ARCHIVED - Chronic Diseases in Canada

 

Volume 29 · Supplement 2 · 2010

Environmental Tobacco Smoke (ETS)

Kenneth C. Johnson

Environmental tobacco smoke, also referred to as second-hand smoke or passive smoking, has been established as a causal risk factor for a number of health problems, principally cardiovascular, respiratory, and cancer outcomes. ETS is a combination of sidestream and mainstream cigarette smoke – sidestream smoke comes from the burning end of the cigarette, while mainstream is exhaled from the smoker. More than 50 studies between 1980 and 2005 have examined the relationship between lung cancer and exposure to ETS, and over the last 20 years, at least eight expert committees have independently concluded that ETS causes lung cancer in never-smokers. A recent meta-analysis (systematic summary) of studies of lung cancer risk in women who had never smoked, but whose spouse smoked, estimated a relative risk (RR) of 1.24 (i.e., a 24% increase in risk compared to women whose spouses had never smoked). Recent meta-analyses of lung cancer risk associated with ETS at work estimated a 19% and 39% increase in lung cancer risk for never-smokers exposed regularly to second-hand smoke in the workplace. Where ETS exposure has been examined for combined residential and workplace exposure, greater exposure results in higher risks; the summary lung cancer risk for women never-smokers in the highest category of combined residential and occupational lifetime ETS exposure was estimated at 1.78 (95% Confidence Interval (CI): 1.49-2.12), and for those women in the highest category of occupational exposures the summary risk was 2.25 (95% CI: 1.81- 2.79).

More recently a literature has developed on breast cancer and ETS and there are now more than 20 published studies. A recent meta-analysis found that regular exposure to ETS among women who were life-long non-smokers was associated with increased breast cancer risk (pooled summary risk estimate of 1.27 (95% CI: 1.11-1.45)). The risk estimate for the 5 studies with more complete exposure assessment (quantitative long-term information on the three major sources of passive smoke exposure, childhood, adult residential and occupational) was 1.90 (95% CI: 1.53-2.37); while estimates for 14 studies with less complete ETS exposure measures was only 1.08 (95% CI: 0.99-1.19). The overall premenopausal breast cancer risk associated with ETS was 1.68 (95% CI: 1.33-2.12), and 2.19 (95% CI: 1.68-2.84) for the 5 studies that incorporated three sources of exposure. For women who had smoked the breast cancer risk estimate was 1.53 (95% CI: 1.22-1.91) when compared to women with neither active nor regular passive smoke exposure; 2.08 (95% CI: 1.44- 3.01) for more complete passive exposure assessment. Although the International Agency for Research on Cancer (IARC) concluded in 2002 that the collective evidence on ETS and breast cancer was not supportive of a causal association, in 2005 the California Environmental Protection Agency became the first agency concerned with environmental health to evaluate the association between premenopausal breast cancer and ETS as conclusive.

The relation of ETS to other types of cancer has been less studied. The evidence from the handful of adult brain cancer studies and ETS is inconclusive. Studies of childhood cancer have been equivocal and are likely subject to important biases from recall, and participation; and there are limited numbers of studies of the relationship for other cancers. ETS may be of particular relevance to Canadians who, because of the cold climate, spend much of the year inside closed spaces with limited ventilation. Because of the large number of individuals who have been regularly exposed, even small increases in individual risk associated with ETS exposure can impact a substantial number of Canadians. Compelling evidence exists to warrant introduction of further measures to reduce exposure to ETS in Canada.

Introduction

The impact of environmental tobacco smoke (ETS) on health has been the subject of a large number of investigations and several in-depth reviews over the last 25 years.1a,2a ETS has been established as a causal risk factor for a number of health problems. The relationship between ETS and lung cancer has been the focus of more than 50 studies3a and a number of expert panels.1b-5 There are now more than 20 published studies of breast cancer and ETS,3b over 30 on childhood cancer and parental smoking,3c and a few on brain cancer.3d Studies have also been reported for cancers of the nasal cavity, head and neck, stomach, cervix, bladder and for adult leukemia.3e,6,7a For all other cancers, there is a dearth of information on the possible relationship with ETS. No association was noted in one study of bladder cancer and one study reported mixed results.8,9

This review begins by examining the importance of studying the relationship of ETS to cancer, the cancer risks associated with active smoking, differences in the constituents of passive and active smoke, measurement of individual exposure, population exposure to passive smoking, and the special importance of ETS given the cold Canadian climate. Next the association of ETS with lung cancer is discussed, highlighting the difficulties in studying the relationship, recent meta-analyses of spousal and of workplace exposure and the smaller subset of primarily recent studies that try to enumerate lifetime exposure to residential and occupational ETS. Third, breast cancer and ETS are examined in depth – an area where the accumulating evidence may prove to be of considerable public health importance and extend our understanding of breast cancer etiology. This is followed by a brief look at the equivocal research on ETS and childhood cancer and the limited research on ETS and adult brain cancer. The chapter concludes with a brief discussion of public health efforts to reduce ETS exposure.

Methods and background on epidemiological studies reviewed

Potential studies for review were identified through a MEDLINE search (terms: passive smoking, second-hand smoke, environmental tobacco smoke and cancer) to find studies of cancer risks in never-smokers with lifetime residential and occupational ETS exposure histories.

There are often challenges in characterizing large, disparate bodies of epidemiologic evidence of varying quality. Results presented here for lung and breast cancer are based on formal published meta-analyses, whereas less rigorous, and more descriptive analyses were available for other cancer sites. The quality of the studies included in a meta-analysis will impact on the quality of the meta-analysis and in this area of study misclassification of ETS exposure has been common in the individual studies and thus impacted the meta-analyses.

Background

Research into ETS and health effects

The first detailed reviews of ETS and health risks were performed independently in 1986 by the US National Research Council,1c and the US Surgeon General.4 Both concluded that ETS could cause lung cancer in persons who had never smoked. In 1993 the United States Environmental Protection Agency produced an extensive report10a with more than twice the number of studies available for analysis as available in 1986. In the next six years, five additional in-depth reviews were published (by the Australian National Health and Medical Research Council,11 the United Kingdom Department of Health,12 the California Environmental Protection Agency (Cal/EPA),2b the World Health Organization,13 and the United States National Toxicology Program14). In 2004 the International Agency for Research on Cancer (IARC) monograph series on the Evaluation of Carcinogenic Risks to Humans published Volume 83, Tobacco Smoke and Involuntary Smoking.5a

In 2005, the Cal/EPA updated their earlier Health Effects Assessment including summaries based on a weight of evidence approach.3f The review concluded that there is sufficient evidence that ETS exposure is causally related to the following non-cancer health effects:

  • developmental effects – reduced fetal growth, low birth weight, sudden infant death (SIDS), and pre-term delivery);
  • respiratory effects – acute lower respiratory tract infections in children (e.g. bronchitis and pneumonia), asthma induction and exacerbation in children and adults, chronic respiratory symptoms in children; eye and nasal irritation in adults, middle ear infections;
  • cardiovascular effects –heart disease mortality, acute and chronic heart disease morbidity and alter vascular properties.3g

The report also concluded that there was suggestive evidence for other risks including: spontaneous abortion, intrauterine growth retardation, adverse impacts on cognition and behaviour, allergic sensitization, elevated decreased pulmonary function growth and adverse effects on fertility or fecundity, elevated risk of stroke, and chronic respiratory symptoms in adults.3h

Active smoking and cancer

The interest in ETS and cancer is not surprising given the demonstrated causal relationships of active smoking to a number of cancers. In their 1986 monograph, IARC identified smoking as causing cancers of the lung, larynx, oral cavity, pharynx, oesophagus (squamous cell carcinoma), pancreas, urinary bladder, and renal pelvis.15 Observed relative risks ranged from about three-fold for pancreatic cancer to twenty fold for lung cancer. In their evaluation in 2002, the IARC expert group concluded that additionally there was now sufficient evidence for a causal association between cigarette smoking and cancers of the nasal cavities and nasal sinuses, oesophagus (adenocarcinoma), stomach, liver, kidney (renal-cell carcinoma), uterine cervix and myeloid leukaemia. Observed relative risks for these additional cancers generally were in the two to threefold range.5b Active smoking is estimated to account for about 45% of male cancer cases and 22% of female cancer cases in the USA.16 In 2002, over 36,000 deaths (16.3%) were attributable to active smoking in Canada.17a These were primarily deaths from cancer and coronary heart disease.

