Tobacco Documents Online

IL ii , I I 17' F L F II Comments by Dr. Guy B. Oldaker III on CHAPTER 6 EXPOSURES TO AIR POLLUTANT> General Comments Chapter 6 is in rough draft (as distinguished from final draft) form. The title of the chapter is inconsistent with the subject of text and the Compendium. Several key references are unavailable. The chapter.relies too much on work done by the Harvard School of Public Health in the residential environmental category and neglects the literature dealing with assessments of exposure in public places, which are the clear interests of the stated, intended users of the Compendium: "decision-makers such as labor and management officials concerned with workplace exposures, public health officials and corporate medical directors who are concerned with making health policy recommendations, educators, industrial hygienists and safety officers, ETS researchers, indoor pollution investigators, and legislators who are considering legislation to restrict smoking in workplaces, restaurants, and puElic access buildings." If this chapter is to be useful, it must be revised and expanded substantially. The chapter needs an Introduction. Specific Comments Time Activity Patterns The time-activity data reported by Quakenboss et a1. and summarized in Tables I and II is, in its present form, of questionable representativeness because roughly half the subjects included are students. 'The authors should provide representati-ve time-activity data or address how their reported data relate to the population.

i_ The paragraph dealing with time-activity patterns of infants that appears on pages 65 and 66 shoul.d be omitted since it provides assentially no information j' useful for the intended users of the Compendium. Figure 1 also should be omitted for the same reason. Environmental Tobacco Smoke 'i co On page 67 the authors define ETS as a mixture of exhaled mainstream smoke and sidestream smoke. The following definitiop is more accurate and therefore is recommended: ETS is the diluted, aged mixture of exhalesi mainstream smoke and sidestream smoke that occurs in spaces occupied by smokers. Composition of ETS The use of sidestream data, and by extension sidestream to mainstream ratios, suffers from too many assumptions to justify inclusion within the chapter. Discussions based upon these concepts should be omitted. The relations between sidestream data and ETS levels are currently unknown. Scientific ignorance reflects two problem areas: (a) methods for sampling and analyzing sidestream smoke components have not been applied consistently, and (b) currently available data from,sidestream measurements are of either unkrown or inadequate quality. Research has shown that the method of sample collection can have a profound effect on results from sidestream analyses. As a final note, the table refers to data for the analysis of nonfilter cigarettes, clearly, a nonrepresentative portion of the U.S. market. Measurement of ETS L L L The authors state correctly that most researchers have measured one or more compounds to estimate total exposure to ETS. The text, however, is inconsistent with this statement inasmuch as it neglects most of the literature where more than one compound was measured. Similarly, the authors point to the need to estimate ETS exposure by measuring more than one indicator, yet they present results from but one investigation involving the measurement of more than one indicator. The authors address "possible" measures of ETS, namely, particles, nicotine, cadmium, and nitrosamine. They fail to identify. that cadmium and nitrosamine 2 R : ~

f are impractical and non-specific. Discussions dealing with these two substances should be omitted. The Compendium needs better coordination among chapters, as these two possible measures are inconsistent with t)ose identified in Chapter 5 by Leaderer. Exposures to Environmental Tobacco Smoke P Overall ETS exposure will depend on the proximity of an individual to the source of the smoke. This statement, which is obvious to the layperson, is seemingly overlooked by Repace in his presentation of (invalidated) equilibrium models. The authors of the two chapters should revise their work to~ address this inconsistency. S The second paragraph of- this section contains information important to the stated users of the Compendium, namely, that :.moking between different demographic groups can vary widely. The authors should elaborate on this issue, particularly with respect to how it might affect'expusures in public places and the workplace. This reviewer recommends that the authors of Chapter 6 work with the author of Chapter 7 in addressing how this issue affects one of the main assumptions of the equilibrium model. MICROENVIRONMENTAL MEASUREMENTS OF CONCENTRATIONS Concentrations of Particles and'ETS I L The authors state that numerous studies have been conducted using RSP as a marker of ETS arrd that the number of cigarettes smoked have shown to be correlated well with RSP. These studies need to be identified and the strengths of correlations and significance levels need to be provided. On page 69, first paragraph, the- last sentence would not be misleading if rewritten to read: "Each smoker in the home was associated with an increased RSP level of 20 ug/m3.° On page 69, second paragraph, the last sentence is speculation; should consider revising it to read: "The cause of this increase however, one hypothetical cause is recirculation of indoor air conditioning system that reduced dilution of cigarette smoke." 3 the authors is unknown; ab by the air CZ) .~ ~ US O a

