In 1985 the Surgeon General of the US Public Health Service reviewed the health consequences of smoking with regard to cancer and chronic lung disease in the workplace. It was concluded that for most US workers, cigarette smoking represents a greater cause of death and disability than their workplace environment. However, the control of smoking and a reduction of the exposure to hazardous agents at the workplace are essential, since these factors often act synergistically with smoking in the induction and development of respiratory diseases. Several occupational exposures are known to induce chronic bronchitis in workers. These include exposures to dust from coal, cement and grain, to silica aerosols, to vapors generated during welding, and to sulphur dioxide. Chronic bronchitis among workers in these occupations is often aggravated by cigarette smoking (US Surgeon General 1985).

Epidemiological data have clearly documented that uranium miners and asbestos workers who smoke cigarettes carry significantly higher risks of cancer of the respiratory tract than non-smokers in these occupations. The carcinogenic effect of uranium and asbestos and cigarette smoking is not merely additive, but synergistic in inducing squamous cell carcinoma of the lung (US Surgeon General 1985; Hoffmann and Wynder 1976; Saccomanno, Huth and Auerbach 1988; Hilt et al. 1985). The carcinogenic effects of exposure to nickel, arsenicals, chromate, chloromethyl ethers, and those of cigarette smoking are at least additive (US Surgeon General 1985; Hoffmann and Wynder 1976; IARC 1987a, Pershagen et al. 1981). One would assume that coke-oven workers who smoke have a higher risk of lung and kidney cancer than non-smoking coke-oven workers; however, we lack epidemiological data that substantiate this concept (IARC 1987c).

It is the aim of this overview to evaluate the toxic effects of the exposure of men and women to environmental tobacco smoke (ETS) at the workplace. Certainly, curtailing smoking at the workplace will benefit active smokers by reducing their consumption of cigarettes during the workday, thereby increasing the possibility that they become ex-smokers; but smoking cessation will also be of benefit to those non-smokers who are allergic to tobacco smoke or who have pre-existing lung or heart ailments.

Physico-Chemical Nature of EnvironmentalTobacco Smoke

Mainstream and sidestream smoke

ETS is defined as the material in indoor air that originates from tobacco smoke. Although pipe and cigar smoking contribute to ETS, cigarette smoke is generally the major source. ETS is a composite aerosol that is emitted primarily from the burning cone of a tobacco product between puffs. This emission is called sidestream smoke (SS). To a minor extent, ETS consists also of mainstream smoke (MS) constituents, that is, those that are exhaled by the smoker. Table 7 lists the ratios of major toxic and carcinogenic agents in the smoke that is inhaled, the mainstream smoke, and in the sidestream smoke (Hoffmann and Hecht 1990; Brunnemann and Hoffmann 1991; Guerin et al. 1992; Luceri et al. 1993). Under “Type of toxicity”, smoke components marked “C” represent animal carcinogens that are recognized by the International Agency for Research on Cancer (IARC). Among these are benzene,β-naphthylamine, 4-aminobiphenyl and polonium-210, which are also established human carcinogens (IARC 1987a; IARC 1988). When filter cigarettes are being smoked, certain volatile and semi-volatile components are selectively removed from the MS by filter tips (Hoffmann and Hecht 1990). However, these compounds occur in far higher amounts in undiluted SS than in MS. Furthermore, those smoke components that are favoured to be formed during smouldering in the reducing atmosphere of the burning cone, are released into SS to a far greater extent than into MS. This includes groups of carcinogens like the volatile nitrosamines, tobacco-specific nitrosamines (TSNA) and aromatic amines.

Table 1. Some toxic and tumorigenic agents in undiluted cigarette sidestream smoke

^a C=Carcinogenic; CoC=co-carcinogenic; T=toxic; TP=tumor promoter.
^b ND=not detected.
^c NNN=N‘-nitrosonornicotine.
^d NNK=4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone.

ETS in indoor air

Although undiluted SS contains higher amounts of toxic and carcinogenic components than MS, the SS inhaled by non-smokers is highly diluted by air and its properties are altered because of the decay of certain reactive species. Table 8 lists reported data for toxic and carcinogenic agents in indoor air samples of various degrees of tobacco smoke pollution (Hoffmann and Hecht 1990; Brunnemann and Hoffmann 1991; Luceri et al. 1993). The air dilution of SS has a significant impact on the physical characteristics of this aerosol. In general, the distribution of various agents between vapor phase and particulate phase is changed in favour of the former. The particles in ETS are smaller (<0.2 μ) than those in MS (~0.3 μ) and the pH levels of SS (pH 6.8 - 8.0) and of ETS are higher than the pH of MS (5.8 - 6.2; Brunnemann and Hoffmann 1974). Consequently, 90 to 95% of nicotine is present in the vapor phase of ETS (Eudy et al. 1986). Similarly, other basic components such as the minor Nicotiana alkaloids, as well as amines and ammonia, are present mostly in the vapor phase of ETS (Hoffmann and Hecht 1990; Guerin et al. 1992).