Toxicity of second-hand smoke compared with mainstream smoke

Because the idling cigarette burns at a much lower temperature (resulting in less complete combustion) and because more tobacco is pyrolised during smouldering than during inhalation (2 second puff profile versus a 60 second puff interval), on a per gram basis, the sidestream smoke from a smoldering cigarette contains higher amounts of over 40 known carcinogens – and dozens of possible or probable carcinogens – than the same volume of mainstream smoke. For example, on a per gram basis undiluted sidestream smoke contains 13 to 30 times as much nickel as undiluted mainstream smoke from a non-filter cigarette, up to 50 times as much formaldehyde, 2.5 to 3.5 times as much benzo[a]pyrene, 7.2 times as much cadmium, etc.10b,18 Most Canadians smoke filtered cigarettes which reduce some of the carcinogen exposures in mainstream smoke, but would have no impact on the quality of the sidestream smoke. Thus if the comparison was to filtered cigarettes, the ratios of carcinogens in sidestream to mainstream smoke would generally be higher. Furthermore, because about 80% of the tobacco in a cigarette burns between puffs, indoor pollution from tobacco smoke comes mainly from sidestream smoke.1d

However, because sidestream smoke is diluted by the room air, the actual concentration and hence exposure to carcinogens from sidestream smoke is considerably lower than that from active smoking. In addition, the concentration of sidestream smoke in the air is dependent upon other factors including: room size, ventilation rates, number of smokers in the room and the number of cigarettes smoked. Typically, non-smokers inhale much less tobacco smoke than smokers and are exposed to much lower concentrations because breathing rates are the same but, smokers inhale 35 ml per puff at a higher concentration, while passive smokers inhale about 1 litre per breath at lower concentrations. Passive smoke exposure in non-smokers is estimated to be, on average, about one percent of the active smoke exposure that an active smoker receives, but this is based primarily on the levels of cotinine (a marker of nicotine exposure) measured in the urine.10c,19 It is much more difficult to measure the relative exposure to carcinogens between smokers and non-smokers. For example, a recent study of the metabolites of 4(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), a tobacco specific carcinogen found that passively-exposed non-smoking wives of husbands who smoked, had on average 5.6% of the levels of the NNK metabolite in their urine that their husbands had as compared to 0.6% their husband’s levels of cotinine20

ETS exposure in Canada

ETS may be of particular relevance to Canadians who, because of the cold climate, spend much of the year indoors. The exposure to indoor air contaminants, such as ETS, is directly affected by the number of air changes per hour in an indoor space.21 Higher air change rates during cold weather increase heating costs, so air changes are kept to the minimum acceptable level, generally limited to control of humidity and odour.22

Nearly 5.0 million Canadians aged 15 years or older (19%) were active smokers in 2005 (16% of women, 22% of men).23a They smoked an average of 15.7 cigarettes per day.23b In addition there were 7.3 million former smokers (28% of the adult population).23c As a result, a large number of non-smoking Canadians are, or have until recently, been exposed regularly to ETS residentially as children or adults, occupationally and/or socially. Because of the large number of individuals who have been regularly exposed, even small increases in individual risk associated with ETS exposure can impact a substantial number of Canadians.

Fewer Canadians are being exposed to ETS at home and at work.24 In 1996-97, one-third of Canadian children under age 12 (nearly 1.6 million children), over 50% of children in the lowest income families, and 85% of children living with a daily smoker in the household were being exposed regularly to ETS at home.25 By 2005, the percent of Canadian children under age 12 regularly exposed at home to ETS was down to 9%.23d

Estimates of the percentage of Canadians that have had regular ETS exposure at some time in their lives are available from the National Enhanced Cancer Surveillance System (NECSS).26 The NECSS collected data from a population sample of over 5000 control subjects aged 20 to 74 from eight Canadian provinces (Newfoundland, Nova Scotia, Prince Edward Island, Ontario, Manitoba, Saskatchewan, Alberta and British Columbia) for the period 1994-1997. Overall, 50 percent of the women had actively smoked at some time, while 25% were still smokers at the time of interview. Of the 50% of women who had never smoked, 84% reported having lived with a smoker as a child or adult, or having worked for at least a year where colleagues regularly smoked in the immediate work area. The median number of years of passive exposure reported among women who never smoked was 27.27a The ETS exposure profile of these participants can differ to that of the Canadian population if participation is influenced by some correlate of smoking behaviour (socio-economic status, age, etc).

ETS exposure in various environments

Many studies have examined ETS exposure levels in different environments, and several summaries of these studies have been published. Nearly 100 studies were examined by Guerin et al. in 1992.28a ETS exposure studies have demonstrated consistently that exposure is particularly high in bars.29a Direct measures of specific air contaminants, such as nicotine, carbon monoxide and particulates, and measurements of serum and urinary cotinine levels in non-smokers who work in bars all suggest high exposure.

In an analysis summarizing published studies of passive smoke exposure in the workplace, Siegel30 built on the existing reviews, and found that measured levels of tobacco smoke in bars were 4.4 to 4.5 times higher than in residences with at least one smoker, and 3.9 to 6.1 times higher than in offices. Each workplace exposure estimate was based on 10 to 22 different studies. A recent analysis of studies examining mean nicotine concentrations found that nicotine levels were generally between 1 and 3 µg/m3 in homes, between 2 and 6 µg/m3 in offices, between 3 and 8 µg/m3 in restaurants and between 10 and 40 µg/m3 in bars.28b (More information on ETS exposure is available at http://www.repace.com/fact_exp.html).

Difficulties in studying ETS and cancer

Several difficulties in studying ETS and cancer contribute to uncertainty about the magnitude of risk caused by ETS; these include sample size, quantifying ETS exposure, the relatively small increases in relative risk, misclassification of ever-smoking status, socio-economic differences between smokers and non-smokers, and the limited number of studies that collected data beyond spousal ETS exposure.

Sample size

Inadequate sample size has been a limiting factor in ETS studies, particularly those of lung cancer. Among never-smokers, lung cancer is a rare disease affecting approximately 12 in 100,000 women per year.31a Conversely, 90% of lung cancers among men and 63% among women were estimated to be attributable to active smoking in Canada in 2002.17b As a result, obtaining a sample of several hundred never-smokers who have developed lung cancer is exceedingly difficult and expensive. First, over 60% of potential candidates will turn out to be ineligible but, generally, will have to be interviewed to establish that. Second, lung cancer is rapidly fatal, so cases must be ascertained and approached without delays and some will have died in the interim. (The use of proxies is unlikely to be suitable for describing historic ETS exposure in the workplace or childhood.) Third, a very large population base must be used to be able to complete data collection within a suitable time frame. The largest sample size to date was assembled in the IARC study (650 cases and 1300 controls) that included 12 study centres in seven countries. Differences in study design between centres (e.g., types of controls), in environment and climate, and in work practices may reduce the consistency of the results.

Quantifying ETS exposure

Quantifying ETS exposure is a complex task dependent on many factors, including duration of exposure, room size, season, ventilation, and exposure source. Researchers have developed questionnaires that provide valid measures of ETS exposure32a that correspond to biomarkers of tobacco smoke exposure, such as urinary cotinine levels. The correlation coefficients for association between the questionnaire-based exposure estimates and the biomarker levels, however, are low – ranging from 0.19 to 0.29. Historic ETS exposure is difficult to measure, as there are no biomarkers available that reflect long term exposure levels.32b Most of the studies have depended upon the simple measure of living with a spouse who smokes. More sophisticated studies have evaluated smoking in residential, social and work environments on a year-by-year basis from childhood. On the other hand, because people who smoke tend to smoke over a long time period and with a regular pattern (generally a number of cigarettes everyday at regular intervals), it may be that simple historic exposure indices (smoker-years of ETS exposure) may capture enough information to discern important risks and differentiate for most of the population relative differences in overall exposure.

Small increases in relative risk

Increased relative risks associated with ETS exposure will typically be modest, reflecting the lower overall average carcinogen exposures of ETS relative to active smoking. With the relative risk (RR) estimates for ETS and lung cancer averaging about 1.2 for spousal exposure among women (nonsmoking women who lived with a spouse who smoked),5c there is concern that even a small bias might explain the increase. For example, a bias could be introduced by misclassification of smoking status.

Misclassification of ever-smoking status

A small percentage of individuals who have been smokers will report that they never smoked. Because active smoking carries a high relative risk for lung cancer, even a small amount of misclassification of this sort would increase the risk of lung cancer among those classified as non-smokers and hence reduce the difference in risk between this group and those classified as smokers. Hackshaw et al. evaluated this and found that observed levels of misclassification (1.9% to 7%) would only reduce the summary odds ratio (OR) from 1.26 to between 1.19 and 1.21.33a In addition, individuals misrepresenting their smoking status tend to have quit many years prior and to have been light smokers,33b both of which limit the risk that their active smoking would contribute.

Socio-economic status and ETS

Individuals of lower socio-economic status have a higher risk of lung cancer34 and several surveys have demonstrated higher ETS exposure in this group as well.35 If another correlate of socio-economic status increased lung cancer risk (for example, air quality or diet), an association between lung cancer and ETS could be, at least partly, the spurious result of the association of both lung cancer and ETS with socio-economic status. A number of studies have found positive associations between lung cancer and outdoor levels of air pollution and measures of traffic density. Risk of breast cancer, on the other hand, is positively associated with socio-economic status; studies which inadequately control for socio-economic status or reproductive characteristics may fail to note a true association. The low SES population has higher rates of smoking and thus a higher likelihood of exposure to ETS.

Spousal smoking and cancer

Much of the focus on ETS and cancer has revolved around the risk associated with spousal exposure. This choice of focus for lung cancer and ETS was necessary in the early days because so few of the early studies had better exposure measures but its continued use is unfortunate because: (1) with a binary exposure measure (spouse smoked or not), all the “exposed” are put into one category even though there is a large gradient of exposure; (2) important sources of exposures are missed, particularly parental and workplace ETS exposures, which may be non-existent, equal to or far greater than spousal exposure. As a result, without childhood and workplace exposure information, many women, who may have had substantial total ETS exposure, will be put into the ETS unexposed referent category for spousal exposure.

ETS and lung cancer

Early studies of ETS and lung cancer focussed on spousal exposure, in part because two important early studies, one in Japan36a and one in Greece37 observed increased lung cancer risk in women with a husband who smoked. Because Japanese wives at the time were unlikely to have been exposed to significant ETS from any other source, a husband’s smoking history was a good proxy for a wife’s ETS exposure. This cohort study found that the rate of lung cancer death in never-smoking women whose husbands smoked was 45% higher when compared to never-smoking women with nonsmoking husbands.