The authors note that results of the [Six City Study] indicated that each smoker in the home raised the mean RSP level by 20 ug/m3. The significance of this increase in terms of expected health effects should be addressed. In addition, descriptive statistics quantifying the quality of these calculated results should be provided. The same follows for discussions deriving from work reported by Dockery and Spengler (1981), Spengler and colleagues (1986), and McCarthy et a1. (1987). C i. Particulate Concentration in Offices The selection of literature citations, which are relai:ively old, shows only that the authors relied on their own publications or ised previously published reviews from 1986. These reviews ignore the review prepared by Sterling and coworkers (1982) which is more complete. In citing work by Weber and Fischer (1980) the au-:hors fail to address the representativeness of the results. Weber and Fischer described an investigation done in Switzerland where heating ventilating and air conditioning. systems differ from those in the U.S. Nor do the authors recognize that the piezoelectric balance used by Weber and Fischer (as well as Quant, 1982) provided biased RSP results because the cut point was 3.5 gm, rather than the 2.5 um defined by the authors of the chapter. The authors summarize results reported by Quant (1982) and provide Figure 4 showing "Aerosol Mass Concentration in R&D Office." The paragraph contains insufficient information to allow interpretation o-F the results and figure. The authors should revise the paragraph to address the significance of the results. They also should consider adjusting results for background levels of RSP; otherwise, the reader is apt to be misled. The authors' presentation of results reported by Miesner (1988) is inconsistent with the treatment given for those reported by Weber and Fischer (1980) and Quant (1982). Thus, ranges are presented rather than means as was the case for the earlier paragraphs. In addition to being inconsistent, this manner of presentation conceals from the reader the general observation that exposure distributions are lognormally distributed and consequently that extreme values are generally rare'occurrences. 4

Particulate Concentration in Offices [sic] [page 70] r r i:< The reviewer assumes here that the title should have read: "Particulate Concentrations in Public Places other than Offices." The authors cite only work by Repace and Lowery (1980) and Miesner et al. (1988). The authors have overlooked most of the pu')lished work in this area. For reasons described above, the RSP results reported by Repace and Lowrey are biased high because they used a piezoelectric balam:e. -(Questions of quality assurance still remain.) In the first paragraph of this section, the last sentence needs to be corrected. , The authors state that particulate levels were low, usually less than 30 ug/m3. In the same paragraph they state that a concentration of 63 ,Lg/m3 is slightly higher. Later, they note that higher concentrations still were found in smoking areas such as bars, restaurants and a public smoking ^oom with a mean integrated measurement of 79 µg/m3 and a standard deviation of 44 µg/m3. These statements are critical to the entire Compendium because they provide quantitative results and statistics along with a discussion that provide readers with some perspective on the interpretation of results. Two issues are important here: (a) quantitatively, what constitutes "low," and (b) the variability shown by real-world measures of ETS constituents. The authors of Chapter 6 should address the results they review based upon levels less than or equal to 30 Ug/m3 being "low" and levels at 63 ,tg/3 being "slightly hiqher." Likewise authors of other chapters containing concentration data should do the same to ensure consistency. Additionally, the authors of Chapter 6 should discuss more fully in this section the-practical significance of the relative standard deviation they find: 56 %. Finally, the author of Chapter 7 should consider the implications of this relative standard deviation on attempts to construct equilibrium models of ETS exposure. Concentration of Other Components of ETS U The authors state that McCarthy et al. (1987) found average nicotine concentrations of 4.2 ug/m3 in smoking households. The authors should address the significance of this result. In its present form the paragraph is 5 X_

