Modeling residential exposure to secondhand tobacco smoke

Abstract

We apply a simulation model to explore the effect of a house's multicompartment character on a nonsmoker's inhalation exposure to secondhand tobacco smoke (SHS). The model tracks the minute-by-minute movement of people and pollutants among multiple zones of a residence and generates SHS pollutant profiles for each room in response to room-specific smoking patterns. In applying the model, we consider SHS emissions of airborne particles, nicotine, and carbon monoxide in two hypothetical houses, one with a typical 4-room layout and one dominated by a single large space. We use scripted patterns of room-to-room occupant movement and a cohort of 5,000 activity patterns sampled from a US nationwide survey. The results for scripted and cohort simulation trials indicate that the multicompartment nature of homes, manifested as inter-room differences in pollutant levels and the movement of people among zones, can cause substantial variation in nonsmoker SHS exposure.

Modeling geographic and demographic variability in residential concentrations of environmental tobacco smoke using national data sets

“…The mean estimated ETS-PM concentration of 16 mg/m 3 for exposed households in our study falls on the lower side of the range of concentrations measured in previous large-scale residential ETS studies (Dockery and Spengler, 1981;Leaderer, 1990;Ozkaynak et al, 1996). Our results are also in agreement with recent modeling studies: Klepeis et al reported means ranging from 6.6-49 mg/m 3 from multiple simulations using a simple box model approach; and Myatt et al reported a mean of 15 mg/m 3 and a median of 17.8 mg/m 3 using the CONTAM model (Klepeis and Nazaroff, 2006;Myatt et al, 2008). These studies incorporated time-activity patterns and ventilation patterns but relied on a hypothetical smoking population, whereas our study is a snapshot characterization of equilibrium concentrations based on parameters estimated from a national survey.…”

“…Therefore, we assumed that our study population spends one half of its waking time inside home, and multiplied the predicted daily number of cigarettes by a factor of 0.5 to obtain the predicted daily number of cigarettes smoked at home (Table 1b). We assumed a constant rate of smoking; although it is possible that more cigarettes are smoked outside of working hours given the increasing smoking restrictions in public places, no information was available on this and we chose to use a linear rate based on previous literature (Klepeis and Nazaroff, 2006).…”

“…Emission and deposition parameters were selected based on previously reported central estimates: we used 10 mg per cigarette for PM emissions and 0.1/h for particle deposition loss-rate coefficient (Klepeis and Nazaroff, 2006). These previously reported central estimates were based on an assimilation of multiple previous studies (Klepeis et al, 2003;Martin et al, 1997;Xu et al, 1994).…”

“…Such an approach does provide reasonable estimates of national-scale health risks, albeit with some uncertainties, but the reliability of this surrogate measure may vary across demographic subpopulations, and the approach lacks sufficient resolution to identify potentially meaningful community differences in exposure. The smoking literature shows significant sociodemographic and geographic variation of smoking prevalence and intensity (Shavers et al, 2005;Shavers et al, 2006;Datta et al, 2006;Osypuk et al, 2006); furthermore, housing characteristics such as home volume and air exchange rate significantly influence exposure implications of smoking (Klepeis and Nazaroff, 2006). Demographic factors correlated with smoking may also be correlated with housing characteristics, influencing ETS concentration patterns and potentially increasing variability within and between communities.…”

“…Given increasing smoking restrictions in public places, the majority of the exposure and risk burden may shift to the residential environment, but no study to date has developed broad-based and generalizable models of ETS exposure inside homes. Previous residential studies have either characterized ETS contributions to residential exposures within hypothetical simulation models or using measurements but without a nationally generalizable model framework (Dockery and Spengler, 1981;Leaderer, 1990;Ozkaynak et al, 1996;Klepeis and Nazaroff, 2006;Myatt et al, 2008). It is therefore unknown how variability in smoking patterns correlates with housing characteristics that influence indoor concentrations of ETS, necessary in determining variability in residential concentrations and risk.…”