Following publication of these results, other researchers with existing lung cancer datasets quickly used them to conduct similar analyses. Often the only question the studies asked about ETS, however, was whether the husband smoked, even though the wives may have had substantial ETS exposure at work or as a child. By 1986, twelve other analyses had been published; these were summarized in the meta-analysis of Wald and colleagues.38

Between 1981 and 1996, 20 case-control and three cohort analyses that examined residential ETS were published. All but three studies included fewer than 50 non- smoking cases and most focussed on spousal smoking only. In 1997, Hackshaw et al. published a second meta-analysis which essentially confirmed the results of the original meta-analysis.33c This time, however, there were seven times as many cases available and the data included several far more rigorous, in-depth and larger studies. Fifteen studies met the three quality criteria for inclusion in the meta-analysis. Hackshaw et al. calculated an unadjusted relative risk for women of 1.24 (95% Confidence Interval (CI) 1.13-1.36) for lifelong non-smokers living with a spouse who currently smoked compared with living with a spouse who had never smoked. An adjusted estimate, taking into account the possible bias that would be introduced if any smokers with lung cancer reported themselves as non-smokers, resulted in an estimate of 1.17 (95% CI: 1.05-1.45). Because women may have ETS exposure other than from their spouse, misclassification of the true ETS exposure status of some women will occur. Adjusting for this, Hackshaw et al. estimated that the OR would have been 1.42 (95% CI: 1.21-1.66) if spousal exposure alone were compared to those truly unexposed.33d

A recent Canadian meta-analysis of ETS and lung cancer found similar risk estimates to previous meta-analyses and found no statistically significant differences in the estimated risks when studies were grouped by study design.39 The meta-analysis conducted for IARC (2004) reported a pooled relative risk for spousal exposure as 1.24 (95% CI: 1.14-1.34) among women based on 46 studies and 1.37 (95% CI: 1.02-1.83) among men based on 11 studies.5d The report concluded adult non-smokers exposed to ETS have a higher risk for lung cancer.

A positive association was found in a large European cohort study published in 2007 which concluded that ETS caused between 16 and 24% of lung cancers, mainly due to the contribution of work-related exposure.40 A meta-analysis of workplace ETS indicated a 24% increase in lung cancer risk (RR 1.24, 95% CI: 1.18-1.29) among workers exposed to environmental tobacco smoke.41

Occupational ETS and lung cancer

By 1994, 14 studies had provided information on the risks associated with occupational ETS exposure. Five meta-analyses of occupational ETS and lung cancer were published between 1994 and 1996, each reporting on these 14 studies. In all five, the summary risk estimate for ever having been exposed to occupational ETS was close to unity. All five meta-analyses were conducted by employees or consultants to the tobacco industry.42a Several of the 14 studies included in the meta-analyses had significant study design deficiencies for addressing occupational ETS exposure.42b Some studies had only current workplace exposure, some relied heavily on proxy respondents (who likely would be unable to provide an accurate long-term occupational ETS exposure history) and some included ex-smokers in the group being analysed. A more recent meta-analysis by Wells,42c established stricter quality criteria and revisited the weighting of individual studies in the summary estimate. Based on five studies meeting six quality criteria, Wells found a summary risk estimate of 1.39 (95% CI: 1.15-1.68). The meta-analysis conducted for IARC (2004) reported a pooled relative risk for workplace exposure as 1.19 (95% CI: 1.09-1.30) among women based on 19 studies and 1.12 among men (95% CI: 0.80-1.56) based on 6 studies.5e

Lifetime occupational and residential ETS exposure and lung cancer risk

Table 1 summarizes recent studies of lung cancer and ETS among women that include measures of lifetime residential and occupational ETS exposure. The Fontham study43a is the largest study in the USA, with detailed exposure measures; the Boffetta study was a large study conducted in 12 European countries through the IARC32c and data from several of the other European studies reported in the table are included in the Boffetta study.

Where smoking is unrestricted in the workplace, measured mean concentrations of nicotine generally exceed those in residences of smokers and, in some work environments, the concentrations can be several times as high as the average levels in homes.29b Table 1 contrasts spousal risks with risks for the highest category (usually the highest quartile) of occupational and total ETS exposure. Exposure to spousal smoking was generally associated with risk increases of up to 25% in the individual studies and a summary risk was calculated as 1.20 (95% CI: 1.01-1.43). In contrast, the individual study risk estimates for the highest quartile of combined occupational and residential exposure – and similarly for the high occupational exposure estimates – were often statistically significant and generally ranged from about a 50% to 200% increase (ORs of 1.5 to 3.0). Results of a recent analysis in Canada44a are consistent with those of the other reported analyses, although the study size was relatively small, making the estimated risks somewhat unstable. A summary risk estimate based on the nine studies for women never-smokers in the highest category of combined residential and occupational lifetime ETS exposure was 1.78 (95% CI: 1.49-2.12). For those women in the highest category of occupational exposure the summary risk was 2.25 (95% CI: 1.81-2.79).

Observing increased risk primarily in those subjects likely to have been most highly exposed is not unexpected. Studies of non-smokers suggest that only those within the top quartile of passive exposure manifest much increase in urinary cotinine.45a

Table 1
Lung cancer risk associated with spousal, occupational and total passive smoking in female never–smokers – population studies with at least adult lifetime residential and occupational assessment of ETS exposure
Study Spousal risk Combined residential and occupational exposure – high exposure category High occupational exposure category
Fontham et al. 1994 (USA)43b 1.23 (0.96–1.57) 1.74 (1.14–2.65) 1.86 (1.24–2.78)
Boffetta et al. 1998 (Europe)32d a 1.11 (0.88–1.39) 1.49 (0.93–2.38) 1.87 (1.10–3.28)
Nyberg et al. 1998 (Sweden)86 1.05 (0.65–1.68) 2.52 (1.28–4.9) 2.51 (1.28–4.9)
Jockel et al. 1998 (Germany)45b 1.12 (0.54–2.32) 3.24 (1.44–7.32) 3.10 (1.12–8.60)
Zhong et al. 1999 (China)87 1.1 (0.8–1.5) 1.8 (1.1–2.8) 2.9 (1.8–4.7)
Kreuzer et al. 2000 (Germany)88 0.96 (0.70–1.33) 1.39 (0.96–2.01) 2.52 (1.12–5.71)
Lee et al. 2000 (Taïwan)89 2.2 (1.5–3.3) 2.8 (1.6–4.8) Not reported
Wang et al. 2000 (China)90 Not reported 1.51 (0.9–2.7) 1.93 (1.04–3.58)
Johnson et al. 2001 (Canada)44b 1.21 (0.6–4.0) 1.82 (0.8–4.2) 1.58 (0.6–4.0)
Summary Risk Estimatesb 1.20 (1.01–1.43) 1.78 (1.49–2.12) 2.25 (1.81–2.79)

a The Nyberg, Jockel and earlier Kreuzer data (173 of 292 cases) are included in this 12 centre, 7 country European study.

b Calculated using the method of DerSimonian and Laird91.

ETS and breast cancer

Breast cancer is the most commonly diagnosed cancer among women in Canada, and incidence rates among women 50 or more rose gradually between 1975 and 1992, but since 1993 have stabilized.46 The established potentially-modifiable risk factors for female breast cancer (primarily reproductive factors and lack of physical activity) account for less than half of breast cancer risk.47 The published studies of breast cancer and passive smoking provide some conflicting evidence regarding the impact of regular long-term exposure to ETS on breast cancer risk.

Historically, most studies which looked at active smoking and breast cancer have not observed an association; some have even suggested a reduced risk.48 Palmer and Rosenberg,49a in reviewing 19 studies of breast cancer meeting specific quality criteria related to their ability to evaluate smoking risk, found that relative risks ranged from 0.93 to 1.3 for women smoking at least one pack of cigarettes per day, compared to never-smokers. They concluded that “the current evidence strongly supports the idea that there is no risk of breast cancer related to smoking.” However, it has been suggested that the failure to note an increased risk for active smoking may lie in the choice of referent group.50a In all 19 studies reporting on active smoking and breast cancer, the referent group included all never-smokers, many of whom invariably were exposed to ETS.

Tables 2 to 4 presents a summary of the published studies of breast cancer and ETS. The basic study characteristics are given in Table 2; Table 3 summarizes the exposure measures, and Table 4, the risk results. The studies had to meet two basic study quality criteria: (1) include some quantitative measure of adult exposure to ETS and (2) confine the analysis to women who had never actively smoked. The studies are briefly described below in historic order, followed by a summary of a recent meta-analysis synthesizing the results.

