuch complaints. We have found that it can be misleading to focusonly upon the symptoms described by building occupants inattempting to determine the actual cause of poor indoor air quality.Through extensive on-site testing, ACVA seeks to discover thesource of the problem and to suggest remedial steps specific to aparticular contaminant and situation.

A.Sick Building Syndrome

1.The Impact of Energy Conservation

The complex of symptoms that I already have mentioned has beentermed the "sick building syndrome," a condition often thought toresult from energy conservation efforts to seal buildings and reducethe infiltration/ exfiltration of air. Stringent energy consersationefforts by building managers have reduced the natural infiltration offresh air that previously existed in many buildings. These stepshave exacerbated the often undiscovered problem of a poorlydesigned and maintained HVAC system.

cI

Without doubt, the desire to reduce operating expenses has been themajor obstacle to increasing ventilation rates. Most companies haveincorporated energy management programs and new operatingbudgets based on saving every energy dollar possible. In fact, inmy address last year to the Association of Professional EnergyManagers, I pointed out that, since their very salaries and bonusesare dependent on reduced costs, it would be an anathema for themto consider increasing energy usage and operating costs byincreasing ventilation.

However, forward thinking companies should look beyond the budgetconstraints of energy managers. Consider the following facts: Theaverage heating, ventilating and air conditioning operating costs of atypical 100,000 square foot building in this area would be $50,000per annum. A commendable target for energy saving by reducingventilation may be 25%, resulting in $12,500 per annum.

Now consider the payroll costs for people in that building. A veryconservative value based on national averages would be $10,000,000per annum per 100,000 square feet. Thus each 1% of absenteeismcosts $100,000 per annum. Typical absentee rates run at 3-7%, andno less than 50% of all absenteeism is estimated to be due to upperrespiratory problems. How many of these absences are due todusts, bacteria, fungi, fibres, chemicals, ETS, carbon monoxide,oxides of nitrogen; i.e., how many are due to the internal pollutioncomplex?

In short, what does it profit a company to realize $12,500 in energysavings if that small saving potentially causes hundreds of thousandof dollars in absenteeism, not to mention lost worker efficiency. Itis not surprising that some European countries, including Denmark,West Germany and Switzerland, have introduced legislation mandatingthat steps be taken to prevent the build-up of internal pollutants.This country is destined to follow that course, either by slowevolution or legislation, and it may result from court actions broughtby individuals or trade unions. Such suits could seek that buildingowners, designers and operators accept responsibility for the healthand welfare of their staff or tenants. In addition to tighteningbuildings and sealing windows, building managers have shut downair conditioning systems at night and on weekends in an effort tolower energy costs.

4

When the air conditioning is shut down in humid climates,condensation builds up and settles inside the ductwork. If dirt ispresent in damp ductwork, spores and microbes can flourish, only tospread throughout the building once the HVAC system is turned onthe next morning. This often results in Monday morning complaintsof building odors or building sickness that disappear during theweek, only to reoccur the following Monday morning. To save moreenergy, automatic temperature controllers are used to cycle fans onand off during the day. Vibrations from the start-up of these fanscan cause dirt and germs trapped inside ductwork to be dislodgedand carried into occupied areas.

Another energy conservation effort that may contribute to sickbuilding syndrome is the recirculation of indoor air, at the expenseof fresh outdoor air. This may be a deliberate policy or it may be aresult of the designer's shortsightedness. This results in thecontinuous redistribution of infectious microbes, allergenic dusts andspores from office to office and floor to floor.

Another cause of sick building syndrome is the installation ofvariable air volume (VAV) air systems that sometimes permit theintroduction of contaminated air into an occupied space. Forexample, we have found that the mixing boxes of some VAV systemsare equipped with louvers that open into the ceiling voids in whichthey are mounted. When specific temperature conditions exist, thelouvers are opened and return/exhaust air can be introduced fromthe void and distributed throughout the system. This operationallows air from the building void, which is often dirty andcontaminated, to return to the building space without treatment bythe system's filtration devices.Fiberglass particles sometimes areintroduced into a building space from this source. During ourstudies from June 1984 through July 1985, three outbreaks ofdermatitis reactions in employees were traced to the introduction offiberglass particles through VAV mixing boxes. In May 1985, ACVAtraced a sickness outbreak in a federal building to the spread ofallergenic fungi from its source in the ceiling void through a VAVheater box. In addition, the substitution of low cost, low efficiencyfilters to reduce pressure drops and save energy seriously reducesthe efficiency of building filtration systems, and can lead to seriousindoor air quality problems (filters may be not only poor, but poorlyinstalled).

5

8

2.Sources of Indoor Air Contamination

Virtually every indoor decoration, building material, or piece offurniture sheds some type of gaseous or particulate pollutant. Whena building is new, some compounds are given off quickly and soondisappear. Others continue "outgassing t ' at a slow pace for years.Clothing, furnishings, draperies and carpets contribute fibers andother particulates. The use of cleaning solvents, detergents andpest control agents, such as chlordane, also may cause persistentproblems in office environments. Sweeping, vacuuming and dustingnormally remove the larger particles, but often increase airborneconcentrations of smaller particulate matter. Common office supplieshave been found to release dangerous chemicals, such asformaldehyde (from paper), into the air. Evaluations by NIOSH ofindoor office spaces have discovered methyl alcohol from spiritduplicators and methacrylate and ozone from copiers •2

Environmental tobacco smoke often is blamed for the symptomsassociated with "sick building syndrome." The reason is obvious -environmental tobacco smoke often is the only visible indoor airpollutant. ACVA has determined, however, that the presence ofhigh concentrations of tobacco smoke indicates a much more seriousproblem of poorly designed and improperly maintained ventilationsystems. In this case ETS, as well as a number of other moresignificant indoor contaminants, are permitted to accumulate in theindoor space, and air quality deteriorates rapidly.

High levels of environmental tobacco smoke have been determined tobe the immediate cause of indoor air problems in only five, or 4% ofthe 125 major buildings investigated by ACVA since 1981.In asimilar study of 203 buildings from 1978-1983, NIOSH found that only4 of the buildings studied (2%) had indoor air quality problemsattributable to high concentrations of ETS. 3 Significantly, wherehigh accumulations of ETS have been found, ACVA has discoveredan excess of fungi and bacteria in the HVAC system. Thesemicroorganisms usually are found to be the primary causes of thecomplaints and acute adverse health effects reported by buildingoccupants.

ACVA determines the extent of bacterial and fungal contamination ofa building through diverse microbiological sampling techniques,classical air sampling in rooms, air sampling from ducts and surfacesampling. While we recognise that such sampling may sometimes bedifficult to interpret quantitatively due to factors such as surfaceirregularities, the discovery of significant concentrations of specificfungi and bacteria is significant. An excess number of thesemicrobial contaminants may be a precursor to future health problems.

Among the bacteria that we frequently have isolated from the HVACsystems of offices are: Staphylococcus aureus, Staphylococcusepidermidis, Coliform species, Dephtheroid baccilli, Eseudomonasspecies, Preteus species, Baccillus species and Serratia species.Eleven, or 9% of the buildings studied by ACVA have exhibited highlevels of potentially pathogenic or allergy causing bacteria. ACVAhas found over 27 different species of fungus contaminating airhandling systems.The most common isolates include Penicillinspecies, Nucor species, Aspergillus species, Alternatia species,Rhizopus species, Rhodotorola species, Candida species,Cladosporium species, Fusarium species and Cephalosporium species.Of the over 125 buildings studied by ACVA since 1981, 39, or 31%,have been found to contain high levels of potentially pathogenic orallergy causing fungi. In several buildings with excessive staffcomplaints, either Aspergillus species and/or Cladosporium species offungus were found. In some instances, colony counts of airbornefungi have run into the thousands per cubic meter of air.Epidemiological tests run by various doctors have confirmed severeallergic reactions to the spores of these fungi in all affected staff.Subsequent cleaning and removal of the sources of these fungalcontaminants have resulted in a complete abatement of complaints.In a similar situation, microbiologist Phil Morey, formerly of NIOSHand now associated with Honeywell, found fungi in a governmentbuilding in Washington with levels of up to 83,700 fungi per cubicmeter, characterizing the building's environment as "sort of like achicken coop . ,,4

fA

Dangerous amounts of fungi and bacteria were discovered in each ofthe five buildings found by ACVA to contain high levels ofenvironmental tobacco smoke. In each of these buildings, once theHVAC systems were cleaned and ventilation rates were modified, theindoor air quality improved dramatically. Tobacco smoking continuedto be permitted throughout the building without any furtherdiscomfort or adverse effects being reported.

HVAC systems also have been found to be poorly designed andnegligently maintained. Excessive dirt accumulations are common inductwork, even in hospitals. Following the inspection of a numberof buildings, hundreds of pounds of fungi, dust and dirt have beenremoved from the ductwork. Dead birds, insects, rodent carcassesand faeces and excess amounts of dust have been found in manybuildings where employees have complained of eye irritation,headaches, allergies and general respiratory problems. In addition,ACVA often discovers that the fresh air dampers of a building'sHVAC system have been closed completely in an effort to saveenergy. During the summer of 1985, the fresh air dampers, wereclosed completely in over 35% of those buildings studied by ACVA.One misguided engineer actually had bricked up the fresh air ventsto save energy. All of these buildings were operating with 100%recycled indoor air. The lack of an adequate fresh air supply,coupled with dangerously low ventilation rates, has led to hazardousventilation conditions in many of the buildings evaluated by ACVA.Fifty percent of the investigations conducted by NIOSH from 1971-1985 attributed the indoor air quality problems to inadequateventilation.5

The American Society of Heating, Refrigerating and Air ConditioningEngineers (ASHRAE) has suggested a minimum ventilation rate of 15cubic feet per minute (CFM) per person, but recommended a rate of15-25 cfm/person in general office spaces (ASHRAE Standard 62-73).As a result of the move to conserve energy in the mid-19701s,ASHRAE sought to revise its existing Standard Natural andMechanical Ventilation with a new standard describing "Ventilationfor Acceptable Indoor Air Quality" (ASHRAE 62-1981). An internalASHRAE Revision Committee is about to release a revised version ofASHRAE Standard 62-1981 incorporating a ventilation rate of 20cfm/person for general office spaces.

8

This is a direct recognition of the potential build-up andaccumulation of internal air contaminants permitted by low ventilationrates. The newly revised ASHRAE standard for ventilation isexpected to be released for public comment during the summer of1986; it contains higher ventilation rates for all indoor spaces, anacknowledgement that indoor air pollution problems often can becorrected simply through improved ventilation and increased airmovement.

The newly revised ASHRAE standard also will incorporate a minimumrecommended ventilation rate of 15 cfm/person when no other rate isspecified. This rate was adopted based primarily upon the need toreduce measurable levels of CO 2 to 1,000 ppm. Japan recentlyadopted this level of acceptable CO 2 concentration, and Canada isseriously considering adopting a rate of 600 ppm as a safety factor.Not surprisingly, a majority of the buildings investigated by ACVAhave had measurable ventilation rates far below 20 or even 15cfm/person.In fact, on many occasions, ACVA has foundventilation rates of 5 cfm/person or lower.

ACVA also conducts measurements to determine the extent ofairborne particulate contamination within a building. ACVA routinelymeasures airborne particulates with electronic light scatteringequipment as well as mass particle weight balances. ACVA hasfound that a major source of high indoor particulate levels is theintroduction of outside air into an indoor space. Frequently, the airis unfiltered, partially filtered, or filtered through grosslyinefficient filters; as a result, dusts and dirt accumulate in the airconditioning ductwork. Another source of contamination is theconstruction dirt and debris that accumulates in the ductwork duringbuilding construction. In fact, the largest single source of airbornecontamination in many buildings is this excessive dirt accumulationinside the air ducts.

9

Increased human activity also can produce high levels of measurableairborne particulates. Movement within an enclosed space can stirup paper dusts, textile fibers and loose fiberglass particles.Dermatitis due to fibrous glass insulation has been a commonproblem, usually after the fibrous glass insulation has beendisturbed during construction activity. Seven of the 125 buildings(5.6%) evaluated by ACVA since 1981 were found to have heavyconcentrations of fiberglass particles in the air.

Recently, during a study of a large bank, ACVA discovered afiltration efficiency of approximately 10% at a particle size of 1.0micron and larger. The average efficiency of a filtration system insimilar buildings ranges from 40 to 50% at this particle size, whereasknowledgeable and concerned building operators can achieve filtrationefficiency in excess of 80% at this particle range. The simplereplacement of grossly inefficient filters with adequate filters,followed by careful maintenance, often results in a dramaticreduction of airborne particulates. Tobacco smoke here acts as thesignal for the existence of a more significant indoor air qualityproblem, primarily poor ventilation, the lack of an efficient exhaustsystem and an inadequate supply of fresh air. An overemphasisupon the assumed particle contribution from environmental tobaccosmoke could lead a building manager to overlook the much morefundamental problem of a poorly designed and maintained HVACsystem. In the five instances in which we have observed highconcentrations of tobacco smoke, the smoke accumulation led thebuilding operator to contact ACVA, at which time an in-depthinvestigation revealed the presence of allergenic fungi that later wasdiscovered to be a major cause of the problems experienced by theoccupants.

A novel source of air contamination is related to the presence of anindoor parking garage in the basement of a building. High levels ofcarbon monoxide in offices have been traced to the exhaust fromautomobiles in parking garages below the building. This gas canpermeate a building through elevator shafts or stairwells. ACVAalso has discovered high amounts of carbon dioxide in manybuildings, as well as measurable levels of ozone and formaldehyde.Organic compounds are released into the air by building materials,adhesives, aerosols, paint, cosmetics, cleaning solvents, pesticidesand burning fuels.

10

Some of the examples of ACVA's findings with respect to specificbuildings may help to illustrate the general points that I haveattempted to make in this presentation:

1. Several years ago, four young infants died tragically at aDutch hospital, and many more had become seriously ill. Whilethe hospital was able to identify the cause of the infection, itcould not trace the source. After an ACVA investigation, twodead bird carcasses were discovered inside the air conditioningductwork. Bacteria from the putrified and decomposingcarcasses multiplied rapidly in the dirty and damp ductwork,and had been spread throughout the building by theventilation system.

2. After the main air handling units were replaced and highefficiency filters were installed in a large government buildingin Washington, ACVA was requested to confirm that the airquality was acceptable. ACVA discovered that theaccumulation of dust, bacteria and fungi in the originalductwork was re-contaminating the air supply from the newHVAC units.

3. In another instance, ACVA investigated complaints in a federalcourthouse. The air handling systems were clean, and thefilter systems were efficient. After extensive testing,however, ACVA discovered that textile fibers were being shedinto the air from the flocked wallpaper and were causinggeneral respiratory problems. The wallpaper was cleaned andsealed, and there were no further complaints.

4. Employees of a major bank were complaining of headaches,nausea and general respiratory problems. In that case, anelectrostatic precipitator - an ultra-efficient filter - had beenpermitted to run when the air supply system was shut down.Ozone generated by the filter was diffusing up the main airsupply ducts and accumulating in specific areas.

11

12

5. In 1983, physicians at the Health and Human Services ParkiawnBuilding in Rockville, Maryland requested ACVA to investigatecomplaints of sick building syndrome. ACVA discovered thatlow efficiency and poorly installed filters had permittedunfiltered air into office areas occupied by 5,000 - 7,000 HHSemployees. In addition, recycled air rates were lower thanpermitted by Federal regulations, and excess fungal growth inthe air handling system was discovered. Filters wereupgraded, the air handling systems were cleaned and stepswere taken to assure an adequate supply of fresh air. Airquality then improved substantially and complaints werevirtually eliminated.

6. Students at a medical college in Maryland were experiencingrespiratory problems. ACVA's investigation revealed excessiveaccumulations of allergenic fungi emanating from the airconditioning ductwork. The source of these fungi later wastraced to sawdust produced in the mechanical room. Thesimple removal of the sawdust and cleaning of the mechanicalroom resulted in improved air quality throughout the building.

7. In another investigation, ACVA discovered excessivecontamination by Cladosporium fungi within the HVAC systemand ceiling voids of a 19th century four-storey brownstonebuilding being used as a law office. A total lack of fresh air,cheap and poorly installed filters and excessive condensationfrom the air conditioning coils created the problem. Theventilation was increased, the system cleaned and sanitized andthe filters were upgraded.There have been no morecomplaints of poor indoor air quality in the building. Thesame problem was discovered in a Pennsylvania high schoolthat EPA had threatened to close due to high instances of staffand student sickness. ACVA took the same remedial stepsdescribed above, and the air quality within the school hasbeen improved dramatically. EPA no longer is threatening toforce the closure of the school.

13

8. In a state government health department, ACVA discoveredthat loose fiberglass insulation in the ceiling voids was beingcarried via the return air system to the air handling units.Poorly installed filters permitted the distribution of unfilteredair throughout the building. The problem was exacerbated byfungi in the condensate trays of the air handling units. Thefilters were upgraded, the air handling units were cleaned andthe ductwork was sanitized. Ninety-nine percent of thecomplaints have disappeared, and there has been a tremendousimprovement of the building's indoor air quality.

9. ACVA investigated one hospital in which the operating roomexhaust air ducts had become clogged from lint and textilefibers. Since no air could leave the room, no fresh air couldenter. Pictures taken inside the exhaust ducts revealed thatthe ducts had bowed inward from the negative pressurecreated by this situation. This operating room should havereceived 20 air changes per hour. The ductwork was cleanedand the problem was resolved.

10. Upon investigating the ventilation system of the headquartersof a large insurance company, ACVA discovered that anexhaust duct carrying paper dust from a computer printerinadvertently had been coupled to the air supply duct for anadjacent room. In addition, excessive fungal growth wasfound in the paper debris in the duct. The fungus later wasanalyzed as Aspergillus. Over 70% of the employees in theadjacent room had become senstized to the fungus andexhibited all the common symptoms upon exposure. ACVAdisconnected the exhaust duct, removed the paper dust anddisinfected the ductwork. There have been no more staffcomplaints.

14

11. ACVA recently completed one of the most interestinginvestigations conducted to date. A large bank requestedACVA to investigate complaints of excessive ETS in a two-yearold building. The building always had been plagued withsickness problems. Another company had conducted a studythrough the use of subjective occupant questionnaires.Following analysis of the questionnaires, the bank was advisedby that company to ban smoking throughout the building. Thebank then sought a second opinion from ACVA. ACVAdiscovered that the ventilation system was designed poorly,and that most of the occupants would receive less than 5cfm/person of fresh air even if the HVAC system were workingto its maximum design potential. ACVA found that the freshair dampers had been closed completely to save energy. Evenif fresh air had been permitted to enter, however, the systemwas so inadequate that there would have been no filtration ofthe fresh outside air.Further, the indoor air was beingrecycled through grossly inefficient filters.Substantialquantities of fungi were found in the ductwork, includingCladosporium species. The bank is now in the process ofcleaning and disinfecting the ductwork, as well as increasingthe ventilation and upgrading the filters.

CONCLUSION

ACVA's investigations have established that a multitude of factorscontribute to poor indoor air quality. The confounding problemsassociated with these factors make their analysis and subsequentresolution difficult. To concentrate upon one element of this matrixenvironmental tobacco smoke, would be inefficient and dangerous. Itwould be naive to assume that the removal or control of ETS, themost visible indoor air pollutant, would solve the indoor air pollutionproblems found in "sick buildings". That step ignores the manyhidden sources of indoor contamination, and can lead - at least inthe short run - to a false sense of security.

ACVA's work and experience, and evaluations conducted by NIOSH,indicate that indoor air quality can be improved significantly withproper maintenance of and attention to the HVAC system. Assuringthe introduction of filtered outside air, maintaining effectivefiltration systems, ongoing monitoring of air quality, cleaning andrepairing ductwork and improving the designs of ventilation systemsare necessary steps for the achievement of acceptable indoor airquality. High concentrations of environmental tobacco smoke are asymptom of the much more dangerous problem of poorly maintainedand designed ventilation systems. Persistent indoor air qualitycomplaints can be resolved only if building managers and operatorsare prepared to focus on those systems in a appropriate manner.

Without positive action to deal with the entire spectrum of indoorpollution, there is a grave danger that one potential irritant maytake the blame for all others. In our experience, these problems arecaused by a combination of factors. In practice, misguided attackson environmental tobacco smoke actually may eliminate the one telltalemarker we now have alerting us to the operation of an inadequateventilation system. In this sense, a fitting epitaph on the tomb of afuture Legionnaires' disease victim might be:

"He lived in a smoke-free environment."

15

REFERENCES

1. Melius, James; Wallingford, Kenneth; Keenlyside, Richard; andCarpenter, James; "Indoor Air Quality - The NIOSHExperience," Ann. Am. Conf. Gov . Ind Hyg., Vol. 10 (1984)

2. "Indoor Air Quality - The NIOSH Experience," supra at 4.

3. "Indoor Air Quality - The NIOSH Experience," supra at 4; seealso Wallingford, K.M.; "NIOSH Indoor Air QualityInvestigations: 1971 through 1985," (representing more than350 building investigations).

4. "Air Pollution in the Great Indoors," State Legislatures,July/August 1985, at 35.

5. "NIOSH Indoor Air Quality Investigations: 1971 through 1985,"supra at 4.

16

ENVIRONMENTAL TOBACCO SMOKEA REVIEW OF THE LITERATURE

OCTOBER 1988

TABLE OF CONTENTS

INTRODUCTION .

ENVIRONMENTAL TOBACCO SMOKE AND HEALTH CLAIMS .............AnOverview ..........................................LungCancer .....................................

Lung Function/Disease: Adults .................."Compromised" Individuals ..................

HeartDisease ..................................."Compromised" Individuals ..................

Lung Function/Disease: Children ................Allergy .........................................Sick Building Syndrome ..........................

ENVIRONMENTAL TOBACCO SMOKE AND INDOOR AIR QUALITY ........

SidestreainSmoke .....................................Constituents of ETS ..................................

Carbon Monoxide .................................Particulates .....................................Nicotine ........................................Nitrosamines ....................................OtherConstituents ..............................Cotinine as a Marker for ETS Exposure ...........Summary .........................................

ENVIRONMENTAL TOBACCO SMOKE AND THE PUBLIC ................Government Regulations and Individual Rights .........What is the Solution' ................................

REFERENCES ................................................

1

5

59

21252728303436

39

394242444649505152

53

5360

61

INTRODUCTION

The claim that exposure to environmental tobacco smoke

(ETS) is harmful to the health of nonsmokers has generated a great

deal of public concern. Although a review of the scientific

literature indicates that this claim has not been proved, it is

used by antismokers in their efforts to make smoking socially un-

acceptable and, thereby, to create a "smoke-free" society. The

ETS/health issue is so critical to the antismoking movement that

the Director of the U.S. Office of Smoking and Health has remarked

that "of all the issues, this [ETS] is the one that will propel

the United States toward a smoke-free society.111

According to one sociologist, the program for a

"smoke-free" society may be characterized as follows:

The strategy, quite overtly, is toprogressively stigmatize smoking,segregating the smoker in all publicplaces, and eventually to eliminatesmoking as a socially acceptable custom.How is this to be made politicallypalatable? The answer is clear: bysuggesting that smoking harms not onlythe smoker, but various categories of"innocent bystanders. ..2

The questionable and often contradictory nature of

studies concerning ETS exposure and health is largely ignored by

the antismoker; the goal is to achieve publicity and arouse

emotion. This strategy is typified by a statement from a commen-

tator on the proceedings of the Fifth World Conference on Smoking

- 1 -

and Health: "Regardless of the ultimate validity of the findings,

which remains to be established, studies such as these have brought

the issue of passive smoking to the public's attention."3

In addition, the issue of nonsmoker exposure to ETS

has become the means by which antismokers seek to enact legislation

regulating smoking in the workplace and other public places. Here

again, the strategy of using the ETS/health argument to modify

smoking behavior is apparent. For example, the chairperson of a

U.S. National Academy of Sciences committee which authored a recent

report on ETS remarked that "legislation could also be justified,

not because it will have much effect on the occurrence of lung

cancer among nonsmokers, but because it may motivate some people

either to stop smoking or never to start." 4 Indeed, one reviewer

observed that there are even instances in the scientific literature

in which calls are made to establish ETS as harmful "as a means to

limit smoking."5

However, the tactic . of using the ETS/health argument

to modify smoking behavior has been criticized in the international

press. One columnist for The (London) Times noted that while she

agreed in principle with the antismoking movement, she could not

condone "the lying that accompanies it." She explained: "The lie

is the claim that the health hazards of second-hand smoke have

been scientifically established." 6 Similarly, a writer for a

German publication called the use of the ETS/health argument an

- 2 -

"unethical" tactic intended to instill "anxiety in the nonsmoker"

and "feelings of guilt in the smoker."7

Certainly, ignorance of scientific procedures and the

inability to recognize faulty research methodology play a role in

the controversy on ETS. As one reviewer of the literature wrote

in 1987:8

A pluralistic society such as ours haslobbies. And it is they who have removedthe subject of passive smoking from thescientific to the political stage. Itwould be wrong to input this to badintentions. Inadequate or biassedoccupation with the problem or generalinsecurity with regard to scientificcomprehension, are quite enough.

Indeed, in another recent review article, a German

scientist, Dr. K. Uberla, cautioned that the argument that ETS

causes lung cancer in nonsmokers may "have serious negative effects

on the credibility of epidemiology" and may become an "obstacle"

to the antismokers' "goal" of the reduction of active sm6king.9

Dr. Uberla based his observation upon a comprehensive review of

the studies to date on ETS and lung cancer. He concluded that, at

best, the data for establishing an association are "inadequate"

and that, consequently, "a causal connection cannot be inferred."

Given the importance of the ETS issue for the antismok-

ing movement, the purpose of this paper is to review the current

scientific literature regarding environmental tobacco smoke and

the health of nonsmokers. It is intended to identify sources of

- 3 -

information and misinformation about ETS in an effort to demon-

strate that the weight of the scientific evidence does not justify

the claim that ETS causes disease in nonsmokers.

-4-

'1

ENVIRONMENTAL TOBACCO SMOKE AND HEALTH CLAIMS

An Overview

Antismokers typically refer to several highly publi-

cized studies as "proof" that tobacco smoke is harmful to non-

smokers. The studies most frequently cited are those by Hirayama10

and Trichopoulos, et al. (1981)11, who reported that nonsmokers

exposed to tobacco smoke have a higher risk of developing lung

cancer, and White and Froeb (1980) 12 , who asserted that nonsmokers

with long-term exposure to tobacco smoke in the workplace have

significantly reduced small-airways function. Still other reports

have suggested that children and individuals with lung and heart

disease are adversely affected by exposure to environmental tobacco

smoke.

In addition, government health officials have issued

reports which antismokers are using to pressure legislative bodies

in the promotion of a "smoke-free" society. In 1986, two such

reports in the U.S., one issued by a National Academy of Sciences

(NAS) committee to the U.S. Environmental Protection Agency 13 , and

the other by the Office of the Surgeon Genera1 14 , concluded that

ETS exposure is harmful to nonsmokers. More recently, the

Independent Scientific Committee on Smoking and Health in the United

Kingdom issued its report with conclusions similar to those of the

U.S. Surgeon General.15

- 5 -

However, critical analyses of these studies and reports

by physicians and scientists from around the world, together with

other studies which reported finding no association between ETS

and human disease, show that such adverse health claims are highly

suspect and not justified on a scientific basis. For example, in

1983, under the sponsorship of the U.S. Tobacco Institute, an inter-

national group of scientists and physicians met at the University

of Geneva to discuss scientific developments relating to ETS. The

organizer of the symposium summarized their discussions as follows:

An overall evaluation based upon availablescientific data leads to the conclusionthat an increased risk (of lung cancer]for nonsmokers from ETS exposure has notbeen established. 16

In May of 1983, scientists and epidemiologists who

attended a workshop on "Respiratory Effects of Involuntary Smoke

Exposure: Epidemiologic Studies" sponsored by the U.S. National

Institutes of Health concluded:

A review of the data from the studieswhich have been carried out or are inprogress which address the effect ofpassive smoking on the respiratory systemsuggest that the effect varies fromnegligible to quite small.17

Similarly, organizers of a symposium on the "Medical

Perspectives on Passive Smoking" held in Vienna in April, 1984 and

co-sponsored by the German and Austrian societies for industrial

- 6 -

medicine in cooperation with the International Green Cross,

concluded:

Should lawmakers wish to take legislativemeasures with regard to environmentaltobacco smoke, they will, for the present,not be able to base their efforts on ademonstrated health hazard fromenvironmental tobacco smoke.18

Indeed, in the preface to the 1986 reprint of those

proceedings published by the International Green Cross, Dr. Joseph

Handler, a former Director of the World Health Organization (WHO),

observed that "the question as to whether the conceivable

theoretical possibility of a risk [from ETS] calls for preventive

measures is not a medical but rather a politico-social one.1119

In October, 1986, an international gathering of scien-

tists in Essen, Germany, considered the experimental and

toxicological findings regarding ETS. 20 Professor J.G. Gostomzyk,

director of the Health Bureau in Augsburg, Germany, concluded in

his review of the proceedings that "so far, even toxicology has

not been able to ascertain with any greater degree of probability

than did epidemiology that there exists a link between damage to

health and passive smoking."21

At the conclusion of the International Conference on

Indoor Air Quality held in Tokyo, Japan, in November 1987, the

chairman of the organizing committee called for more research on

the ETS issue, while cautioning against the use of restrictive

measures prior to establishing an adequate scientific basis.22

- 7 -

In May of 1988, organizers of a symposium on ETS in

Austria entitled "Illness Due to Passive Smoking?" issued a press

release in which they concluded that "a causal relationship between

ETS and illness cannot be established" and that "there is no

positive evidence that cancer and other such illnesses are caused

by passive smoking."23

Other scientists recently have reached similar con-

clusions. Dr. Paul Witorsch, Professor of Medicine at George

Washington University in Washington, D.C., concluded that "a

thorough and critical examination of the relevant literature fails

to provide compelling evidence that exposure to ambient tobacco

smoke produces adverse chronic health effects." 24 Similarly,

Dr. Domingo Aviado, a physician and former professor of

pharmacology, wrote that his review of the literature "indicates

that there is no substantial evidence to support the view that

exposure to environmental tobacco smoke presents a significant

health hazard to the nonsmoker. ,,25

Moreover, in a study published in 1986, researchers in

the United Kingdom who had investigated what impact ETS has on the

health of nonsmokers concluded: "Amongst lifelong non-smokers,

passive smoking was not associated with any significant increase

in risk of lung cancer, chronic bronchitis, ischaemic heart dis-

ease or stroke in any analysis."26

Given such findings, it appears remarkable that anti-

smokers continue to claim that exposure to environmental tobacco

- 8 -

smoke is a proven cause of disease in nonsmokers. Obviously, this

claim carries a great deal of emotional weight even without

appropriate scientific support. Therefore, the charges are seri-

ous enough to warrant close examination.

Lung Cancer

Two highly publicized articles published in 1981 are

still among the most frequently cited to support the claim that

ETS exposure increases the nonsmoker's risk of lung cancer. A

study of Japanese women by Dr. Takeshi Hirayama reported that

nonsmoking wives of heavy smokers have a much greater risk of

developing lung cancer than nonsmoking wives of nonsmokers. 10 In

a study of Greek women, Trichopoulos, et al., concluded that a

nonsmoking woman whose husband smokes has twice the risk of

developing lung cancer as a nonsmoking woman married - to a

nonsmoker.