Table 2. Some toxic and tumorigenic agents in indoor environments polluted by tobacco smoke

^a NNN=N‘-nitrosonornicotine.
^b ND=not detected.
^c NNK=4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone.

Biomarkers of the Uptake of ETS by Non-Smokers

Although a significant number of non-smoking workers are exposed to ETS at the workplace, in restaurants, in their own homes or in other indoor places, it is hardly possible to estimate the actual uptake of ETS by an individual. ETS exposure can be more precisely determined by measuring specific smoke constituents or their metabolites in physiological fluids or in exhaled air. Although several parameters have been explored, such as CO in exhaled air, carboxyhaemoglobin in blood, thiocyanate (a metabolite of hydrogen cyanide) in saliva or urine, or hydroxyproline and N-nitrosoproline in urine, only three measures are actually helpful for estimating the uptake of ETS by non-smokers. They allow us to distinguish passive smoke exposure from that of active smokers and from non-smokers who have absolutely no exposure to tobacco smoke.

The most widely used biomarker for ETS exposure of non-smokers is cotinine, a major nicotine metabolite. It is determined by gas chromatography, or by radioimmunoassay in blood or preferably urine, and reflects the absorption of nicotine through the lung and oral cavity. A few millilitres of urine from passive smokers is sufficient to determine cotinine by either of the two methods. In general, a passive smoker has cotinine levels of 5 to 10 ng/ml of urine; however, higher values have occasionally been measured for non-smokers who were exposed to heavy ETS over a longer period. A dose response has been established between duration of ETS exposure and urinary cotinine excretion (table 3, Wald et al. 1984). In most field studies, cotinine in the urine of passive smokers amounted to between 0.1 and 0.3% of the mean concentrations found in the urine of smokers; however, upon prolonged exposure to high concentrations of ETS, cotinine levels have corresponded to as much as 1% of the levels measured in the urine of active smokers (US National Research Council 1986; IARC 1987b; US Environmental Protection Agency 1992).

Table 3. Urinary cotinine in non-smokers according to the number of reported hours of exposure to other people’s tobacco smoke within the previous seven days

^a Trend with increasing exposure was significant (p<0.001).

Source: Based on Wald et al. 1984.

The human bladder carcinogen 4-aminobiphenyl, which transfers from tobacco smoke into ETS, has been detected as a haemoglobin adduct in passive smokers in concentrations up to 10% of the mean adduct level found in smokers (Hammond et al. 1993). Up to 1% of the mean levels of a metabolite of the nicotine-derived carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), which occurs in the urine of cigarette smokers, has been measured in the urine of non-smokers who had been exposed to high concentrations of SS in a test laboratory (Hecht et al. 1993). Although the latter biomarker method has not as yet been applied in field studies, it holds promise as a suitable indicator of the exposure of non-smokers to a tobacco-specific lung carcinogen.

Environmental Tobacco Smoke and Human Health

Disorders other than cancer

Prenatal exposure to MS and/or ETS and early postnatal exposure to ETS increase the probability of complications during viral respiratory infections in children during the first year of life.

The scientific literature contains several dozens of clinical reports from various countries, reporting that children of parents who smoke, especially children under the age of two years, show an excess of acute respiratory illness (US Environmental Protection Agency 1992; US Surgeon General 1986; Medina et al. 1988; Riedel et al. 1989). Several studies also described an increase of middle ear infections in children who had exposure to parental cigarette smoke. The increased prevalence of middle ear effusion attributable to ETS led to increased hospitalization of young children for surgical intervention (US Environmental Protection Agency 1992; US Surgeon General 1986).

In recent years, sufficient clinical evidence has led to the conclusion that passive smoking is associated with increased severity of asthma in those children who already have the disease, and that it most likely leads to new cases of asthma in children (US Environmental Protection Agency 1992).

In 1992, the US Environmental Protection Agency (1992) critically reviewed the studies on respiratory symptoms and lung functions in adult non-smokers exposed to ETS, concluding that passive smoking has subtle but statistically significant effects on the respiratory health of non-smoking adults.

A search of the literature on the effect of passive smoking on respiratory or coronary diseases in workers revealed only a few studies. Men and women who were exposed to ETS at the workplace (offices, banks, academic institutions, etc.) for ten or more years had impaired pulmonary function (White and Froeb 1980; Masi et al. 1988).

Lung cancer

In 1985, the International Agency for Research on Cancer (IARC) reviewed the association of passive tobacco smoke exposure with lung cancer in non-smokers. Although in some studies, each non-smoker with lung cancer who had reported ETS exposure was personally interviewed and had supplied detailed information on exposure (US National Research Council 1986; US EPA 1992; US Surgeon General 1986; Kabat and Wynder 1984), the IARC concluded:

The observations on non-smokers that have been made so far, are compatible with either an increased risk from ‘passive’ smoking, or an absence of risk. Knowledge of the nature of sidestream and mainstream smoke, of the materials absorbed during ‘passive’ smoking and of the quantitative relationship between dose and effect that are commonly observed from exposure to carcinogens, however, leads to the conclusion that passive smoking gives rise to some risk of cancer (IARC 1986).