Table 2
Published studies of passive smoking and breast cancer risk
Never–smokers
Study Place Years Study type Outcome Age range # of cases # of controls
Hirayama 199251a, a Japan 65–81 Prospective Death 40+ 115 91,540
Sandler et al. 198553a, a USA – North Carolina 79–81 Case/control Diagnosis 15–59 32 177
Smith et al.199454a, a United Kingdom 85–88 Case/control Diagnosis <36 94 100
Morabia et al. 199655a Switzerland 92–93 Case/control Diagnosis <75 126 620
Millikan et al. 199863a USA – North Carolina 93–96 Case/control Diagnosis 20+ 247 253
Lash et al.199956a USA – Massachusetts 83–86 Case/control Diagnosis all 120 406
Zhao et al. 199957a China – Chengdu 94–97 Case/control Diagnosis 26–82 252 259
Jee et al. 199958a Korea 94–97 Prospective Diagnosis 18–65 138 157,298
Johnson et al. 200027b Canada – 8 provinces 94–97 Case/control Diagnosis 20–74 608 727
Wartenberg et al. 200059a USA 82–94 Prospective Death 30–70+ 669 146,488
Delfino et al. 200066a USA ND Case/control Diagnosis 40+ 64 60b
Marcus et al. 200062a USA – North Carolina 93–96 Case/control Diagnosis 20–74 445 423
Nishino et al. 200161a Japan 84–92 Prospective Diagnosis 40+ 67 9,671
Egan et al. 200260a USA 82–96 Prospective Diagnosis 36–61 1,138 78,206
Kropp et al. 200265a Germany 92–95 Case/control Diagnosis <51 197 454
Lash et al. 200264a USA – Massachusetts 87–93 Case/control Diagnosis all 305 249
Gammon et al. 200467a USA – Long Island NY 96–97 Case/control Diagnosis all 443 457
Reynolds et al. 20043i,68a USA – California 95–2000 Prospective Diagnosis all 1,174 76,534
Shrubsole et al. 20043j,70b China – Shanghai 96–98 Case/control Diagnosis 25–64 1,013 1,117
Hanaoka et al. 200571b Japan 90–99 Prospective Diagnosis 40–59 162 20,169

a Risk estimates were obtained by Wells through personal communication with the authors.52a

b Delfino et al. Cases and controls were selected from 391 women with suspicious breast masses detected clinically or mammographically. Cases were the 113 women where histopathology indicated breast cancer, while controls were the 278 women with benign breast disease. Controls for this review were the 107 “low risk controls” with “normal breast or benign breast disease histopathologies with non-proliferative disease

Table 3
Published studies of passive smoking and breast cancer risk

 

Passive smoking exposure assessment
Study Summary of exposure measures Childhood exposure Adult residential exposure Occupational exposure Other Exposure
Hirayama 1992a Husband’s smoking history No Husband’s smoking history No  
Sandler et al. 1985a Childhood and husband’s history Yes Husband’s smoking history No  
Smith et al.1994a Lifetime residential and occupational Detailed history Detailed history Detailed history  
Morabia et al. 1996 Lifetime residential and occupational and social Detailed history Detailed history Detailed history Social
Millikan et al. 1998 Residential Years with smoker at home Lived with a smoker No  
Lash et al.1999 Lifetime residential Yes Yes No  
Zhao et al. 1999 Lifetime passive smoking history Yes Yes Yes Yes
Jee et al. 1999 Husband’s smoking history No Husband’s smoking history No  
Johnson et al. 2000 Lifetime residential and occupational # of smokers in each residence # of smokers in each residence # of smokers in each job/immediate work area  
Wartenberg et al. 2000 Husband’s smoking history No Husband’s smoking history Nob  
Delfino et al. 2000 Adult residential No Adult residential No No
Marcus et al. 2000 Residential Yes Lived with a smoker No  
Nishino et al. 2001 Living with a smoker in 1984 No Living with a smoker in 1984 No No
Egan et al. 2002 Maternal/paternal smoking, years as adult living with smoker, current (1982) work and home exposure Maternal and/or paternal smoking Years lived with smoker, current, 1982 Current, in 1982 only No
Kropp et al. 2002 Childhood, residential and occupational history Detailed history Detailed history Yes No
Lash et al. 2002 Years of exposure; age first lived with a smoker Yes Years lived with smoker No No
Gammon et al. 2004 Parental and spousal exposure Yes Yes No No
Reynolds et al. 2004 Residential Yes Yes No No
Shrubsole et al. 2004 Husband’s smoking history and occupational No Husband’s smoking history Hours per day over past 5 years No
Hanaoka et al. 2005 Residential and occupational Yes Yes Categorical No

a Risk estimates were obtained by Wells through personal communication with the authors.52b

b For main analysis no occupational exposure

See Table 2 for citation

Table 4
Published studies of smoking and breast cancer risk
Passive smoking Active smoking
Overall exposure Higher exposure category Overall exposure
Study Overall Preb Postb Overall Pre Post Overall Pre Post
Hirayama 199236ca 1.32
(0.83–2.09)
1.50
(0.5–4.2)
1.0
(0.3–3.6)
  1.53
(0.78–3.02)
Age 50-69
1.97
(1.07–3.6)
1.59
(1.01–2.52)
   
Sandler et al. 1985a 1.62
(0.76–3.44)
7.1
(1.6–31.3)
0.9
(0.4–2.2)
      1.21
(0.58–2.51)
   
Smith et al.1994a 2.53
(1.19–5.36)
2.69
(0.85–9.3)
NA NA 2.24
(0.75–6.68)
NA   2.00
(0.98–4.12)
 
Morabia et al. 1996 2.3
(1.5–3.7)
3.6
(1.6–8.2)
  2.5
(1.5–4.2)
    3.0
(1.9–4.8)
3.5
(1.5–7.8)
 
Millikan et al. 1998 1.3
(0.9–1.9)
1.5
(0.9–2.8)
1.2
(0.7–2.2)
      1.1
(0.7–1.7)
1.4
(0.8–2.6)
1.7
(1.1–2.6)
Lash et al.1999 2.0
(1.1–3.7)
2.1
(1.0–4.1)
2.0
(1.1–3.7)
2.1
(1.0–4.1)
    2.0
(1.1–3.6)
   
Zhao et al. 1999 2.36
(1.66–3.66)
2.56
(1.63–4.01)
2.38
(1.66–3.40)
      3.54
(1.36–9.18)
   
Jee et al. 1999 1.3
(0.9–1.8)
    1.7
(1.0–2.8)
         
Johnson et al. 2000 1.48
(1.06–2.07)
2.3
(1.2–4.6)
1.2
(0.8–1.8)
  2.9
(1.3–6.6)
1.4
(0.9–2.3)
  2.3
(1.2–4.5)
1.5
(1.0–2.3)
Wartenberg et al. 2000 1.0
(0.8–1.2)
1.1
(0.8–1.6)
  1.0
(0.8–1.4)
    1.0
(0.8–1.2)
   
Delfino et al. 2000 1.78
(0.77–4.11)
          1.25
(0.27–5.82)
   
Marcus et al. 2000 0.8
(0.6–1.1)
          1.2
(0.8–1.6)
   
Nishino et al. 2001 0.58
(0.32–1.1)
               
Egan et al. 2002 1.07
(0.88–1.30)
    1.03
(0.86–1.24)
    1.15
(0.98–1.34)
   
Kropp et al. 2002 1.61
(1.08–2.39)
    1.83
(1.16–2.87)
    1.45
(0.96–2.19)
   
Lash et al. 2002 0.85
(0.63–1.10)
    0.75
(0.47–1.2)
    0.90
(0.8–1.0)
   
Gammon et al. 2004 1.04
(0.81–1.35)
1.21
(0.78–1.90)
0.93
(0.68–1.29)
1.22
(0.90–1.66)
    1.33
(0.97–1.83)
0.98
(0.54–1.78)
1.08
(0.72–1.62)
Reynolds et al. 2004c 0.94
(0.82–1.07)
0.93
(0.71–1.22)
    1.27
(0.84–1.92)
0.87
(0.73–1.03)
1.25
(1.02–1.53)
0.96
(0.55–1.68)
1.21
(0.95–1.54)
Shrubsole et al. 2004d 1.02
(0.81–1.29)
1.10
(0.83–1.46)
             
Hanaoka et al. 2005 1.1
(0.8–1.6)
2.6
(1.3–5.2)
0.6
(0.4–1.0)
           

Blank = Not reported
NA= Not applicable

a Risk estimates were obtained by Wells through personal communication with the authors.52c

b Pre = Premenopausal, Post = Postmenopausal

c Reynolds et al. 2004 passive smoking higher exposure categories from Reynolds et al. letter (2006).69a

d Shrubsole et al. (2004) combined husband or workplace only and husband and workplace exposure.

See Table 2 for citations

Initial interest in the issue of breast cancer and ETS arose from a cohort study in Japan51b which reported that breast cancer deaths were elevated by 32% among women whose husbands smoked.52d A case-control study in North Carolina,53b noted a 62% increase in breast cancer risk among women exposed to ETS, primarily among premenopausal women.52e A British case-control study of breast cancer in women under age 37 found more than a doubling of risk associated with ETS in the subset of cases and controls for whom passive smoking exposure was known.54b

Prompted by these findings, Morabia mounted a detailed case-control study in Switzerland to directly evaluate the impact of ETS on breast cancer.55b The study collected detailed year-by-year histories of passive and active smoking in residential, workplace and social environments from 244 women with breast cancer and 1032 population controls. Of these, 126 cases and 620 controls had no exposure to active smoking. Measurement of cotinine levels in study subjects’ urine augmented the validation work. In all four of these earlier studies, the ORs associated with the highest levels of exposure were close to or over 2.0. The Swiss study, which most accurately assessed the level of ETS exposure and restricted the passively exposed category to at least one hour per day for at least one year found ORs of 3.1 (95% CI: 1.3-7.5) and 3.2 (95% CI: 1.5-6.5) for less than 50 and greater than 50 hours/day-years of passive exposure. Wells calculated a four-study combined summary relative risk of 1.83 (95% CI: 1.40-2.40) for passive smoking and 2.17 (95% CI: 1.63-2.88) for ever having actively smoked.52f

Lash and Aschengrau56b in a case-control study in Massachusetts of 265 cases and 765 controls, also found a doubling of risk with passive exposure and with active smoking. Women exposed to ETS before age 12 had higher risks. The sample was primarily postmenopausal. A study by Zhao et al.57b in Chengdu, China found more than a doubling of breast cancer risk for passive (OR 2.36, 95% CI: 1.66-3.66) and active smoking (OR 3.54, 95% CI: 1.36-9.18).