Both studies have been widely criticized in the scien-

tific literature. Numerous inadequacies and inconsistencies in

the Hirayama study have been pointed out by noted scientists and

physicians who have questioned the validity of its conclu-

sions. 8 ' 2743 Dr. Ragnar Rylander, the organizer of the 1983 Uni-

versity of Geneva symposium on ETS, noted that the study had been

criticized for its lack of questionnaire reliability, absence of

histological diagnosis, questionable statistical treatment, and

-9--

failure to examine such factors as air pollution from heating and

cooking. 44 A participant at the International Conference on

Indoor Air Quality held in Tokyo in 1987 noted, in a published

report, that several recent critiques had questioned Hirayama's

study for bias in statistical processing, as well as for failures

to measure ETS exposure levels and to evaluate and report "dietary

and occupational factors known to influence susceptibility to lung

cancer." 27 Moreover, two studies presented at the Conference on

Indoor and Ambient Air Quality, held at Imperial College in London

in 1988, reported that Hirayama had not divided his study population

into appropriate age groups. The presenters noted that the risk

he reported for nonsmoking wives of smokers disappears when such

age bias is removed.28'33

Criticisms of the Trichopoulos, et al., study were ac-

knowledged by the authors in the 1983 update of the study. 45 They

noted that their research had been "criticized (by ourselves and

others) because of the small number of subjects, because several

tumors lacked histological confirmation, and because controls and

cases were from different hospitals."

Moreover, their findings were inconsistent with another

study published that year. In late 1981, Lawrence Garfinkel, an

official with the American Cancer Society (ACS), reported the

results from a follow-up of the group's long-term lifestyle study,

involving nearly 180,000 nonsmoking American women divided into

categories based on the amount their husbands smoked. 46By

- 10 -

comparing the lung cancer mortality rates of women reportedly

exposed to different levels of tobacco smoke, he determined that

none of the differences observed were statistically significant

and that "compared to nonsmoking women married to nonsmoking

husbands, nonsmokers married to smoking husbands showed very little,

if any, increased risk of lung cancer." Moreover, Garfinkel

cautioned that any study based solely upon classifications of the

husbands' smoking habits could not account for total exposure, and

that, therefore, the results could be misleading.(See also

Garfinkel et al., 1985.)

Several subsequent studies have claimed to support the

reported findings of Hirayama and Trichopoulos, et a l. 47-52,54

but a close examination of the methods and conclusions of these

studies reveals that their claims are scientifically unfounded.

For example, a report by Knoth, et al., which identified 39

nonsmoker lung cancer cases, asserted that exposure to ETS was the

"most plausible explanation" for the reported lung cancers.52

However, the study failed to provide a comparison group as a control

population. One reviewer consequently characterized the report as

containing "only tentative conclusions based on poor data analyzed

by unacceptable methods."53

In a 1983 study, Correa, et al., reported an increased

risk of lung cancer for nonsmokers married to smokers in a south-

ern U.S. state. 54 However, the results were based on only 30 cases

and did not take into account occupational, indoor and outdoor

- 11 -

13

exposures or other confounding variables. Furthermore, one of the

co-authors of this study later issued a report with contradictory

findings for a similar population in a neighboring state.55

In 1985, a published article by two antismoking advo-

cates, Repace and Lowrey, estimated that exposure to ETS is

responsible for 500 to 5,000 lung cancer deaths a year among

nonsmokers in the United States. 56 The risk estimates were cited

widely in the press prior to the report's publication, perhaps

because it was incorrectly identified as a study by the U.S.

Environmental Protection Agency (EPA). In reality, the report was

neither approved nor endorsed by that agency. Indeed, Repace now

faces a full-scale ethics inquiry into charges that he was paid

for his involvement in antismoking activities while he was an EPA

employee.57

The report itself presents two highly theoretical

models for estimating risks of lung cancer from ETS exposures.

The first model relies upon a "reinterpretation" of the

epidemiological studies of lung cancer in nonsmokers, and the second

provides an estimate for lung cancer mortality from a single study

dealing with the Seventh Day Adventists, a religious group known

for its vigorous opposition to smoking.

Critics of this approach have pointed out that the

estimates are based on errors and "unrealistic assumptions" which

result in overestimations of exposure. 58 ' 59 For example, one

recent analysis of their model showed that, depending upon the

- 12 -

assumptions and input data used, such model estimates are inherently

inaccurate and may vary by as much as 300-fold. 60 Another scientist

noted that the exposure and dose levels they used are not based on

actual measurements, and that actual measurements reported by other

researchers range from "ten-to-one-hundred-fold less than that in

the Repace and Lowrey model." 61 Still other scientists have

questioned the methods of analysis used in their report. 6264 They

suggest that Repace and Lowrey failed to control for other

confounding factors, and that their model did not provide "the

very statistical bases of estimation procedures." 61

Repace and Lowrey's estimate of nonsmoker lung cancer

risks in the workplace has been criticized by another group of

scientists who pointed out that none of the epidemiological studies

which examine the relationship between ETS exposure and disease in

nonsmoking working women reports a statistically significant

increase in risk.58

Additional studies have claimed that the risk for all

cancers is increased among nonsmokers exposed to ETS. One such

study examined a population which was selected from death notices

in a northeastern U.S. newspaper. 65 However, analysis of the data

reveals that the reported risk disappears when age comparisons of

cases and controls are considered.66

In a series of papers, Sandler, et al., reported in-

creased risks for lung cancer and cancer of other sites in non-

smokers exposed to ETS. 6769 Yet published criticisms cast seri-

- 13 -

ous doubt on these conclusions. One commentator challenged the

authors' selection of controls and their use of questionnaire

information. 70 Another scientist observed, that under one inter-

pretation, the data suggest that exposure to ETS had no effect on

overall cancer incidence. 772 A third critic called the studies

"seriously flawed" and questioned the authors' methods of data

collection. 73 Still others challenged their conclusions because

the reports had not considered potential confounding variables

such as diet, hygiene and lifestyle.7475

With regard to the issue of ETS and overall cancer

mortality, a 25-year follow-up study of over 1,000 married couples

in Holland reported that "passive smoking was not associated with

an increase in total mortality." 76 Although the authors cautioned

that the results do not necessarily contradict those which suggest

such an association, they added that their findings "were

reassuring to the extent that the possible absolute risk carried

by passive smoking is probably small."

The complex nature of the scientific data in this

area is further illustrated by the mixed results of some studies;

that is, certain data in a study may suggest a relationship be-

tween ETS and lung cancer while other data from the same study may

not. For example, in a report on the preliminary results of a

health survey conducted in two urban communities in Scotland,

Gillis, et al., noted that of the six lung cancer deaths observed

in male nonsmokers, four were among those who had reported expo-

- 14 -

sure to ETS. 77 However, no dose-response relationship was noted for

females in the study.

Kabat and Wynder, in a 1984 study of 25 male and 53

female nonsmoker lung cancer cases, reported a greater amount of

ETS exposure at work for male nonsmoking cases. 78 However, j

significant effect was reported for males exposed to ETS at home

or for female cases exposed either at work or at home.

In 1985, Garfinkel, et al., published a study of 134

nonsmoking women with lung cancer who were selected from hospital

records from 1971 to 1981. Garfinkel, et al., provided two

analyses of the same data set which produced apparently contra-

dictory results. While a statistically significant dose-response

relationship was reported between nonsmoking female lung cancer

patients and the number of cigarettes smoked per day by their

spouses, no significant relationship was observed between the

occurrence of lung cancer in nonsmoking women and the total number

of hours per day of exposure to ETS over the last five or 25 years.

Still other studies have observed no significant

association between ETS and lung cancer. These include a 1982

study from Hong Kong which reported fewer "passive smokers" among

lung cancer patients than among controls. 80 The authors noted

that "this finding is at variance with that of Dr. Hirayama's."

Similarly, Koo, et al., studied 200 female lung cancer

cases among Chinese women and concluded that ETS exposure in the

home or at work had no statistically significant impact on lung

- 15 -

cancer incidence. 81In a second study published in 1987, Koo

et al., calculated total ETS exposures at home and at work for their

population of nonsmoking female patients. They observed no

significant disease trends for any of the lifetime measurements of

exposure to ETS.82

In two papers published in 1988, Dr. Koo re-analyzed

her data set on nonsmoking wives of smokers. 83 ' 84 . Her results

indicated that wives of nonsmoking husbands had "healthier"

lifestyles than wives with smoking husbands; they exhibited higher

socio-economic status and had better indices of family cohesiveness

and lower frequencies of selected health problems and complaints.

An important and statistically significant difference was found in

the diets of the two groups. Wives with smoking husbands consumed

more processed and spicy foods and ate fewer fresh fruits and

vegetables than did wives of nonsmoking men. 1(00 concluded that

such correlates of smoking status "act as important confounders

when evaluating health risks among families with smoking husbands."

In 1985, Wu, et al., reported on data from a study of

female lung cancer cases in a large metropolitan area in the U.S.85

They reported that the lung cancer risk among the 31 nonsmokers in

the study population was not affected by ETS exposure.

Researchers have recently published results of a study

undertaken to investigate the occurrence of lung cancer in an

industrial area in Scotland. 86 The authors reported that ETS

exposures were not significantly different between the cases and

- 16 -

controls in the study; in other words, ETS appeared to play no

role in the reported increased incidence of lung cancer in the

study population.

In 1986, researchers investigated the potential role

of ETS on the incidence of nonsmoker lung cancer in England. 26 The

researchers observed no significant trends in increased risk for

lung cancer patients who reported exposure to ETS. In fact, in

some disease categories, more controls than cases reported exposure

to ETS.

Perhaps as a result of the failure of individual

studies to report consistently significant results, certain

researchers have recently begun using another method of statis-

tical analysis which combines the reported results from numerous

studies. With this method, called "meta-analysis", .they claim to

have calculated estimated excess risks for nonsmokers exposed to

ETS that are 10 to 50 percent greater than those for nonexposed

nonsmokers. In basic terms, they have reached these broad con-

clusions or "generalizations" by calculating an average of the

relative risks for nonsmoker lung cancer cases reported by those

studies.15'8789

However, this method of generalization has been crit-

icized for its questionable application to those studies, due to

the wide variety in their study designs, population selections,

techniques of analysis and their results. 9092 This method also

ignores basic inethodologic weaknesses which, according to one recent

- 17 -

governmental report, are characteristic of each published study on

ETS and lung cancer. 93 It further ignores the fact that the

majority of studies considered in the various "meta-analyses" do

not report statistically significant increases in lung cancer risks

for nonsmokers. For example, only two of the thirteen studies

considered by the 1986 NAS Report achieved statistical

significance. 94 (Although the authors of several studies report

no statistically significant risks, they report "positive" trends

in their data. However, trends which fail the test for statis-

tical significance are of questionable scientific validity).

In a recent analysis, Letzel and co-workers considered

only those epidemiological studies on ETS which met minimum

requirements of scientific quality. The researchers then performed

their own meta-analysis on those studies with corrections for

misclassification (see discussion below). They concluded that the

computed excess risk was negligible and not statistically

significant.92

Peter Lee, a statistician and epidemiologist from the

United Kingdom, has argued that the increased risks reported in

various "meta-analyses" are the result of an inherent bias in study

design rather than the result of any genuine effect from exposure

to ETS. 95100 Lee . presents data which indicate that the reported

risks cannot be explained on the basis of either ETS exposure or

dose for the nonsmoker. It is Lee's contention that the reported

- 18 -

"risks" are the result of bias caused by a small number of smokers

who are misreported in the studies as nonsmokers.

Other kinds of misclassification may contribute to the

reported increase in lung cancer risks among nonsmokers, according

to several scientists. For example, none of the studies on ETS

and lung cancer provides direct observational information on ETS

exposures. Instead, spouses, next-of-kin or friends are asked to

estimate the amount of ETS to which they think the subject was

exposed. Such estimates may lead to a kind of misclassification,

called exposure misclassification, 101 which has been shown by

Garfinkel 79 and Friedman102 and others 10306 to -lead to improper

indices of exposure and incorrect estimations of risk. In

Garfinkel's study, for example, relative risks varied from 0.83

and 0.77 when the women with lung cancer or the husband was the

respondent, to a risk of 3.57 when a son or daughter responded.91

Dr. S. James Kilpatrick, a biostatistician from the

Medical College of Virginia, has analyzed another form of mis-

classification, called differential misclassification, which

results "from-the tendency of respondents to inflate the amount' of

ETS exposure for lung cancer cases and deflate the report of

exposure for controls. 11101 Similarly, Dr. Ernst Wynder, President

of the American Health Foundation, notes that "relatives of' a

nonsmoking lung cancer patient are more likely to report passive

inhalation exposure on the part of their relative than are rela-

tives of a control patient."107

- 19 -

A more subtle form of potential bias is known as

"publication bias", which is generated by the failure to publish

studies which report negative or weakly positive results.108'109

Scientists have recently expressed concern over the growing trend

in the literature which tends to overemphasize (and hence publish)

only those studies which report increases in risk. 110 , 3]l Published

studies which are thus combined for meta-analyses may not truly

represent all investigations on the topic of ETS exposures and

lung cancer.

Most of the epidemiological studies on ETS and lung

cancer have failed to consider age differences, diet, occupation

and exposures to indoor or outdoor pollution as potential

confounding elements. The importance of such factors is

underscored by four recently published reports, one, from Japan

and three from China. 2115 The reports suggest that indoor

pollution generated by kerosene heaters, coal stoves, liquified

petroleum gas and exposures to cooking oil vapors may be responsi-

ble for the increased risk of lung cancer among Oriental women.

In addition, relatively few studies have considered ETS exposures

outside the home; only four studies sought information about total

exposure to ETS from various sources. 26,79,81,82 Interestingly,

none of these studies report a significant association between

total ETS exposure and lung cancer.

Given these difficulties in interpretation, it is

therefore not surprising that an eminent statistician should

- 20 -

conclude that "it is unlikely that any epidemiological study has

been, or can be, conducted which could permit establishing that

the risk of lung cancer has been raised by passive smoking.

Whether or not the risk is raised remains to be taken as a matter

of faith according to one's choice."108

Lung Function/Disease: Adults

The issue of ETS and respiratory disease in nonsmokers

was raised in a 1980 study by White and Froeb.- 2 They measured

the small-airways function of smokers and nonsmokers and concluded

that nonsmokers exposed to tobacco smoke at work for 20 or more

years had reduced function of the small airways compared to

nonsmokers not so exposed. Similarly, in 1983, French researchers

reported that nonsmoking spouses over 40 years of age who were

married to smokers exhibited decreases in pulmonary function

compared to nonsmoking spouses of nonsmokers.116

The White and Froeb study has been criticized for a

number of reasons. For example, a physician at a U.S. medi-

cal school questioned the authors' use of carbon monoxide as an

index of smoke exposure, contending that they "do not have reli-

able estimates of the smoke exposure in the environment of their

nonsmokers" because it is not unique to tobacco smoke.120

Based upon his own analysis of the White and Froeb

study, a British reviewer concluded that the authors' findings

- 21 -

"relate to an index which is contentious and certainly not an

accepted reliable indicator of an increased health risk." 34 White

and Froeb themselves noted that the average values of the pulmo-

nary tests of nonsmokers exposed to tobacco smoke "were not notably

different" from the values suggested as normal by a specialist in

this area.121

Perhaps the most telling criticism of the study was

voiced by Dr. Michael Lebowitz of the University of Arizona at an

annual joint meeting of the American Lung Association/American

Thoracic Society and in a subsequent letter published in the U.S.

Congressional Record. 122123 During a forum at the ALA meeting,

Dr. Lebowitz stated that he had concluded, from his own extensive

review of the study and from an interview with White, that the

study was "improperly designed" from an epidemiological point of

view. He said that there were problems "inherent in the study,"

including the selection of the study group and the proper mea-

surement of smoke in the workplace. Dr. Lebowitz also expressed

concern that the statistical significance of the data appeared to

depend on the unexplained omission of data for 3,000 people who

were originally included in the study. Based upon these con-

siderations, Dr. Lebowitz urged that the study not be used to

support the claim that environmental tobacco smoke affects the

lung function of adults in the workplace.

In a paper presented to the 1984 Vienna Symposium,

- 22 -

Dr. Lebowitz again took issue with the White and Froeb study. He

wrote:

Even with a biased population, poor studydesign, and incorrect statisticalevaluation, there were no clear-cut,consistent, medically meaningfuldifferences between passive smokers andgroups of nonsmokers; a correctedstatistical analysis strengthened thisconclusion. 124

Dr. Lebowitz also questioned the finding of the 1983

French study which reportedly showed significant differences in

lung function in one part of the study population, although

significant differences between exposed and nonexposed nonsmokers

were reported in the population as a whole. Dr. Lebowitz noted

that since the"healthiest" population in the study lived in the

most polluted areas, the study may have been flawed due to biased

population selection and testing or other confounding factors.124

In addition, neither the White and Froeb nor the French

study is consistent with other research on lung disease and lung

function in nonsmokers. For example, a 1984 study of 1,351 German

office workers reportedly found no effect on pulmonary function

among exposed nonsmokers. 125 In a 1988 update of the study, the

investigators noted that "there is no evidence that average

everyday passive smoke exposure in the office or at home leads to

an essential reduction of lung function in healthy adults."126

Canadian researchers reported that a group of healthy young adult

nonsmokers showed "no consistent reaction" of static lung volumes

- 23 -

and insignificant changes of other lung function measurements when

exposed to tobacco smoke.127

Numerous studies of individuals exposed to ETS in the

home also present conclusions which are contrary to both the Whiteand Froeb and the French study. In 1981, several epidemiologists

at the Johns Hopkins University School of Medicine reported that

in a group of 1,724 adults, the frequency of respiratory symptoms

in nonsmokers was not associated with the number of smokers in

the household. 128 Nor was the frequency of impaired ventilatory

function significantly higher if there were smokers in the home.However, their analysis did show that "among men who never smoked

cigarettes, gas cooking was definitely associated with impaired

ventilatory function, even when corrected for multiple comparisons."

Other studies on pulmonary function, respiratory dis-

ease and environmental tobacco smoke deserve special mention. In

1983, for example, Jones, et al., reported that in a case-control

study of several hundred nonsmoking women from a U.S. study popu-

lation, there was no significant association between decreases in

lung function and exposure to ETS in the home. 129 Coordinators of

an epidemiologic study of obstructive lung disease in Arizona

reported finding no effects from ETS exposures in their adult study

population. 130

In a study of 376 families, Yale University scientists

reportedly found no evidence that environmental tobacco smoke

affected either lung function or symptoms in adults. 131 More

- 24 -

recently, in a Canadian study on the effect of indoor and outdoor

pollutants on the lung function of housewives, researchers con-

cluded that ETS exposure did not influence lung function but that

gas stove usage played a role in average lung function decline.132

It is not surprising, therefore, that the participants

at the U.S. National Institutes of Health workshop on ETS exposure

concluded, after a review of the data from on-going studies on

lung function and disease, that a possible effect from environ-

mental tobacco smoke "varies from negligible to quite small."17

"Compromised" Individuals: Lung Disease

Many believe asthmatics to be particularly vulnerable

to various environmental influences, including ETS. A 1981 study

by Dahms, et al., for example, reported decreases in the pulmonary

function of several asthmatics exposed to environmental tobacco

smoke. 133 However, the study suffers from several obvious

limitations. The subjects, for example, were challenged under the

unrealistic experimental conditions of exposure to high levels of

smoke in an enclosed smog chamber. In addition, as the authors

themselves noted, their experiment lacked proper controls and the

effects observed may have been due to psychological, not physical,

factors.

Contrary to the reported findings of Dahins, et al., a

Canadian group who examined the reactions of asthmatics to levels

- 25 -

of tobacco smoke typically found in public places observed no

systematic lung changes after such exposure. 134 Later research by

the co-authors of this study supports this conclusion. They re-

ported that respiratory data collected from a group of asthmatic

volunteers exposed to tobacco smoke "do not suggest that asthmatic

subjects have an unusual sensitivity" to such exposure.135

Although several volunteers claimed that they experienced wheezing

and tightness in the chest due to the exposure, the researchers

found that the "physiological data give little support to the

concept of a subgroup with particular sensitivity." They noted

that these reactions probably were due to the "suggestibility" of

the subjects.

The difficulty in determining the impact of psycho-

logical responses in such studies is clearly demonstrated by the

results of two reports from Australia. Although a 1985 report

suggests that short-term. ETS exposures are capable of inducing

reactions in asthmatics, 136 an earlier study tends to support the

theory that psychological reactions may partially explain symptoms

observed during such exposures. In the latter study, Ing, et al.,

reported that while asthmatics exposed to tobacco smoke complained

of subjective symptoms, no significant objective evidence of airways

obstruction was observed.137

Another report on results from a large-scale epidemio-

logical study in the U.S. suggests that environmental tobacco smoke

in the home does not affect symptoms or pulmonary function in either

- 26 -

children or adult asthmatics, although dust and pollen in the home

apparently can provoke such effects.138

A group of researchers from Yale University, in a

recently published study on this issue, studied the short-term

effects of ETS exposure on a group of young asthmatic patients.139

They observed no changes in lung flow rates and concluded that

such exposures presentacute respiratory risk" to asthmatics.

Heart Disease

Epidemiologic studies assessing ETS and the risk of

cardiovascular disease which appear in the literature show no clear

trends. Although two studies reported a small, but statistically

significant, increase in heart disease risk among nonsmoking wives

of smokers, 140 ' 14l another generated both statistically significant

and insignificant results. 142 A fourth study reported a

statistically insignificant increased risk, 143 and two additional

reported no significant association between heart disease and ETS

exposures.26'77

In 1984, two German researchers reviewed the available

studies on the roles of carbon monoxide (CO), nicotine and other

tobacco smoke components in the possible etiology of coronary heart

disease in nonsmokers. The authors Concluded that "there is little

evidence" to suggest that substances found in ETS "may adversely

affect the cardiovascular system." They also stated that "neither

- 27 -

Co nor nicotine is likely to play a role in the development and

progression of coronary heart disease in those concentrations

normally found in passive smokers." 144 In addition, scientists

and physicians who examined the issue of carbon monoxide at the

1983 Geneva Symposium on ETS agreed that "carbon monoxide from

environmental tobacco smoke is not important from a health point

of view."16

"Compromised" Individuals: Heart Disease

The issue of ETS exposure for individuals with pre-

existing heart disease arises primarily from a 1978 study by Aronow

who concluded, on the basis of examining ten heart patients, that

such exposure hastens the onset of heart pain during exercise.145

However, the Aronow experiment has been repeatedly and

severely criticized by a number of authorities, including the U.S.

Surgeon General. 146 Dr. Edwin Fisher, a professor of pathology at

the University of Pittsburgh, stated that the study must be

"evaluated in light of the fact that the end-point of the study

was highly subjective, that the stress factor was not controlled,

and that a sham smoke or other environmental impingement was not

used. In other words, not only was the sample small, but the

scientific design was exceedingly poor. ,,147

After reviewing the Aronow study, a Canadian professor

of medicine who otherwise maintains that tobacco smoke exposure

- 28 -

may be harmful wrote that although the patients showed some in-

crease in carboxyhemoglobin (COHb) or CO blood levels, "the end-

point (the reported onset of anginal pain) is necessarily subjec-

t ive." 148 He explained, "it is difficult to imagine that enclo-

sure in a very smoky room did not have some emotional impact upon

patients who were liable to angina, and the psychological dis-

turbance may have done more to hasten the onset of symptoms than

the increase of blood COHb."

In addition, other researchers in the area have been

unable to replicate or duplicate Aronow's reported findings. For

example, researchers from Great Britain measured oblective

responses of a group of heart patients to ETS and found no adverse

physical response to any level of exposure.149

Other serious questions have been raised regarding the

underlying validity of Aronow's work. In 1983, after one U.S.

governmental agency reported that Aronow had falsified data in a

study it had sponsored, another agency, the Environmental Protec-

tion Agency (EPA), conducted an independent review of several of

Aronow's studies, including his work on heart disease and CO. The

EPA concluded that, because of several irregularities in the study,

it could no longer rely on his research to formulate environmental

standards.150151

- 29 -

Lung Function/Disease: Children

Perhaps no claim regarding ETS is as capable of provok-

ing strong feelings as the charge that parents who smoke may harm

the health of their children. While the issue of parental smoking

is laden with emotion, the scientific basis for the claim is

difficult to • interpret. For example, while one study examines

respiratory symptoms or illness such as coughs and colds by ques-

tionnaire responses from parents, 152 another measures lung func-

tion with special equipment at a school or health faci1ity. 153 In

the United States alone, according to one report, this has led, to

a situation in which studies of ETS and the respiratory system are

being "carried out by at least three different groups, are

employing different populations and methodologies and have led to

varying conclusions. i,l7

Such studies, each with a different sample size, data

collection method and analysis, tend to yield factually incom-

patible and contrary conclusions. For instance, although certain

studies have reported adverse findings, 13 ' 14 ' 154170 others have

observed no significant relationship between parental smoking and

respiratory illness in children. 131,152,171-181 After a five-year

study of over 400 children, for example, Dutch researchers

concluded there was "no evidence" that parental smoking had an

appreciable effect on respiratory symptoms in school children.176

A similar conclusion was reached by a group of U.S. researchers,

- 30 -

including a critic of smoking, who found "no significant relation"

between parental smoking and respiratory symptoms in a study of

nearly 400 families with 816 children in three cities.131

Moreover, the authors of studies reporting adverse

effects from ETS exposures among children concede that their

conclusions must be viewed with caution because of numerous con-

founding factors. For example, one group of British researchers

acknowledged the possible influence of factors such as cross-

infection in the home and genetic susceptibility to childhood

respiratory illness and symptoms. 158,159 More recently, researchers

in Hong Kong reported "a highly significant correlation between the

frequency of respiratory illnesses" of mothers and their

children. 182

Others have conceded that the reliance on

questionnaires for information about respiratory symptoms casts

doubt on the findings. In one study which reported a significant

association between parental smoking and respiratory symptoms, for

example, it was noted that even "slight changes" in the way the

questions were phrased could result "in substantial differences in

the type of responses one obtains." 160 Similarly, another study

observed that there was a significant difference in the respiratory

symptoms reported depending on which parent completed the

questionnaire. 168

The importance of such confounding factors was given

special consideration in the report from the workshop on ETS

- 31 -

sponsored by the U.S. National Institutes of HealthJ 7 Among the

many factors listed in the report were types of heating used,

socio-economic status and other variables affecting household

conditions, including the number of residents, and demographic and

medical characteristics of the study population, such as age,

parental symptoms and annoyance responses. The report cautioned

"that any study which ignores them will be seriously flawed." The

relevance of such factors in affecting the outcome of research on

parental smoking is supported by a number of reports which have

shown that the use of gas stoves in the home may be independently

associated with childhood respiratory diseasè.138'174'18387

Other confounding factors independent of parental smok-

ing have been reported recently in the medical literature. For

example, studies in the United Kingdom have identified damp hous-

ing and father's occupation189 as potential explanatory

factors for the occurrence of respiratory illness in children.

Other recent studies have identified outdoor air pollution,190'191

infections transmitted during day-care attendance 192 and the use

of kerosene heaters in the home 193 as risk factors for childhood

respiratory disease.

The contradictory nature of findings on the issue of

parental smoking is also apparent in the growing number of studies

examining the relationship between parental smoking and children's

respiratory or lung function. Although several reports have

claimed that parental smoking results in decreased pulmonary lung

- 32 -

function in children, 1-56,l6O,]70,194200 others have not,131'201

204 including those of a group of researchers who have published a

series of studies on this subject. 130,138,153,205 In 1982, for

example, the latter research group showed that a comparison of

body size with lung function eliminated any reported correlation

between parental smoking and children's lung function. 153 Two

years later, a reanalysis of families from their study group again

showed that "parental smoking did not have a significant effect on

children's pulmonary function; smoking habits of others in the

household (predominantly siblings) did not have any effect

either. ,,l30

Thus, claims that parental smoking plays a causal role

in the development of respiratory symptoms and reduced lung

function in children are not scientifically justified. Such claims

are typically based upon a single study of a selected symptom (such

as cough or wheeze). These kinds of studies invariably fail to

consider nutrition, health habits of the family or other lifestyle

variables. Similarly, studies which report reduced lung function

in children of smoking parents fail to address the issue of socio-

economic status or the potential role of genetic and family traits

in pulmonary function capabilities. 204 Moreover, the reductions

reported in the literature are small and of uncertain clinical or

biological significance, and are contradicted by a number of studies

which reportedly have observed no effect of parental smoking on

children's lung function.

- 33 -

Allergy

Although some individuals are annoyed by the sight and

smell of tobacco smoke and a few even report experiencing irrita-

tion, the existence of human allergens in tobacco smoke has

been established scientifically. This is an extremely important

point in the ETS debate because those seeking to ban smoking in

public places often do so by claiming that nonsmokers are "allergic"

to tobacco smoke.

However, a number of research groups have been unable

to conclude that a tobacco smoke allergy in humans exists.206209

In 1980, for example, a group of researchers noted that "direct

evidence that tobacco smoke is immunogenic (capable of evoking a

specific response) in man is yet to be documented." 21° A recent

report by this same group affirms this conclusion.211

Claims about tobacco allergy stem primarily from

studies in which tobacco leaf extract has been reported to cause

allergic skin responses in some people, usually in those who already

experience allergic reactions to other substances such as

weeds. 212214 However, as an English immunologist pointed out,

there are "great difficulties" in determining whether positive

reactions to tobacco leaf extracts are relevant to clinical re-

sponses to tobacco smoke. 209 Although he noted that there may be

substances in tobacco smoke which could "theoretically" act as

- 34 -

such agents, he concluded that "there is no proof that the spe-

cific sensitization to tobacco smoke exists." Recent research

affirms this point. Scientists reported in 1988 that tobacco leaf

sensitivity was not associated with decreased pulmonary function

in allergic asthmatics.215

It has also been hypothesized that tobacco smoke is

capable of provoking asthma as an allergic reaction.216217

However, a Swedish specialist concluded that such results are not

proof of a tobacco allergy because the studies, which used tobacco

extracts, did not differentiate between non-specific and true al-

lergic reactions in evaluating the results of skin tests and

bronchial provocation. 218 Consequently, he stated, "for the

present, the question as to whether allergy to cigarette smoke

exists or not should be kept open."

Certainly, there appear to be people who may be sen-

sitive to tobacco smoke, but personal annoyance and emotional

reactions should not be confused with genuine allergic reactions.

In many cases, the individual may be responding to high

room temperatures, lack of ventilation, or even to the mere sight

of tobacco smoke. Indeed, Dr. John Salvaggio, the director of an

allergic disease center in the U.S., has suggested that reported

reactions to tobacco smoke may be irritative rather than allergic.

After reviewing the studies on the allergy question, he concluded

that "there is no proof that tobacco smoke is allergenic in

man."219

- 35 -

Accordingly, it is not surprising that researchers at

the Mayo Clinic failed to find any evidence of tobacco smoke

allergies in their tests of subjects who considered themselves

allergic to tobacco and tobacco smoke. 220 It has been pointed out

in other studies that people who claim to be "smoke sensitive" did

not react more frequently to tobacco leaf or smoke extract than

those who are "smoke resistant."221

On the basis of these data, it must be questioned

what people really mean when they say they are "allergic" to

tobacco smoke. They may simply not like the sight and smell of

tobacco smoke and interpret this to mean that they are "allergic"

to environmental tobacco smoke. But, as has been pointed out,

such personal reactions should not be regarded as true tobacco

smoke allergies.

Sick Building Syndrome

Advocates of smoking restrictions in the workplace

commonly argue that ETS exposure gives rise to a number of corn-

plaints, including headaches, nausea, coughs, sore eyes and

breathing difficulties. However, recent research indicates that

this complex pattern of symptoms, the so-called "sick building

syndrome", commonly occurs in modern office buildings whether or

not smokers are present.222224

- 36 -

Canadian researchers attending the 1983 University of

Geneva symposium on ETS reported on their extensive review of over

150 indoor air quality evaluations of office buildings compiled by

U.S. government agencies, universities and others. 223 After an

examination of the data, the authors concluded that smoking did

not significantly affect either indoor atmospheres or the frequency

of worker complaints and symptoms:

The review of available studies does notprovide any objective evidence that eitherpollution levels or patterns of health re-lated complaints differ in some remarkableway between locations with or withoutsmoking restrictions.