ambiguous; thus, if 0.1 Kg/m3 is "low," what does a concentration of 4.2 Ag/3 imply? The authors provide a "selection of ... studies [where integrated readings were used to determine nicotine levels in offices and puElic buildings] in Table 7. This "selection," because it contains relatively old citations and ignores most of the research done in this area, provides an inaccurate view of results. The authors should revise this section and the Table to provide the reader with useful information. In addition, Table 6 should Ee revised to exclude some results reported. The results reported by Hinds and First (1975) and Weber and Fischer (1980) rely on invalid methods (Badre et a1., 1978). The results reported by Badre et a1. (1978) for the "Room" and the "Car" were not obtained under realistic conditions. Following discussions dealing with determinations of' ETS nicotine, the authors provide three paragraphs touching upon measurements of cadmium, nitrosamines, and benzene. These paragraphs do not address the subjects adequately and speculate on the validity of the substances as indicators of ETS; because of these deficiencies, the paragraphs should be omittei. The work of Lebret and coworkers (1987) figures prominently in the section as well as the Chapter; however, the literature citations contain no mention of this work. The significance of the results from determinations of cadmium are never stated nor is any information given relative to the reliability of this marker for assessing exposure to ETS. A similar situation exists with respect to nitrosamines (Brunnemann et al., 1978), the methodDlogy for which has yet to be used to assess exposures in the context of surveys. The work of Wallace et al. (1987) is cited in connection with exposures to benzene. Although the text indicates that the exposures of nonsmokers during the fall and winter (but not during the spring and summer) were elevated relative to nonsmokers not reporting ETS exposure, the magnitude of this elevation and the biological significance of this.elevation are not addressed. Furthermore, ro information is presented regarding the practical utility of benzene as an indicator of ETS exposure. Personal Exposures L The authors note that results reported by McCarth;f et al. (1987) show that children from nonsmoking families show mean exposure to RSP 27 µg/m3 higher than that for those of Smoking households. The significance of this difference is 6

L:J I I L L not addressed;•-thus, is this mean exposure "low," "medium," or "high."? Based upon the authors' previous discussions, this 27 µg/m3 exposure might be presumed to be "low." A similar situation exists for the case of nicotine exposures. This reviewer notes that the RSP to nicotine rati3s for these two exposure categories are 97:1 and 22:1, respectively. These ratio values are inconsistent with speculations made in the Compendium to the effect that RSP might be predicted from nicotine and vice versa. The author cif Chapter 7 evaluates data such as these assuming a 13:1 ratio between RSP and nicotine (Repace, Environ. Sci. Technol. 1988). Applying this evaluation approiLch to the results reported by McCarthy et al. would lead to the following conclusions: (a) the RSP method is biased high, causing RSP results to overestimate ETS exposure substant•ially; (b) the nicotine method is biased low; or (c) the 13:1 ratio, which is derived from measurements of unrealistically high levels of ETS in an environmental chamber, does not generally apply to real-world settings. This reviewer recommends strongly that the authors of chapters a.ddressing RSP to nicotine ratios revise their work to produce discussions consistent with experimental results. The authors cite work by Spengler et al. (1985;1 in connection with 101 nonsmoking volunteers and state that results showed that personal exposure to RSP was not correlated with outdoor concentrations but that ETS significantly increased an individuals [sic] personal concentration profile. The authors should provide quantitative results for the measurements and the statistics. The concluding sentence of this paragraph is probably more accurate if phrased: "... reported exposure to ETS was associated with a statistically significant increase in personal concentration profile." The authors also should define "personal, concentration profile." The authors present results in connection with deti!rminations of cotinine in body fluids. Because the subject is treated superficially, users of the Compendium can be expected to derive little benefit from its current inclusion in the chapter. The authors should either expand on this subject or omit it from the text. The authors note that they measured cotinine in urine and saliva of ch.ildren and found a high correlation with reported exposure. The magnitude and strength of this association need to be provided. They follow this presentation with results which cause ambiguity: Coultas et al. (1987) measured 7 d

cotinine in a significant number of subjects claiming no ETS exposure. What does this mean? Summary In summarizing the chapter, the authors conclude that ETS can be a substantial contributor to the indoor air pollution concentratioi of benzene, acrolein, N- nitrosamine, pyrene, and carbon monoxide; however, their text supports the "substantial contribution" of none of these. They also conclude that measured exposures to RSP are higher for nonsmokers who report exposure to ETS; however, they address neither the magnitude nor the significance of this higher exposure. REFERENCES The text does not cite all the references in this chapter. L