Johnson et al. reported the results of a large Canadian case-control study (805 premenopausal and 1512 postmenopausal women with newly diagnosed primary breast cancer, and 2438 population controls).27c Among premenopausal women who were never active smokers, regular exposure to ETS was associated with an adjusted breast cancer OR of 2.3 (95% CI: 1.2-4.6). ETS exposure showed a strong dose-response trend (test for trend p < 0.001) with an OR of 2.9 (95% CI: 1.3-6.6) for more than 35 years of ETS residential and/or occupational exposure. When premenopausal women who had ever actively smoked were compared with women never regularly exposed to passive or active smoke, the adjusted OR for breast cancer was also 2.3 (95% CI: 1.2-4.5). At the same time, a direct comparison of women who had actively smoked with women who had never actively smoked, without controlling for passive smoking, showed no increase in premenopausal breast cancer risk, consistent with the active smoking meta-analysis of Palmer and Rosenberg.49b

Among postmenopausal women who were never active smokers in the Canadian study, regular exposure to ETS was associated with an adjusted breast cancer OR of 1.2 (95% CI: 0.8-1.8) and an OR of 1.4 (95% CI: 0.9-2.3) for the most highly exposed quartile of women. The adjusted OR for postmenopausal breast cancer risk for women who had ever actively smoked compared with women never regularly exposed to passive or active smoke was 1.5 (95% CI: 1.0-2.3). Statistically significant dose-response relationships were observed with increasing number of years of smoking, increasing number of pack-years and decreasing number of years since quitting. Women with 35 or more years of smoking had an adjusted OR of 1.7 (95% CI: 1.1-2.7). Passive and active smoking were associated with a 50% to 90% increase in risk among the younger half of the postmenopausal women (age up to 62). Risk was near null for the older women (age 63 to 75).

A prospective Korean cohort study found results quite similar to the Canadian study. The cohort study of 165,000 Korean civil servants and their spouses included a total of 138 pre- and postmenopausal breast cancer cases. Jee et al. found an overall relative risk of 1.2 for wives of ex-smokers, 1.3 for wives of current smokers, and 1.7 (95% CI: 1.0-2.8) for wives of current smokers with at least 30 years of smoking.58b An extended follow up of the cohort now includes 506 incident breast cancer cases. Preliminary analyses of these data suggest that women who lived with men who were smoking 20 or more cigarettes per day had a relative risk of 2.1 (95% CI: 1.5-3.0) for breast cancer under age 50 and 1.6 (95% CI: 1.0-2.6) for breast cancer at age 50 or higher. (Personal communication with the author, July 2000).

Two large American cohort studies, however, have not found an association between ETS and breast cancer risk. A cohort study using the American Cancer Society’s CPS-II (Cancer Prevention Study 2) cohort,59b examined breast cancer in a 12-year follow up of 147,000 never-smoking wives and found no overall increase in the risk of death from breast cancer associated with living with a husband who smoked (RR=1.0). An analysis of the Nurses Health Study cohort found a relative risk of breast cancer for regular passive exposure at work and at home (in 1982) of 0.90 (95% CI: 0.67-1.22), while the relative risk for active smoking was 1.04 (95% CI: 0.94-1.15).60b A Japanese cohort study also did not observe any increased risk (RR=0.6).61b

Two case-control studies from North Carolina reported by Marcus et al.62b, and by Millikan et al.63b did not observe an increased risk with either adolescent or adult ETS. However, although childhood exposure to passive smoking was quantified, these studies asked only one question on adult exposure to ETS, namely, had the subjects lived with a smoker when they were 18 years of age or older.

Based on a new set of cases and controls, Lash and Aschengrau64b were unable to replicate their earlier findings of an increased risk of breast cancer with exposure to ETS, with no effect observed with either active smoking (OR 0.72, 95% CI: 0.55-0.95), or passive smoking (OR 0.85, 95% CI: 0.63-1.1). However, a recent German case-control study65d reported an increased risk for ETS (OR 1.59, 95% CI: 1.06-2.39) and for active smoking (OR 1.45, 95% CI: 0.96-2.19).

Delfino and colleagues66b examined the issue of N-acetyltransferase 2 (NAT2) genotype as it relates to smoking and breast cancer risk. While finding no evidence that NAT2 was either a risk factor for breast cancer, or that it altered susceptibility to tobacco smoke, this study did note modest increases in risk to women exposed to ETS. The study by Gammon and colleagues67b, while not observing an association between ETS and breast cancer overall (OR 1.04, 95% CI: 0.81-1.35), observed a significantly increased risk for women who lived with a smoking spouse for more than 27 years (OR 2.10, 95% CI: 1.47-3.02).

Reynolds et al. (2004) conducted a study of passive and active smoking in the California Teachers Study cohort.68b An elevated risk for current smokers was reported. Relative to never-smokers not exposed to household ETS the hazard ratio was 1.25 (95% CI: 1.02-1.53). ETS exposure was limited to the question of ever having lived with a smoker as a child or adult. There was no association reported between ETS exposure and breast cancer among never-smokers, although a revised analysis using women age under 50 at diagnosis found a risk of 1.27 (95% CI: 0.84-1.92) for women exposed residentially both in childhood and as adults.69b

Shrubsole et al. (2004) using case-control data from the population-based Shanghai Breast Cancer Study reported no association with spousal smoking.70c There was some evidence for elevated risk for ETS exposure in the workplace of five hours or more per day (OR=1.6, 95% CI: 1.0-2.4) with a significant dose-response trend (p=0.02). Hanaoka et al. conducted a study of active and passive smoking in a cohort of Japanese women ages 40-59 with ten years follow up.71c An elevated risk was reported for active and passive smoking for premenopausal women but not postmenopausal women. Among women premenopausal at baseline with a reference of never-active smokers without ETS exposure, the RR for ever-smokers was 3.9 (95% CI: 1.5-9.9). Among premenopausal women at baseline the RR for residential or occupational/public exposure to ETS among never-active smokers was 2.6 (95% CI: 1.3-5.2).

Biological Plausibility

It is biologically plausible that cancer sites not directly in contact with tobacco smoke can be affected by it. For instance, pancreatic, cervical and bladder cancers have higher incidence among smokers.72 Petrakis et al.73a,74a report cigarette smoke mutagens in the breast fluid of non-lactating women, and nicotine has been found in greater concentrations in the breast fluid of smokers than in the plasma.75

Because mutagens in cigarette smoke accumulate in the breast tissue of non-lactating women,73b,74b it is biologically plausible that exposure to tobacco smoke is related to breast cancer. Compounds similar to those found in tobacco smoke (e.g., 7, 12-dimethylbenz(a)anthracene (DMBA)) are powerful breast carcinogens in animals.76

A number of studies have suggested that both passive and active smoking were stronger risk factors for premenopausal than for postmenopausal breast cancer, suggesting that there may be a subgroup of women at increased susceptibility for breast cancer when exposed to tobacco smoke (passive or active exposure), who tend to express their risk after relatively low exposures (and thus primarily at younger ages). Recent studies have focussed on the possibility that N-acetylation phenotype may affect breast cancer risk. The acetylator enzyme acts on common carcinogens such as those in cigarette smoke. It may also impact on risks observed in women with second-hand smoke exposure. Approximately half of white women have defective acetylator enzymes and are referred to as “slow acetylators”.77 Evidence to date is mixed regarding the role, if any, of acetylation status and risk of breast cancer, or how it might relate to passive smoking.

As in all observational epidemiological studies, there could be risk factors that correlate with tobacco smoke exposure and breast cancer that are unknown or not collected adequately and that might provide an alternative explanation for the results.78 Arguing against a causal relationship is the failure of positive studies to note any difference in the magnitude of risk between active and passive smoking, which seems counter-intuitive. A number of positive studies have not found an exposure-response relationship.

Breast Cancer Meta-analyses

To explore the risk heterogeneity among the 20 studies published above, a recent meta-analysis explored the impact on observed breast cancer risk of: study design (case-control or cohort); menopausal status (pre-menopausal or postmenopausal); publication date (before or after January 1, 2000), the study location (Europe, North America or Asia), the degree of confounder control; and the quality (completeness) of the (lifetime) ETS exposure assessment.79a Based on risk estimates from 19 studies meeting basic quality criteria, the meta-analysis found a pooled summary risk estimate of 1.27 (95% CI: 1.11-1.45) associated with regular long-term exposure to ETS among women who were life-long non-smokers. The quality of the exposure assessment best differentiated the level of observed risk. The five studies with more complete exposure assessment (quantitative long-term information on the three major sources of ETS exposure, childhood, adult residential and occupational), yielded a higher pooled risk estimate for ETS-exposed non-smokers of 1.90 (95% CI: 1.53-2.37). Studies with less complete ETS exposure measures resulted in little increase in risk (1.08). For pre-menopausal breast cancer, the overall risk associated with ETS exposure was 1.68 (95% CI: 1.33-2.12). Studies with better exposure measures yielded a premenopausal risk estimate of 2.19 (95% CI: 1.68-2.84). Figure 1 summarizes the individual study premenopausal risk estimates for all exposed women. For women who had smoked the breast cancer risk estimate was 1.53 (95% CI: 1.22-1.91) when smokers were compared to women who had neither active nor regular passive smoke exposure; 2.08 (95% CI: 1.44-3.01) for more complete and 1.15 (95% CI: 0.98-1.35) for less complete ETS exposure assessment.