They did observe that "inadequate" ventilation creates conditions

"where discomfort and illness result irrespective of whether or

not smoking is permitted." The researchers reaffirmed their

findings in a report published in 1987.225

Similarly, in a review of 203 air quality investiga-

tions of government and business offices, schools and health care

facilities by the U.S. National Institute of Occupational Safety

and Health (NIOSH), U.S. government researchers concluded that

tobacco smoke played a contributing role in only four of the

buildings investigated. 222 A large majority of all complaints

were traced to general building contamination and inadequate

ventilation.

In 1988, a representative of a firm specializing in the

maintenance of office air conditioning and heating systems reported

- 37 -

on 223 individual indoor air quality investigations of public and

privately owned office buildings. 224 As in the NIOSH

investigations, ETS was implicated in only 4 percent of the

buildings investigated. He stated that the majority of indoor air

quality problems in modern office buildings may be traced to

inadequate fresh air circulation and to poorly maintained

ventilation systems which act as breeding grounds for fungi,

bacteria and other contaminants. He also suggested that visible

tobacco smoke ought to be considered a symptom, not a cause, of

general indoor air quality problems, in that ETS is often the only

visible sign that a ventilation problem exists.

It is therefore understandable, given the easy

recognition of ETS, that persons experiencing sick building symptoms

may blame ETS as the cause. Indeed, Canadian researchers have

verified that the mere visibility or pesence of tobacco smoke

may provoke claims that ETS is the cause of reported symptoms and

complaints.226

Thus, not only is ETS erroneously labeled the "cause"

of worker discomfort in "building illness" situations, its perceived

presence may even heighten anxiety about the safety of an

individual's environment. This, of course, adds impetus to the

movement for smoking restrictions in the workplace. However,

removal of ETS through smoking bans may serve only to divert

attention from more basic, underlying indoor air quality problems.

- 38 -

ENVIRONMENTAL TOBACCO SMOKE AND INDOOR AIR QUALITY

Antisinokers often claim that ETS is a major source of

indoor pollution in order to support their argument that ETS expo-

sures harm the health of nonsmokers.However, the scientific

literature indicates that, except under experimental or other

extraordinary conditions ! ETS does not have a significant influ -

ence on the quality of indoor air.227

The claim that ETS is a significant source of indoor

air pollution (and, as such, a hazard to health) is based upon

several unwarranted assumptions and misrepresentations of the scien-

tific data. This section will provide an analysis of the ETS/in-

door air quality question. The nature of ETS, its constituents, its

contribution to indoor air, and the methods employed to measure

exposure to it will be addressed.

Sidestream Smoke

It is frequently claimed that sidestream smoke, or

the smoke from the burning end of the cigarette, contains much

higher amounts of certain constituents than mainstream smoke, or

the smoke inhaled by smokers. Such a claim implies that a non-

smoker has an increased risk of disease because of his exposure to

sidestream smoke, which has more allegedly "toxic" or harmful

constituents than smoke inhaled by the smoker.However, this

- 39 -

argument is extremely misleading for several reasons, but most

notably because it fails to take into account that sidestream smoke

is immediately diluted in the surrounding air. This diluted

sidestreani smoke is more accurately called "ETS".

The impact of dilution on sidestreáni smoke in room air

cannot be overemphasized. Scientists have estimated that under

normal, realistic conditions, the quantity of various constituehts

in ETS are only a tiny fraction--from 1/100th to 1/1000th--of those

found in mainstream smoke.228

The potential toxicity of ETS was recently assessed

in a number of studies. 229232 These studies typically tested the

body fluids of nonsmokers exposed to ETS for mutagens, or substances

capable of altering the genetic structure of cells. Although it can

be argued that the scientific literature on this subject is very

limited, and that the laboratory tests employed are crude, the

studies reported no mutagenic activity in the body fluids of non-

smokers exposed to ETS.

Animal inhalation experiments using sidestream smoke

or constituents of sidestreain smoke are also inconclusjve.233234

German scientists exposed rats and hamsters to very high levels of

sidestream smoke during a 90-day inhalation experiment. 233 The

researchers reported no significant physiological effects on the

tissues of the animals. In his comprehensive review of the

literature on suspected pulmonary carcinogens, Dr. Domingo Aviado

observed that none of the constituents in sidestream smoke which

- 40 -

have been identified as potentially carcinogenic has induced

pulmonary cancer in animals under experimental conditions.234

Antismokers often refer to comparisons of sidestream

and mainstream smoke constituents, or sidestream to mainstream smoke

ratios, to support their claims about ETS. (A ratio here means

the number obtained by dividing the amount of a constituent in

sidestream smoke by the amount of that same constituent in

mainstream smoke. For example, a ratio of 10 for nicotine would

indicate that there is 10 times as much nicotine in sidestream as

in mainstream smoke.) However, such a ratio is derived from

laboratory experiments which usually employ smoking machines and

experimental chambers to measure samples taken from the burning

ends of cigarettes. Given these conditions, it is not surprising

that some levels of sidestreain smoke constituents may be greater

than those in mainstream smoke. Moreover, as mentioned above,

these experimental techniques do not take into account the impact

of dilution which occurs under more realistic settings. Nor do

they take into account the fact that even slight variations in the

moisture content of tobacco or the porosity of the cigarette paper

may greatly affect the reported ratios.228

Thus, it is both inappropriate and misleading to sug-

gest that sidestream smoke is equivalent to ETS and that, under

ordinary conditions, the nonsmoker is exposed to the levels of

constituents reported in sidestream smoke.

- 41 -

Constituents of ETS

A number of constituents typically have been cited in

the literature to demonstrate the contribution of ETS to the indoor

air.These include carbon monoxide, particulates, nicotine,

nitrosamines and others. However, the use of one or any

combination of such constituents to determine ETS levels presents

many problems. For example, although analytic and sampling meth-

ods continue to improve, there is at present no completely satis-

factory and uniform procedure for measuring ETS. 235236 Moreover,

although the findings of studies which measure constituents in

experimental conditions (e.g., in unventilated smog chambers) are

cited to dramatize the potential effect of ETS0n indoor air, they

have little, if any, similarity to those studies which measure ETS

in realistic settings. And finally, with the exception of nicotine,

none of the constituents which have been used as substitute

measures for ETS are characteristic of ETS alone. Other sources,

such as heaters, stoves and furnishings, generate substances at

levels greater than those found in ETS.

Carbon Monoxide

The scientific literature on carbon monoxide (CO) from

environmental tobacco smoke contains numerous examples of studies

conducted under unrealistic conditions. For instance, in one such

- 42 -

experiment, 80 cigarettes and 2 cigars were burned during a

78-minute period in a small unventilated simulated conference

room. 237 In another case, 9 cigarettes were smoked one after

another in an unventilated automobile. 238 Not surprisingly, the

levels of Co measured in these unusual situations were extraordi-

narily high. Regrettably, studies of that sort are often cited as

conclusive evidence of the impact of ETS on ordinary room air.

On the other hand, when carbon monoxide from ETS has

been measured under realistic conditions, its contributions have

been determined to be minimal. 239246 Most recently, scientists

in the United Kingdom monitored Co and other constituents in over

3,000 locations throughout the country. 246 They reported that

carbon monoxide levels did not differ significantly between smoking

and nonsmoking environments. Putting such studies into perspective,

Dr. Duncan Hutcheon, a professor of medicine who has reviewed the

literature, noted: "Environmental studies suggest that tobacco

smoke has little impact on the CO content of the room air except

under highly artificial conditions."247

What then are the sources of CO in public places? Re-

search has shown that the main sources in the outdoor urban en-

vironment are motor vehicles and industrial processes, 248 and that

indoor levels are affected by these outdoor sources, mainly through

ventilation, and by numerous activities such as cooking and heating.

In fact, studies have indicated that gas stoves in kitchens and

heating units are often major sources of CO in homes.249251

- 43 -

After their review of such studies, participants at the

Geneva symposium on ETS concluded that Co from environmental

tobacco smoke "is not important from a health point of view."16

Similarly, researchers from the Lawrence Berkeley Laboratory in

the U.S., who are otherwise critical of tobacco smoke, concluded

that "based on theoretical and empirical results, CO sidestream

emissions from cigarettes have often been overemphasized."252

Particulates

A paper published in a 1980 issue of Science magazine,

in which the authors reported the results of their efforts to mea-

sure particles or particulates in the air of smoking and nonsmok-

ing areas, is often cited to support the claim that ETS is a major

indoor pollutant. 253 The authors, Repace and Lowrey, contend that

the levels of particles they observed in the smoking areas were

much higher than in the nonsmoking areas. However, their study

results are inconsistent with many others. For example, the

average particle count attributed to ETS in their study was nearly

twice as high as that determined in a study of 44 offices, 254 and

nearly three to five times higher than the average levels reported

in other studies of office buildings, restaurants and resi-

dences.225'246'255260

There are a number of explanations for the authors'

apparent overestimation of ETS exposure. First, they selectively

- 44 -

sampled environments such as meeting and game rooms, bars and

sandwich shops which did not represent normal occupancy conditions

and where particulate levels would likely be high regardless of

the presence or absence of tobacco smoke. Second, through

inappropriate testing methods, they incorrectly assumed all

particles in the air arose from ETS.However, as several

researchers have noted, ETS typically contributes less than half

to overall particle levels in indoor spaces. 26263 Moreover,

particles also are generated by people and their everyday routine

activities such as movement and cooking.264265

In addition to ignoring potential. indoor sources of

particles, the study has been criticized on other grounds. For

example, Repace and Lowrey did not measure ventilation rates or

humidity and they failed to calibrate properly their instrument

prior to testing. 265 Indeed, the testing device they used "is no

longer recommended" for tobacco smoke measurements.235236

Repace and Lowrey implied that the particle levels at-

tributed to ETS are potentially harmful to the health of non-

smokers. However, an earlier study by Bouhuys, et al.,, found "no

evidence that high TSP (total suspended particulates) levels in

homes with smokers were associated with increased symptoms or lung

function loss among nonsmokers in the same home." 266 Binder, et

al., found that respirable particulate levels in homes were

"independent of the presence or. absence of respiratory disease.11171

More recently, Lebowitz, et al., reported on data which showed

- 45 -

that respiratory symptoms in asthmatics were associated with total

particle levels indoors (e.g., dust), but not with the presence of

tobacco smoke. 138,205

Nicotine

Since environmental nicotine is produced exclusively

by burning tobacco, it is considered to be a more reliable indicator

of the amount of tobacco smoke in the environment than other smoke

constituents. Studies which have used nicotine in this way suggest

that tobacco smoke contributes little to the indoor atmo-

sphere.242 ,246, 257260, 267-272

For example, Drs. William Hinds and Melvin First of the

Harvard School of Public Health found very small amounts of nico-

tine in the atmospheres of bars, bus and airline terminals, res-

taurants, and cocktail and student lounges. 269 In a 1980 publica-

tion, Dr. , Hinds wrote that in public places, "the typical, average

airborne concentration" of the tobacco smoke to which the nonsmoker

is exposed is equivalent to a small fraction of a cigarette per

hour.273

French researchers, using a different method of nico-

tine measurement to assess the amount of tobacco smoke in the

atmosphere, reported higher concentrations of nicotine than Hinds

and First. 267 However, they still concluded that "smoking does

not present a risk to nonsmokers."

- 46 -

In 1984, Japanese researchers tested a personal nico-

tine monitor in a number of public places, including offices,

restaurants, lobbies, terminals and public transportation. They

reported levels of nicotine exposure equivalent to one one-

thousandth (1/1000) to four one-hundredths (4/100) of a cigarette

per hour. 271 Their findings were repeated in 1987.272

In 1986, scientists from the R.J. Reynolds Tobacco

Company developed a portable sampler for nicotine, particulates

and CO. In conjunction with ACVA International, a firm specializing

in the assessment of indoor air quality, samples of nicotine,

particulates and CO were taken in 23 office buildings and 48 res-

taurants in New York City. The scientists reported such low levels

that a typical New Yorker would have to work 550 uninterrupted

hours, or dine continuously for 400 hours, in order to be exposed

to the nicotine equivalent of one cigarette. 257 In 1987, scien-

tists from IT Corporation, a firm specializing in the assessment

and reduction of environmental substances, repeated the New York

study in offices and restaurants in Dallas, Texas and in Ottawa,

Canada. 258 ' 259 They reported average nicotine exposure levels

equivalent to three one-hundredths (3/100) of a cigarette per

eight-hour workday, and three one-thousandths (3/1000) of a

cigarette during a one-hour meal.258

In a recent nationwide sampling survey in the United

Kingdom, researchers monitored nearly 3000 sites in travel, work,

home and leisure locations for ambient nicotine, CO and particle

- 47 -

levels. 246 Smoking was known to have occurred at almost half of

those sites, yet in three-fourths of the samples, nicotine levels

were too low to be detected. Canadian researchers also reported

levels of nicotine at or below levels of detection even in locations

with recirculated air from smoking designated areas.260

There is little to suggest that the small amount of

nicotine to which a nonsmoker may be exposed is related to human

disease. For example, two German scientists monitored several

physiological responses in nonsmokers exposed to tobacco smoke

under laboratory conditions. 274 They concluded that the amount of

nicotine to which their subjects were exposed was too small to

alter sensitive test measurements of heart rate, heart muscle

tracings (EKG), blood pressure or skin temperature.Similar

findings were reported in 1982 by a Japanese physiologist. 275 More

recent research tends to support these conclusions. In a 1983

study of nonsmoking flight attendants, researchers measured

exposures to nicotine during transpacific flights. The researchers

concluded that the concentrations of nicotine achieved were so

small that they were "unlikely to have physiologic effects.11276

In 1986, researchers monitored levels of nicotine, CO and

particulates in 66 commercial flights in the U.S. They concluded

that the non-smoking sections of commercial aircraft were

essentially free of ETS.277

- 48 -

Nitrosamines

Nitrosamines have been described as substances present

everywhere in the environment, and, consequently, as detectable

"wherever they have been searched for." 278 This includes the air

we breathe, the food we eat and the water we drink. One scientist

stated that nitrosamines occur everywhere "independent of any type

of tobacco combustion, in which case concentrations are often

reached in respiratory air that are far above those that may be

produced by tobacco smoke." 279 Another scientist, Dr. Helmut

Schievelbein, Director of the Institute of Clinical Chemistry in

Munich, observed that concentrations of nitrosamines either con-

sumed or generated daily by individuals are "many times greater"

than those found in environmental tobacco smoke.280

The claim has been made, largely based on reports by

Brunnemann, et al., that nitrosamines from tobacco smoke represent

a cancer risk to nonsmokers. 281-282 It is important to note,

however, that Brunnemann and his co-workers have emphasized that

there are at present "no epidemiological data linking human

respiratory cancers to volatile nitrosamines." 283 Even the 1979

U.S. Surgeon General's Report raised questions about the relevance

of their findings.146

It is sometimes claimed that nitrosamine levels in

sidestream smoke exceed those found in mainstream smoke, with the

implication that those levels are also found in ETS. This, how-

- 49 -

ever, is not the case. In 1982, Austrian scientists reported that

measurements of nitrosamine levels in ordinary public places were

far below the values reported earlier by Brunnemann and his

co-workers. They emphasized that the levels of nitrosamines which

may occur in environmental tobacco smoke "are quite different"

from those found in sidestreaxn smoke.284

A reviewer of the literature in this area noted that

the "experimental basis for carcinogenic effect on human beings has

not been established for the very low concentrations" of

nitrosamines to which nonsmokers are exposed. 242 A German scien-

tist who discussed this point concluded that, as a consequence,

any claimed implications of health risks from nitrosamines in

tobacco smoke "clearly move into the range of speculation."279

Other Constituents

Other constituents which have been identified in ana-

lyses of sidestream smoke are often assumed to contribute sub-

stantially to indoor air in the form of ETS. The most commonly

mentioned constituents include nitrogen dioxide (NO2 ), formalde-

hyde and volatile organic compounds. However, these compounds are

present in very small amounts in ETS. 285287 Recent studies report

no correlations between ETS and levels of NO 2 and volatile organic

compounds in residences and offices. 288 ' 289 Indeed, research

indicates that levels of these components generated from other

- 50 -

ordinary sources (e.g., cooking stoves, heaters and building

materials) are much greater than those contributed by cigarette

smoking. 285,290

Cotinine as a Marker for ETS Exposures

In 1984, Japanese researchers reported that cotinine,

a substance converted from nicotine by the body, could be used to

measure nonsmoker exposure to ETS. 291 While a number of other

reports appear to endorse the claim that cotinine is a reliable

marker for total exposures to ETS, 292297 others do not.298304

For example, researchers have found that individuals metabolize

nicotine in different ways at different times and that elimination

rates for cotinine vary among individuals. 298 ' 299 ' 303 ' 304 In addi-

tion, recent research indicates that diet may contribute to levels

of nicotine and cotinine found in the body.305

Scientists have also noted that different methods of

analysis may influence final recorded levels of cotinine.301'303'306

The 1984 Japanese study was criticized because the researchers

used a method of analysis which was not specific to cotinine and

which subsequently yielded cotinine estimates ten times higher

than those reported by other researchers for the same levels of

exposure. 306,307

Thus, scientific research casts serious doubt on the

use of cotinine as a marker for total ETS exposure. A uniform

- 51 -

method of analysis of cotinine must be developed and the extent of

individual variability in metabolism must be determined before any

pronouncements concerning the efficiency of cotinine as a marker

for ETS can be made.

Summary

The foregoing literature survey demonstrates that, at

most, tobacco smoke has only a minor influence upon ordinary indoor

environments compared to common household or workplace sources,

such as stoves, heaters and furnishings. Yet tobacco smoke, because

it is so easily identifiable, has become a target for those who

apparently want a quick, simple solution to the problem of indoor

air pollution. However, as one Canadian scientist observed, tithe

claim that smoking is responsible for indoor air pollution is an

oversimplification of a complex, multi-source problem. it265

- 52 -

ENVIRONMENTAL TOBACCO SMOKE AND THE PUBLIC

Exposure to environmental tobacco smoke under extreme

conditions may provoke complaints of irritation and annoyance.

However, the mere sight and smell of tobacco smoke may be enough

to produce strong reactions in individuals who intensely dislike

smoking. Although the extent to which environmental tobacco smoke

may bother or annoy individuals under normal conditions has not

been established scientifically 16 , it has been suggested that

tobacco smoke is at most a minor annoyance for many individuals. 308-

310 It has also been suggested that extreme reactions to tobacco

smoke exposure may have a strong emotional or psychological

basis. 31312 Whatever the reasons for such reactions, the

literature on this subject suggests it is misleading to claim, as

antismokers frequently do, that such subjective or emotional

reactions have a physical basis.

Governmental Regulations and Individual Rights

Although antismoking organizations claim to seek

protection for the health and welfare of the nonsmoker, their

efforts to make smoking socially unacceptable are part of a broader

objective -- the severe restriction or even prohibition of smoking

by governmental legislation. Of course, such regulations restrict

the freedom of the smoker, and antismokers are compelled to justify

this kind of political control of individual behavior. They have

- 53 -

accomplished this, in large measure, by exploitation of the so-

called "health" argument involving ETS exposures to nonsmokers.

For example, the most recent U.S. Surgeon General's

Report on this subject contends that ETS causes lung cancer in

adult nonsmokers and respiratory conditions in children, and that

separation of smokers and nonsmokers is not effective in minimizing

the nonsmoker's exposure to ETS 14 . The Report concludes that

smoking bans will not only reduce ETS exposures, but will "alter

smoking behavior and public attitudes about tobacco use." The

Report further suggests that "over time, this may contribute to a

reduction of smoking." Thus, the underlying motivation for the

use of the ETS/health argument is to attain a "smoke-free society

by the year 2000."

The Surgeon General's tactics, however, have been ques-

tioned by a number of critics. One reviewer suggested that the

Surgeon General's conclusions were based on "flimsy" evidence

presented in an effort to "divert attention" from important health

concerns such as the "poisoning of the environment. ,, 313 A U.S.

Congressman, in a letter to the Congressional Record, wrote that

"the conclusions in the Surgeon General's Report are not supported

by the research in his own report." 314 Moreover, recent scientific

studies aboard commercial aircraft and in offices indicate, contrary

to the Surgeon General's Report, that separation of smokers and

nonsmokers effectively minimizes ETS exposures for

nonsmokers. 260,277

- 54 -

Five U.S. labor union presidents criticized the Surgeon

General's Report for minimizing the risk imposed by workplace

toxins, for shifting the burden of cleaning up the workplace to

workers, and for dismissing the importance of engineering controls

(e.g., ventilation) in the maintenance of a safe -and comfortable

workplace. 315 The U.S. Tobacco Institute, in a comprehensive review

of the issue, observed that premature scientific conclusions about

ETS, together with increasing political pressure to accept the

Surgeon General's conclusions, have "brought scientific integrity

to a crossroads."316

However, some governmental bodies have refused to give

in to the emotional rhetoric of antismoking organizations and

public health officials. For example, courts in the United States

have, with few exceptions, denied the claims of individuals who

contend they have a "right" to a smoke-free workplace. A law

journal article described this trend as representing "substantial

judicial reluctance to supplant the roles of employer and employee

in addressing smoking in the private wrkplace." 317-

In 1987, the Supreme Court of Canada refused an appeal

by a worker who sought a smoke-free environment on the grounds

that ETS is a "dangerous" workplace substance. 318 Similarly, in

Japan, a court judgment denied a plaintiff's request that more

than half of all the railway cars on passenger trains be designated

nonsmoking because, the Court ruled, ETS exposures do not "exceed

the limits of normal tolerance."319

- 55 -

In 1984, the U.S. Civil Aeronautics Board (CAB), the

federal agency responsible at that time for regulating smoking

aboard aircraft, considered a petition calling for a ban on smok-

ing on flights lasting two hours or less. The CAB assessed the

health claims advanced by many antismoking organizations, but

denied the petition, arguing that there was little scientific basis

to the claims.320

In 1987, the U.S. Department of Transportation

rejected a recommendation from the National Academy of Sciences

to prohibit smoking aboard commercial airliners 321 because it was

not supported by data associating health effects and ETS exposures

aboard aircraft. 322 Nonetheless, the U.S. Congress voted to ban

smoking aboard domestic flights of two hours or less until 1990

when the law expires. The Department of Transportation subsequently

issued a request for proposals for in-flight monitoring of ETS

and other indoor air constituents. 323The results will be

presented to Congress for its deliberations in 1990. In the

meantime, other research on this issue continues. In 1988, for

example, a review of the relevant data concluded that "the available

scientific evidence does not support the prohibition of smoking on

commercial aircraft." It also stated that the available data

"suggest that factors or substances other than ETS may be major

contribution to subjective complaints of discomfort by passengers

and flight crew."324

- 56 -

Other governmental agencies, on the other hand, have

done little to clarify the issue for the public. In the United

States, for instance, two separate government-sponsored reports

recommended restricting tobacco smoking on the basis of risks that

were characterized as "biologically plausible". 93,321 Another

report sought to identify the known facts and areas of uncertainty

regarding the ETS issue and observed that "many more of the latter

were found than the former." The report nevertheless concluded

that exposures to ETS harm the health of nonsmokers. 13 Similarly,

the most recent monograph on tobacco smoke from the WHO's

International Agency for Research on Cancer (IARC) warns of

possible health consequences from exposures to ETS while conceding

that the scientific evidence is consistent "either with an increase

or with an absence of risk."325

In the United Kingdom, the Fourth Report of the

Independent Scientific Committee on Smoking and Health published

in 1988 endorsed the heavily criticized claim that nonsmoker

exposure to ETS increases the risk of lung cancer from 10 to. 30

percent compared with nonsmokers who are not exposed 15 , despite

the fact that, as one commentator noted, "exposure data are very

unreliable, and there is a proven basis for bias in the direction

of exaggeration of the risk."326

In 1985, the Supreme Insurance Court of Sweden awarded

compensation for occupational injury to the estate of a nonsmoker

who worked with smokers and who subsequently died of lung cancer.

- 57 -

Although the media described the Court's decision as a causal

indictment of ETS exposure, one commentator noted that the deci-

sion was based upon the structure of occupational injury regula-

tions in Sweden and not upon any scientific evidence of a causal

relationship between lung cancer and ETS exposures. 327 This author

explained that the Swedish insurance system operates under a reverse

burden-of-proof provision, namely, that the relationship between a

suspected factor and an injury is assumed to be causal unless

substantially stronger reasons are produced to suggest otherwise.

Consequently, this provision does not require the claimant to

present scientific proof of a causal relationship between a

suspected factor and an injury.

Three years later, in Australia, a city bus driver

who contended his lung cancer was caused, at least in part, by

exposure to ETS on the job was awarded a claim by the insurer of

the Melbourne Transit Authority. 328 This settlement was interpreted

by some as establishing legal precedent for the claim that ETS

exposure has been proven to cause lung cancer. However, the bus

driver's claim regarding ETS was only part of his complaint and

the tribunal before which the proceeding was held heard only part

of the evidence in the case before the insurer made a settlement

offer. 329 Thus, this settlement established no formal legal

precedent regarding ETS health claims.328'329

- 58 -

An Australian writer recently summarized the issue of

governmental influence on the ETS question. 330 He wrote:

Those involved in government anti-smokingactivities should be aware of the tenuousnature of the data on passive smoking andhealth effects. The deliberate use offear inducing tactics by misrepresentingsuch data by dubious extrapolations ispropaganda and not health education.

It is clear that once the health claims regarding en-

vironmental tobacco smoke are placed in their proper scientific

perspective, the issue of prohibiting smoking in public places

becomes a social and political one. As a former director of the

WHO wrote recently, "the question as to whether the conceivable

theoretical possibility of a risk calls for official preventive

measures is not a medical but a politico-social one.1119

The issue thus involves whether or not public policy

should mandate prohibition of smoking because it is seen by some

as a nuisance or annoyance. If the decision is reached in favor

of regulation, then one must decide whether the same line of

reasoning applies to other kinds of "annoyances" encountered in

everyday life and whether those "annoyances" should also be banned.

The implication of such reasoning should be obvious to-anyone who

is opposed to unwarranted governmental intrusion into people's

lives.

- 59 -

VIWhat is the Solution?

A recent article on the principles of public policy

relating to smoking suggests an alternative to governmental regu-

lations. According to this article, "the appropriate form of

public policy is the promotion of courtesy and cooperation between

smokers and nonsmokers, rather than outright prohibition."331

This solution relies upon common sense, courtesy and the free

marketplace for decisions concerning smoking. More importantly,

however, this solution avoids unnecessary governmental interference

with its subsequent loss of individual freedom.

- 60 -

REFERENCES*

1. Shopland, D., Quoted in "The Campaign Against Smoking GainsMomentum," New Engi314(16): 1059-1064, 1986.

2. Berger, P., "A Sociological View of the Antismoking Phenome-non," Smoking and Society , ed. R. Tollison (Lexington:D.C. Heath and Co., 1986): 225-240.

3., Warner, K., "Toward a Global Strategy to Combat Smoking:The 5th World Conference on Smoking and Health," J PublicHealth Policy 5(1): 28-39, 1984.

4. Hulka, B., Quoted in "Other People's Smoke May Be Risky forYou," . NRC News Report 37(1): 8-11, 1986-1987.

5. Witorsch, P., "Passive Smoking," New Zealand Med : 258-9,April 22, 1987.

6. Ainiel, B., "Smoke in Their Eyes," The (London) Times: 15,March 13, 1987.

7. Kris, M., "High Risk of Cancer for Passive Smokers. MoreThan Much Ado About Nothing?," Aertze Zeitin g 5(220): 1-2,December 11, 1986.

8. Adikofer, Franz, "The Problems of Passive Smoking," DerKassenarzt 27(51/52): 29-39, 1987. Translation.

9. Uberla, K., "Lung Cancer from Passive Smoking: Hypothesisor Convincing Evidence?," Int Arch Occu p Environ Health59: 421-437, 1987.

10. Hirayama, T., "Non-Smoking Wives of Heavy Smokers Have aHigher Risk of Lung Cancer: A Study from Japan," Br Med JI, 282: .183-185, 1981.

11. Trichopoulos, D., et al., "Lung Cancer and Passive Smoking,"mt J Cancer 27(1): 1-4, 1981.

12. White, J. and H. Froeb, "Small-Airways Dysfunction inNonsmokers Chronically Exposed to Tobacco Smoke," New EngiJ Med 302(13): 720-723, 1980.

*All translations used in preparing this document are "unofficial"and may be subject to modification.

- 61 -

13. Committee on Passive Smoking, Board on Environmental Studiesand Toxicology, National Research Council, EnvironmentalTobacco Smoke: Measuring Exposures and Assessing HealthEffects (Washington: National Academy Press, 1986).

14. U.S. Department of Health and Human Services, The HealthConsequences of Involuntary Smoking :A Report of theSurgeon General (U.S. DHHS: Washington, 1986).

15. Department of Health and Social Security, Fourth Re port ofthe Independent Scientific Committee on Smoking and Health(London: Her Majesty's Stationery Office, 1988).

16. Rylander, R., "Workshop Perspectives," ETS-EnvironmentalTobacco Smoke: Report from a Workshop on Effects andExposure Levels, eds. R. Rylander, et al., Eur J RespirDis, Suppi. 133(65): 143-145, 1984.

17. U.S. Department of Health and Human Services, Public HealthService, National Institutes of Health, "Report of Workshopon Respiratory Effects of Involuntary Smoke Exposure: Epi-demiologic Studies, May 1-3, 1983," December, 1983.

18. Valentin, H. and E. Wynder, "Health Danger Through PassiveSmoking Not Proven: Physicians View on Passive Smoking,"(Press release for Symposium on "Passive Smoking from aMedical Point of View," April 9-12, 1984, Vienna, Austria).

19. Handler, J., Preface, Passivrauchen aus medizinischer Sight[Medical Perspectives on Passive Smoking) (Geneva:IRL

-Imprimeries Reunies Lausanne, 1986): A-B.

20. "Proceedings of the International Experimental ToxicologySymposium on Passive Smoking," ed. E. Mohtashamipur E.,Toxicology Letters 35(l): 1-168, 1987.

21. Gostomzyk, G.J., "Passive Smoking-Report on an InternationalSymposium (23-25 October 1986) ," Public Health 49: 212-215,1987.

22. Kasuga, Hitoshi, et al. , "International Conference on IndoorAir Quality, Tokyo, 1987," Nikkei Medical 1(10): 168-175,1988. Translation.

23. Lehnert, G. and E. Wynder, "Illness Due to PassiveSmoking?," (Press Release, Vienna, Austria: May 3, 1988).Translation.

24. Witorsch, P., "Passive Smoking," New Zealand Med : 865,November 12, 1986.

- 62 -

25. Aviado, D., "Health Issues Relating to 'Passive Smoking',"Smoking and Society , Chapter 4: 158, 1985.

26. Lee, P., et al., "Relationship of Passive Smoking to Riskof Lung Cancer and Other Smoking-Associated Diseases,"Brit J Cancer 54: 97-105, 1986.

27. Aviado, D.M., "Toxicological Basis for Regulation of IndoorAir Quality," J Applied Toxicol 8(3): 155-157, 1988.

28. Ahiborn, W. and K. Uberla, "Passive Smoking and Lung Cancer:Reanalyses of Hirayama's Data," Indoor and Ambient AirQuality , eds. R. Perry and P.W. Kirk (London: Selper Ltd.,1988) : 169-178.

29. Diamond, G. and J. Forrester, "Clinical Trials and Statis-tical Verdicts: Probable Grounds for Appeal," Ann InternMed 98(3): 385-394, 1983.

30. Gostomzyk, J., "Krank dürch Passivrauchen? (Disease Due toPassive Smoking?)," Munch med Wschr 124(4):11, 1982.Translation.