Thus, there is an apparent dichotomy between experimental data which support the concept that ETS gives rise to some risk of cancer, and epidemiological data, which are not conclusive with respect to ETS exposure and cancer. Experimental data, including biomarker studies, have further strengthened the concept that ETS is carcinogenic, as was discussed earlier. We will now discuss how far the epidemiological studies that have been completed since the cited IARC report have contributed to a clarification of the ETS lung cancer issue.

According to the earlier epidemiological studies, and in about 30 studies reported after 1985, ETS exposure of non-smokers constituted a risk factor for lung cancer of less than 2.0, relative to the risk of a non-smoker without significant ETS exposure (US Environmental Protection Agency 1992; Kabat and Wynder 1984; IARC 1986; Brownson et al. 1992; Brownson et al. 1993). Few, if any, of these epidemiological studies meet the criteria of causality in the association between an environmental or occupational factor and lung cancer. Criteria that fulfil these requirements are:

1. a well-established degree of association (risk factor≥3)
2. reproducibility of the observation by a number of studies
3. agreement between duration of exposure and effect
4. biological plausibility.

One of the major uncertainties about the epidemiological data lies in the limited reliability of the answers obtained by questioning cases and/or their next-of-kin with regard to the smoking habits of the cases. It appears that there is generally an accord between parental and spousal smoking histories provided by cases and controls; however, there are low agreement rates for duration and intensity of smoking (Brownson et al. 1993; McLaughlin et al. 1987; McLaughlin et al. 1990). Some investigators have challenged the reliability of the information derived from individuals about their smoking status. This is exemplified by a large-scale investigation carried out in southern Germany. A randomly selected study population consisted of more than 3,000 men and women, ranging in age from 25 to 64 years. These same people were questioned three times in 1984-1985, in 1987-1988 and again in 1989-1990 as to their smoking habits, while each time urine was collected from each proband and was analysed for cotinine. Those volunteers who were found to have more than 20 ng of cotinine per ml of urine were considered to be smokers. Among 800 ex-smokers who claimed to be non-smokers, 6.3%, 6.5% and 5.2% had cotinine levels above 20 ng/ml during the three time periods tested. The self-proclaimed never-smokers, who were identified as actual smokers according to cotinine analyses, constituted 0.5%, 1.0% and 0.9%, respectively (Heller et al. 1993).

The limited reliability of the data obtained by questionnaire, and the relatively limited number of non-smokers with lung cancer who were not exposed to carcinogens at their workplaces, point to the need for a prospective epidemiological study with assessment of biomarkers (e.g., cotinine, metabolites of polynuclear aromatic hydrocarbons, and/or metabolites of NNK in urine) to bring about a conclusive evaluation of the question on causality between involuntary smoking and lung cancer. While such prospective studies with biomarkers represent a major task, they are essential in order to answer the questions on exposure which have major public health implications.

Environmental Tobacco Smoke and the Occupational Environment

Although epidemiological studies have thus far not demonstrated a causal association between ETS exposure and lung cancer, it is nevertheless highly desirable to protect workers at the site of employment from exposure to environmental tobacco smoke. This concept is supported by the observation that long-term exposure of non-smokers to ETS at the workplace can lead to reduced pulmonary function. Furthermore, in occupational environments with exposure to carcinogens, involuntary smoking may increase the risk of cancer. In the United States, the Environmental Protection Agency has classified ETS as a Group A (known human) carcinogen; therefore, the law in the United States requires that employees be protected against exposure to ETS.

Several measures can be taken to protect the non-smoker from exposure to ETS: prohibiting smoking at the worksite, or at least separating smokers from non-smokers where possible, and assuring that the smokers’ rooms have a separate exhaust system. The most rewarding and by far the most promising approach is to assist employees who are cigarette smokers in cessation efforts.

The worksite can offer excellent opportunities for implementing smoking cessation programmes; in fact, numerous studies have shown that worksite programmes are more successful than clinic-based programmes, because employer-sponsored programmes are more intense in nature and they offer economic and/or other incentives (US Surgeon General 1985). It is also indicated that the elimination of occupationally related chronic lung diseases and cancer frequently cannot proceed without efforts to convert the workers into ex-smokers. Furthermore, worksite interventions, including smoking cessation programmes, can produce lasting changes in reducing some cardiovascular risk factors for the employees (Gomel et al. 1993).

We greatly appreciate the editorial assistance of Ilse Hoffmann and the preparation of this manuscript by Jennifer Johnting. These studies are supported by USPHS Grants CA-29580 and CA-32617 from the National Cancer Institute.

Back