Figure 1
Meta-analysis of passive smoking and premenopausal breast cancer risk among women who never smoked.79b.
Individual and summary risk estimates for women ever regularly exposed to passive smoking, stratified by the completeness of the passive smoking exposure assessment and study design. For Reynolds et al., 2004, new risk estimate from Reynolds et al. letter (2006)69c presented for women exposed in childhood and adulthood (risk for all exposed women not reported).

Contour levels for the estimated monthly mean ozone concentration field for June 1985, southern British Columbia
Text equivalent, Figure 1

Figure 1
Meta-analysis of passive smoking and premenopausal breast cancer risk among women who never smoked

Breast Cancer and ETS: Summary and Conclusions

There is a considerable degree of heterogeneity in risk estimates amongst the studies of breast cancer and ETS. In general, cohort studies have noted lower risks than case-control studies. For those studies which report data according to menopausal status, relative risks tend to be higher in pre-menopausal as compared to post-menopausal women. Studies also varied in how well confounders were controlled, and how completely exposure to passive smoking was assessed.

Prospective cohort studies are generally seen as methodologically superior to case-control studies, as they are generally free of concerns of response bias, proxy data and poor response rates, which are at least hypothetical problems with case-control studies. If case subjects were more likely than control subjects to recall times they lived or worked with smokers, then this could create an artificial increase in risk. The fact that three Asian cohort studies observed increased risk and dose-response relationships (where the case-control concerns do not apply); that both case-control and cohort studies with poorer exposure measures observe similar (lower) summary risk estimates; that this kind of bias does not appear to have substantively impacted on case-control studies of ETS and lung cancer or heart disease; and that premenopausal risk is fairly consistently higher than postmenopausal risk in these case-control studies all argue against these case-control specific potential biases as the explanation for the observed increases in risks.

An alternate explanation may be the inability of many cohort studies (and the case-control studies with poorer exposure assessment) to adequately identify the unexposed comparison group. For example, in the main analysis of the CPS-II American cohort study, ETS exposure information was limited to a history of spousal smoking and workplace and household exposure in 1982 only. The study did not collect information on the history of workplace, childhood or non-husband residential ETS exposure for the women. In a North American study, missing these ETS exposures is likely to result in important misclassification of exposure status.80 In their dose-response analysis only 50% of women were categorized as exposed to ETS from their husbands.59c Other studies that examined major sources of ETS exposure, including residential, workplace and sometimes social exposure to ETS, have found 80 to 95% of the women were exposed to ETS.27d,43c,56c The Nurses Cohort Study, the second large American cohort study, also only collected current workplace exposure in 198260c and the third study on California teachers has reported only on residential exposure.68c This misclassification may seriously dilute risk estimates.44c The results are particularly divergent for the women with higher ETS exposure.29c

The IARC concluded in 2002 that the collective evidence on ETS and breast cancer was not supportive of a causal association.5f Although four of the ten case-control studies reviewed found significant increases in risk, prospective studies as a whole did not report increased risk. The lack of a positive dose-response relationship also weighted against an association. The IARC evaluation was limited to the 15 studies of ETS and breast cancer available to mid 2002, and therefore didn’t include several cohort studies published since 2002 that have suggested active smoking risks. As well, the IARC document presented only descriptive data on the individual studies and a non-systematic evaluation of the quality of each study. No meta-analyses were performed to try to synthesize the 15 studies or examine the impact of study characteristics or quality, sub-populations or menopausal status on observed risk.

In contrast, the California EPA became the first agency concerned with environmental health to evaluate the association between premenopausal breast cancer and ETS as conclusive,3i based partly on the meta-analysis published in 2005 which reported an overall premenopausal breast cancer risk for ETS3j among life-long non-smokers of 1.68 (95% CI 1.33-2.12).79c One factor was the additional studies available including five of ETS and breast cancer. An additional factor in the California EPA conclusion was that using a referent population of never-smoking women not exposed to ETS, while there continued to be some heterogeneity in study results, the studies reviewed provided evidence of a role for active smoking in causation of breast cancer and included evidence of a dose-response relationship. A summary and extension of the Cal/EPA 2005 review also concluded ETS was causally related to breast cancer in premenopausal women.81a Others have concluded that the “jury is still out” on the subject of ETS and breast cancer.82a

A relationship between ETS and breast cancer has significant public health implications. Over 90% of the subjects in the large Canadian population-based study reported regular exposure to tobacco smoke at some time. Over 50% had been regular smokers at some time in their life, and another 40% of the women (all never-smokers) had been regularly exposed in some period of their life to ETS. Unlike most other established risk factors for breast cancer, exposure to ETS is modifiable through public policy. More study is warranted to clarify the exposure specifics of the relationship of ETS to breast cancer.

ETS and brain cancer

Evidence on passive smoking and brain cancer risk in adults is based on four studies. The cohort mortality study in Japan by Hirayama,36b with only 34 deaths, found the strongest association, with a more than threefold increase in brain cancer mortality among non-smoking wives of husbands who smoked. Risk varied by the number of cigarettes the husband smoked per day: the relative risk was 3.0 (95% CI: 1.1-8.6) for one to 14 cigarettes per day, 6.3 (95% CI: 2.0-19.4) for 15 to19 cigarettes, and 4.3 (95% CI: 1.5-12.2) for 20 or more cigarettes. A case-control study in the United States, with only 11 non-smoking cases,53c found increased risks for some types of cancer, including a non-significant increase in brain cancer risk related to husbands’, but not wives’, smoking. In a case-control study of intracranial meningioma and smoking in the United States results for active smoking were not consistent, but among never active smokers, passive smoking from a spouse was associated with increased risk in both sexes (n=95 cases, 202 controls OR 2.0, 95% CI: 1.1-3.5).83a

The Adelaide Adult Brain Cancer Study84a was one of ten case-control studies with a common protocol, coordinated through IARC and including data on passive exposure to parental, spousal and co-worker smoking. It found increased risk estimates associated with lifetime passive exposure for meningioma (OR 2.5, 95% CI: 1.0-6.0) and glioma (OR 1.35, 95% CI: 0.6-2.7). Unfortunately the study did not separate smokers from non-smokers, making it difficult to separate ETS effects in smokers from those in non-smokers.

ETS and childhood cancer

Over 30 published studies have examined maternal and/or paternal exposure to tobacco smoke and childhood cancer. For a review of these see Chapter 7 of the California EPA reports.3k For all cancers combined the evidence was considered inconclusive for an association with maternal smoking and suggestive for paternal smoking based on relatively small risks. Findings were considered inconclusive for childhood leukemia, and suggestive for childhood lymphomas and brain cancer, although the suggested association may be with pre-conceptual smoking rather than ETS. An earlier meta-analysis by Boffetta et al.85a found a small overall increased risk of childhood cancer in association with maternal smoking in a summary of 12 studies (RR 1.10, 95% CI: 1.03-1.19), but not for specific neoplasms. The summary RR for paternal smoking and childhood brain cancer from ten studies was 1.22 (95% CI: 1.02-1.40), and for lymphoma, the summary risk RR from four studies was 2.08 (95% CI: 1.08-3.98). However, there is no clear evidence of dose-response relationships. As well, childhood cancer studies are invariably case-control in design; such studies have the potential for recall bias.

Public health efforts to reduce ETS exposure

Much progress has been made in reducing exposure to ETS in Canada, the United States, Australia, and increasingly in Europe as well (in particular Ireland went smoke-free for virtually all public places in March, 2004). Many non-governmental organizations have lobbied for smoking restrictions, and various levels of government have instigated media campaigns to raise awareness of the dangers of ETS and have enacted legislation to restrict smoking in public places. Smoking restrictions in larger workplaces and federal buildings in Canada have existed since the late 1980s. By 2004, 91% of Canadians reported that they worked in an environment in which there were at least some restrictions on smoking.

In 2000, 27% of children under the age of 18 were regularly exposed to ETS. In 2003, only 16% were regularly exposed. (http://www.hc-sc.gc.ca/hl-vs/tobac-tabac/ research-recherche/stat/index-eng.php).

The State of California banned smoking in all restaurants in 1998 and in bars in 1999. Massachusetts has eliminated smoking from restaurants, as has New York City. (For a short overview of the laws and impact on restaurant revenues see http://www.repace.com/ fact_rest.html). There have been a wide variety of restrictions in local municipalities across Canada. The nation’s capital, Ottawa, brought in a total ban in indoor public places in the summer of 2002. (See Tobacco Control By-laws in Canada at http://www.hc-sc.gc.ca/hl-vs/pubs/ tobac-tabac/tcbc-rmtc/index_e.html and Canadian Law and Tobacco at http://www.cctc.ca/cctc/ EN/lawandtobacco).

Provincially, in British Columbia (BC), through an initiative of the BC Workers Compensation Board to protect workers, smoking in bars and restaurants was banned on January 1st, 2000. In 2001, these regulations were modified to allow the construction of smoking rooms which do not have to be enclosed, and into which staff may volunteer to serve. An important boost to smoke-free environments in Canada occurred on June 1st 2006, when both Ontario and Quebec brought in province-wide bans on smoking in all indoor public places including bars and restaurants. Thus, smoking is now banned in these two provinces in virtually all venues and there are no provisions for designated smoking rooms accessible to the public. Over 95% of Canadians now live in communities with 100% protection from ETS in public places.92a

Conclusions

Over the last 25 years, ETS has been implicated in delayed childhood development, childhood respiratory problems, adverse reproductive outcomes, cardiovascular disease and cancer. ETS has been established as a causal agent in lung cancer. The California EPA has recently become the first agency concerned with environmental health to come to the conclusion that there is a causal relationship between regular long-term ETS exposure and breast cancer in younger, primarily premenopausal women.3l Our understanding of individual susceptibility could be refined through further genetic epidemiological studies, and the dose-response paradigm of carcinogenicity for tobacco smoke in relation to breast cancer may need to be reconsidered to include thresholds and susceptible subgroups.