31. Grimmer, G., "Kausalittsbeziehung Fraglich (Causality-Relationship Questionable)," Munch med Wschr 124(4): 16,1982. Translation.

32. Grundmann, E., et al., "Non-Smoking Wives of Heavy SmokersHave a Higher Risk of Lung Cancer," Br,Med J I 282: 1156,1981.

33. Kilpatrick, S.J. and J. Viren, "Age as a Modifying Factorin the Association Between Lung Cancer in Non-smoking Womenand Their Husbands' Smoking Status," Indoor and Ambient AirQuality , eds. R. Perry and P.W. Kirk (London: Selper Ltd.,1988) : 195-202.

34. Lee, P., "Passive Smoking," Food Chem Toxicol 20: 223-229,1982.

35. Lehnert, G., "Krank durch Passivrauchen? (Disease Due toPassive Smoking?)," Munch med Wschr 123(40): .1485-1488,1981. Translation.

36. MacDonald, E., "Non-Smoking Wives of Heavy Smokers Have aHigh Risk of Lung Cancer," Br Med J II 283:915-916,1981.

- 63 -

37. MacDonald, E., "Non-Smoking Wives of Heavy Smokers Have aHigher Risk of Lung Cancer," Br Med J II 283: 1465, 1981.

38. Mantel, N., "Epidemiologic Investigations: Care in Conduct,Care in Analysis, and Care in Reporting," J Cancer Res ClinOncol 105: 113-116, 1983.

39. Rutsch, M., "Mangeihafte Statistische Absicherung (Inade-quate Statistical Substantiation) ," Munch med Wschr 124(4):15, 1982. Translation.

40. Rutsch, M., "Non-Smoking Wives of Heavy Smokers Have aHigher Risk of Lung Cancer," Br Med J I 282: 985, 1981.

41. Schievelbein, H., "Kein Beweis fUr Gesundheitsschdlichkejt(No Proof of Injuriousness to Health) ," MunchWschr124(4): 13-15, 1982. Translation.

42. Schmh1, D., Quoted in "Lungenkrebs durch Passivrauchen?Studie aus Japan bestatigt Verdacht/Widerspruch aus Deutsch-land (Lung Cancer from Passive Smoking? Study from JapanConfirms Suspicion/Contradictory Opinion, from Germany),"Stlddeutsch Zeitung, March 18, 1981. Translation.

43. Sterling, T., "Non-Smoking Wives of Heavy Smokers Have aHigh Risk of Lung Cancer," Br Med J I 282: 1156, 1981.

44. Rylander, R., "Environmental Tobacco Smoke and Lung Cancer,"ETS -Environmental Tobacco Smoke: Report from a Workshop onEffects and Exposure Levels, eds. R. Rylander, et al.,J Respir pj, Suppl.133(65): 127-133, 1984.

45. Trichopoulos, D., et al., "Lung Cancer and Passive Smoking:Conclusions of Greek Study," Lancet II: 677-678, 1983.

46. Garfinkel, L., "Time Trends in Lung Cancer Mortality AmongNonsmokers and a Note on Passive Smoking," JNCI 66(6):1061-1066, 1981.

47. Lam, T.H., et al., "Smoking, Passive Smoking andHistological Types in Lung Cancer in Hong Kong ChineseWomen," Br J Cancer, 56(5) : 673-678, 1987.

48. Pershagen, G., et al., "Passive Smoking and Lung Cancer inSwedish Women," Am J Epideiniol 125(1): 17-24, 1987.

49. Akiba, S., et al., "Passive Smoking and Lung Cancer AmongJapanese Women," Cancer Res 46: 4804-4807, 1986.

- 64 -

50. Humble, C.G., et al., "Marriage to a Smoker and Lung CancerRisk," AJPH 77(5): 598-602, 1987.

51. Brownson, R.C., et al., "Risk Factors for Adenocarc jnoxna ofthe Lung," Am J Epidemiol 125(1): 25-34, 1987.

52. Knoth, A., et al., "Passive Smoking as a Causative Factorof Lung Cancer in Nonsmoking Women," Med Klin Prax 78(2):56-59, 1983. Translation.

53. Heller, W., "Lung Cancer and Passive Smoking," Lancet II:1309, 1983.

54. Correa, P., et al., "Passive Smoking and Lung Cancer,"Lancet II: 595-597, 1983.

55. Buffler, P., et al., "The Causes of Lung Cancer in Texas,"Lung Cancer: Causes and Prevention, eds. Mizell and Correa(Deerfield Beach: Verlag Chemie International, 1984):83-99.

56. Repace, J. and A. Lowrey, "A Quantitative Estimate ofNonsmokers' Lung Cancer Risk from Passive Smoking," EnvironInt 11(1): 3-22, 1985.

57. Marshall, E., "Tobacco Science Wars," Science 236: 250-251,1987.

58. Arundel, A., et al., "Nonsmoker Lung Cancer Risks fromTobacco Smoke Exposure:An Evaluation of Repace andLowrey's Phenomenological Model," J Environ Sci HealthC4(1): 93-118, 1986.

59. Anonymous, "The Murky Hazards of Secondhand Smoke," ConsumerReports: 81-84, February, 1985.

60. Arundel, A., et al., "Never Smoker. Lung Cancer Risks fromExposure to Particulate Tobacco Smoke," Environ mt 13:409-426, 1987.

61. Lebowitz, M., "The Potential Association of Lung Cancer withPassive Smoking," Environ Int 12: 3-9, 1986.

62. Johnson, C., "Letter to the Editors," Environ mt 12:21-22, 1986.

63. Burch, P., "Health Risks of Passive Smoking: Problems ofInterpretation," Environ Int 12: 23-28, 1986.

- 65 -

64. Kilpatrick, S.J., Jr., "Letter to the Editors," Environ12:29-31, 1986.

65. Miller, G., "Cancer, Passive Smoking and Nonernployed andEmployed Wives, West J Med 140(4): 632-635, 1984.

66. Arundel, A. and T. Sterling, "Review of the EpidemiologicalStudies on Lung Cancer and Nonsmoker's Exposure toEnvironmental Tobacco Smoke," Private Communication to theSpecial Projects Office, U.S. OTA Health Program, April,1986.

67. Sandier, D., et al., "Cancer Risk in Adulthood from EarlyLife Exposure to Parents' Smoking," Am J Public Health75(5): 487-492, 1985.

68. Sandier, D., et al., "Cumulative Effects of Lifetime PassiveSmoking on Cancer Risk," Lancet II: 312-314, 1985.

69. Sandier, D., et al., "Passive Smoking in Adulthood andCancer Risk," Am J Epidemiol 121(1): 37-48, 1985.

70. Lee, P., "Lifetime Passive Smoking and Cancer Risk," LancetII:1444, 1985.

71. Burch, P., "Lifetime Passive Smoking and Cancer Risk,"Lancet II: 866, 1985.

72. Burch, P., "Passive Smoking in Adulthood and Cancer Risk,"Am J Epidemiol 123(2) : 368-9, 1986.

73. Higgins, I., "Lifetime Passive Smoking and Cancer Risk,"Lancet II: 866-867, 1985.

74. Friedman, G., "Passive Smoking in Adulthood and CancerRisk," Am J Epidemiol 123(2): 367, 1986.

75. Mantel, N., "Passive Smoking in Adulthood and Cancer Risk,"Am J Epidemiol 123(2) : 367-8, 1986.

76. Vandenbroucke, J., et al., "Active and Passive Smoking inMarried Couples: Results of 25 Year Follow Up," Br MedI: 1801-1802, 1984.

77. Gillis, C., et al., "The Effect of Environmental TobaccoSmoke in Two Urban Communities in the West of Scotland,"ETS -Environmental Tobacco Smoke. Report from a Workshop onEffects and Exposure Levels, eds. R. Rylander, et al., EurJ Respir Dis, Suppl. 133(65): 121-126, 1984.

- 66 -

78. Kabat, G. and E. Wynder, "Lung Cancer in Nonsmokers,"-Cancer 53(5): 1214-1221, 1984.

79. Garfinkel, L., et al., "Involuntary Smoking and Lung Cancer:A Case-Control Study," JNCI 75(3): 463-469, 1985.

80. Chan, W. and S. Fung, "Lung Cancer in Non-Smokers in HongKong," Cancer Campaign Vol. 6, Cancer Epidemiology , ed. E.Grundmann (Stuttgart: Gustav Fischer Verlag, 1982):199-202.

81. Koo, L., et al., "Is Passive Smoking an Added Risk Factorfor Lung Cancer in Chinese Women?," J Exp Clin Cancer Res3(3):277-283, 1984.

82. Koo, L., et al., "Measurements of Passive Smoking andEstimates of Lung Cancer Risk Among Non-Smoking ChineseFemales," Int J Cancer 39: 162-169, 1987.

83. Koo, L., et al., "Life-History Correlates of EnvironmentalTobacco Smoke: A Study on Nonsmoking Hong Kong ChineseWives with Smoking Versus Nonsmoking Husbands," Soc SciMed 26(7): 751-760, 1988.

84. Koo, L., "Dietary Habits and Lung Cancer Risk Among ChineseFemales in Hong Kong Who Never Smoked," Nutr Cancer 11:155-172, 1988.

85. Wu, A., et al., "Smoking and Other Risk Factors for LungCancer in Women," JNCI 74(4): 747-751, 1985..

86. Lloyd, 0., et al., "Respiratory Cancer in a Scottish Indus-trial Community: A Retrospective Case-Control Study,"Occup Med 36: 2-8, 1986.

87. Wald, N.J., et al., "Does Breathing Other People's TobaccoSmoke Cause Lung Cancer?," Br Med J 293: 1217-1222, 1986.

88. Blot, W. and J. Fraumeni, Jr., "Passive Smoking and LungCancer," JNCI 77: 993-100, 1986.

89. Doll, R., "Lung Cancer: Observed and Expected Changes inIncidence from Active and Passive Smoking," Presentationto UICC Conference, Budapest, August, 1986.

90. Katzenstein, A., Environmental Tobacco Smoke (ETS) and Riskof Lung Cancer--How Convincing is the Evidence?,(Katzenstein Associates, Larchmont, NY, March, 1987.)

- 67 -

91. Mantel, N., "Passive Smoking and Disease," StatisticsColloquium, The American University, May 4, 1988.

92. Letzel, H., et al., "Meta-Analyses on Passive Smoking andLung Cancer Effects on Study Selection and Misclassificationof Exposue," Indoor and Ambient Air Qualit y , eds. R. Perryand P.W. Kirk (London: Selper Ltd., 1988): 293-302.

93. Staff Paper, Special Projects Office of the Health Program,OTA, "Passive Smoking in the Workplace: Selected Issues,"Office of Technology Assessment, United States Congress,May, 1986.

94. Schwartz, S.L. and N.J. Baiter, "ETS-Lung CancerEpidemiology: Supportability of Misclassification andRisk Assumptions," Indoor and Ambient Air Quality, eds. R.Perry and P.W. Kirk (London: Selper Ltd., 1988): 159-168.

95. Lee, P., "Passive Smoking and Lung Cancer: Problems inInterpreting the Epidemiological Data," Presentation toToxicology Forum, Washington, D.C., February, 1987.

96. Lee, P., "Does Breathing Other People's Tobacco Smoke CauseLung Cancer?," Br Med J 293: 1503-1504, 1986.

97. Lee, P., "Misclassification as a Factor in Passive SmokingRisk," Lancet II: 867, 1986.

98. Lee, P., "Lung Cancer and Passive Smoking: Association anArtefact Due to Misclassification of Smoking Habits?,"Toxicology Letters 35: 157-162, 1987.

99. Lee, P., "Passive Smoking and Lung Cancer Association: AResult of Bias?," Human Toxicol 6: 517-524, 1987.

100. Lee, P., "An Alternative Explanation for the IncreasedRisk of Lung Cancer in Non-Smokers Married to Smokers,"Indoor and Ambient Air Quality , eds. R. Perry and P.W. Kirk(London: Selper Ltd., 1988): 149-158.

101. Kilpatrick, S.J., Jr., "Misclassification of EnvironmentalTobacco Smoke Exposure: Its Potential Influence on Studiesof Environmental Tobacco Smoke and Lung Cancer," ToxicologyLetters 35: 163-168, 1987.

102. Friedman, G., et al., "Prevalence and Correlates of PassiveSmoking," Am J Public Health 73(4): 401, 1983.

103. Pron, G.E., et al., "The Reliability of Passive Smoking

- 68 -

104.

105.

106.

107.

108.

109.

110.

Ill.

112.

113.

114.

115.

Histories Reported in a Case-Control Study of Lung Cancer,"Am J Ep idemiol 127(2): 267-273, 1988.

Schenker, M., et al., "Assessment of Environmental TobaccoSmoke Exposure in Epidemiologic Studies," Chest 91(2):313-314, 1987. Abstract.

Lerchen, L. and J.M. Sainet, "An Assessment of the Validityof Questionnaire Responses Provided by a Surviving Spouse,"Am J Ep iderniol 123(3): 481-489, 1986.

Sandier, D.P. and D.L. Shore, "Quality of Data on Parents'Smoking and Drinking Provided by Adult Offspring," Am JEp idemiol 124(5): 768-778, 1986.

Wynder, E., "Guidelines to the Epidemiology of WeakAssociations," Prey Med 16: 211-212, 1987.

Mantel, N., "Lung Cancer and Passive Smoking," Br MedJ 294:440, 1987.

Lee, P., "Deaths in Canada from Lung Cancer Due toInvoluntary Smoking," CMAJ 137: 372-373, 1987.

Gordis, Leon, "Challenges to Epidemiology in the NextDecade," Am J Ep idemiol. 128(1): 1-9, 1988.Newcombe, Robert G., "Towards a Reduction in PublicationBias," Br Med J 295: 656-659,.1987.

DeSerres, F. J. and T. Matsushima, "Meeting Report: Muta-genesis and Carcinogenesis by Nitropyrenes and Cancer Cheino-therapeutics," Mutat Res 164: 4, 1986.

Anonymous, "Lung Cancer in South China," Oncology Times,VIII (6): 33, 1986.

Mumford, J.N., et al., "Lung Cancer and Indoor Air Pollutionin Xuan Wei, China," Science 235: 217-220, 1987.

Gao, Y.T., et al., "Lung Cancer AxnongChinese Women,"J Cancer 40: 604-609, 1987.

116. Kauffmann, F., et al., "Adult Passive Smoking in the HomeEnvironment: A Risk Factor for Chronic Airflow Limitation,"AmEpideiniol 117(3): 269-280, 1983.

117. Adlkofer, F., et al., "Small-Airways Dysfunction in PassiveSmokers," New Engl J Med 303(7): 392, 1980.

- 69 -

118.

119.

120.

121.

Huber, G., "Small Airways Dysfunction in Passive Smokers,"New Engi J Med 303(7): 392, 1980.

Freedman, A., "Small-Airways Dysfunction in Passive Smok-ers," New Engl J Med 303(7): 393, 1980.

Aviado, D., "Small-Airways Dysfunction in Passive Smokers,"New Encil J Med 303(7): 393, 1980.

White, J. and H. Froeb, "Small-Airways Dysfunction in Pas-sive Smokers," New Enql J Med 303(7): 393-394, 1980. -

122. Lebowitz, M., "Forum: SecondHandSmoke -- How Dangerous?,"Presentation to American LungAssociation/American ThoracicSociety Annual Meeting, Los Angeles, California, May 17,1982.

123. Fountain, L.H., "White-Froeb Study Discredited by Scien-tists," U.S. Congressional Record, 97th Cong., 2nd Sess.,December 16, 1982 (Washington: Government Printing Office,1982) : E5252-5254.

124.

125.

Lebowitz, M., "Influence of Passive Smoking on PulmonaryFunction: A Survey," Prey Med 13(6): 645-655, 1984.

Kentner, M., et al., "The Influence of Passive Smoking onPulmonary Function - A Study of 1,351 Office Workers," PreyM. 13(6): 656-669, 1984.

126. Kentner, M. and D. Weltle, "Passive Tobacco Smoke Inhalationand Lung Function in Adults," Indoor and Ambient AirQuality , eds. R. Perry and P.W. Kirk (London: Selper Ltd.,1988) : 232-241.

127. Shephard, R., et al., "Responses of Exercising Subjects toAcute 'Passive' Cigarette Smoke Exposure," Environ Res19(1):279-291, 1979.

128.

129.

130.

Comstock, G., et al., "Respiratory Effects of HouseholdExposures to Tobacco Smoke and Gas Cooking," Am Rev RespirDis 124(2): 143-148, 1981.

Jones, J., et al., "Effects of Cooking Fuels on Lung Func-tion in Nonsmoking Women," Arch Environ Health 38(4):219-222, 1983.

Lebowitz, M., et al., "Family Aggregation of Pulmonary Func-tion Measurements," Am Rev Respir Dis 129: 8-11, 1984.

- 70 -

131.Schilling, R.,and Smoking in1977.

et al., "Lung Function, Respiratory Disease,Families," Am J E piderniol 106(4): 274-283,

132. Hosein, H. and P. Corey, "Domestic Air Pollution and Respi-ratory Function in a Group of Housewives," Can J PublicHealth 77: 44-50, 1986.

133.

134.

135.

136.

Dahins, T., et al., "Passive Smoking -- Effects on BronchialAsthma," Chest 80(5): 530-534, 1981.

Pilum, P., et al., "Physiological Effects of Acute PassiveExposure to Cigarette Smoke in Asthmatics," Fed Proc 36(3):606, 1977. Abstract.

Shephard, R., et al., "'Passive' Exposure of Asthmatic Sub-jects to Cigarette Smoke," Environ Res 20: 392-402, 1979.

Knight, A., and A. Breslin, "Passive Cigarette Smoking andPatients With Asthma," Med J Aust 142(3): 194-5, 1985.

137. Ing, A. and A. Breslin, "The Effect of Passive CigaretteSmoking on Asthmatic Patients," Proceedings, The ThoracicSociety of Australia, October, 1983: 543. Abstract.

138. Lebowitz, M., "The Effects of Environmental Tobacco SmokeExposure and Gas Stoves on Daily Peak Flow Rates in Asth-matic and Non-Asthmatic Families, " ETS-Environmental TobaccoSmoke: Report from a Workshop on Effects and ExposureLevels, Eur J Respir pj, eds. R. Rylander, et al., Suppl.133(65): 90-97, 1984.

139. Weideinann, H., et al., "Acute Effects of Passive Smoking onLung Function and Airway Reactivity in Asthmatic Subjects,"Chest 89(2): 180-185, 1986.

140. Hirayama, T., "Lung Cancer in Japan: Effects ofNutritionand Passive Smoking," Lun g Cancer: Causes and Prevention,eds. Mizell and Correa (Deerfield Beach: Verlag ChemieInternational, 1984): 175-195.

141.

142.

Helsing, K.J., et al., "Heart Disease Mortality inNonsmokers Living with Smokers," Am J E piderniol 127(5):915-922, 1988.

Svendsen, K.H., et al., "Effects of Passive Smoking in theMultiple Risk Factor Intervention Trial," Am J EpideTniol126(5): 783-795, 1987.

- 71 -

143. Garland, C., et al., "Effects of Passive Smoking on IschemicHeart Disease Mortality of Nonsmokers," Am J Epiderniol121(5):645-50, 1985.

144. Schievelbejn, H. and F. Richter, "The Influence of PassiveSmoking on the Cardiovascular System," Prey Med 13(6):626-644, 1984.

145. Aronow, W., "Effects of Passive Smoking on Angina Pectoris,"New Engi299(1): 21-24, 1978.

146. U.S. Public Health Service, "Chapter 11, Involuntary Smok-ing," Smoking and Health: A Report of the Surgeon General,Department of Health, Education and Welfare, Washington,DHEW Publication No. 79-50066: 1-41, 1979.

147. Fisher, E., Statement, U.S., Congress, House, Committee onAgriculture, Subcommittee on Tobacco, Effect of Smoking onNonsmokers, Hearing, 95th Cong., 2nd Sess., September, 7,1978 (Washington: Government Printing Office, 1978):35-40.

148. Shephard, R., The Risks of Passive Smoking (London: Croom-Helm Ltd., 1982): 73.

149. McNicol, N., and J. A. McM Turner, "Oxygen Uptake at theOnset of Angina Pectoris: Effects of Nicotine and CarbonMonoxide," din Sci 65(3): 24, 1983. Abstract.

150. Mintz, M., "FDA, Citing Phony Evidence, Bars Drug Tests byResearcher," The Washington Post, March 23, 1983.

151. Peterson, C., "EPA Probe Criticizes a Study Used in AirQuality Standard," The Washington Post, June 7, 1983.

152. Camacho, E., et al., "Pulmonary Symptoms and Pulmonary Func-tional Tests Among Children in Relation to the Area of Resi-dence,"-Eur Respir Dis 63(2): 165-166, 1982.

153. Lebowitz, M., et al., "The Effect of Passive Smoking onPulmonary Function in Children," Environ Intl 8: 371-373,1982.

154. Bland, M., et al., "Effect of Children's and Parents' Smok-ing on Respiratory Symptoms," Arch Dis Child 53(2): 100-105, 1978.

155. Bonham, G. and R. Wilson, "Children's Health in Familieswith Cigarette Smokers," Am Public Health 71(3): 290-293,1981.

- 72 -

156. Burchfiel, C., et al., "Passive Smoking in Childhood:Respiratory Conditions and Pulmonary Function in Tecumseh,Michigan," Am Rev Respir Dis 133(6): 966-73, 1986.

157.

158.

159.

160.

161.

Cameron, P. and D. Robertson, "Effect of Home EnvironmentTobacco Smoke on Family Health," J Appl Psychol 57(2):142-147, 1973.

Colley, J., "Respiratory Symptoms in Children and ParentalSmoking and Phlegm Production," Br Med J II:201-204,1974.

Colley, J., et al., "Influence of Passive Smoking and Par-ental Phlegm on Pneumonia and Bronchitis in Early Child-hood," Lancet II: 1031-1034, 1974.

Ekwo, E., et al., "Relationship of Parental Smoking and GasCooking to Respiratory Disease in Children," Chest 84(6):662-668, 1983.

Fergusson, D., et al., "Parental Smoking and Lower Respira-tory Illness in the First Three Years of Life," J EpidetniolCommunity Health 35(3): 180-184, 1981.

162. Fergusson, D., et al., "Parental Smoking and RespiratoryIllness During Early Childhood," Ped Pul 1(2):99-106,1985.

163. Ferris, B., et al., "The Six-City Study: Examples of Prob-lems in Analysis of Data," Environ Health Perspect 52:115-123, 1983.

164.

165.

166.

Gortmaker, S., et al., "Parental Smoking and the Risk ofChildhood Asthma," Am J Public Health 72(6): 574-579, 1982.

Harlap, S. and A. Davies, "Infant Admissions to Hospital andMaternal Smoking," Lancet I: 529-532, 1974.

Kraemer, N., et al., "Risk Factors for Persistent Middle EarEffusions: Otitis Media, Catarrh, Cigarette Smoke Exposure,and Atopy," JANA 249(8): 1022-1025, 1983.

167. Ogston, S.A., et al., "Association of Infant Alimentary andRespiratory Illness with Parental Smoking and OtherEnvironmental Factors," J Epidemiol Commun Health 41(1):21-25, 1987.

168. Schenker, M., et al., "Risk Factors for Childhood Respira-tory Disease: The Effect of Host Factors and Home Environ-

- 73 -

mental Exposures," Am Rev Res pir Dis 128: 1038-1043, 1983.

169. Tsimoyianis, G.V., et al., "Reduction in Pulmonary Functionand Increased Frequency of Cough Associated With PassiveSmoking in Teenage Athletes," Pediatrics 80(1): 32-36,1987.

170. Ware, J., et al., "Passive Smoking, Gas Cooking, and Respi-ratory Health of Children Living in Six Cities," Am RevRespir Dis 129: 366-374, 1984.

171. Binder, R., et al., "Importance of the Indoor Environmentin Air Pollution Exposure," Arch Environ Health 31(6):277279, 1976.

172. Bouhuys, A., "Lung Diseases in Rural and Urban Communities,"Report Sponsored by the Division of Lung Diseases, NationalHeart, Lung and Blood Institute, U.S. National Institutesof Health, pp. 77-89, June 10,. 1977.

173. Gardner, G., et al., "Effects of Social and Family Factorson Viral Respiratory Infection and Illness in the First Yearof Life," J Epidemiol Community Health 38: 42-48, 1984.

174. Harrington, W. and A. J. Krupnick, "Short-Term NitrogenDioxide Exposure and Acute Respiratory Disease in Children,"J Air Pollut Control Assoc 35: 1061-1067, 1985.

175. Holma, B. and 0. Winding, "Housing, Hygiene, and Health:A Study in Old Residential Areas in Copenhagen," ArchEnviron Health 32(2): 86-93, 1977.

176. Kerrebijn, J., et al., "Chronic Nonspecific RespiratoryDisease in Children, A Five Year Follow-Up Study," ActaPaediatr Scand, Suppi. 261: 11-72, 1977.

177. Lebowitz, N. and B. Burrows, "Respiratory Symptoms Relatedto Smoking Habits of Family Adults," Chest 69(1): 48-50,1976.

178. Love, G., et al., "The Incidence and Severity of AcuteRespiratory Illness in Families Exposed to Different Levelsof Air Pollution, New York Metropolitan Area, 1971-72," ArchEnviron Health 36(2) : 66-73, 1981.

179. Martin, C., et al., "Housing Conditions and Ill Health," Br294: 1125-1127, 1987.

180. Salzman, M.B., et al., "Passive Smoking and Croup," ArchOtolaryngol Head Neck Surg 113: 866-868, 1987.

- 74 -

181. Shy, C., et al., "The Chattanooga School Children Study:Effects of Community Exposure to Nitrogen Dioxide. II.Incidence of Acute Respiratory Illness," J Air PollutControl Assoc 20(9): 582-588, 1970.

182. Koo, L.C., et al., "A Comparison of the Prevalence ofRespiratory Illnesses Among Nonsmoking Mothers and TheirChildren in Japan and Hong Kong," Am Rev Res pir Dis 138:290-295, 1988.

183. Melia, R., et al., "Association Between Gas Cooking andRespiratory Disease in Children," Br Med J II: 149-152,1977.

184. Melia, R., et al., "Childhood Respiratory Illness and theHome Environment II. Association Between Respiratory Ill-ness and Nitrogen Dioxide, Temperature and Relative Humid-ity," mt J Epidemiol 11(2): 164-169, 1982.

185. Melia, R.J.W., et al., "Respiratory Illness and HomeEnvironment of Ethnic Groups," Br Med J 296: 1438-1441,1988.

186. Melia, R., et al., "The Relation Between Respiratory Illnessin Primary School Children and the Use of Gas for Cooking:I -- Results from a National Survey," mt j Epidemiol 8(4):333-338, 1979.

187.

188.

189.

Speizer, F. E., et al., "Respiratory Disease Rates and Pul-monary Function in Children Associated with NO 2 Exposure."Am Rev Respir Dis 121: 3-10, 1980.

Strachan, D. and R. Elton, "Relationship Between RespiratoryMorbidity in Children and the Home Environment," FamilyPractice 3(3): 137-142, 1986.

Watkins, C., et al., "Patterns of Respiratory Illness in theFirst Year of Life," Br Med J 293: 794-796, 1986.

190. Kerigan, A., et al., "A Three-Year Cohort Study of the Roleof Environmental Factors in the Respiratory Health ofChildren in Hamilton, Ontario: Epidemiologic Survey,Design, Methods, and Description of Cohort," Am Rev RespirDis 133: 987-993, 1986.

191. Goren, A.I. and S. Hellmann, "Prevalence of RespiratorySymptoms and Diseases in Schoolchildren Living in a Pollutedand in a Low Polluted Area in Israel," EnvironRes 45(1):28-37, 1988.

- 75 -

192. Fleming, D., et al., "Childhood Upper Respiratory TractInfections: To What Degree is Incidence Affected by DayCare Attendance?," Pediatrics 79(1): 55-60, 1987.

193. Berwick, N., et al., "Respiratory Illness in ChildrenExposed to Unvented Combustion Sources," Indoor Air: Radon,Passive Smoking , Particulates and Housin g Epidemiology,eds. B. Berglund, et al. 1984: 255-260.

194.

195.

196.

197.

Berkey, C. S., et al., "Indoor Air Pollution and PulmonaryFunction Growth in Preadolescent Children," Am J Epidemio].123(2): 250-260, 1986.

Chen, Yue and W. X. Li, "The Effect of Passive Smoking onChildren's Pulmonary Function in Shanghai," Am J PublicHealth 76(5): 515-518, 1986.

Corbo, G.M., et al., "Maternal Smoking and BronchialResponsiveness in Children," Am Rev Respir Dis 137(1):245, 1987.

Hasselbiad, V., et al., "Indoor Environmental Determinantsof Lung Function in Children," Am Rev Respir Dis 123(5):479-485, 1981.

198. Tager, I., et al., "Effect of Cigarette Smoking on the Pul-monary Function of Children and Adolescents," Am Rev RespirDis 131(5): 752-759, 1985.

199.

/200.

201.

Tager, I., et al., "Longitudinal Study of the Effects ofMaternal Smoking on Pulmonary Function in Children," jEngl.J Med 309(120): 699-703, 1983.

Tashkin, D., et al., "The UCLA Population Studies of ChronicObstructive Respiratory Disease. VII. Relationship BetweenParental Smoking and Children's Lung Function," Am RevRespir Dis . 129: 891-897, 1984.

Dodge, R., "The Effects of Indoor Pollution on Arizona Chil-dren," Arch Environ Health 37(3) ; 1982.

202. Hosein, R. and P. Corey, "Multivariate Analysis of NineIndoor Factors on FEy1 of Caucasian Children", AmericanThoracic Society Meeting, Miami Beach, FL, May 20-23, 1984.Abstract.

203. Mandi, A., et al., "Relationship Between Lung FunctionValues and Air Pollution Data, in Budapest School Children,"Pneumonologie 150(2/4): 217-225, 1974.

- 76 -

204. Witorsch, R. and P. Witorsch, "Maternal Smoking andPulmonary Performance in Children: A Critical Analysis,"Indoor Air 1 87, proceedings of the 4th InternationalConference on Indoor Air Qualit y and Climate, Berlin,August 17-21, 1987: 175-179.

205. Lebowitz, M., et al., "Respiratory Symptoms and Peak FlowAssociated with Indoor and Outdoor Air Pollutants in theSouthwest," J Air Pollut Control Assoc 35: 1154-1158,1985.

206. Justus, D. and D. Adams, "Evaluation of Tobacco Hypersen-sitivity Responses in the Mouse: A Potential Animal Modelfor Critical Study of Tobacco Allergy," mt Arch AllergyAppi Inununol 51(6): 687-695, 1976.

207. Lehrer, S., et al., "linmunochemical Properties of TobaccoSmoke and Leaf Components," Fed Proc 37(6): 1456, 1978.

208. Lehrer, S., et al., "Immunogenic Properties of TobaccoSmoke," J Allerg Clin Inununol 62(6): 368-370, 1978.

209. Taylor, G., "Tobacco Smoke Allergy -- Does It Exist?,"Environmental Tobacco Smoke Effects on the Nonsmoker:Report From a Workshop , ed. R. Rylander, Scand J Respirpj . Suppl. 91: 50-55, 1974.

210. Lehrer, S., et al., "mnununogenicity of Tobacco Smoke Com-ponents in Rabbits and Mice," Int Arch Allergy Appl Immunol62:16-22, 1980.

