Tobacco Documents Online

Comments on Chapter 7: Exposure Assessment in Passive Smoking By: Dr. S. Keith Cole The statement is made that the RSP concentration is directly proportional to the product of the number of smokers, - smoking rate, and emmissions per tobacco product and inversely proportional to the product of the space volume and the removal rate. This is correct only in terms of the single compartment equilibrium model which assumes the gases are thoroughly and uniformly mixed. Th:Ls assumption has not been proven or even tested. Flow-dynam:Lc models might be more appropriate (see Horstman or Siurna and Bragg). It is stated that RSP and nicotine are i.he two most promising markers for ETS. Unless all sources of RSP can be accounted for RSP is not a good or even legitimate marker of ETS. Other sources may be present and iaay vary in time, creating unrealistic measures of RSP. Recent experiments have shown nicotine to decay rapidly in smoking environments due to adsorption to surfaces then to desorb over long periods of time. This makes measuremeni: of nicotine possible when there may not be any other ETS_components in the air, thus overestimating ETS exposure. The statement that RSP is the single largest component of ETS by weight is incorrect. Carbon dioxide and carbon monoxide are generated in larger quantities. The model proposed for predicting ETS exposures is only valid in the limit of a uniformly mixed gas. This limits application of the model to conditions where the residence time of the particular component (which depends upon details of the ventilation system, space geomet:_y, and physical characteristics of the component such as diffusion coefficient, etc.) is long compared to the decay time, whether this be due to chemical reaction, exhaust through the ventilation system, or a physical p:rocess such as adsorption or evaporation. The basic model is: Ceq = G / [m(Nv+Ns)V] where Ceq is the equilibrium concentration, G is the rate of generation of RSP in micrograms per hour, Nv is the ventilation or infiltration rate in air changes per hour (ACH), Ns is removal rate of RSP due to adsorption in ACH, and m is a mixing factor. A condensed Eorm of this.equation is presented as: Ceq = 650 Ds/Nv

where Ds is the density of active smokers per 100 m3 volume. The constant term is derived from average values assumed for RSP emission rates, building ventilatior. rates, mixing factors, and sink rates. Other derivative equations are provided which are supposed to adjust for intermittant smoking rates. Several questions arise from this presentation. What are these average values? Do any of the ave:rage values represent a realistic environment? For instance, the smoking rate used is presumably 2 cigare:ttes per hour per smoker as the author has used this value: in most previous publications. This is derived from the total number of cigarettes sold and the total number of smokers in the US. The author implies that since 90% of the: average persons time is spent indoors that 90% of all'ci.garettes are consumed indoors. Is this verifiable? If eight hours of sleep are allocated then the ammount of time indoors awake is reduced to 58% of a 24 hour day. Also, the•impression is given that all blue collar and white co].lar employment is indoors. Is it? How many cigarettes are smoked outdoors or in automobiles? Since the model is linear in all the variables it is extremely sensitive to such factors as smoking rate. In field studies conducted in aircraft, smoking rates of at most 0.5 to 0.6 cigZirettes per smoker per hour were observed (Drake and Johnson, submitted for publication to Aviation, Space, and Environmental Medicine, 1989). This lower smoking rate would reduce the estimated RSP concentration by 75%.' Perhaps more care should be taken in estimating or measuring smoke generation rates. In the field studies referenced such parameters as mixing factor and ventilation rate were not measured but were assumed. Since these factors are critical to the "predictive" power of the model they should be measured. The only validation of this model has been in controlled settings, i.e. chamber studies, where the air is mixed by fans. To my knowledge the model has noi: predicted RSP concentrations using accurately measured input parameters in the field. Regarding pathways for distribution of ]:TS through building ventilation systems the author is refer::ed to a publication by James Axley from the National Bureau of Standards that was prepared for the EPA and DOE (Indoo:: Air Quality Modeling - Phase II Report, NBSIR 87-3661). The treatment of recirculated air is presented for homes and could be extended to other mechanically ventilated structures. The following references are reported i;z the text but are missing from the reference list at the and of the chapter: Persily and Grot, 1986; Williams et al., 1985, IARC (1987).

In section C the author reports measured concentrations of RSP from ETS and states that the size r<inge of particles is less than.2.5 um. In the publications referenced (Repace and Lowrey, Science, 1980) the piezoelec:tric balance used was stated to have 100% efficiency for particles from 0.1 to 3.0 um diameter, 50% efficiency at 3.5 um, and 10% efficiency at 4.0 um. Was a device usec[ to restrict - measurements to particles less than 2.5 um diameter? If so, state what device was used and how it w<<s implemented. The author states that the weighted average RSP levels (taken from various publications) in buildings where smoking is permitted is 262 ug/m while in buildings with no smoking allowed it is 36 ug/m . The suggestion is then made that 85% of RSP in the buildings where smoking is allowed is due to smoking. First, what is the weighting procedure used to calculate the average, i.e. by occupancy or volume? Secondly, comparison of RSP levels between various buildings does not allow for such correlations as to the,origin of RSP. The author states that "under extreme conditions of indoor pollution, it has been calculated that a nonsmoker would inhale volatile nitrosamines equivalent to 10 nonfilter cigarettes or 35 filter cigarettes". The reference, which is missing from the reference list, is F[offmann and Hoffmann, Significance of Exposure to Sidestream Tobacco Smoke, in IARC, vol. 9 (1987), p.6. The more appropriate reference is to the original work of Brunnemann and Hoffmann, IARC Scientific Publications, No. 19, pp. 343-356. Review of this paper brings the statement referred to above into doubt. The only volatile nitrosamine detected in ambient air in the study by Brunnemann <<nd Hoffmann was NDMA. This compound is not specific to tobacco and is found in various substances in the environment. No other "ETS markers" were measured in the study. Also, the exposure to NDMA was given as a range of cigarette equivalents (9-10 nonfilter cigarettes, 17-35 filter ciga3-ettes) which depended on respiratory rate. Presenting only the upper bound of this range is misleading.