Exposure to tobacco smoke has been epidemic in many developed countries for at least the last half century. Fortunately the landscape is changing rapidly regarding smoking in public places in North America, in particular. There has been a major shift in public attitudes towards the social acceptability of cigarette smoking in public. Effective measures to control exposure have included legislated bans in workplaces and public places and no smoking policies where bans have not been implemented (homes and cars). However, there are still many children, spouses and workers being exposed to tobacco smoke daily. In light of the risks associated with lung cancer, breast cancer, and nasal cancer, as well as heart disease and asthma, it is clearly time to redouble efforts to reduce non-smokers exposure to ETS in all environments.

References

  1. a,b,c,d National Research Council. Environmental tobacco smoke: Measuring exposures and assessing health effects. 1986. Washington, D.C., National Academy Press.
  2. a,b National Institutes of Health. Health Effects of Exposure to Environmental Tobacco Smoke: Report of the California Environmental Protection Agency. 1999. Washington, U.S.A. NIH Pub No. 994645. Smoking and Tobacco Control Monograph Number 10.
  3. a,b,c,d,e,f,g,h,i,j,k,l California Environmental Protection Agency, Air Resources Board and Office of Environmental Health Hazard. Proposed Identification of Environmental Tobacco Smoke as a Toxic Air Contaminant. Available at: http://www.arb.ca.gov/regact/ ets2006/ets2006.htm (accessed 21-02-2006). 2005.
  4. ^ U.S. Department of Health and Human Services. Public Health Service. Office of the Assistant Secretary for Health. The health consequences of involuntary smoking: A report of the Surgeon General. 1986. Washington, D.C., DHHS (PHS).
  5. ^ International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 83, Tobacco Smoke and Involuntary Smoking. 2004. Lyon.
  6. ^ Yuan JM, Wang XL, Xiang YB, Gao YT, Ross RK, Yu MC. Non-dietary risk factors for nasopharyngeal carcinoma in Shanghai, China. Int J Cancer 2000;85:364-369.
  7. ^ Kasim K, Levallois P, Abdous B, Auger P, Johnson KC, Canadian Cancer Registries Epidemiology Research Group. Environmental tobacco smoke and risk of adult leukemia. Epidemiology 2005;16:672-680.
  8. ^ Zeegers MP, Goldbohm RA, van den Brandt PA. A prospective study on active and environmental tobacco smoking and bladder cancer risk (The Netherlands). Cancer Causes Control 2002;13:83-90.
  9. ^ Alberg AJ, Kouzis A, Genkinger JM et al. A prospective cohort study of bladder cancer risk in relation to active cigarette smoking and household exposure to secondhand cigarette smoke. Am J Epidemiol 2007;165:660-666.
  10. a,b,c National Institutes of Health, National Cancer Institute, and Environmental Protection Agency. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders: The Report of the U.S. Environmental Protection Agency. 933605. [4]. 1993. Washington, USA., NIH. Smoking and Tobacco Control Monograph.
  11. ^ National Health and Medical Research Council. The health effects of passive smoking. 1997. Australia.
  12. ^ Department of Health. Report of the Scientific Committee on Tobacco and Health. 1998. London, United Kingdom, The Stationery Office.
  13. ^ World Health Organization. International Consultation on Environmental Tobacco Smoke (ETS) and Child Health: Consultation Report. WHO Technical Document. 1-1-1999.
  14. ^ U.S. Department of Health and Human Services. Public Health Service. National Toxicology Program. 9th Report on Carcinogens. 2000. Washington, D.C., U.S. Department of Health and Human Services.
  15. ^ International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Volume 38, Tobacco Smoking. 1986. Lyon.
  16. ^ Shopland DR, Eyre HJ, Pechacek TF. Smoking-attributable cancer mortality in 1991: Is lung cancer now the leading cause of death among smokers in the United States? J Natl Cancer Inst 1991;83:1142-1148.
  17. a,b Baliunas D, Patra J, Rehm J, Popova S, Kaiserman M, Taylor B. Smoking-attributable mortality and expected years of life lost in Canada 2002: Conclusions for prevention and policy. Chronic Dis Can 2007;27:154-162.
  18. ^ Kaiserman MJ, Rickert WS. Carcinogens in Tobacco Smoke: Benzo[a]pyrene from Canadian Cigarettes and Cigarette Tobacco. Am J Public Health 1992;82:1023-1026.
  19. ^ Thompson SG, Stone R, Nanchahal K, Wald NJ. Relation of urinary cotinine concentrations to cigarette smoking and to exposure to other people’s smoke. Thorax 1990;45:356-361.
  20. ^ Anderson KE, Carmella SG, Ye M et al. Metabolites of a tobacco-specific lung carcinogen in nonsmoking women exposed to environmental tobacco smoke. J Natl Cancer Inst 2001;93:378-381.
  21. ^ Repace JL, Jinot J, Bayard S, Emmons K, Hammond SK. Air nicotine and saliva cotinine as indicators of workplace passive smoking exposure and risk. Risk Anal 1998;18:71-83.
  22. ^ Reardon JT, Shaw CY, Chown GA. Ventilation Strategies for Small Buildings. 1990. National Research Council of Canada.
  23. a,b,c,d Health Canada. Canadian Tobacco Use Monitoring Survey (CTUMS). Summary of Results for 2005 (February to December). Health Canada website. 2007.
  24. ^ Health Canada. Canadian Tobacco Use Monitoring Survey (CTUMS). Summary of Results for 2002 (February to December). Health Canada website. 2006.
  25. ^ Stephens, T. Smoking in Canadian homes. Special Report. 1999. Toronto, ON, Ontario Tobacco Research Unit.
  26. ^ Johnson KC, Mao Y, Argo J et al. The National Enhanced Cancer Surveillance System: A case-control approach to environment-related cancer surveillance in Canada. Environmetrics 1998;9:495-504.
  27. a,b,c,d Johnson KC, Hu J, Mao Y, The Canadian Cancer Registries Epidemiology Research Group. Passive and active smoking and breast cancer risk in Canada, 1994-97. Cancer Causes Control 2000;11:211-221.
  28. a,b Guerin MR, Jenkins RA, Tomkins BA. The chemistry of tobacco smoke: composition and measurement. Chelsea, MI: Lewis Publisher; 1992.
  29. a,b,c Hammond SK. Exposure of U.S. workers to environmental tobacco smoke. Environ Health Perspect 1999;107 Suppl 2:329-340.
  30. ^ Siegel M. Smoking and leukemia: evaluation of a causal hypothesis. Am J Epidemiol 1993;138:1-9.
  31. ^ U.S. Department of Health and Human Services. Public Health Service. Reducing the health consequences of smoking: 25 years of progress. A Report of the Surgeon General. 1989. Washington, D.C., DHHS (PHS).
  32. a,b,c,d Boffetta P, Agudo A, Ahrens W et al. Multicenter case-control study of exposure to environmental tobacco smoke and lung cancer in Europe. J Natl Cancer Inst 1998;90:1440-1450.
  33. a,b,c,d Hackshaw AK, Law MR, Wald NJ. The accumulated evidence on lung cancer and environmental tobacco smoke. BMJ 1997;315:980-988.
  34. ^ Epidemiology of lung cancer. (Lung Biology in Health and Disease; 74). Samet, J. M. 1994. New York, Marcel Dekker.
  35. ^ Curtin F, Morabia A, Bernstein MS. Relation of environmental tobacco smoke to diet and health habits: variations according to the site of exposure. J Clin Epidemiol 1999;52:1055-1062.
  36. a,b,c Hirayama T. Cancer mortality in nonsmoking women with smoking husbands based on a large scale cohort study in Japan. Prev Med 1984;13:680-690.
  37. ^ Trichopoulos D, Kalandidi A, Sparros L. Lung cancer and passive smoking: conclusion of Greek study. Lancet 1983;2(8351):677-678.
  38. ^ Wald NJ, Nanchahal K, Thompson SG, Cuckle HS. Does breathing other people’s tobacco smoke cause lung cancer? Br J Med 1986;293(6556):1217-1222.
  39. ^ Zhong L, Goldberg MS, Parent ME, Hanley JA. Exposure to environmental tobacco smoke and the risk of lung cancer: a meta-analysis. Lung Cancer 2000;27:3-18.
  40. ^ Vineis P, Hoek G, Krzyzanowski M et al. Lung cancers attributable to environmental tobacco smoke and air pollution in non-smokers in different European countries: a prospective study. Environ Health 2007;6:7.
  41. ^ Stayner L, Bena J, Sasco AJ et al. Lung cancer risk and workplace exposure to environmental tobacco smoke. Am J Public Health 2007;97:545-551.
  42. a,b,c Wells AJ. Lung cancer from passive smoking at work. Am J Public Health 1998;88:1025-1029.
  43. a,b,c Fontham ET, Correa P, Reynolds P et al. Environmental tobacco smoke and lung cancer in nonsmoking women. A multicenter study. JAMA 1994;271:1752-1759.