211. Lehrer, S., et al., "Tobacco Smoke Sensitivity: A Resultof Allergy?," Ann Allergy 56: 1-10, 1986.

212. Gleich, G., et al., "Allergy to Tobacco Antigens," J AllergClin Iminunol 63(3): 211-212, 1979.

213. Harkavy, J., "Tobacco Allergy in Cardiovascular Disease:A Review," Ann Allerq 26: 447-459, 1968.

214. Speer, F., "Tobacco and the Nonsmoker: A Study of Subjec-tive Symptoms," Arch Environ Health 16(3): 443-446, 1968.

215. Stankus, R.P. and S.B. Lehrer, "Inhalation Challenge Studieson Tobacco Smoke Sensitive Asthmatics," Indoor and AmbientAir Quality , eds. R. Perry and P.W. Kirk (London: SelperLtd., l988): 303-304.

- 77 -

216. Stahie, I. and L. Tibbling, "Tobaksallergi hos Patienter MedAsthma Bronchiale (Tobacco Allergy Among Patients with Bron-chial Asthma)," Lakartidningen 75(17): 1711-1713, 1978.

217. Tibbling, K., "Elicitation of Bronchial Asthma with TobaccoExtract," Abstracts of the 4th World Conference on Smokingand Health, Stockholm, Sweden, June 18-21, 1979.

218. Bylin, G., "Tobaksallergi -- finns den?(TobaccoAllergy --Does It Exist?)," Lakartidningen 77(16):1530-1532, 1980. Translation.

219. Salvaggio, J., Statement, U.S. Congress Committee on Agri-culture, Subcommittee on Tobacco, Effect of SmokingNon-

-smokers, Hearing, 95th Cong., 2nd Sess., September 7, 1978.(Washington: Government Printing Office, 1978): 46-49.

220. McDougall, J. and G. Gleich, "Tobacco Allergy -- Fact orFancy?," J Allerg din Immunol 57(3): 237, 1976. Abstract.

221. Lehrer, S., et al., "Immunological Evaluation of TobaccoSmoke 'Sensitive' Individuals by Skin, Radioallergosorbentand Precipitin Tests," ! Allerg din Immunol 67: 49, 1981.Abstract.

222..Melius, J., et al., "Indoor Air Quality -- the NIOSHExperience," Ann Am Conf Gov Ind Hyg 10: 3-7, 1984.

223. Sterling, T. and E. Sterling, "Investigations on the Effectof Regulating Smoking on Levels on Indoor Pollution and onthe Perception of Health and Comfort of Office Workers,"ETS -Environmental Tobacco Smoke: Re port from a Workshop onEffects and Exposure Levels, eds. R. Rylander, et al.,

Respir pj, Suppi. 133(65): 17-31, 1984.

224. Robertson, G., "Source, Nature and Symptomology on IndoorAir Pollutants," Indoor and Ambient Air Quality , eds. R.Perry and P.W. Kirk, (London: Selper Ltd., 1988): 311-319.

225. Sterling, T., et al., "Environmental Tobacco Smoke andIndoor Air Quality in Modern Office Work Environments,"Occup Med, 29(1): 57-62, 1987.

226. Pixnm, P., et al., "Physiological Effects of Acute PassiveExposure to Cigarette Smoke," Arch Environ Health 33(4):201-213, 1978.

227. Sterling, T., et al., "Indoor Byproduct Levels of TobaccoSmoke: A Critical Review of the Literature," J Air PollutControl Assoc 32(3): 250-259, 1982.

- 78 -

228. Nystrom, C. and C. Green, "Assessing the Impact of Environ-mental Tobacco Smoke on Indoor Air Quality: CurrentStatus," Presentation, ASHRAE Indoor Air Quality Conference,Atlanta, GA, April 21, 1986.

229. Hoepfner, I., et al., "Hydroxy-Phenanthrenes in the Urineof Non-smokers and Smokers," Toxicology Letters 35(1):67-71, 1987.

230.

231.

Scherer, G., et al., "Urinary Mutagenicity After ControlledExposure to Environmental Tobacco Smoke (ETS)," ToxicologyLetters 35(1): 135-140, 1987.

Sonnenfeld, G. and D. Wilson, "The Effect of Smoke Age andDilution on the Cytotoxicity of Sidestream (Passive) Smoke,"Toxicology Letters 35(1): 89-94, 1987.

232. Husgafvel-Pursiainen, K., "Sister-Chromatid Exchange andCell Proliferation in Cultured Lymphocytes of Passively andActively Smoking Restaurant Personnel," Mutation Research190: 211-215, 1987.

233. Adlkofer, F., et al., "Exposure of Hamsters and Rats toSidestream Smoke of Cigarettes: Preliminary Results of a90-Day-Inhalation Study," Indoor and Ambient Air quality,eds. R. Perry and P.W. Kirk (London: Selper Ltd., 1988):252-258.

234. Aviado, D.M., "Suspected Pulmonary Carcinogens inEnvironmental Tobacco Smoke," Indoor and Ambient AirQuality, eds. R. Perry and P.W. Kirk (London: Selper Ltd.,1988) : 141-146.

235. First, M., "Environmental Tobacco Smoke Measurements: Re-trospect and Prospect," ETS -Environmental Tobacco Smoke:Report from a Workshop on Effects and Ex posure Levels, eds.R. Rylander, et al.,. Eur J Res pir Qj, Suppi. 133(65):9-16, 1984.

236. Jenkins, R. and M. Guerin, "Analytical Chemical Methods forthe Detection of Environmental Tobacco Smoke Constituents,"ET5-Environmental Tobacco Smoke: Re port from a Workshop onEffects and Exposure Levels, eds. R. Rylander, et al.,J Respir Dis, Suppl. 133(65):. 33-46, 1984.

237. Russell, M., et al., "Absorption by Non-Smokers on CarbonMonoxide From Room Air Polluted by Tobacco Smoke," LancetI: 576-579, 1973.

- 79 -

238. Harke, H., et al., "ZUIn Problem des Passivrauchens II.Untersuchungen uber den Kohl enmonoxidgehalt der Luft liiiKraftfahrzeug durch das Rauchen von Zigaretten (The Problemof Passive Smoking. II. Investigations of Co Level in theAutomobile After Cigarette Smoking) ," Int Arch Arbeitsmed33(3): 207-220, 1974. Translation.

239. Anderson, G. and T. Dalhamn, "Halsoriskerna vid passiv rök-fling (Health Hazards from Passive Smoking) ," Lakartidningen70: 2833-2836, 1973. Translation.

240. Bridge, D. and M. Corn, "Contribution to the Assessment-ofExposure of Nonsmokers to Air Pollution From Cigarette andCigar Smoke in Occupied Spaces," Environ Res 5(2): 192-209,1972.

241. Duncan, D. and P. Greavey, "Passive Smoking and Uptake ofCarbon Monoxide in Flight Attendants," JAMA 251(20): 120-21, 1984.

242. Hugod, C., "Passive Smoking," Ugeskr Laeger 143(34): 2181-2184, 1981. Translation.

243. Jones, R. and R. Fagan, "Carboxyhemoglobin in Nonsmokers:A Mathematical Model," Arch Environ Health 30(4): 184-189,1975.

244. Klosterkötter, W. and E. Gono, "Zum Problem des Passiv-rauchens (The Problem of Passive Smoking),"Bakt Hy,Iabt Orig 162: 51-69, 1976. Translation.

245. Szadkowski, D., et al., "Kohlenmonoxidbelastung durch Pas-sivrauchen in BtlrorUumen (Body Burden of Carbon Monoxide

-.From Passive Smoking in Offices)," Inn Med 3(6): 310-313,1976.

246. Kirk, P.W., et al., "Environmental Tobacco Smoke in IndoorAir," indoor and Ambient Air Quality, eds. R. Perry andP.W. Kirk (London: Selper Ltd., 1988): 99-112.

247. Hutcheon, D., Statement, State of New Jersey Public HealthCouncil, Public Hearing on Regulation of Smoking in CertainPublic Places, Trenton, New Jersey, December 12, 1978.

248. Stewart, R., "The Effects of Low Concentrations of CarbonMonoxide in Man," Environmental Tobacco Smoke Effects onNonsmoker: Report from a Workshop, ed. R. Rylander, ScandJ Respir Dis, Suppi. 91: 56-62, 1974.

- 80 -

249. Wade, W., "A Study of Indoor Air Quality," J Air PollutControl Assoc 25(9): 933-939, 1975.

250. Yocom, J., "Indoor-Outdoor Air Quality Relationships," J AirPollut Control Assoc 32(5): 500-520, 1982.

251. Cox, B. D. and M. J. Whichelow, "Carbon Monoxide Levels inthe Breath of Smokers and Nonsmokers: Effect of DomesticHeating Systems," J E pideiniol Community Health 39: 75-78,1985.

252. Girman, J. and G. Traynor, "Indoor Concentrations,"Pollut Control Assoc 33(2): 89, 1983.

253. Repace, J. and A. Lowrey, "Indoor Air Pollution, TobaccoSmoke, and Public Health," Science 208: 464-472, 1980.

254. Weber, A. and T. Fischer, "Passive Smoking at Work," mtArch Occup Environ Health 47: 209-221, 1980.

255. Gieseke, J., et al., "Collection and Characterization ofAirborne Particulate Matter in Buildings," ASHRAE Transac-tions 84(1): 572-589, 1978.

256. Spengler, J., et al., "Long-Term Measurements of RespirableSulfates and Particles Inside and Outside Homes," AtinosEnviron 15: 23-30, 1981.

257. Press Release, "Study of Air Quality in 100 N.Y.C.Restaurants, Offices Shows Tobacco Smoke is InsignificantFactor," The (U.S.) Tobacco Institute, December 10, 1986.

258. IT Corporation, Final Re port: Environmental Tobacco SmokeSurvey , Dallas, Texas, Submitted to Tobacco Institute,Washington, D.C., August 4, 1987.

259. Carson, J.R. and C.A. Erikson, "Results from Survey ofEnvironmental Tobacco Smoke in Offices inOttawa, Ontario,"Environ Technol Letters 9: 501-508, 1988.

260. Sterling, T., "ETS Concentrations Under Different Conditionsof Ventilation and Smoking Regulation," Indoor and AmbientAir Quality, eds. R. Perry and P.W. Kirk (London: SelperLtd., 1988): 89-98.

261. Ogden, M.W. and K.C. Maiolo, "Collection and Analysis ofSolanesol as a Tracer of En'Ifronmenta1 Tobacco Smoke (ETS) ,"Indoor and Ambient Air Quality, eds. R. Perry and P.W. Kirk(London: Selper Ltd., 1988): 77-88.

- 81 -

262. Proctor, C.J., "The Analysis of the Contribution of ETS toIndoor Air," Indoor and Ambient Air Quality , eds. R. Perryand P.W. Kirk (London: Selper Ltd., 1988): 57-66.

263. Eatough,D.J., et al., "Assessing Exposure to EnvironmentalTobacco Smoke," Indoor and Ambient Air Quality, eds. R.Perry and P.W. Kirk (London: Selper Ltd., 1988): 131-140.

264. First, M., "Constituents of Sidestream and Mainstream Tobac-co Smoke and Markers to Quantify Exposure to Them," IndoorAir and Human Health (Chelsea, Michigan: Lewis Publishers,1985): 195-203.

265. Sterling, T., "Discussion -- Tobacco Smoke, Ventilation, andIndoor Air Quality," ASHRAE Transactions 88(1): 911-912,1982.

266. Bouhuys, A., et al., "Do Present Levels of Air PollutionOutdoors Affect Respiratory Health?," Nature 276(5687):466-471, 1978.

267. Badre, R., et al., "Pollution Atmospherique par la Fumee deTabac (Atmospheric Pollution by Smoking) ,"fl Pharm36(9-10): 443-452, 1978.. Translation.

268. Harke, H., et al., "Passivrauchen: Abhngigkeit der Konzen-stration von Rauchinhaltsstoffen in der Luft verschiedengrosser Raumer von der Zahi der verrauchten Zigaretten undder Zeit (Passive Smoking: Concentration of Smoke Constitu-ents in the Air of Large and Small Rooms as a Function ofNumber of Cigarettes Smoked and Time) ," Int Arch Arbeitsmed29(4): 323-339, 1972. Translation.

269. Hinds, W. and M. First, "Concentrations of Nicotine andTobacco Smoke in Public Places," New En gl !292(16):844-845, 1975.

270. Jenkins, R.A. et al., "Development and Application of aThermal Desorption-Based Method for the Determination ofNicotine in Indoor Environments," Indoor and Ambient AirQuality, eds. R. Perry and P.W. Kirk (London: Selper Ltd.,1988) : 557-566.

271.

272.

Murumatsu, M., et al., "Estimation of Personal Exposure toTobacco Smoke with a Newly Developed Nicotine Personal Moni-tor," Environ Res 35: 218-227, 1984.

Muramatsu, M., et al., "Estimation of Personal Exposure toAmbient Nicotine in Daily Environment," Arch Occup EnvironHealth 59: 545-550, 1987.

- 82 -

273. Hinds, W., "The Lung and the Environment," Semin Res pir Med1(3) :197-210, 1980.

274. Harke, H. and A. Bleichert, "Zum Problem des Passivrauchens(On the Problem of Passive Smoking)," Int Arch Arbeitsmed29: 312-333, 1972. Translation.

275. Ohkubo, C., "Some Acute Cardiopulmonary Effects of Main-stream and Sidestream Cigarette Smoke in Man," Prey11:173-186, 1982.

276. Foliart, D., et al., "Passive Absorption of Nicotine inAirline Flight Attendants," New Engl J Med 308(18): 1105,1983.

277. Oldaker, G.B. and F.C. Conrad, "Estimation of Effect ofEnvironmental Tobacco Smoke on Air Quality Within PassengerCabins of Commercial Aircraft," Environ Sci Technol 21:994-999, 1987.

278. Ember, L., "Nitrosamines: Assessing the Relative Risk,"Chem Eng News 58(13): 20-26, 1980.

279. Lehnert, G., "Schlusswort: Zum Bereich des Spekulativen(Conclusion: In the Realm of Speculation)," Munch med Wschr124(4): 19-20, 1982. Translation.

280. Schievelbein, H., "Are There Really Important New FindingsAbout Passive Smoking?," Oeff Gesundheitswes 44(0): 454-456, 1982. Translation.

281. Brunnemann, K., et al., "Assessment of Carcinogenic VolatileN-Nitrosamines in Tobacco and in Mainstream and SidestreamSmoke from Cigarettes," Cancer Res 37: 3218-3222, 1977.

282. Brunnemann, K., et al., "The Influence of Tobacco Smoke onIndoor Atmospheres. II. Volatile and Tobacco SpecificNitrosam jnes in Main- and Sidestream Smoke and Their Con-tribution to Indoor Pollution," 4th Joint Conference onSensing of Environmental Pollutants, 1977 (Washington, D.C.American Chemical Society, 1978): 876-880.

283. Brunnexnann, K. and D. Hoffmann, "Chemical Studies on TobaccoSmoke. LIX. Analysis of Volatile Nitrosamines in TobaccoSmoke and Polluted Indoor Environments," EnvironmentalAspects of. N-Nitroso Compounds, eds. E. A. Walker, et al.(Lyon: IARC Scientific Publications, 1978): 343-355.

- 83 -

284. Stehlik, G., et al., "Concentration of Dimethylnitrosaminein the Air of Smoke-Filled Rooms," Ecotoxicol Environ Safety6: 495-500, 1982.

285. Good, B., et al., "Effect of Cigarette Smoking on Residen-tial NO2 Levels," Environ Intl 8: 167-175, 1982.

286. Hugod, C., et al., "Exposure of Passive Smokers to TobaccoSmoke Constituents," lilt Arch Occu p Environ Health 42:21-29, 1978.

287. Holzer, G., et al., "Gas Chromatographic-Mass SpectrometricEvaluation of Exhaled Tobacco Smoke," J Chroinatogr 126:771-785, 1976.

288. Sega, K. and M. Fugas, "Nitrogen Dioxide Concentrations inResidences," Indoor and Ambient Air quality , eds. R. Perryand P.W. Kirk (London: Selper Ltd., 1988): 493-496.

289. Bayer, C. W. and M. S. Black, "Thermal Desorption/GasChromatographic/Mass Spectrometric Analysis of VolatileOrganic Compounds in the Offices of Smokers and Nonsmokers,"Biomed Environ Mass S pect 14(8): 363-367, 1987.

290. Hollowell, C. and R. Miksch, "Sources and Concentrations ofOrganic Compounds in Indoor Environments," Bull jy Acad Med57(10): 962-977, 1981.

291. Matsukura, S., et al., "Effects of Environmental TobaccoSmoke on Urinary Cotinine Excretion in Nonsmokers",Engl J Med 311(13): 828-832, 1984.

292. Greenberg, Robert A., et al., "Measuring the Exposure ofInfants to Tobacco Smoke," New Encil J Med 310(17), 1075-1078, 1984.

293. Hoffmann, Dietrich, et al., "Tobacco Sidestream Smoke:Uptake by Nonsmokers," Prey Med 13: 608-617, 1984.

294. Jarvis, M. J., et al., "Passive Exposure to Tobacco Smoke:Saliva Cotinine Concentrations in a Representative Popula-tion Sample of Non-Smoking School Children," Br Med J 291:

-927-929, 1985.

295. Jarvis, Martin J. and Michael A. H. Russell, "Measurement-and Estimation of Smoke Dosage to Non-Smokers From Environ-

mental Tobacco Smoke," Environmental Tobacco Smoke Reportfrom a Workshop on Effects and Exposure Levels, eds. R.Rylander, et al., Eur J. Res pir Dis, Suppi. 133 (65): 68-75, 1984.

- 84 -

296.Wald, Nicholas J., et al., "Urinary Cotinine as Marker ofBreathing Other People's Tobacco Smoke," Lancet I, 230-231,1984.

297.Jarvis, M., et al., "Passive Smoking in Adolescents:One-Year Stability of Exposure in the Home," The Lancet I:1324-1325, 1987.

298, Benowitz, Neal L., "The Use of Biologic Fluid Samples InAssessing Tobacco Smoke Consumption," Measurement in theAnalysis and Treatment of Smokin g Behavior, U.S. NationalInstitute on Drug Abuse Research Monograph 48: 6-25, 1983.

299. Benowitz, Neal L., et al., "Cotinine Disposition and Ef -fects," Clin Pharmacol Ther34(5): 604-611, 1983.

300. Johnson, L. C., et al., "Passive Smoking Under ControlledConditions," mt Arch Occup Environ Health 56, 1985: pp.99-110.

301.

302.

303.

Letzel, H., et al., "Measuring Problems in Estimating theExposure to Passive Smoking Using the Excretion ofCotinine," Toxicolo gy Letters 35(1): 35-44, 1987.

Biber, A., et al., "Determination of Nicotine and Cotininein Human Serum and Urine: An Interlaboratory Study,"Toxicology Letters 35(1): 45-52, 1987.

Schwartz, S., et al., "Mathematical Modellingof Nicotineand Cotinine as Biological Markers of Environmental TobaccoSmoke Exposure," Toxicology Letters 35(1): 53-58, 1987.

304. Baiter, N.J., et al., "Application of PhysiologicalPharmacokinetic Modeling to the Design of Human ExposureStudies with Environmental Tobacco Smoke," Indoor andAmbient Air quality , eds. R. Perry and P.W. Kirk (London:Seiper Ltd., 1988): 179-188.

305.

306.

307.

Castro, A. and N. Monji, "Dietary Nicotine and Its Signifi-cance in Studies on Tobacco Smoking," Bio Arch, 2(2): 91-97, 1986.

Adlkofer, F., et al., "Passive Smoking," N Engl312(11): 719-720, 1984.

Pittenger, D., "Passive Smoking," N En gl J Med 312(11):720, 1984.

- 85 -

308. U.S. Civil Aeronautics Board, "Airline Consumer ComplaintsDown 31% in First Half of 1982," CAB News, September 20,1982.

309. Cohen, R., Statement, U.S., Congress, House, Committee onAgriculture, Subcommittee on Tobacco, Effect of Smokin g onNonsmokers, Hearing, 95th Cong., 2nd Sess., September 7,1978 (Washington: Government Printing Office, 1978):69-75.

310. Response Analysis Corporation, Overall Re port: Smoking andProductivity in the Work Place, April, 1984.

311. Feinhandler, S., Statement, U.S., Congress, House, Committeeon Agriculture, Subcommittee on Tobacco, Effect of Smokingon Nonsmokers, Hearing, 95th Cong., 2nd Sess., September 7,1978 (Washington: Government Printing Office, 1978):56-63.

312. Lehrer, S., et al., "Tobacco Smoke 'Sensitivity' -- Is Therean Immunologic Basis?," Q Allergy din Iminunol 73(2): 240-245, 1984.

313. Fettner, A., "Where There's Smoke There's Ire--The New'Passive Smoking' Rules:Sidestepping Issues byScapegoating Smokers," Voice 5-7, March 30, 1987.

314. Jones, W., "Inconclusive Evidence on the Harmful Effects ofSmoking," U.S. Congressional Record - Extensions of Remarks,February 18, 1987: E489-E490.

315. Press Release, "Coalition of Labor Unions React to SurgeonGeneral's Report on Smoking," (December 16, 1986).

316. Tobacco Institute, Tobacco Smoke and the Nonsmoker:Scientific Integrity At the Crossroads, (Washington, D.C.:The Tobacco Institute, 1986).

317. Ashe, R. and D. Vaughn, "Smoking in the Workplace:AManagement Perspective, Emp 3j Law ! 11(3):383-406,1985.

318. "Court Upholds Appeal Ruling on ETS," Tobacco Review 114(6):12, 1987.

319. "Summary of Court Judgement Relating to Rights forAnti-Smoking," District Court (Tokyo, Japan) Ruling, 1-10,1987.

- 86 -

320. U.S. Civil Aeronautics Board, "Smoking Aboard Aircraft,Final Rule," Federal Re gister 49(120):Part V, 14 CFR,Part 252 1 June 20, 1984.

321. Committee on Airliner Cabin Air Quality, Board on Environ-mental Studies and Toxicology, Commission on Life Sciences,National Research Council, The Airliner Cabin Environment:Air Quality and Safety, (Washington: National AcademyPress, 1986).

322. U.S. Department of Transportation, Airline Cabin AiQuality : Report to Congress, (U.S. DOT, February, 1987).

323. U.S. Department of Transportation, Solicitation, Offer andAward, August 5, 1988.

324. Holcomb, L.C., "Impact of Environmental Tobacco Smoke onAirline Cabin Air Quality," Environ Technol Letters 9: 509-514, 1988.

325. WHO/IARC, Evaluation of the Carcinogenic Risk of Chemicalsto Humans: Tobacco Smoking, IARC Monograph Series 38, Lyon,1985.

326. Roe, F., "Passive Smoking", Lancet I: 888, 1988.

327. Zethraeus, S., "Arbetsskade - försäkringen at förmãnhig mendoinen bevisor inte att passiv rökning ger cancer [Occupa-tional Injury Insurance is Beneficial but the JudgmentDoes Not Prove that Passive Smoking Leads to Cancer"], jgAvtal 1: 31-32, 1986. Translation.

328. "No Jobs If You Smoke: Bosses," The Daily (Sydney,Australia) Telegraph, July 22, 1988.

329. Press Release, "Sean Carroll Settlement," Tobacco Instituteof Australia, July 21, 1988.

330. Crawford, W.A., "Health Effects of Passive Smoking in theWorkplace," Indoor and Ambient Air Quality, eds. R. Perryand P.W. Kirk (London: Selper Ltd., 1988): 203-210.

331. Littlechild, S. and J. Wiseman, "Principles of Public PolicyRelevant to Smoking," Policy Studies, Volume 4, Part 3, ed.Sir Charles Carter (London, Thetford Press, 1984): 1-14.

- 87 -

RESPONSE TOTHE NEW ZEALAND DEPAJTMENT OF HEALTH PROPOSAL

CREATING SMOKE FREE INDOOR ENVIRONMENTSOPTIONS FOR ACTION

LARRY C. HOLCOMB, Ph. D.

HOLCO.MBENVIRONMENTAL SERVICES17375 Garfield Road/Olivet, Michigan 49076

616 - 763-9217 IJus!ness/Residenceor 616 - 966-7329 Answering Service

Larry C. Holcomb, Ph.D.Environmental RiskAss esmentEnvironmental AuditsHazardous MterllManagementWildlife ManagementR,jource Recovery

A RESPONSE TO

THE NEW ZEALAND DEPARTMENT OF HEALTH PROPOSAL

"CREATING SMOKEFREE INDOOR ENVIRONMENTS

OPTIONS FOR ACTIONS"

This paper is being written in response to a proposal written by theNew Zealand Department of Health titled "Creating Smokefree IndoorEnvironments, Options for Action". The proposal discusses what theNew Zealand Department of Health considers to be health risks dueto peoples' exposure to Environmental Tobacco Smoke (ETS) andproposes legislation to help control or eliminate that exposure.

Because of the health problems ascribed publicly to active smoking,it has been assumed that the same types of problems are associatedwith ETS exposure. As a result, public demand in some areas forlegislated control of ETS exposure has outdistanced the scientificcommunities' ability to provide research to confirm or refute theirfears. Because of this demand, preliminary and inconclusive dataare being treated as definite proof that health problems do occur asa result of exposure to ETS.

The New Zealand proposal cites a 1987 poll that very effectivelydemonstrates the public demand in some areas for control over ETSexposure. The proposal then goes on to describe the popularlyassumed health hazards associated with ETS exposure and states thatthe only way to protect the public from ETS is to regulate smokingby means of restrictive legislation.

This paper will assess the health risks reported in the proposal andprovide a more complete view of the data that exists in relation toETS exposure and health effects. It will also discuss healthy indoorair in a more complete manner, not focusing on just one aspect of it,and provide alternative suggestions to the ones given in theproposal.

Chemistry of ETS

The statements made comparing the chemistry of Sidestream Smoke(SS) to Mainstream Smoke (MS) are not the same as comparing ETSwith MS. These statements do not take into consideration the factthat when SS mixes with exhaled MS and ambient air to form ETSsubstantial dilution and several chemical changes take place.

There is considerable dilution of both the SS and exhaled MS whenETS is formed. As a result the figures which show that somecompounds in SS are present in much greater concentrations than inMS become meaningless when one is discussing ETS. There are fewstudies which take the dilution effects of the air in the room or thesmoke components' tendency to adhere to surfaces into consideration(Baiter, Kilpatrick and Witorsch: 1986). One that has shows a rapiddecay of the toxic effects of ETS within a short time (Sonnenfeldand Griffith: 1987).

Other Sources of Toxic Chemicals

Exposure to chemicals in our society is unavoidable. With theexception of nicotine and its derivatives, all of the compoundsmeasured in tobacco smoke are also produced by other sources(Baiter, Kilpatrick and Witorsch: 1986).Some examples of theseare

Carbon Monoxide - This is also found in vehicle exhaust and can beproduced by many sources of indoor combustion such as spaceheaters and gas stoves.

Sulfur Dioxide and Nitrogen Oxides - These are formed wheneverfossil fuels are burned and can come from vehicle exhaust and anynearby coal-fired power plants.

Vinyl Chloride, Formaldehyde and other organic vapors - Organicsolvents are very common in households and offices. Organic vaporscan come from the use of adhesives, cleaning fluids, or paints.They can be given off from insulation and carpeting.

Radionuclides - This type of contaminant has recently beendocumented by the U.S. E.P.A. to occur naturally in virtually everyhome in the United States in the form of radon gas.

Hydrogen Cyanide - Cyanide exposure is so common from severalfood sources that researchers have been unable to documentexposure to ETS by attempting to measure thiocyanate levels in theblood.

Respiribie Suspended Particulates (RSP) - Particles suspended in theair that are small enough to be inhaled can come from insulation(asbestos and fiberglass), shedding from fabrics and the build up ofdust and bacteria in ventilation ductwork (Robertson: 1988).

Ammonia - Ammonia is a commonly used ingredient in cleaningcompounds. It is also used in several types of printing operations.

Eliminating ETS will not eliminate peoples' exposure to many of thechemicals found in ETS. It will be shown later in this paper thatthere are more effective ways to protect people from the healthaffects of breathing contaminated air.

2

Chemical Exposure and Toxic Effects

One of the steps necessary to prove that a chemical has caused acertain effect in a person is to determine that the person wasactually physiologically exposed to the chemical. In the case of ETSthe only biological marker that has been found to demonstrateexposure to tobacco products is Nicotine or its biological derivativeCotinine. While measuring Cotinine levels in a person's body fluidswill demonstrate that that person was exposed to tobacco smoke itwill not determine that that person absorbed any of the compoundsof concern in tobacco smoke. Cotinine levels in body fluids have notbeen associated with any other tobacco products in body fluids.

Another drawback in using Cotinine levels to demonstrate exposureto ETS and thus extrapolate to a health risk is the length of timethat Cotinine stays in the body. Cotinine has a half life of about 20hours in the bloodstream (NRC; 1986). Most of the health effects ofconcern are the type which develop over a period of years.Measuring cotinine in a persons blood at a given point in time willnot show that he was exposed over a long enough period to developadverse health effects due to that exposure.

Appendix II of the proposal discusses carcinogens and sidestreamsmoke. It suggests in the penultimate paragraph that passivesmokers exposed to tobacco smoke for 20 hours or more per week areexposed to at least 2% of the smoke ingested by active smokers.However, Arundel,Sterling and Weinkam (1988) reportedconsiderably different results using retention figures of 11% forneversmokers and 80% for active smokers. They calculated 310mg/day Respirible Suspended Particulate (RSP) retention for activesmokers versus 0.07 mg/day for male neversmokers and 0.03 mg/dayfor female neversmokers. This would lead to a 4,429 fold reductionfor the passive smoker compared to an active smoker or 0.02% of thesmoke ingested by an active smoker.

One has to be careful not to exaggerate the exposure to theparticulate component from ETS, that is the portion of totalrespirible particulates. There have been two studies done by Healthand Welfare Canada (1985a; 1985b) which measured the amount ofRSP generated in office buildings due to smoking. The first studydetermined that under conditions of normal ventilation andoccupancy, smoking contributed about 7 ug/m3 of the total RSP inthe building. The second determined that recirculated air fromdesignated smoking areas contributes less than 8 ug/m3 RSP in thenonsmoking areas of the same building. A similar study by Sterling(1988) reported that nonsmoking areas of a building which receivedrecirculated air from smoking areas in the same building had RSPlevels indistinguishable from background levels.

Holcomb (1988) reported on the impact of ETS in airlines and showedthat the particulates in smoking sections of commercial aircraft havebeen reported as a mean of 30-40 ug/m 3 in the smoking sections and12-14 ug/m3 in the nonsmoking sections. This points out thatseparation of smoking and nonsmoking passengers is very effectiveand that if ventilation of the aircraft is reasonable, indoor airquality can be protected.

Epidemiological Studies

There are no biomarkers that have been found to demonstrateexposure to ETS over the time frame required for chronic healtheffects to develop. As a result researchers have been required touse epidemiological studies to try to show adverse health effects dueto exposure to ETS. By their very nature, epidemiological studieshave weaknesses in them which make it very difficult if notimpossible to demonstrate a cause and effect relationship betweenETS exposure and disease.

ETS exposure has been determined by questionnaire inepidemiological studies. Since there is no way to independentlyconfirm the answers provided, there is room for inaccuracy in thereporting that will not be discovered or corrected for. In twostudies (Garfinkel, Auerbach and Joubert; 1985; Coultas, Samet andPeake; 1987) it was shown that the quantity of exposure to ETS thatwas reported could vary with both the person who was asked andwhen they were asked.

People filling out questionnaires regarding ETS exposure are oftenrequired to remember how much they were exposed to many yearsago. (In the case of cancer, it is believed that relevant exposurescould have occurred as much as 20 years prior to the onset of thedisease.) This type of documentation of exposure does not lead toaccurate quantification of actual exposure.

Most of the studies determine ETS exposure by asking whether ornot a spouse or parent was a smoker. This only documents whetheror not a spouse or parent was a smoker not whether or not theperson of interest was actually exposed to ETS. Friedman et al(1983) has shown that whether or not a woman reports that herhusband is a smoker does not correlate well with whether or not shereports being exposed to ETS.

There are also several confounding factors associated withhouseholds where smoking occurs. Many of these can affect thehealth of the people living in that household and are difficult toseparate from ETS exposure. These include differences in lifestyle,diet and nutrition, socio economic factors, and health attitudes andhabits.

The U. S. Surgeon General's Report

One of the major sources of information cited when discussing thehealth effects of ETS exposure is the 1986 U.S. Surgeon General'sreport on "The Health Consequences of Involuntary Smoking".Unfortunately what is often cited are statements made in thesummary of the report. These statements are in direct contradictionto the conclusions made in the body of the report. Below areexcerpts from the body of the report which state the conclusionsdrawn regarding ETS exposure and various health effects. Theseexcerpts are very brief but they are not taken out of context. Onecan go to the page listed in the excerpt and read the entireconclusion.

4

P.60Acute Respiratory Illness

"There are no studies -".

P.60Pulmonary Function

Kentner et al (1984) "-S found no influence of involuntarysmoking on pulmonary function.

P.60Cough, Phlegm and Wheezing

Schilling et al (1977) found no relationship of occurrenceof cough, phlegm and wheezing to smoking habits of thespouse. Schenker et al (1982) confirmed this.

P.63Bronchoconstriction

When small changes in air flow were noted in nonsmokersexposed to ETS "- it is unlikely that this change in airflow, per se, results in symptoms."

P. 65Asthmatics

"Studies of large numbers of individuals - will benecessary to address -" ETS effects on asthmatics.

"Acute exposure in a chamber may not adequatelyrepresent exposure in the general environment. Biasesin observation and the selection of subjects and thesubjects' own expectations may account for widelydivergent results."

P.65Ear, Nose and Throat

"- no studies -"

P.104Cancers other than Lung Cancer

Several studies have investigated the relationship(epidemiological) between ETS exposure and cancersother than lung cancer. The studies are plagued byinadequate methodology, failure to investigateconfounding factors and lack of biological plausibility.Due to the findings, the Surgeon General's studyconcluded: "There are at present insufficient data toadequately evaluate the role of involuntary smoking inadult cancers -".

P.106-107 Cardiovascular Diseases

More detailed research "- needed before an effect ofinvoluntary smoking on the etiology of cardiovasculardisease can be established".

ETS and Respiratory Diseases In Children

There has been some documented increase of respiratory diseases inchildren of parents who smoke. These studies have all beenepidemiological and as a result may include some confounding factorsthat are being measured instead of simply ETS exposure. In studiesof young children, parents may be more likely to report respiratorysymptoms that they themselves show. Parents who smoke may haverespiratory illnesses more frequently than those who do not and maypass those respiratory infections on to their children. In studiesthat measured the symptoms of older children there is the problem ofmisclassifying an actively smoking youth as a nonsmoker and thusbiasing the results (Surgeon General; 1986; NRC; 1986).

ETS and Cancer

The disease most often mentioned in conjunction with ETS exposureis lung cancer. It is probably the most frequently studied diseaseand the most hotly debated. All of the studies associating anincreased risk of lung cancer to ETS exposure are epidemiological.They all have the weaknesses that were previously listed inassociation with epidemiological studies. They all report results thatare not reproducible in other studies.

The overall relative risk for a nonsmoker to contract lung cancerdue to ETS exposure that is most often cited is 1.34 (NRC; 1986).This translates into a 34% increased risk. This number was obtainedby taking thirteen epidemiological studies, only three of whichshowed a statistically significant increase in risk, and pooling all thedata to obtain one result. Uberia (1987) states in his review of theepidemiological studies that "All studies with positive associations canjust as well be explained by change, bias, confounding ormisclassification. Such poorly conducted and inconclusive studiescannot be added or pooled to get convincing evidence."

The IARC (1985) has stated that "the observations made onnonsmokers so far are compatible with either an increased risk frompassive smoking or an absence of risk". In the opinion of anorganisation whose job is to study the causes of cancer, the jury isstill out.

While epidemiological studies have not been able to show a consistentrelationship between ETS exposure and lung cancer there is someevidence to show that ETS exposure does not cause changes in thetissues of the respiratory tract that are associated with smoking.Garfinkel (1981), while reviewing microscope slides of lung cancer,found none of the cell changes that occur in the tissues of smokers.A study exposing hamsters to high levels of SS was run by Adlkoferet al (1988). At the end of the study the only way to find any cellchanges caused by the SS exposure was to examine tissue samplesunder an electron microscope. Slight tissue changes found in thisway were all reversible.

Conclusions Regarding Health Effects

When one steps back and looks at the evidence supporting the beliefthat ETS exposure is a significant health risk it can be observedthat the proof is just not there. The belief that ETS exposurecauses adverse health effects is fuelled primarily by the fact thatsome people expect exposure to ETS to cause the same symptomsattributed to direct smoking. They then pick out those pieces ofdata that confirm their fears and ignore those that would modifytheir conclusions.

Chemical Exposure, Indoor Air Quality and Ventilation

As stated earlier in this paper, exposure to chemicals has become afact of life in our society. In poorly ventilated buildings the amountof chemicals one is exposed to can be several times as high asimmediately outside the same building. ACVA, an internationalcompany specialising in indoor air quality has found that poor indoorair quality can dramatically effect the health of the people workingin that environment.

Though there are several different sources of indoor aircontaminants, ETS is the only one that is visible. This makes it aneasy scapegoat and can often be cited as the cause of poor qualityindoor air when it is only a symptom of the problem. Both ACVA(Robertson; 1988) and NIOSH (Melius et al; 1983) after reviewing airpollution problems in several hundred buildings concluded thatproblems arising from indoor air quality are caused by cigarettesmoke in only 2% - 4% of the cases. The rest of the time theproblem has been inadequate ventilation or air contamination fromother sources.

Figures compiled by ACVA suggest that the costs of improvingventilation in a building to provide good quality air are more thanoffset by the reduction in absenteeism that is achieved.

If one takes a critical look at the literature available on the healtheffects of ETS and on indoor air quality several conclusions becomeevident.

1. It is not possible to extrapolate from SS or MS figures todetermine the concentrations or components of ETS. The fewstudies that have been done on the chemistry of ETS suggestthat there are significant chemical differences between ETS andSS or MS.

2. There are other sources of chemical contamination in indoorair. Many of these produce compounds which are also foundin tobacco smoke.

3. Actual absorption of tobacco smoke components due to exposureto ETS is very poorly documented. There is currently no wayto associate any observed health effects with exposure to ETSin this manner.

7

8

4. Though it is tempting to try to calculate how much ETS apassive smoker is exposed to in cigarette equivalents this hasled to grossly exaggerated numbers. Recent studies havedetermined that in terms of particulates, passive smokers areexposed to 1/4,429, or about 0.02% of the amount of smokethat active smokers are exposed to. Other studies have shownthat active smoking in an area actually produces as little as 7uglm3 to 20 ug/m of the total RSP in the air and that thisexposure can be effectively controlled by separating smokingand nonsmoking areas.

5. Due to the lack of evidence regarding biological exposure toETS, studies regarding the health effects of ETS exposureneed to rely on epidemiological evidence. These types ofstudies, by their very nature, have problems in them thatmake it impossible to state a cause and effect relationshipbetween ETS exposure and health risks. The US SurgeonGeneral, in the body of his report does not conclude that ETSexposure affects the frequency of any diseases exceptchildhood respiratory infections and lung cancer.

6. A positive relation between parental smoking and childhoodrespiratory infections has been found but both the US SurgeonGeneral's report and the NRC report state that there areconfounding factors in these studies which cannot be correctedfor.

7. It has become very popular lately to cite a 34% increase in riskfor lung cancer due to ETS exposure. However this numberhas come from combining the results of several differentstudies. All of these studies have weaknesses of their ownand only a few of them actually showed a statisticallysignificant increase in overall risk. One does not come upwith a good figure by combining several questionable figures.

8. Studies of several hundred buildings by ACVA and NIOSHhave determined that though air quality is a problem in manybuildings, ETS is the cause of the problem in a very smallminority of the cases.Banning smoking in a building toimprove air quality does not cure the problem. It onlyremoves the symptom. The initial costs to improve a building'sventilation and provide good quality air have been shown toprovide a very rapid return on investment.

If one is truly interested in achieving a healthy indoor environmentone must look at all the aspects of clean indoor air and determinethe best solution for the overall problem. Legislating smoking inpublic places and thus trying to control exposure to ETS will notsolve a problem with indoor air quality and would provide negligiblehealth benefits if any.

LITERATURE CITED

Adikofer F., Scherer G., Wenel-Hartung R., Brune H. and ThomasC. 1988. Exposure of Hamsters and Rats to Sidestream Smoke ofCigarettes: Preliminary results of a 90-Day Inhalation Study. Proc.Indoor Amb. Air Qual. Conf. London, pp. 252-258.

Arundel A., Sterling T. and Weinkam J. 1988. Exposure and ofSmoking Lung Cancer Deaths in the US in 1980 from Exposure toETS. Proc. Indoor Ambient Air Qual. Conf. London, pp. 242-251.

Baiter N.J., Kilpatrick S.J. and Witorsch P. 1986. CausalRelationship between Environmental Tobacco Smoke and Lung Cancerin Nonsmokers: A Critical Review of the Literature. 79th Meeting ofthe Air Pollution Control Association, June 22-27.

Coultas D . B., Samet J . M. and Peake G . T. 1987. Reliability andValidity of Questionnaire Assessment of Involuntary Tobacco SmokeExposure. Proc. 4th Conf. Indoor Air Quality and Climate, WestBerlin, 2: 121-125.

Friedman G . D., Petitti D. B. and Bawol R . D. 1983. Prevalence andCorrelates of Passive Smoking. Amer. J. Pubi. Health, 73: 401.

Garfinkel L. 1981. Time Trends in Lung Cancer Mortality AmongNonsmokers and a Note on Passive Smoking. JNCL, 6: 1061-1066.

Garfinkel L., Auerbach 0. and Joubert L. 1985.InvoluntarySmoking and Lung Cancer: A Case Control Study. JNCL, 75: 463-469.

Health and Welfare Canada. 1985a. Investigation on the Impact of aNew Smoking Policy on Office Air Quality: Office of the AuditorGeneral Ottawa, Ontario. Occupational Health Unit Medical ServicesBranch.

Health and Welfare Canada. 1985b. Investigation on the Impact of aNew Smoking Policy on Office Air Quality: National Library ofCanada Hull, Quebec. Occupational Health Unit Medical ServicesBranch.

Holcomb L.C. 1988.Impact of Environmental Tobacco Smoke onAirline Cabin Air Quality. Environmental Toxicity Letters, 9: 509-514.

International Agency for Research on Cancer, 1985.TobaccoSmoking, Monograph 38, Lyon France. IARC.

Melius J., Wallingford K., Keenlyside R. and Carpenter J. IndoorAir Quality - The NIOSH Experience.

National Research Council. 1986.Environmental Tobacco Smoke,Measuring Exposures and Assessing Health Effects.NationalAcademy Press. Washington D. C. 337 pp.

Robertson G. 1988. Source, Nature and Symptomology of Indoor AirPollutants. Proc. Indoor Air Qual. Conf. London, pp. 311-319.

9

Sonnenfeld G., Griffith R . B. 1987. The Effect of Smoke Age andDilution of the Cytotoxicity of Sidestream (Passive) Smoke. Toxicol.Lett., 35: 89-94.

Sterling T. 1988. ETS Concentrations Under Different Conditions of-Ventilation and Smoking Regulation. Proc. Indoor Amb. Air Qual.

Conf. London, pp. 89-98.

Surgeon General. 1986.The Health Consequences of InvoluntarySmoking.U.S. Department of Health and Human Services.Rockville, MD. 332 pp.

from Passive Smoking: Hypothesis or Convincing Evidence? Int.Arch. Occup. Environ. Health, 59: 421-437.

10

TOBACCO OBSERVER - OCTOBER

INDOOR AIR QUALITY

CAUSE EFFECT

EPA FIGHTS DIRTY IN-HOUSE AIR

or the past five months, the

EJ government agency chargedwith cleaning up America'sair has fought its pollutionbattles somewhat closer to

=home: within its own officein Southwest Washington,

D.C. You might say, in fact, that theheadquarters of the Environmental Pro-tection Agency (EPA) has become oneof the toxic waste sites of the 1990's—asick building.

SICK BUILDING SYNDROMEThe EPA office suffers from an alarmingbut increasingly common phenomenonnow widely known as "sick buildingsyndrome," a problem frequently blamedon tobacco smoke. Ironically, tobaccohas been virtually banned at the ER& formore than a year, under a policy thatallows smoking in only a few, separatelyventilated areas.

The problem of sick building syn-drome first drew national attention in1976, when 29 participants at anAmerican Legion Convention in Phil-adelphia died from exposure to badlycontaminated air. Since then, expertshave estimated that occupants in asmany as half of all modern Americanbuildings may be subject to indoorcontamination.

Sick building syndrome usually resultsfrom inadequate or unhealthy air flowcaused by building ventilation systemsthat are inefficient, poorly designed, orclogged and dirty. Without proper clean-ing, these systems serve as perfect in-cubators for infectious bacteria, fungi

and a host of other contaminants that thesystems then spread throughout the office.If the systems are simply inefficient,they can allow many existing officepollutants to accumulate.

Victims of sick building syndromecommonly complain of such healthproblems as chronic headache, ear, eyeand throat irritation, fatigue, nausea andother symptoms associated with allergiesor flu. For chemically sensitive workers,the effects can be much more serious.

IMPROVING FRESH AiRSTANDARDSMost experts trace today's indoor airpollution problem to the energy crisis ofthe 1970's, when buildings were designedto be energy efficient and airtight.

So pervasive is the problem today thatthe ventilation industry's standard-bearer,the American Society of Heating,Refrigerating , and Air ConditioningEngineers, has proposed a significant in-

crease in air circulation rates for majorbuildings. The increased cost of betterventilation, the Society believes, will bemore than offset by a parallel increase inworker productivity.

SYMPTOM, NOT A CAUSEThat the nation's top pollution police—the EPA—fell victim to sick buildingsyndrome provides a good indication ofthe severity of indoor air pollution prob-lems. That the incident occurred despitesevere smoking restrictions within theEPA office demonstrates what manyexperts have been saying for years: envi-ronmental tobacco smoke is a symptom,not a cause, of poor indoor air.

According to ventilation expert GrayRobertson, president of ACVA Atlantic,a leading indoor air quality consultingfirm, the buildup of cigarette smoke in aroom "doesn't mean you have a cigaretteproblem—it almost always proves that youhave a ventilation and hygiene problemthat a simple smoking ban will not cure.

"Because of its visibility, cigarettesmoke often becomes the scapegoat forall indoor pollution. But the fact is'Robertson points out, "tobacco smoke isthere because it's a symptom of badventilation.

"Once you realize that bad ventilationis trapping the smoke, you now mustalso realize that any invisible pollutants

continued on page 4

"By the time'I had been there 15mminutes,, y voice sta,led craclanga little, then my eyes, my throat and'ears started burning and Icouldn't breathe

"Me budding] should be declared a Superfund toxic waste site Ihave known sites that impact fewer people' than that building."

Bobbie Lively-DieboldEPA Environmental Protection Specialist

Quoted in 'The Washington Post

CAUSES OF SICK BUILDING SYNDROMEINDOOR AIR QUALITY:A LABOR PERSPECTIVE

s public awareness of indoor\ pollution has increased in

recent years, workplace airquality has become one of the

J

major issues faced by em-ployees and their representa-tives. Organized labor has

recognized the health and safety implica-tions of poor indoor air quality and hasbrought a new and authoritative voice tothe workplace air quality debate.

The Executive Council of the AFL-CIO was among the first groups to takea strong stand on the issues of work-place air quality and workplace smok-ing, addressing the problem throughadoption of two formal resolutions overthe last three years. Taken together, thestatements reflect concern that govern-ment and industry resolve the real issueof workplace air quality—not just thesmokescreen issues of employee lifestyledecisions and workplace smoking.

The stake that organizations like theAFL-CIO have in improving overallworkplace air quality is significant. Inissuing its 1986 statement, the AFL-CIOrecognized that indoor pollution is agrowing problem, and that cigarettesmoke may provide an excuse foremployers to avoid cleaning up theworkplace. Regarding the 1985 SurgeonGeneral's Report on the Health Conse-quences of Smoking, the statementnotes, "Rather than shedding light andcontributing new information on theoccupational health risks faced by manyworkers, the report minimizes the risksposed by workplace toxins.

"The AFL-CIO believes that employerswill attempt to use the report to shirktheir responsibility to clean up the work-place and to place blame for occupa-tional disease on workers who smoke,"the statement continues. "Employersshould not be allowed to shift theburden to individual workers."

The 1988 AFL-CIO statement addressesthe broader concerns of overall indoorair quality and methods of reducingworkplace pollutants. In that statement,the Council noted that proper ventilationis essential to improved indoor air quality.The Council also urged government andindustry to recognize that "identificationof pollutants, cleaning and replacementof poor ventilation systems, and buildingmodifications to provide an adequatesupply of fresh air are essential steps toassure all workers a safe job site."

05%

505

55%

55%

45%

40%

35%

35%

25%

35%

15%

111%

5%

0%

obacco smoke plays a minorr

7 '40A role in air quality problems inthe indoor environment.

That's the conclusion of a

flnumber of scientists whospoke in June before the In-

door Ambient and Air Quality Sym-posium sponsored by Imperial Collegein London. Assessing the contribution ofenvironmental tobacco smoke (ETS) tooverall indoor air quality, severalresearchers determined that ETS levelsin a variety of settings typically are low.

One scientist concluded, "An exclusivereliance on regulating smoking whileignoring other sources of indoor pollu-tion. . . may accomplish little in meetingindoor air quality problems?'

U.S. environmental expert LarryHolcomb, formerly senior scientist withthe Michigan Toxic Substance ControlCommission and currently president ofhis own environmental consulting firm,reviewed studies of the contribution ofETS to air quality on board commercialaircraft.

Holcomb found that ETS levels inboth smoking and nonsmoking sectionsof aircraft are extremely low. He alsoidentified a number of substances andconditions not associated with ETS aslikely factors in cabin air quality com-plaints. Ozone levels and relativehumidity were cited as potentially im-portant causes of in-flight discomfortoften blamed on ETS.

Similar results were reported from anextensive study of indoor air quality innearly 3,000 locations across the United

Kingdom. Led by P. W. Kirk, professorof public health engineering at ImperialCollege, a team of British researchersmonitored ETh levels in home, office,travel and leisure environments over afour-week period. The study demon-strated that ETS components in all envi-ronments were low in comparison withU.K. recommended limits. Carbon mon-oxide levels, commonly used to measureETS, were not significantly different insmoking and nonsmoking situations.

Two other studies presented at theconference indicate that indoor air qualityis largely a function of factors other thantobacco smoke.

Reporting the results of two fieldstudies of ETS in adequately ventilatedCanadian offices, Theodor Sterling, pro-fessor of applied sciences at SimonFraser University in British Columbia,found low levels of ETS in air recir-culated from smoking to nonsmokingareas. Since these levels were not appre-ciably different than in separately ven-tilated nonsmoking areas, Sterling con-cluded that separate ventilation of smok-ing areas is unnecessary.

U.S. researchers John Carson andCarol Erikson, also studying Canadianworkplaces, found equally low levels ofETh in office air. These scientists con-cluded that the average office workerwould have to spend 260 hours in anoffice to be exposed to the nicotineequivalent of just one cigarette, and1,000 hours in an office to be exposedto the particulate matter equivalent of asingle cigarette.

CIGARETTE SMOKE A MINOR PLAYER IN INDOORPOLWTION, RESEARCHERS SAY

Ventilation Key toIndoor Air Quality

ray Robertson is president and co-founder of ACVA Atlantic, a Faitfax,

Virginia firm specializing in diagnosisand treatment of indoor air pollutionproblems. ACVA Atlantic has inspectedover 45 million square feet of commer-cial office space, including the US.Supreme Court, the Federal Reserve andthe United Nations, and is consideredone of the leading indoor air qualityfirms in the country.

U The Observer: When most peoplethink of air pollution, they think ofdirty outdoor air. What is indoor airpollution, and is it really a seriousproblem in the modern workplace?

Robertson: Indoor air pollution isthe accumulation within a confined areaof chemicals, mists, smoke, dusts, gasesand vapors, microbes, including fungiand bacteria, fibers and other materialswhich can pose a threat to the health ofthose who work, live or play within thatarea. Based on ACVA's investigations, weestimate that as many as half of allmodern buildings may have some degreeof indoor pollution.

U The Observer: Where do thesesubstances come from?

Robertson: People themselves andvirtually every product we use indoorsshed some particles or gases. Commonoffice supplies and equipment releasedangerous chemicals. Clothing, furnish-ings, draperies and carpets contributechemicals and various fibers. Cooking,broiling, grilling, gas and oil burning,smoking, coal and wood fires alsogenerate vast numbers of particles whichpollute the air.

U The Observer: Doesn't ventilationtake care of the problem?

El Robertson: Ventilation is the key toclean indoor air. The problem is thatmany buildings don't have proper oreven adequate ventilation systems. Infact, more than half of the buildingswe've inspected simply don't providesufficient air.

Even more alarming is the fact that 35percent of the buildings we inspect areoperating with no fresh air at all. Theproblem results from the way a building'sventilation system is maintained. We've

gone into buildings where maintenanceengineers have completely closed the airintake dampers to conserve energy,thereby preventing fresh air from enter-ing the building. In other words, peoplewho work or live in these buildings arebreathing 100 percent recycled air.

U The Observer: You mentionedearlier that tobacco smoke is one ofmany constituents in indoor air. Isn'ta smoking ban the most importantstep building managers can take incleaning up indoor air?

Robertson: That's a popular miscon-ception. When building managers oremployers get complaints about scratchythroats, watery eyes, headaches or achi-ness—all common symptoms of sickbuilding syndrome—they often think theycan alleviate the problem by banningcigarette smoking. They believe thattobacco smoke is the problem becauseunlike other indoor air pollutants, it isvisible. Cigarette smoke therefore fre-quently becomes the scapegoat for allindoor pollution.

If smoke accumulates in the air of anoffice, you're seeing a symptom of badventilation. And if you have bad ventila-tion, you can also be sure that the in-visible pollutants—the bacteria, fungi,fibers, chemicals, and gases—are trappedin your building as well. Smoking banswill not reduce these pollutants, whereascorrecting or improving ventilation willreduce the levels of all contaminants.

• The Observer: Suppose your ven-tilation system does circulate enoughair. Can you assume then that youdon't have indoor pollution?

El Robertson: Not always. There aretwo other main components that canplay a role in what we now term "sickbuilding syndrome." They are poorfiltration of the air in a building's airsystem and general uncleanliness of abuilding's ductwork.

Specifically, filtration systems must be

• The Observer: Why has indoorpollution become such a problem par-ticularly in recent years?

El Robertson: A huge demand exists to-day for "tighter," more energy-efficientbuildings—structures that will costowners and building managers less tomaintain and operate. These energy con-servation efforts have resulted inelaborate, complex ventilation systemsand have reduced the natural ventilationand filtration .that we've relied on forcenturies. For instance, in most newbuildings, the windows don't even open.

• The Observer: For large officebuildings, isn't improving ventilationand filtation systems too expensive asolution to the indoor air pollutionproblem? What advice can you giveemployers and building managers whoare concerned about improving indoorair quality?

Robertson: First, it doesn't take aninordinate amount of money, in mostcases, to make sure indoor air is cleanand healthy. In fact, improving indoorair quality can save money for anemployer in the long term, because itcan reduce employee absenteeism andincrease productivity.

Think of it this way. Ventilationrestrictions designed to conserve energyreally save the equivalent of $50 perperson per year. But just one day ofemployee illness due to indoor pollutionfar exceeds that yearly expense.

Building managers can improve indoorair quality by following a few simplesteps to insure that air-conditioning andheating systems are properly maintained.Inspect the air-conditioning and heatingunits, including the ductwork, and cleanwhen necessary. Use good filters, andchange them frequently. Keep the build-ing's fresh air dampers open to bring inthe optimal amount of air from outdoors.

In many cases, good building hygieneis all that is necessary to alleviate in-door air quality complaints.

maintained, cieaneuano upgraded wnennecessary. Dirty

4^4ZACis a perfect breeding groundilbr all sorts

of fungi, bacteria, und other pollutantswhich can build up\aiid easily spreadthroughout thesystem. In many of the buildings we've c

looked at, we've seen a virtual petcemetery of dead animals—cockroaches "v

cans and othê'iáteials left inside the-ductwork by construction workers. -

EPA Fights...continued from page 1are also going to be trapped at the sametime—and they often are harmful enoughto make you ill?'

This analysis is borne out by govern-ment tests which show that the vastmajority of indoor air quality complaintscan be traced to poor ventilation. Ciga-rette smoke is responsible in only twopercent of cases, according to a recentgovernment study of air quality prob-lems in federal and private buildings.

Robertson's advice in treating sickbuildings, then, is to "go to the cause.

"Hanging up a no-smoking sign mightstop people smoking, but it won't cure

INDOOR AIR UUAUTY CHECKLIST - • Use quality air filters on air-conditioning andheating units and replace them regularly. Pleated

Indoor air quality experts recommend the followingpanel filters are fine for most homes however,steps to help insure cleaner air. Many of these if someone in your home suffers from allergies,suggestions apply to the office as welt ask yourupgrade your filtration system by installing highemployer or building manager what can be doneefficiency filters or electronic air cleanemto provide clean indoor air at your workplace. • Insure that appliances using gas, oil or other• Clean all air-conditioning and heating units Eachcombustible fuels are adequatoly vented to the

season the coils should be inspected, cleaned outdoors. Wood burning stoves and fireplacesand clisinfoctet .. .....,should be operated only with.open vents andior

clean chimneys.• Inspect ductwork bi.annuahy. If dirt, dust, insects :-'•- .or molds are visible, ductwork should be cleaned .• Purthase a radon Aktedcn kit if you suspectand sanitized. .- .: 'your home could be affected by rddon gas Kits

are easy to install, inexpensive and available at• Leave one or two windows open to iitreaso drug and convenience stores, and manyventilation whenever temperatures outside permitsupermarkets.In extreme temperatures,open windows or doorsfor a minimum of 5-10 minutes daily. • Inspect your home and office for asbestos,

which can be found in many older structures as• Follow all label directions when using them,cal a component of plaster, tiles, surface finishesbased household products, such as furniture and as an insulation material on furnaces, boilerspolishes, paint thinners and cleaning fluids. In or pipes. When materials age and fragment. .many instances, ventilation rates may need to befibers can be released. These fragments shoulddramatically improved to reduce toxins releasedbe analyzed for asbestos.by most of these products. .

I

fl,Published monthly and available free frontThe Tobacco Institute1875 I Street Northwest Suite 800Washington, DC 20000

The Obsenw presents information and commenton public events of interest to the tobaccoindustry. It recognizes that there is diversity ofopinion about tobacco use and that chargesagainst tobacco are widely publicized while lessattention is given to differing view which areincluded in its columns. Its aim is to aid full,free and informed discussion in the pubicinterest in the conviction that the smoking andhealth controversy must be resolved by scientificresearch.

Copyright et 1988. The Tobacco InstituteAll rights reserved.October 1988

the indoor problem of all the bacteria,fungi, gases, chemicals, and vapors. Sogo to the cause, which is the bad ven-tilation, and put your ventilation right."

EPA'S "POISONED" EMPLOYEESAccording to reports in The WashingtonPost about the EPA's pollution battle,agency official Bobbie Lively-Dieboldleft the headquarters after her initialbout of sick building syndrome, but shereturned some weeks later only to sufferfour additional attacks.

Since then, The Washington Timesreported, more than 75 employees havecomplained of health effects directly

related to the building's air. Sevenworkers—who call themselves the Com-mittee of Poisoned Employees—becameso ill that they were forced to workelsewhere. In June, the union represen-ting federal employees joined these sevenand more than 100 other EPA employeesin picketing the agency's headquarters todraw attention to the workers' plight.

Specialists called in to examine thebuilding determined, not unexpectedly,that the building's ventilation system wasinadequate to disperse a variety of fumesemitted by new carpet and furnishings.These experts also found, however, thatthe problem was not new: EPA employeeshave complained of unhealthy air in thebuilding for a number of years.

For now, the employees' complaintsappear to have worked. EPA officialshave removed the materials that causedthe immediate crisis, but they are stilldetermining how to clean up the air inthe clean air headquarters of the nation.

As a smoker, or even as a nonsmoker whobelieves that workplace smoking restrictions don'tsolve the problem of indoor pollution, take thisopportunity to get involved.

The Tobacco Institute can provide informationon the issues of indoor air quality and workplacesmoking. Detailed information on these subjects,as well as many others is available from TheInstitute upon request, free of charge. Pleasecontact us at:

The Tobacco institute1875 I Street NorthwestWashington, DC 20006(202) 457-4800Toll Free: 1-800424-9876

Some tobacco manufacturing employees receive TheTobacco Observer through their companies. If youregularly receive more than one copy, please mail youraddress label to Circulalior Tobacco Observer

First ClassU.S. Postage

PAIDThe Tobacco Institute

M8

-J

TOBACCO OBSERVER - NOVEMBER

SMOKING AND THE STATE

and

SOCIAL COSTS - "SETTING THE RECORD STRAIGHT"

Volume 13 #9

AAAA

RECtI Vn16 NOV

1:Liriittempts to regulate behaviorand restrict personal choices

/are nothing more thanefforts to mandate someoneelse's idea of the "ideal"

•society. In the case oftobacco products, these

efforts are seen daily with proposals toincrease excise taxes, limit smoking inpublic places and ban tobacco advertising.

A potentially dangerous scenarioemerges when "activists" seek to man-date social policy and control personallifestyle decisions using legislation as avehicle. Dr. Richard Wagner, co-authorof Smoking and the State, calls it"policy by preference."

"Policy by preference' . is a realdanger," Wagner says. "It is a case ofpeople trying to use the legislative arenafor stopping or preventing things theydon't like, whether it be smoking, bearwrestling, hang gliding or rock climbing.That is where we come back to thequestion of 'social costs'."

In fact, this skewed logic is now beingever more widely applied today and mayresult increasingly in limits or restric-tions of everyday activities. The notion,"smoking today, bear wrestling tomorrow"may sound farfetched, but it illustrates

• an undercurrent of questionable public-policy decisions on the horizon.

For example, in suburban Washington,D.C., county council officials recentlystrengthened an already stiff publicsmoking ban by extending it to includebowling alleys, barber shops and beautysalons.

The latest target for the Council'sattention? Gas-powered leaf blowers usedby lawn care companies. Regulation isthought to be necessary because, as onecouncil member stated, "I find theextremely loud and unpleasant noisemade by these machines upon the peaceand quiet of the community to be onemore instance of being captive toanother's activities."

Gas-powered leaf blowers may seemto be a humorous example, but themessage is serious and the threat is real.This thinking may prompt examinationof other aspects of our lives and invitethe scrutiny of a host of other productsused every day.

Dictating when and where one canenjoy a cigarette may open the door toregulation of the use of other productsthought by some to be an annoyance.Thus, public-policy decisions based onpersonal preferences begin to resemblean Orwellian "big brother." It is doubt-ful that society benefits from regulatoryefforts of those in pursuit of "utopia."

"'SOCIAL COSTS:"SF11/NC THERECORD STRAIGHTterm that is being applied more

j often in the debate on tobaccot. issues is "social costs."

"Social costs" is an area of economicstudy, but you don't have to be aneconomist to understand it. A lot of"non-economists" toss it around. Unfor-tunately, many are using it incorrectly.

"ECON 101."A "social cost" results when an indi-vidual's actions, behavior, or use of aproduct or service, impose a cost onsociety as a whole. "Social costs," there-fore, are different from "private costs,"which are assumed solely by the indivi-dual as a result of his or her actions.

"Social costs" can be economic,such as worker productivity, or finan-cial, such as medical expenses.

This "social cost" theory is com-plex. It is said to involve a number ofcomponents including "incometransfers," productivity and potentialearnings losses, and medical andhealth-care costs.

"Social costs" is a valid economicconcept. In the case of tobacco, how-

continued on page 4

According to a recent government study, softball slidinginjuries cost society nearly $2 billion each year. This isn'tthe only activity that is claimed to "cost" society money.

The theory of "social costs" is often misunderstood.(See "'Social Costs' Setting the Record Straight" for adiscussion of "social costs.") By broadly applying thiseconomic approach, virtually any activity, product or serv-ice can be said to have an enormous "cost to society."

The following "findings," reported in various publishedstudies and other sources, could some day be used tolimit or regulate any number of products and activities.c The cost of obesity is estimated at about $27 billion.o Participating in sports cost more than $2 billion a

year in lost productivity. Spectator sports account foranother $4 billion in lost productivity annually.

• Vacationing by car has a social cost of $114billion—not including road repairs.

• In 1986, delays of airline flights cost $3.2 billion inlost lime.

• Fuel wasted during traffic jams cost society nearly$3 billion annually.

Io.r

/

iVbSTr-,

CH-4R-

PERSONAL CHOICES IN AFREE SOCIETY:SMOKING AND THE STATE

One argument used by supporters ofsmoking regulations is the claim thatsmokers impose "costs" on society.

Robert Tollison and Richard Wagner,economists with the Center for Study ofPublic Choice, George Mason University,

examine "social costs" in a recentlypublished book, Smoking and the State.The authors put economic analysis andpublic policy discussions in perspectivein an effort to lend balance to the"social cost" debate.

"Even to a nonsmoker as myself,Tollison and Wagner expose some of theloose logic in the fervor of the anti-smoking campaign' notes Brian L.Goff, assistant professor of economics atWestern Kentucky University.

Other economists have respondedsimilarly. Recently, Ryan C. Amacher,Dean of the College of Commerce andIndustry at Clemson University, reviewedSmoking and the State for a Columbia,S.C., newspaper:

2

T6et tteColumbia, South Colin - Sunday, Ano,( 14, 1988

Economists explore thedangerous aspects ofgovernment protection

Reviewed by Ryan C. Amacher

In the last decade, political action,regulation and tax activity against smok-ing and smokers has accelerated andmany smokers have begun to feel socialstigma associated with smoking.

One airline has even made prohibitionof smoking a major advertising andpublic relations gimmick. The health tollexacted by smoking has been estimatedby some scholars to be in excess of $100billion. The families of cancer victimshave sued tobacco companies, claimingthey bear responsibility for the death ofloved ones.

Into this jumble of politics, publicrelations and legal maneuvering enterssome refreshing analysis, Smoking andthe State by Robert D. Tollison, directorof the Center for the Study of PublicChoice at George Mason University andRichard E. Wagner, professor of eco-nomics at Florida State University.

The authors do not examine the scien-tific case against smoking. Instead, theyaddress the public policy issues asso-ciated with the debate over smoking.These are important questions becausethey are at the center of decision-makingin a democratic society.

As economists, Tollison and Wagnerare comfortable viewing consumers asrational decision-makers. If these con-sumers cannot make rational and in-formed decisions about smoking, howcan they make "correct" decisions about

nutrition, child bearing, job selection,gardening or house painting? More im-portantly, if government should have thepower to prevent people from engagingin allegedly high-risk activity, why stopwith tobacco? What about skiing, autoracing, mountain climbing, working longhours in stressful occupations?

"If consumers cannotmake rational andinformed decisions aboutsmoking, how can theymake 'correct' decisionsabout nutrition, garden-ing or house painting?"

What emerges is a powerful examina-tion of the role of government. Should itinform or should it protect? The differ-ences between these two roles arefundamental.

The central issue in this debate isover who bears the alleged costs ofsmoking. If smokers bear the costs,there is no need for governmentalregulation. If nonsmokers bear signifi-cant portions of the cost, there may bearguments for governmental action tomake smokers more responsible.

First, Tollison and Wagner argue thatthe claim that there are direct socialcosts borne by nonsmokers is incorrect.Such health and productivity costs, theysay, are paid by smokers as lost pay, jobstatus and promotions. The second issueconcerns environmental tobacco smoke

(ETS) and the costs it purportedly im-poses on nonsmokers. The authors acceptfor discussion the fact that ETS can bea nuisance to some nonsmokers, but theysuggest the real issue is how best todeal with the policy issue of these costs.

They conclude that a competitive,private economy will solve the ETS"problem' Firms will have sufficientincentives to establish smoke-free work-places and smoke-free environments forconsumers if these workers and con-sumers view the El'S costs as signifi-cant. Restaurants will establish smoke-free dining areas if consumers demandit. The same will hold for the work en-vironment. In other words, the authorswould applaud the action of NorthwestAirlines in banning smoking as a marketexperiment in action, but would criticizegovernmental prohibitions on smoking inplanes as an inefficient intrusion into themarketplace.

Tollison and Wagner also raise theissue of taxation of tobacco products.Economists have long pointed to the in-efficiencies resulting from selectiveexcise taxation, not to mention theextremely regressive nature of tobaccotaxes.

Smoking and the State clears the airon much of the debate over smokingand political action to solve its perceivedills. But it is more than a simple look attobacco issues. It draws attention to therole of the state in a free society andthe dangers inherent, for all of us, whenthe state is used as a vehicle to protectus from ourselves.

The Stale, Columbia, S.C.. Sunday, August 14, 198&Reprinted with permission.

An Economist's Perspective on Social Costs

110 smokers "cost" society as a wholemore than nonsmokers? Proponents.

of ad bans, smoking restrictions andhigher excise taxes often say so, tojustify increased regulation. The Tobacco

Observer asked Dwight R. Lee, professor,Department of Economics, University ofGeorgia, Athens, to comment on theeconomic theory of "social costs" andits application.

3

0U N I

N I

iim

The Observer: The Office ofchnology Assessment, Washington,.C., has estimated that smokers costciety $65 billion each year. Where) these numbers come from. . .and

now realistic are they?

Li Dr. Lee: The social cost of smokingcalculations are deeply flawed. Groupsthat have made such claims have notproperly taken account of all factors.Not being economists, they have madesome glaring and fundamental errors.

Many of the "costs" they claim to besocial costs are really private costs, ifthey exist at all. For example, ifsmokers as a group are absent fromwork more often than nonsmokers—adebatable proposition—most economistswould agree that it is the smoker whowould pay the cost for such behavior,through lower raises and slower jobadvancement. Smokers are not stickingsociety with the costs.

When it comes to medical costs thatsome attribute to smoking, research bythe National Bureau of EconomicResearch has shown that smokers do notgenerate additional social costs fornonsmokers. Smokers pay nearly $10billion each year in excise taxes ontobacco alone. They also pay sales, in-come, social security and other taxes.

When these taxes are taken into ac-count, along with the use of public pro-grams by smokers, economic researchshows that smokers do not impose addi-tional costs on society.

The Observer: Aren't smokersabsent from work more often and lessproductive than nonsmokers? Doesn'tthat translate into a loss to businessand to society?

LI Dr. Lee: There are statistics thatattempt to show that smokers as a grouphave higher absentee rates than non-smokers. However, the same data alsoindicate that males who smoke less than15 cigarettes a day have lower absenteerates than nonsmokers.

The problem with such simple statisticalinterpretations is that they do not takeinto consideration a host of importantdemographic factors. This makes thedata on absenteeism misleading. For ex-ample, a greater percentage of smokersare blue collar workers who work inless attractive environments compared towhite collar workers. Therefore, anyhigher absenteeism rates by smokersmay be due to factors associated withthe blue collar environment.

Also, compared to nonsmokers,smokers as a group are younger, con-sume more alcohol, and are employed inmore transitory occupations. Absenteeismis related more strongly to thesefactors—not smoking.

• The Observer: Because smokersallegedly "cost" society, isn't it impor-tant to discourage the consumption ofthese products, perhaps by increasingexcise taxes?

Li Dr. Lee: There is little convincingevidence that increased excise taxes willhave an effect. Higher cigarette taxes,however, represent a severe burden onlow income individuals. Tax increasesare an unfair way to influence the activi-ties of one segment of the population.

• The Observer: What about the so-called slippery slope? Is there a riskthat raising cigarette taxes or passinglaws that restrict smoking to deal withsmoking's claimed "social costs" couldsignal lawmakers to regulate otherproducts as well?

LI Dr. Lee: There is no doubttrend is in the making. Legislaconstantly looking for sourcesrevenues. As a consequence, nthem are eager to jump on thebandwagon simply as at medrevenue. In the process, they,,.

tingly joining reformers and zealots whowant to crush individual freedom bypassing laws to discourage the consump-tion of products they do not like.

The end result of this process is whatthe recent Nobel Prize winner in econo-mics, James Buchanan, describes as"regulatory anarchy'—a state of affairswhereby many products and individualpreferences become regulated by thestate. In the end, most individuals loseout, including many of the reformers,since a whole range of products becomemore expensive and difficult to use.When economic freedoms are tamperedwith in one area, the door is opened forregulation in other areas. The SurgeonGeneral, for example, shortly after hisadmonitions concerning tobacco use,released a nutrition report advisingagainst the prevailing eating preferencesin the U.S.

Finally, if the anti-smoking forceswant to persist with their social costcalculations, they should be consistentand apply them to other commodities aswell. In fact, an entire social cost "in-dustry" could be set at work to calculatethe "social cost" of eating red meat,skydiving, skiing, leisure driving, choles-terol intake, auto racing, team sports,sedentary lifestyles and many othereveryday activities.

The "social costs" of these activitieswould run well over $200 billion peryear, but what does it prove? Should wecontrol people's lives through federalregulations to squelch individual choiceand voluntary risk-taking? And, what arethe social costs of depriving individualsof their freedom of choice?

• 0 - .v.

Social Costs continued from page 1

ever, it is increasingly misused andmisapplied. The current anti-tobaccomovement relies, in part, on this con-cept to support moves to raise cigaretteexcise taxes, impose smoking restric-tions-and control the use of a legalproduct.

"Social cost" theory has become atool for some "activists" who claimthat society picks up the tab for allegedlosses they say result from smoking.

A 1985 "social cost" report by theOffice of Technology Assessment(OTA) is frequently cited by anti-smokers to justify increased regula-tion. OTA, an agency of Congress,reviewed the existing literature on thissubject and estimated that the U.S.economy may lose billions of dollarsannually because of cigarette smoking.

However, researchers dispute theseconclusions, noting that the body ofliterature used as the basis for the reportis inconclusive, contains inaccuraciesand relies on incorrect assumptions.

For example, some noted that OTAconfused "private costs" with "socialcosts," and ignored other considera-tions, such as demographics.

ON THE .1013Economists and non-economists alikecontinue to debate the accuracy andapplication of a host of theories andstatistics. Meanwhile, "social cost"miscalculations are being used everyday to limit or restrict workers' actionsboth on and off the job.

Advocates of smoking restrictions inthe workplace claim that workers whosmoke are not as productive as theirnonsmoking cp-workers. In addition to

establishing "no smoking" policies,further discriminatory measures arepromoted—such as bans on hiringsmokers.-.--even if that individual is thebest candidate for the job.

According to a study by ResponseAnalysis Corp., such biases are patentlyunfair and cannot be substantiated ineconomic terms.

Response Analysis examined produc-tivity in the workplace in a survey oftop-line managers in business andindustry, federal and state government,and local unions.

According to those questioned,smoking is seldom a factor when com-paring productive and unproductiveemployees. Good performers weredescribed as dependable and consis-tent, self-motivated and fast learners.On the other hand, it was theunreliable, unmotivated employee whowas called unproductive.

Robert Toll ison, professor ofeconomics at George Mason Univer-sity, States, "There is no convincingeconometric evidence that smokers, asa group, are less productive thannonsmokers. When any worker, forwhatever reason, is less productivethan another, that worker is going toearn less and not be promoted. Societydoesn't pay the cost of that worker'slower productivity, the worker does, byearning less money."

Another charge anti-tobacco forceslevel is that smokers are absent fromwork more often than nonsmokers.Again, this claim is questionable. Anumber of outside factors ranging fromjob satisfaction to commuting problemsare overlooked or not considered whenassessing the rates and causes ofabsenteeism.

PICKING SMOKERS' POCKETSProponents of raising cigarette excisetaxes argue that smokers get sick moreoften, generating greater medical ex-penses. Therefore, they seek to "ear-mark" portions of the revenue fromcigarette taxes to support varioushealth programs.

Expenses related to any lifestyledecision are "private costs" borne bythe individual, not society. It is the in-dividual who chooses to use sugar, eatfoods high in fat, or engage in ac-tivities such as skiing or riding amotorcycle. That same person assumesany risks or "costs" associated withthose choices.

Imposing a "user fee" on selectgroups is unfair. Smokers, in additionto paying taxes on cigarettes, pay theirfair share in income, sales, socialsecurity and other taxes. "Earmark-ing" excise taxes is discrimination bytaxation.

"Most current social economic analy-ses serve political rather than economicends," Robert Tollison says. "Thenumbers are at best broad projectionsbased on erroneous assumptions."

As a smoker, or even as a nonsmoker whobelieves individuals should be free to make theirown lifestyle choices, take this opportunity to getinvolved.

The Tobacco Institute can provide informationon the issues affected by "social costs" such aspublic smoking, excise taxes and advertisingbans. Detailed information on these subjects, aswell as many others is available from TheInstitute upon request, free of charge. Pleasecontact us at: The Tobacco Institute

1875 I Street NorthwestWashington, DC 20006(202) 4574800Toll Free: 1-800424-9876

on public reofntc

copiniona .tsbacç9 usagainstJoco;feidattention is given to diffe

dided i its .columns.9""free 'and informddiscu

ntefes1nthe'co1wictionhealth ccntIovesy must

.r)..:COpght@I988.The1iM rights reserved.4

Novemberl988

AN APPENDIX TO

"CLEARING THE AIR" - DR ROBERT D TOLLISON

THE RESEARCH EVIDENCE ON ENVIRONMENTAL TOBACCO SMOKE

Appendix

Research Evidence onEnvironmental Tobacco Smoke

The risk of reporting chronic cough, chronic wheeze, lower respiratory ill-ness, bronchitis in the last year, chest illness in the last year and doctor diag-nosed respiratory illness before age two. was consistently higher amongchildren living in homes positive in one or more of the moisture charac-teristics. These associations generally remained after adjustment for paren-tal respiratory illness, current maternal smoking and parental education.

B. Brunekreef, D.W. Dockery, F.E. Speizer, J.H. Ware, J.D. Spengler, andB.G. Ferris, Jr. (Harvard School of Public Health and Harvard MedicalSchool, Boston, and the Agricultural University of Wagenin gen, the Neth-erlands), "An Association between Moisture in the Home and RespiratorySymptoms in Primary School Children," American Review of RespiratoryDisease Supplement 135 (4, Pt. 2) (1987): A340 (abstract).

The criticisms of the Japanese (Hirayarna) and Greek (Trichopoulos) studieswere basically related to thegeneral lack of reliability of questionnaires,inadequate histological diagnoses, uncertain accuracy of reported smok-ing habits,. and failure to take into account general urban pollution andindoor pollutions emanating from cooking and heating practices.

W. Allan Crawford, (consultant in environmental and occupational health,Sydney, Australia), "Environmental Tobacco Smoke: The Use of Mathe-matical Models to Predict Health Effects," Environment International 13(1987): 151-154.

120 Clearing the Air

Thus, the Repace and Lowrey tentative mathematical modelling with itspreliminary results has as its basic input significantly flawed data with regardto environmental factors. It is based on epidemiologic data of a highlysuspect nature which have been severely criticized by expert bodies. Theirpublications have engendered unwarranted anxiety in sections of the publicwho are concerned about the alleged health effects of environmental to-bacco smoke.

Ibid.

The [National Academy of Sciences: The Airliner Cabin Environment: AirQuality and Safety] Committee's report provides no citation or data to sup-port the statement that high transient concentrations of tobacco smokecomponents occur in the nonsmoking section of the cabin.

U.S. Department of Transportation and Related Agencies AppropriationsAct, 1988, Speech of Hon. Tom DeLay of Texas, House of Representatives,July 13, 1987, in Congressional Record, July 15, 1987, pp. E2904—E2908.

It may be said that so far even toxicology has not been able to ascertainwith any greater degree of probability than did epidemiology that thereexists a link between damage to health and passive smoking.

G.J. Gostomzyk (director, Health Bureau, Augsburg, Germany), "PassiveSmoking—Report on an International Symposium (23-25 October, 1986,Essen)," Public Health 49 (1987): 212-215.

The Surgeon General, like myself, is a politician. I have learned that whendealing with politicians it is better to watch what they do as opposed tolistening to what they say. This is as true for the Surgeon General as itis for me. The simple fact of the matter is that the conclusions in theSurgeon General's report are not supported by the research in his ownreport.

Hon. Walter B. Jones of North Carolina, House of Representatives, February18, 1987, "Inconclusive Evidence on the Harmful Effects of Smoking,"Congressional Record, February 18, 1987, pp. E489-E490.

The present study supported the Conclusions of Rogers et al. (1984) thatexposure to cigarette smoke in the home was not a risk factor for middleear problems in children. Rogers et al. (1984) reanalyzed previous data

from Kraemer et al. (1983) by controlling for the influence of nasal con-gestion. They suggested that it was the nasal congestion that influencedKraemer's significant results, not exposure to household cigarette smoking.

The relative elimination of air pollution in the present investigation sug-gested that tobacco smoke exposure alone might not be as great a riskfactor for middle ear problems in children as previously assumed.

Ken J. Kallail (University of Kansas School of Medicine, Wichita) HarryR. Rainbolt (Kansas State University), and Melvin D. Bruntzel (KansasDepartment of Education), "Passive Smoking and Middle Ear Problemsin Kansas Public School Children," Journal of Communication Disorder 10(3) (1987): 187-196.

A total of 14 studies on lung cancer risk for the non-smoker in relationto spousal smoking survive critical assessment despite their technical flaws.Of the 6 studies based on American subjects, none shows a statisticallysignificant increase in risk for the nonsmoker with a spouse who smokes.

Based on the evidence to date, the concern about the risk of lung cancerfor nonsmoking Americans appears to be overstated and unsupported.

Alan W Katzenstejn (Katzenstein Associates, Larchmont, New York), "En-vironmental Tobacco Smoke ETS and Risk of Lung Cancer—How Con-vincingls the Evidence' Washington, D.C., Tobacco Institute, March 1987.

In summary, when a spouse's smoking status is used to estimate a non-smoker's ETh exposure, "a considerable amount of misclassification" mayresult. Since selection bias and confounding must also be considered, ex-treme caution is required in the interpretation of these studies. This isespecially so when the literature as a whole contains several studies report-ing no significant association between ETS exposure and lung cancer aswell as various inconsistencies, both among and within studies.

S.J. Kilpatrick, Jr. (Medical College of Virginia, Virginia CommonwealthUniversity, Richmond Virginia), "Misclassification of Environmental To-bacco Smoke Exposure Its Potential Influence on Studies of Environmen-tal Tobacco Smoke and Lung Cancer," Toco/ers 35(1987): 163-168.

Lifetime exposures to environmental tobacco smoke from the home orworkplace for 88 "never-smoked" female lung cancer patients and 137"never-smoked" district controls were estimated in Hong Kong to assess

122 Clearing the Air

the possible causal relationship of passive smoking to lung cancer risk.Relative risks based on the husband's smoking habits, or lifetime estimatesof total years, total hours, mean hours/day, or total cigarettes/day smokedby each household smoker did not show dose-response results. Similarly,when such categories as mean hours/day, or earlier age of initial exposure,were combined with years of exposure, there were no apparent increasesin relative risk.

Linda C. Koo, John H-C. Ho, Daisy Saw, and CFiing-yee Ho (Depart-ment of Community Medicine, University of Hong Kong, "Measurementsof Passive Smoking and Estimates of Lung Cancer Risk among Non-Smoking Chinese Females," International Journal of Cancer 39 (1987): 162-169.

Bias caused by misclassification of smoking habits coupled with between-spouse smoking habits concordance can completely explain reported ap-parent excesses in lung cancer risk in non-smokers married to smokers.

Peter N. Lee, "Lung Cancer and Passive Smoking: Association an ArtefactDue to Misclassification of Smoking Habits," Toxicology Lsflem 35 (1987):157-162.

In the event, whether the true relative risk is 1.05 or 1. 14, it is unlikelythat any epidemiological study has been, or can be, conducted whichcould permit establishing that the risk of lung cancer has been raisedby passive smoking. Whether or not the risk is raised remains to be takenas a matter of faith according to one's choice.

Nathan Mantel (professor of mathematics, statistics, and computer science,American University, Bethesda, Maryland), "Lung Cancer and PassiveSmoking," British Medical Journal 294 (1987): 440.

Other possible more serious biases in the studies conducted were notconsidered. (These include publishing bias: if an investigator got a weaklyor insignificantly negative result for the role of passive smoking in lungcancer would he bother submitting it for publication? And if he did,would it be accepted? There seems to be a tendency toward acceptinguncritically or less critically manuscripts which are on the right side ofthe fence on the issue of passive smoking.)

Ibid.

-- Ti -

Children living in damp houses, especially where fungal mold was pres-ent, had higher rates of respiratory symptoms, which were unrelated tosmoking in the household, and higher rates of symptoms of infection andstress. Housing should remain an important public health issue, and theeffects of damp warrant further investigation.

Claudia J. Martin, Stephen D. Platt, and Sonja M. Hunt (Research Unitin Health and Behavioural Change, University of Edinburgh), "Hous-ing Conditions and Ill Health," British Medical Journal 294 (1987): 1125—1127.

A nonsmoking employee in a typical office would have to work morethan six and a half 40-hour weeks in a row to be exposed to nicotine"equivalent" of one cigarette, according to a new scientific study releasedtoday.

The study also shows that it would take a marathon eating session last-ing 398 continuous hours for a diner in a restaurant to be exposed tothe nicotine "equivalent" of a single cigarette.

The study fimdings by an independent scientific team, cover actual scien-tific measurement of nicotine—an indicator of cigarette smoke in the air—in38 offices and 36 restaurants randomly selected in Dallas.

"The levels of nicotine found in the study indicate that smoking regula-tions are unnecessary in order to assure adequate indoor air quality," saidDavid Weeks, M.D., an expert on substances in indoor air.

Tobacco Institute, news release, New Study of Actual Air Quality inRestaurants, Offices Shows Tobacco Smoke an Insignificant Factor (June 30,1987).

Research efforts on indoor air quality should consider fungi as an integraland important parameter to study.

Health and Welfare Canada Working Group on Fungi and Indoor Air,Significance of Fungi in Indoor Air: Report of a Working Group. Distributedat Indoor Air '87, the Fourth International Conference on Indoor Air Qual-ity and Climate, West Berlin, August 1987.

Prohibition of smoking has not been shown to have any measurable ef-fect on either indoor air quality or associated health and comfort symp-toms of sick building syndrome. Ventilation required to remove indoorcontaminants produced by the occupants themselves, specifically CO2

124 Clearing the Air

and body odor, will also remove the constituents of ETS. On the otherhand, if adequate ventilation rates are not provided, then indoor-generatedsubstances and dusts and chemicals infiltrating the building envelope fromoutdoors increase in concentration to unacceptable levels, even if ETSshould be entirely absent.-

In modern office buildings under normal use and occupancy,• • . ETS does not appear to contribute significantly to a build-up of con-taminants in offices.

Theodore D. Sterling, Chris W. Collett, and Ella M. Sterling, "Environmen-tal Tobacco Smoke and Indoor Air Quality in Modern Office Work En-vironments," Journal of Occupational Medicine 29 (1)(1987): 57-62.

None of the six case control studies yielding a positive relationship wasconducted according to the state of art of epidemiological research, givingreasonable and sound evidence which cannot be explained by chance, bias,confounding or misclassification.

The available studies, while showing some evidence of association, donot exclude chance, bias or confounding. They provide, however, a serioushypothesis. Further studies are needed, if one wants to come to an ade-quate and scientifically sound conclusion concerning the question as towhether passive smoking causes lung cancer in man.

Of the three prospective studies, only one shows a moderate risk in-crease of 1.74.

Of the 12 case-control studies, two contribute nothing to the evidence,six show a moderate risk increase, but do not sufficiently exclude chance,bias and confounding, four studies show a moderate risk decrease or norisk change.

All studies with positive associations can just as well be explained bychange, bias, confounding or misclassification. Such poorly conducted andinconclusive studies cannot be added or pooled to get convincing evidence,as has been attempted in serious efforts to evaluate the situation. Scienceshould disregard poor studies. False plus false does not equal true.

K. Uberla, "Lung Cancer from Passive Smoking: Hypothesis or Convinc-ing Evidence," International Archives of Occupational and Environmental Health59 (1987): 421-437.

The FAA/PHS 1970-71 study of tobacco smoke contamination in air-craft has been labeled by some s "inadequate" HO\VCVCr, the limited data

1-ippenaix 125

• that have been collected in subsequent less-structured testing, prior to theNAS study, have done little to refute the conclusion reached in the earlystudy, i.e., that, based on environmental levels and expected dose-responserelationships of contaminants, tobacco combustion products do not repre-sent a hazard to the nonsmoking passengers.

U.S. Department of Transportation, Report to Congress Airline Cabin AirQuality, Report of the Department of Transportation to the United StatesCongress (Washington, D.C.: Government Printing Office, February 1987).

The NAS Committee recommends a ban on smoking on all domesticcommercial flights, for four major reasons: to lessen irritation and discom-fort to passengers and crew, to reduce potential health hazards to cabincrew associated with ETS (environmental tobacco smoke), to eliminatethe possibility of fires caused by cigarettes, and to bring the cabin air qual-ity into line with established standards for other closed environments.

The Committee stated that "Empirical evidence is lacking in qualityand quantity for a scientific evaluation of the quality of airliner cabin airor of the probable health effects of short or long exposure to it."

We agree that exposure to ETS could be viewed as a problem by somecrew and passengers. However, we believe that further study is neededbefore the Department can propose a definitive response to this recom-mendation.

Ibid.

None of the four epidemiological studies which specifically examine theeffect of exposure of women to ETS at work finds a statistically signifi-cant increase in risk.

Anthony Arundel, Ted Irwin, and Theodor Sterling (Faculty of AppliedSciences, School of Computing Science, Simon Fraser University, Burnaby,British Columbia, Canada), "Nonsmoker Lung Cancer Risks from TobaccoSmoke Exposure: An Evaluation of Repace and Lowrcy's PhenomenologicalModel:' Journal of Environmental Science and Health C4 (1) (1986): 93-118.

There are three problems with Repace and Lowrey's phenomenologicalmodel. (1) The lung cancer risk estimates derived from the SDA studyare based on very few observed deaths and are unstable. (2) The apparentdifferences in lung cancer mortality between SDA and nonSDA never

126 Clearing the Air

smokers may be due to a variety of factors other than ETS exposure.(3) A number of assumptions and calculations are made that are clearlyincorrect.

For several age groups, the phenomenological model predicts more lungcancer deaths among never smokers as a result of exposure to ETh alonethan a reasonable estimate of the total number of lung cancer deathsfrom all causes.

Repace and Lowrey's phenomenological estimate of the lung cancerrisk for nonsmokers from ETS exposure is unstable and inaccurate Theevidence presented by Repace and Lowrey to support the plausibility oftheir estimate is based on several errors and unrealistic assumptions.Without corroborating data from independent sources, little confidencecan be placed in their results.

Ibid.

Of the four studies that have considered ETS exposure outside the home,three have attempted to define some composite measure of exposure fromseveral sources. None of these analyses yielded a significant relationshipbetween composite ETS exposure and lung cancer. The case-control studyof Garfinkel, et al., which represents the most rigorous effort to dateto document total ETh exposure, found no dose-response relationshipbetween the hours/day of ETS exposure from all sources and lung cancerrisk.

The epidemiological data concerning any cause and effect relationshipbetween ETS and lung cancer is, at best, equivocal. The most persistentweakness is the absence of adequate exposure information. This precludesthe evaluation of dose-response relationships—a critical element of thetest for causality. Moreover, problems of sampling bias and misclassifica-tion complicate interpretation of studies done to date.

Nancy J. Baiter (Department of Biology, Georgetown University,Washington, D.C.), S. James Kilpatrick (Department of Biostatistics,Medical College of Virginia, Richmond, Virginia), Philip Witorsch (Depart-ment of Medicine, George Washington University, Washington, D.C.),and Sorell L. Schwartz (Department of Pharmacology, Georgetown Univer-sity, Washington, D.C.), "Causal Relationship bet\veen EnvironmentalTobacco Smoke and Lung Cancer in Non-smokers: A Critical Reviewof the Literature," Proceedings of the Air Pollution Control Association,April 11, 1986.

Risk extrapolation based on the presence of carcinogens, in the absenceof appropriate epidemiological data, does not establish causation. It presumescausation.

Ibid.

The conclusion that exposure to ambient tobacco smoke produces . . . about5,000 lung cancer deaths per year in U.S. nonsmokers aged 35 years . . ."belongs more to speculation than reality. When we take into account theoutcome of the randomized trials together with the paradoxical implica-tions of the findings of Sandier et al. (1985), the best estimate of LCDsper year is approximately zero.

P.R.J. Burch (Department of Medical Physics, University of Leeds, GeneralInfirmary, Leeds, United Kingdom), "Health Risks of Passive Smoking:Problems of Interpretation," Environment International 12 0 986): 2 3-28.

A total of 16 scientific contributions and working papers were presentedat the Vienna meeting that closed with a round-table discussion duringwhich important aspects of the symposium were gone into once more,in greater depth. With the exception of Dr. Hirayama, all the participantsagreed that so far there was no definite proof of a causal relationship be-tween passive smoking and the risk of lung cancer. None of the epi-demiological studies considered satisfied the essential criteria of scientificmethodology.

Joseph Handler (former director of the World Health Organization), Prefaceto Passiverauche'n aw niedizinLcher Sicbt [Medical pempectives on passive Smoking](Geneva: IRL Imprimeries Reunies Lausanne, 1986), p. A.

The question as to whether the conceivable theoretical possibility of riskcalls for official preventive measures is not a medical but a politico-social one.

Ibid., p. B.-

After a detailed scientific evaluation of all the major buildings we havestudied, we have determined that high levels of environmental tobaccosmoke were the medical cause of indoor pollution problems in only 4percent of those buildings.

This result has also been corroborated by the work of NIOSH. In asimilar study of 203 buildings over a 5-year period, they found that

128 Clearing the Air

environmental tobacco smoke was the cause of problems in only 2 per-cent of their buildings.

It is also significant that in every single building we found environmen-tal tobacco smoke had accumulated we also found that high levels of fungalspores were also present in the air inside those buildings.

The reason environmental tobacco smoke often takes the blame for thesesymptoms is obvious: it is the only visible indoor pollutant. However,we have determined that the presence of high cdn cent ration of tobaccosmoke indicate the much more serious problem of poorly designed andan improperly maintained ventilation system.

U. S. Congress, House of Representatives, Hearings before the Subcom-mittee on Health and the Environment of the Committee on Energy andCommerce, Gray Robertson, statement, Designation of Smoking Areas inFederal Buildings, 99th Cong., 2d sess., June 12, 27, 1986.

The lung function of non-smokers who lived in homes with (a) no smokers,(b) a smoker who smoked one pack per day, and (c) a smoker who smokedtwo or more packs per day, were compared. In this comparison all membersof the household were included and this total group was separated intothose who lived in homes with electric stoves and those who lived in homeswith gas stoves. In general, the lung function of non-smokers who livedin homes with electric stoves was not influenced by passive smoke.

The lung function of non-smokers who lived in homes with gas stovesdid not appear to be affected by passive smoke, but the numbers in eachsmoking category was small.

H. Roland Hosein and Paul Corey (Occupational and Environmental HealthUnit, University of Toronto), "Domestic Air Pollution and Respiratory Func-tion in a Group of Housewives," Canadian Journal of Public I-Iealth 77 (1986):44-50.

Our examination has shown that the figures derived by the authors [RepaceandLowrey] are based upon incorrect theoretical assumptions and inflatedempirical estimates. The other calculations, contained in their paper, sug-gest to us that the tendency to choose inflated estimates with regard toexposure in the home was consistently followed. As a result, one must-conclude that the estimate of 5,000 lung cancer deaths per year in theUnited States, due to exposure to ambient tobacco smoke, does not repre-sent an accurate assessment of the problem.

P1pJJt_I1L i.' 19Clark Johnson (Heinz Letzel, Munich, Germany), "Letter to the Editors,"Environment International 12 (1986): 21-22.

In the examination of the relationship between levels of air pollutantsand respiratory health, it is very important that any confounding effectof covariables be distinguished from the effect of air pollution itself. Wehave shown that the industrial area, which has the highest level of [totalsuspended particulates], has also the highest prevalence of domestic smok-ing, parental respiratory symptoms, and gas cooking.

Anthony T. Kerigan, Charles H. Goldsmith, and L. David Pengelly, "AThree-Year Cohort Study of the Role of Environmental Factors in theRespiratory Health of Children in Hamilton, Ontario. Epidemiologic SurveyDesign, Methods, and Description of Cohort," American Re-view of Respi-ratory Disease 133 (1986): 987-993.

The null hypothesis that passive smoking and lung cancer mortality are caus-ally unrelated still stands. Until it is rejected, I consider it irresponsible toapply risk management techniques, even if they had been applied correctly.

ID

S. James Kilpatrick (Medical College of Virginia, Richmond, Virginia) "Let-ters to the Editors," Environment International 12 (1986): 29-3 1.

The potential association of lung cancer with passive smoking has beenstudied epidemiologically, clinically, and with mathematical models. Therehave been both positive and negative studies, such that the associationis still considered a potential rather than probable one.

Michael D. Lebowitz (professor, Division of Respiratory Sciences ) Univer-sity of Arizona College of Medicine, Tucson, Arizona), "The PotentialAssociation of Lun g Cancer with Passive Smoking," Environment Inte-rna-tional 12 (1986): 3-9.

-

Only mathematical models, such as that of Repace and Lowrey (1985),

appear to estimate dosage to humans from environmental tobacco smokethat may be carcinogenic. These models are based on mathematical as sump-tions and calculations that appear to include overestimations. The doseso derived does not reflect estimates based on actual controlled laboratorystudies. Actual estimates imply a dosage to the passive smoker of less than2 cigarettes/day, except in unusual circumstances; these estimates are ten-to one-hundred-fold less than that in the Repace and Lowrey model.

130 Clearing the Air

The model and methods used by Repace and Lowry (1985) and othersdo not provide estimates of precision, confidence intervals, or Consistencyestimates, the very statistical bases of estimation procedures.

Ibid.

If misclassification of smokers and non-smokers and other possible biasesare not taken into account the result is likely to -be an estimateof lungcancer deaths attributed to passive smoking that is incompatible with theamounts of smoke to which non-smokers are exposed.

Peter N. Lee, "Misclassification as a Factor in Passive Smoking Risk," Lancet2(8511) (1986): 867.

Amongst lifelong non-smokers, passive smoking was not associated withany significant increase in risk of lung cancer, chronic bronchitis, ischaemjcheart disease or stroke in any analysis. From all the available evidence,it appears that any effect of passive smoke or risk of any of the majordiseases that have been associated with active smoking is at most small,and may not exist at all.

P.N. Lee, J. Chamberlain, and M.R. Alderson (Institute of Cancer Research,United Kingdom), "Relationship of Passive Smoking to Risk of Lung Cancerand Other Smoking-Associated Diseases," &itish Journal of Cancer 54(1986):97-105.

Our analyses showed no significant effect of passive smoking on lifelongnon-smokers as regards risk of chronic bronchitis, ischaemjc hear diseaseor stroke. In all the analyses relating the various indices of passive smokeexposure to these diseases, no significant differences were seen and slightdecreases in risk were as common as slight increases.

Ibid.

Overall the results showed no evidence of an effect of passive smokingon lung cancer incidence among lifelong non-smokers. In male patients,relative risks were increased from some of the indices but numbers ofcases were small and none of the differences approached statisticalsignificance In females, where numbers of cases were larger, such trendsas existed tended to be negative and indeed were marginally significantlynegative for passi smoking during travel and during leisure. For the

Appendix 13'

combined sexes no differences or trends were statistically significant at the95% confidence level; such trends as existed tending to be slightly negative.

Ibid.

It is not known if any tobacco incineration products per se are immunogenicin man.

To date, the role of hypersensitivity is as yet undefined. In spite ofthe considerable number of previous studies and their supporting evidenceof an allergic response to tobacco smoke, the obvious pitfalls of thesestudies (undefined study populations, inappropriate antigens, and inade-quate documentation of smoke sensitivity) undermine any significantconclusions.

Samuel B. Lehrer, Richard P. Stankus, and John E. Salvaggio (Clinical Im-munology Section, Department of Medicine, Tulane Medical Center, NewOrleans), "Tobacco Smoke Sensitivity: A Result of Allergy?" Annals ofAllergy 56(1986): 1-10.

With social class not allowed for in the analysis [regarding parent's smok-ing], any social class effect can readily be misinterpreted as a passive smok-ing effect, inasmuch as the social classes differ so much in their smokingrates.

N. Mantel, "Letter to the Editor: Does Passive Smoking Stunt the Growthof Children," International Journal of Epidemiology 15(3) (1986): 427-428.

Three studies have shown a small reduction in pulmonary function in nor-mal adults exposed to ETS. Interpretation of these findings is difficultbecause pulmonary effects in normal adults are likely to reflect thecumulative burden of many environmental and occupational exposures andother insults to the lung. Thus, the effects of ETS on the lungs of adultsare likely to be confounded by many other factors, making it difficultto attribute any portion of the effect solely to ETS.

National Academy of Sciences, Environmental Tobacco Smoke: MeasuringExposures and Assessing Health Effects (Washington, D.C.: National AcademyPress, 1986), p. 10.

Reports have noted an excess risk of cardiovascular disease in ETh-exposednonsmokers; however, methodological problems in the designs and analyses

-132 Clearing the Air

of these studies preclude any firm conclusions about the results. Studiesreporting that ETh can precipitate the onset of angina pectoris amongpeople who already have this condition are subject to the same precau-tionary note. Exposure to ETS produced no statistically significant effectson heart rate or blood pressure in school age children or healthy adultsubjects, either during exercise or at rest. Data are not available as to possibleadverse cardiovascular effects in susceptible populations, such as infants,elderly, or diseased individuals.

Ibid.

A study released today shows that a passenger seated in a no-smokingsection of a U.S. commercial airliner would have to complete eight con-tinuous New York-to-Tokyo round trips to be exposed to the nicotineequivalent of one cigarette.

The amount of cigarette smoke in the no-smoking sections of commer-cial airline cabins is so small that it would take the equivalent of 224 hours,or more than nine days, of non-stop flying to reach the exposure level,according to Guy Oldaker, senior research chemist of the Rj. ReynoldsTobacco Company.

All four systematic in-air studies—including . a 1971 report from theDepartment of Health, Education and Welfare and the Federal AviationAdministration, a 1983 study by a team from San Francisco GeneralHospital Medical Center and a 1984 study by Japanese researchers—reachedthe conclusion that the amount of cigarette smoke present in actual com -mercial flights is extremely small, and that the levels do not indicate ademonstrated risk to passengers or flight personnel.

Tobacco Institute, news release, "Study Shows Additional Airline Smok-ing Curb Proposed by National Academy of Sciences Is Not Justified"(August 12, 1986).

A 1985 poll by Tarrance and Associates of a representative sample of 1,000frequent flyers found that 83 percent—including 79 percent of ex-smokersand 8 1 percent of those who have never smoked—support the currentsmoking rules on commercial airlines. Department of Transportation (DOT)complaint files indicate only 2-3 percent of complaints from airline pas-sengers have to do with smoking.

Ibid.

Appendix 133

In this population highly significant associations emerged between dampand mold in the house and respiratory morbidity in children, at least asreported by their parents.

Parental smoking emerged as a less significant factor than might havebeen supposed, but the analysis excluded the possibility that either smok-ing or gas fumes could account for the observed association between damp,moldy housing and lower respiratory morbidity in this sample of primaryschoolchildren.

David P. Strachan and Robert A. Elton, "Relationship between RespiratoryMorbidity in Children and the Home Environment," Family Practice 3(3)(1986): 137-142.

Further studies on the relationship between involuntary smoking and car-diovascular disease are needed in order to determine whether involuntarysmoking increases the risk of cardiovascular disease.

Surgeon General's Report (Washington, D.C.: Government Printing Office,1986), p. 14.

The physiologic and clinical significance of the small changes in pulmonaryfunction found in some studies of adults remains to be determined. Thesmall magnitude of effect implies that a previously healthy individual wouldnot develop chronic lung disease solely on the basis of involuntary to-bacco smoke exposure in adult life.

Ibid., p. 62.

There are no studies of acute respiratory illness experience in adults ex-posed to environmental cigarette smoke.

Ibid.

Sevefal conclusions -arise from this study. Episodes of lower respiratoryillness, defined as those in which there were one or more consultationsat which adventitious lung sounds were recorded, are particularly fre-quent in the children of manual workers. This cannot be explained bythe many social and family variables examined in this study such as over-crowding, smoking habits, parents' respiratory symptoms and breastfeeding.

C.J. Watkins, Y. Sittampalam, D.C. Morrell, S.R. Lceder, and E. Tritron(Department of General Practice, United Medical and Dental Schools of

1 34 Clearing the Air

Guy's and St. Thomas's Hospitals, London) "Patterns of Respiratory Ill-ness in the First Year of Life," British Medical Journal 293 (1986): 794-796.

We studied the acute effects of one hour of passive cigarette smoking onthe lung function and airway reactivity of nine young adult asthmaticvolunteers. Passive smoking produced no change in. expiratory flow rates.We conclude that passive smoking presents no acute respiratory risk toyoung asymptomatic asthmatic patients.

Herbert P. Wiedemann, Donald A. ' Mahler, Jacob Loke, James A. Virgulto,Peter Snyder, and Richard A. Matthay (Department of Medicine, YaleUniversity School of Medicine, New Haven), "Acute Effects of Passive Smok-ing on Lung Function and Airway Reactivity in Asthmatic Subjects," Chest89(2) (1986): 180-185.

A thorough and critical examination of the relevant literature fails to pro-vide compelling evidence that exposure to ambient tobacco smoke pro-duces adverse chronic health effects.

P. Witorsch (physician and professor of medicine, George WashingtonUniversity, Washington, D.C.), "Passive Smoking," New Zealand MedicalJournal November 12, 1986, p. 865.

A review of the literature indicates that there is no substantial evidenceto support the view that exposure to environmental tobacco smoke presentsa significant health hazard to the nonsmoker.

D. Aviado (physician and former professor of medicine) "Health IssuesRelating to 'Passive Smoking,' "in R.D. Toll ison, ed., Smoking and Society(Lexington, Mass.: Lexington Books, 1985), p. 158.

The results suggest that many people, both smokers and nonsmokers,may be at risk from CO generated by certain domestic heating systemsand that nonsmokers are far more likely to be exposed to high levelsof CO from these sources than from being in a room with a heavysmoker. Poor ventilation associated with the current trend towards ex-cluding all draughts is likely to exacerbate the situation for both smokersand nonsmokers.

B.D. Cox and Margaret J . Whichelow (Office of the Regius Professor ofPhysic, Cambridge University School of Clinical Medicine, Cambridge),"Carbon Monoxide Levels in the Breath of Smokers and Nonsmokers:Effect of Domestic Heating Systems," Journal of Epidemiology and Com-munity Health 39 (1985): 75-78.

Other covariates had virtually no explanatory power, and were ratherunstable across subpopulations: educational levels of head of household,sex, and mother's smoking status. In particular we found that a mother'ssmoking in the home was unrelated to acute respiratory disease incidenceof her children. However, this should not be too surprising in view ofthe contradictory findings on the health effects of passive smoking.

Winston Harrington and Alan J. Krupnick, "Short-Term Nitrogen Diox-ide Exposure and Acute Respiratory Disease in Children," Journal of theAir Pollution Control Association 35 (1985): 1061-1067

There were no significantly increased risks for having a mother, a father,or spouse(s) who smoked or for being exposed at work.

Our data are not consistent with the findings with regard to nonsmokersobtained by Hirayama and Trichopoulos et al.

The association of lung cancer risk with exposure to coal heating orcooking warrants further investigation.

Anna H. Wu, Brian E. Henderson, Malcolm C. Pike, and Mimi C. Yu (Depart-ment of Family and Preventive Medicine, University of Southern Cali-fornia School of Medicine), "Smoking and Other Risk Factors for Lung Cancerin Women," Journal of the National Cancer Institute 74(4) (1985): 747-751.

Children exposed to air pollution in the home from a cigarette-smokingparent show no adverse lung performance, a University of Toronto specialistsaid Tuesday.

Roland Hosejn told an environmental concerns conference that whilehe cannot fully explain it, tests in children exposed to cigarette smokealone showed no deterioration in their lung function as measured by theirability and rate of exhaling.

"Conference Told Smoke Didn't Hurt Children's Lungs," Regina Leader-Post, September 27, 1984, Blo.

Because of concern among flight attendants about passive exposure to ciga-rette smoke during work on commercial aircraft, a preliminary investigation

136 Clearing the Air

was conducted to search for an increase in expired air (end tidal) carbonmonoxide in flight attendants after work.

All flights were "turnaround" flights from Los Angeles to Honoluluand back. These flights were of about five hours' duration in each direction.

There was no increase in the concentration of carbon monoxide in theexpired air (end tidal) of these flight attendants during the flights in thisstudy. In fact, their exhaled air carbon monoxide levels decreased by aninsignificant amount.

These results indicate that the concentration of smoke to which flightattendants are passively exposed is too low to alter significantly their ex-pired air carbon monoxide levels.

Douglas B. Duncan and Peter P. Greaney (University of California, Irvine,Southern Occupational Health Center), "Passive Smoking and Uptake ofCarbon Monoxide in Flight Attendants," Journal of the American MedicalAssociation 25 1(20) (1984).

No convincing differences for viral infection or respiratory illness wereseen with parental smoking as an isolated factor.

Gregory Gardner, Arthur L. Frank, and Larry H. Taber (Influenza ResearchCenter, Department of Microbiology and Immunology, and Baylor Col-lege of Medicine, Houston, Texas), "Effects of Social and Family Factorson Viral Respiratory Infection and Illness in the First Year of Life," jour-nal ofEpidemiology and Community Health 38 (1984): 42-48.

Taking our own results into consideration, it seems that the passive in-halation of tobacco smoke at home or in the workplace by healthy in-dividuals probably does not lead to any essential impairment of pulmonaryfunction.

Michael Kentner, Gerhard Triebig, and Dieter Weltie (Institute of Occupa-tional and Social Medicine and Polyclinic of Occupational Diseases of theUniversity of Erlangen-Nuremberg), "The Influence of Passive Smokingon Pulmonary FunctionA Study of 1,351 Office Workers," PreventiveMedicine 13(1984): 656-669.

200 female lung cancer patients and 200 healthy district controls wereinterviewed to identify and quantify the various sources of passive smok-ing among Chinese females in Hong Kong. For the ever-smokers, passiveexposure from external sources do not appear to add to their risk. Forthe never-smokers, qualitative assessments (Sill bke exposure categories, age

Ii- ' 3/

When passive exposure started), and quantitative assessments (hours, years,intensity) showed no significant differences between the data for patientsand controls. Moreover, higher relative risks were not associated withhigher levels of passive smoking for the ever- or never-smokers. Thus,our findings would seem to indicate that passive smoking, as an isolatedfactor, did not have an influence on female lung cancer incidence in HongKong.

L.C. Koo, J.H-C. Ho, and D. Saw (Department of Community Medicine,University of Hong Kong, Hong Kong), "Is Passive Smoking an AddedRisk Factor for Lung Cancer in Chinese Women?" Journal of Experimentaland Clinical Cancer Researcb 3(3) (1984): 277-283.

We did not find any significant interaction between the smoking habitsof either parent smoking and their spouses' lung function, similar to Corn-stock and coworkers and Schilling and Associates, but different from Kauff-mann and coworkers.

Michael D. Lebowitz, Rj. Knudson, and B. Burrow (University of ArizonaCollege of Medicine, Tucson, Arizona), "Family Aggregation of PulmonaryFunction Measurements," American Review ofRespiratoiyDisea.se, 129 (1984):8-11.

White and Froeb studied self-selected volunteers with regard to workplaceexposure and reported some effects of passive smoking in a subset of theirvolunteers. . . . Even with a biased population, poor study design, andincorrect statistical evaluation, there were no clearcut, consistent, medicallymeaningful differences between passive smokers and groups of nonsmokers;a corrected statistical analysis strengthened this conclusion.

Michael D. Lebowitz, "Influence of Passive Smoking on Pulmonary Func-tion: A Survey," Preventive Medicine 13(1984): 645-655.

From a few studies on occupational groups exposed to carbon monoxide(CO) and from experiments with animals chronically treated with COor nicotine, the conclusion can be drawn that neither CO nor nicotineis likely to play a role in the development and progression of coronaryheart disease in those concentrations normally found in passive smokers.

H. Schievelbein and F. Richter (Institute of Clinical Chemistry, GermanHeart Center, West Germany), "The Influence of Passive Smoking on theCardiovascular System," Preventive Medicine 13 (1984): 626-644.

138 Clearing the Air

Should lawmakers wish to take legislative measures with regard to environ -mental tobacco smoke, they will, for the present, not be able to base theirefforts on a demonstrated health hazard from environmental tobacco smoke.

H. Valentin (Executive Committee, German Society of OccupationalMedicine) and E. Wynder (director, American Health Foundation), "HealthDanger through Passive Smoking Not Proven: Physicians' View on PassiveSmoking," press release from International Symposium, "Passive Smok-ing from a Medical Point of View," April 9-12, 1984, Vienna, Austria.

Passive smoking was not associated with an increase in total mortality.

J.P. Vanderbroucke (epidemiologist, Department of Epidemiology, ErasmusUniversity, the Netherlands), J.H.H. Verheesen, A. De Bruin, B.J. Mauritz(students, Department of Environmental and Tropical Health, AgriculturalUniversity, Wageningen), C. Van Der Heide-Wessel (retired communityphysician, Department of Health, City of Amsterdam), and R.M. van DerHeide (infectious disease physician, Academic Hospital, University ofAmsterdam), "Active and Passive Smoking in Married Couples: Resultsof 25 Year Follow Up," British Medical Journal 288 (1984): 1801-1802.

We conclude that there is passive absorption of nicotine from tobaccosmoke by flight attendants during a transoceanic flight but that the quan-tity consumed (equivalent to one cigarette) is relatively small comparedwith that consumed by cigarette smokers, and the concentrations achievedare unlikely to have physiologic effects.

Donna Foliart, Neal L. Beno\vitz, and Charles E. Becker (San FranciscoGeneral Hospital Medical Center), "Passive Absorption of Nicotine inAirline Flight Attendants," New England Jouriwl ofMedicine 308(18) (1983):1105.

Based on theoretical and empirical results, CO sidestream emissions fromcigarettes have often been overemphasized.

John R. Girman and Greg W. Traynor (staff scientists, Building Ventila-tion and Indoor Air Quality Program, Lawrence Berkeley Laboratory), "In-door Concentrations," Journal of the Air Pollution Control Association 33(2)(1983): 90.

There were no significant changes in the pulmonary function parametersmeasured in any of the 5U1)jCCCS when compared with baseline values.

Thus, passive exposure to cigarette smoke in these subjects producedmarked symptoms described as usual asthma but not significant objectiveevidence of airways obstruction.

Alvin J . Ing and B.X. Breslin (Chest Unit, Concord Hospital, Sydney,Australia), 'The Effect of Passive Cigarette Smoking on Asthmatic Patients."Paper presented at the Annual Scientific Meeting of the Thoracic Societyof Australia, May 1983, pp. 541-543.

The present study showed no significant effect on FEy10 from exposureto smokers in the home.

Jeffrey R. Jones, Ian TI Higgins, Millicent W. Higgins, and Jacob B. Keller-(University of Michigan, School of Public Health, Department of Epidemi-ology, Ann Arbor), "Effects of Cooking Fuels on Lung Function in Non-smoking Women,> ' Archives of Environmental Health 38(4) (1983): 2 19 — 2 2 2.

There were no effects of ETh on [pulmonary function] or symptoms inchildren or adults, asthmatics or others.

Michael D. Lebowitz, "The Effects of Environmental Tobacco Smoke Ex-posure and Gas Stoves on Daily Peak Flow Rates in Asthmat

leportfrwn and Non-

Asthmatic Families," in ETS—Environmental Tobacco Smoke. a Workshop on Effects and E.posure Levels, ed. Ragnar Rylander et al. (Geneva:University of Geneva, March 15-17, 1983).

We measured oxygen uptake at the onset of pain in patients with stableangina pectoris, and found it to be reproducible.

There was no change in the mean oxygen uptake at the onset of anginawith any intervention.

These results are in contrast with previous reports of the effects of smok-ing and carbon monoxide on exercise performance in angina pectoris.

M.W. McNicol andJ.A. MCM Turner (Central Middlesex Hospital, Lon-don), "Oxygen Uptake at the Onset of Angina Pectoris: Effects of Nicotineand Carbon Monoxide," Clinical Science 65(3) (1983): 24.

A scientific study that the Environmental Protection Agency used to seta major air-quality standard was seriously flawed, and there is "considerableconcern about the validity of the results reported."

At issue is the work of Dr. Wilbert S. Aronow, whose research into thepollutant's effect on heart patients was instrumental in setting the current

140 Clearing the Air

standard. The EPA formed a team of government and outside scientiststo look at Aronow's work last April, after the Food and Drug Administra-tion found he had falsified data on an experimental heart drug.

In their report, released yesterday, the scientists said they "could notresolve the issue of possible falsification of data. However, we had con-siderable concern about the validity of the results reported. Raw data werelost or discarded, adequate records were not maintained, available datawere of poor quality, quality control was nonexistent or inadequate. . .

"We conclude that EPA cannot rely on Dr. A's data due to the con-cerns we have noted," the report said.

Cass Peterson, "EPA Probe Criticizes a Study Used in Air-Quality Stan-dard," Washington Post, June 7, 1983, p. A15.

A review of the data from the studies which have been carried out orare in progress which address the effect of passive smoking on the respiratorysystem suggests that the effect varies from negligible to quite small. Fromthis review, it was not possible to determine whether there is a specificgroup which is at increased risk or what the mechanism of the effect (ifany) may be.

U.S. Department of Health and Human Services, Report of Workshop onRespiratory Effects of Involuntary Smoke Exposure: Epide-miologic Studies May1-3 1983, Public Health Service, National Institutes of Health (Washington,D.C.: Government Printing Office, December 1983).

I will agree that contributions to indoor levels of CO by smoking in "nor-mally occupied residences" are probably quite small.

John E. Yocorn (TRC Environmental Consultants, Inc., East Hartford, Con-necticut), "Indoor Concentrations. Author's Reply," Journal of the Air Pollu-tion Control Association 33(2) (198 3): 89.

There are less passive smokers among patients than the controls; and morenon-smoking patients have non-smoking spouses. This finding is at variancewith that of Dr. Hirayama's (1981).

W.C. Chan and S.C. Fung, "bing Cancer in Non-Smokers in Hong Kong,"Canter Epidnioi, ed. E. Gindmin (Nv York: Gustav Fischer \rl,1982).

The pulmonary function testing showed that neither parental smoking norgas home cooking fuel adversely aflcCue(l lung fUflCUiOn or yearly ILI MY

Appendix 141

The results of the present study are good evidence that these factorsdo not affect the lung function of children living in the southwestern UnitedStates.

Russell Dodge (assistant professor of medicine, University of Arizona, Tuc-son), "The Effects of Indoor Pollution on Arizona Children," Archives ofEnvironmental Health 3 7(3) (1982): 151-155.

Given these and other criticisms of the White-Froéb study, it would ap-pear that the New England Journal of Medicine has, perhaps unwittingly,performed a disservice to its readership. It is extremely unfortunate thata study so fraught with methodological problems, as indicated throughnumerous criticisms by scientists in the United States and elsewhere, shouldhave been published in such a reputable journal of medicine. The White-Froeb study should, therefore, not be relied upon by the Congress, Federalagencies or other legislative or policymaking bodies when consideringrestrictions on smoking in public places.

Hon. L.H. Fountain of North Carolina, House of Representatives, December16, 1982, "White-Froeb Study Discredited by Scientists," CongressionalRecord, December 16, 1982, pp. E5252-E5254.

It must be concluded that passive smoking in the family, usually due toparental smoking habits, does not seriously affect permanent markers ofrespiratory disease such as pulmonary function.

Michael D. Lebowitz, David B. Armet, and Ronald Knudson (Division ofRespiratory Sciences, Westend Research Laboratories, Arizona HealthSciences Center, Tucson, Arizona), "The Effect of Passive Smoking onPulmonary Function in Children," Environment International 8 (1982):371-373.

The 1979 U.S. Surgeon-General's Report (US Public Health Service, 1979)devoted a chapter to the subject of allergy and tobacco smoke. It con-cluded that the existence of such an allergy was not clearly establishedbut that those with a history of allergies to other substances, especiallythose with rhinitis or asthma, were more likely to report the irritatingeffects of tobacco smoke. Whether this was psychological, rather than aphysiological, response is open to question.

P.N. Lee (independent consultant in statistics, Surrey, England), "PassiveSmoking," Food and Chemical Toxicology 20 (1982):223-229.

LLi

In

142 Clearing the AirV.

It is difficult to imagine that enclosure in a very smokey room did nothave some emotional impact upon patients who were liable to angina,and the psychological disturbance may have done more to hasten the onsetof symptoms than the increase of blood COHb.

R. Shephard, The Risk of Passive Smoking (London: Groom-Helm Ltd., Pub.,1982), p. 73.

The frequency of major respiratory symptoms among subjects showed littleevidence of an association with the presence of some one else in the house-hold who smoked cigarettes.

George W. Comstock, Mary B. Meyer, Knud J . Helsing, and Melvyn S.Tockman, "Respiratory Effects of Household Exposures to Tobacco Smokeand Gas Cooking," American Review of Respiratory Disease 124 (1981):143-148.

The methods used by White and Froeb are open to criticism.

Franz Adikofer, Gerhard Scherer, and H. Weimann (ForschungsgesellschaftRauchen und Gesundheit mbH, Hamburg, Federal Republic of Germany),"Small-Airways Dysfunction in Passive Smokers," New England Journal ofMedicine 303(7) (1980): 392-393.

Suspicions of such an allergy against tobacco smoke have existed for along time, but have never been substantiated.

For present, the question as to whether allergy to cigarette smoke existsor not should be kept open.

Gunnar Bylin (Allergy Section, Department of Internal Medicine, Hud-dinge, Sweden), "Tobacco Allergy—Does It Exist?" Lakartidningen 77(16)(1930): 1530-1532 (translation).

White and Froeb do not have reliable estimates of the smoke exposurein the environment of their nonsmokers.

Gary L. Huber (Harrd Medical School, Boston), "Small-Airways Dysfunc-tion in Passive Smokers," New England Journal of Medicine 30 3(7) (1980):392-393.

Our data thus do not suggest that asthmatic subjects have an unusual sen-sitivity to cigarette smoke.

II

Appendix 143

We would thus conclude that the specific sensitivity of asthmatic sub-jects is not a major consideration when determining air quality criteriafor rooms contaminated by cigarette smoke.

Roy J. Shephard, R. Collins, and F. Silverman (Department of PreventiveMedicine and Biostatistics, University of Toronto and the Gage ResearchInstitute), " 'Passive' Exposure of Asthmatic Subjects to Cigarette Smoke:'Environmental Research 20 (1979): 392-402.

The Aronow study must be evaluated in light of the fact thai the end-point of the study was highly subjective, that the stress factor was notcontrolled, and that a sham smoke or other environmental impingementwas not used. In other words, not only was the sample small, but thescientific design was exceedingly poor.

E. Fisher, statement, U.S., Congress, House, Committee on Agriculture,Subcommittee on Tobacco, Effect of Smoking on Nonsmokers, Hearing, 95thCong., 2d sess., September 7, 1978, pp. 35-40.

The influence of parents' tobacco smoking on the morbidity of the youngestchildren of 1-4 years of age, found by D.J. Hammer et al., could notbe confirmed in our study.

Bo Holma, Ole-Winding (Institute of Hygiene, University of Copenhagen),"Housing, Hygiene, and Health. A Study in Old Residential Areas inCopenhagen," Archives of Environmental Health (1977): 86-93.

In accordance with the findings reported by Colley (14,15) and un-like those reported by a number of other investigators, there was noevidence that so-called passive smoking due to parental smoking hadany appreciable effect on the appearance of respiratory symptoms inschoolchildren.

There is a definite relationship between respiratory symptoms in parentsand those in children.

K.F. Kerrebijn, H.C.A. Hoogeveen-Schroot, and M.C. van der Wa!,"Chronic Nonspecific Respiratory Disease in Children, A Five Year Follow-Up Study," Ada .Paediatrica Scandinavica 261 (Suppl.) (1977): 7-72.

Respiratory symptoms, disease and lung function were studied in 376families with 816 children who participated in a survey in three USAtowns. Parental smoking had no effect on children's symptoms and lung

I'll'

Clearing the Air

function. Also, there was no evidence that passive smoking affected eitherlung function or symptoms of adults.

We conclude that exposure to low levels of smoke produced by cigarettesmokers does not result in chronic respiratory symptoms or loss of lungfunction among children or among adults.

R.S.F. Schilling, A.D. Lctai, S.L. Hui, G.J. Beck,J.B. Schoenberg, and A.Bouhuys, "Lung Function, Respiratory Disease, and Smoking in Families:'American Journal of Epidemiology 106(4) (1977): 274-283.

In general, pollutant exposure was similar for control and disease subjects.The significant difference observed for particulate exposure was due todomestic smoking and was independent of the presence or absence of re-spiratory disease. We found no evidence to support a cause and effectrelationship between pollutant exposure and respiratory disease.

Ralph E. Binder, Charles A. Mitchell, H. Roland Hosein, and ArendBouhuys (Yale University Lung Research Center, New Haven), "Impor-tance of the Indoor Environment in Air Pollution Exposure:' Archives ofEnvironmental Health 3 IN (1976): 277-279.

A study of the effects of family smoking habits on the symptoms of otherfamily members has shown that symptoms of household members, espe-cially children, are related to smoking habits within the households butare not significantly so when symptoms in adults are controlled.

Michael D. Lebowitz and Benjamin Burrows, "Respiratory SymptomsRelated to Smoking Habits of Family Adults," Chest 69 (1976): 48-50.

A close association was found between parents' and children's respiratorysymptoms that was independent of parents' smoking habits. There wasno suggestion that exposure to the cigarette smoke generated when parentssmoked had any more than a small effect upon the child's respiratory symp-toms. While the sharing of genetic susceptibility between parents andchildren is a factor, therefore, cross infection, particularly in the familieswhere parents smoke, is an important element in the association.

J.R.T. Colley, "Respiratory Symptoms in Children and Parental Smokingand Phlegm Production," British Medical Journal 2 (1974): 201-204.