  44. a,b,c Johnson KC, Hu J, Mao Y, Canadian Cancer Registries Epidemiology Research Group. Lifetime residential and workplace exposure to environmental tobacco smoke and lung cancer in never-smoking women, Canada 1994-97. Int J Cancer 2001;93:902-906.
  45. a,b Jockel KH, Pohlabeln H, Ahrens W, Krauss M. Environmental tobacco smoke and lung cancer. Epidemiology 1998;9:672-675.
  46. ^ Canadian Cancer Society/National Cancer Institute of Canada. Canadian Cancer Statistics 2006. 2006. Toronto, Canada.
  47. ^ Madigan MP, Ziegler RG, Benichou J, Byrne C, Hoover RN. Proportion of breast cancer cases in the United States explained by well-established risk factors. J Natl Cancer Inst 1995;87:1681-1685.
  48. ^ Gammon MD, Schoenberg JB, Teitelbaum SL et al. Cigarette smoking and breast cancer risk among young women (United States). Cancer Causes Control 1998;9:583-590.
  49. a,b Palmer JR, Rosenberg L. Cigarette smoking and the risk of breast cancer. Epidemiol Rev 1993;15:145-156.
  50. ^ Wells AJ. Breast cancer, cigarette smoking, and passive smoking. Am J Epidemiol 1991;133:208-210.
  51. a,b Hirayama, T. Lung cancer and other diseases related to passive smoking: a large-scale cohort study. Gupta PC, Hamner III JE, and Murti PR, eds. Control of tobacco related cancers and other diseases, international symposium 1990. Bombay: Oxford University Press, 1992. 129–137. 1992.
  52. a,b,c,d,e,f, Wells AJ. Re: “Breast cancer, cigarette smoking, and passive smoking”. Am J Epidemiol 1998;147:991-992.
  53. a,b,c Sandler DP, Wilcox AJ, Everson RB. Cumulative effects of lifetime passive smoking on cancer risk. Lancet. 1985;1(8424):312-315.
  54. a,b Smith SJ, Deacon JM, Chilvers CE. Alcohol, smoking, passive smoking and caffeine in relation to breast cancer risk in young women. UK National Case-Control Study Group. Br J Cancer 1994;70:112-119.
  55. a,b Morabia A, Bernstein M, Heritier S, Khatchatrian N. Relation of breast cancer with passive and active exposure to tobacco smoke. Am J Epidemiol 1996;143:918-928.
  56. a,b,c Lash TL, Aschengrau A. Active and passive cigarette smoking and the occurrence of breast cancer. Am J Epidemiol 1999;149:5-12.
  57. a,b Zhao Y, Shi Z, Liu L. [Matched case-control study for detecting risk factors of breast cancer in women living in Chengdu]. Zhonghua Liu Xing Bing Xue Za Zhi 1999;20:91-94.
  58. a,b Jee SH, Ohrr H, Kim IS. Effects of husbands’smoking on the incidence of lung cancer in Korean women. Int J Epidemiol 1999;28:824-828.
  59. a,b,c Wartenberg D, Calle EE, Thun MJ, Heath CW, Jr., Lally C, Woodruff T. Passive smoking exposure and female breast cancer mortality. J Natl Cancer Inst 2000;92:1666-1673.
  60. a,b,c Egan KM, Stampfer MJ, Hunter D et al. Active and passive smoking in breast cancer: prospective results from the Nurses’ Health Study. Epidemiology 2002;13:138-145.
  61. a,b Nishino Y, Tsubono Y, Tsuji I et al. Passive smoking at home and cancer risk: a population-based prospective study in Japanese nonsmoking women. Cancer Causes Control 2001;12:797-802.
  62. a,b Marcus PM, Newman B, Millikan RC, Moorman PG, Baird DD, Qaqish B. The associations of adolescent cigarette smoking, alcoholic beverage consumption, environmental tobacco smoke, and ionizing radiation with subsequent breast cancer risk (United States). Cancer Causes Control 2000;11:271-278.
  63. a,b Millikan RC, Pittman GS, Newman B et al. Cigarette smoking, N-acetyltransferases 1 and 2, and breast cancer risk. Cancer Epidemiol Biomarkers Prev 1998;7:371-378.
  64. a,b Lash TL, Aschengrau A. A null association between active or passive cigarette smoking and breast cancer risk. Breast Cancer Res Treat 2002;75:181-184.
  65. a,b Kropp S, Chang-Claude J. Active and passive smoking and risk of breast cancer by age 50 years among German women. Am J Epidemiol 2002;156:616-626.
  66. a,b Delfino RJ, Smith C, West JG et al. Breast cancer, passive and active cigarette smoking and N-acetyltransferase 2 genotype. Pharmacogenetics 2000;10:461-469.
  67. a,b Gammon MD, Eng SM, Teitelbaum SL et al. Environmental tobacco smoke and breast cancer incidence. Environ Res 2004;96:176-185.
  68. a,b,c Reynolds P, Hurley S, Goldberg DE et al. Active smoking, household passive smoking, and breast cancer: evidence from the California Teachers Study. J Natl Cancer Inst 2004;96:29-37.
  69. a,b,c Reynolds P, Hurley S, Goldberg D. Accumulating evidence on passive and active smoking and breast cancer risk. Int J Cancer 2006;119:239.
  70. a,b,c Shrubsole MJ, Gao YT, Dai Q et al. Passive smoking and breast cancer risk among non-smoking Chinese women. Int J Cancer 2004;110:605-609.
  71. a,b,c Hanaoka T, Yamamoto S, Sobue T, Sasaki S, Tsugane S. Active and passive smoking and breast cancer risk in middle-aged Japanese women. Int J Cancer 2005;114:317-322.
  72. ^ U.S. Department of Health and Human Services. Public Health Service. Centres for Disease Control, Prevention and Health Promotion Office of Smoking and Health. The health consequences of smoking: cancer. A report of the Surgeon General. 1982. Washington, D.C., DHHS (PHS).
  73. a,b Petrakis NL, Maack CA, Lee RE, et al. Mutagenic activity in nipple aspirates of human breast fluid. (Letter). Cancer Res 1980;40:188-189.
  74. a,b Petrakis NL. Nipple aspirate fluid in epidemiologic studies of breast disease. Epidemiol Rev 1993;15:188-195.
  75. ^ Petrakis NL, Gruenke LD, Beelen TC, et.al. Nicotine in breast fluid of nonlactating women. Science 1978;199:303-304.
  76. ^ Hollingsworth AB, Lerner MR, Lightfoot SA et al. Prevention of DMBA-induced rat mammary carcinomas comparing leuprolide, oophorectomy, and tamoxifen. Breast Cancer Treat 1998;47:63-70.
  77. ^ Ambrosone CB, Freudenheim JL, Graham S et al. Cigarette smoking, N-acetyltransferase 2 genetic polymorphisms, and breast cancer risk. JAMA 1996;276:1494-1501.
  78. ^ Koo LC, Kabat GC, Rylander R, Tominaga S, Kato I, Ho JH. Dietary and lifestyle correlates of passive smoking in Hong Kong, Japan, Sweden, and the U.S.A. Soc Sci Med 1997;45:159-169.
  79. a,b,c Johnson KC. Accumulating evidence on passive and active smoking and breast cancer risk. Int J Cancer 2005;117:619-628.
  80. ^ Johnson KC. Re: Passive smoking exposure and female breast cancer mortality. J Natl Cancer Inst 2001;93:719-720.
  81. ^ Miller MD, Marty MA, Broadwin R et al. The association between exposure to environmental tobacco smoke and breast cancer: a review by the California Environmental Protection Agency. Prev Med 2007;44:93-106.
  82. ^ Elwood JM, Burton RC. Passive smoking and breast cancer: is the evidence for cause now convincing? Med J Aust 2004;181:236-237.
  83. ^ Phillips LE, Longstreth WJ, Koepsell T, Custer BS, Kukull WA, van Belle G. Active and passive cigarette smoking and risk of intracranial meningioma. Neuroepidemiology 2005;24:117-122.
  84. ^ Ryan P, Lee MW, North B, McMichael AJ. Risk factors for tumors of the brain and meninges: results from the Adelaide Adult Brain Tumor Study. Int J Cancer 1992;51:20-27.
  85. ^ Boffetta P, Tredaniel J, Greco A. Risk of childhood cancer and adult lung cancer after childhood exposure to passive smoke: A meta-analysis. Environ Health Perspect 2000;108:73-82.
  86. ^ Nyberg F, Agrenius V, Svartengren K, Svensson C, Pershagen G. Environmental tobacco smoke and lung cancer in nonsmokers: does time since exposure play a role? Epidemiology 1998;9:301-308.
  87. ^ Zhong L, Goldberg MS, Gao YT, Jin F. A case-control study of lung cancer and environmental tobacco smoke among nonsmoking women living in Shanghai, China. Cancer Causes Control 1999;10:607-616.
  88. ^ Kreuzer M, Krauss M, Kreienbrock L, Jockel KH, Wichmann HE. Environmental tobacco smoke and lung cancer: a case-control study in Germany. Am J Epidemiol 2000;151:241-250.
  89. ^ Lee CH, Ko YC, Goggins W et al. Lifetime environmental exposure to tobacco smoke and primary lung cancer of non-smoking Taiwanese women. Int J Epidemiol 2000;29:224-231.
  90. ^ Wang L, Lubin JH, Zhang SR et al. Lung cancer and environmental tobacco smoke in a non-industrial area of China. Int J Cancer 2000;88:139-145.
  91. ^ DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177-188.
  92. ^ Physicians for a Smoke-Free Canada. Newsletter. November 2007. Available at: http://www.smoke-free.ca/pdf_1/ fall2007-web.pdf, http://www.smoke-free.ca/ pdf_1/ Q&A-smokefreecommunities.pdf
Report a problem or mistake on this page
Please select all that apply:

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: