Distribution analysis of airborne nicotine concentrations in hospitality facilities

A number of publications report statistical summaries for environmental tobacco smoke (ETS) concentrations. Despite compelling evidence for the data not being normally distributed, these publications typically report the arithmetic mean and standard deviation of the data, thereby losing important information related to the distribution of values contained in the original data. We were interested in the frequency distributions of reported nicotine concentrations in hospitality environments and subjected available data to distribution analyses. The distribution of experimental indoor airborne nicotine concentration data taken from hospitality facilities worldwide was fit to lognormal, Weibull, exponential, Pearson (Type V), logistic, and loglogistic distribution models. Comparison of goodness of fit (GOF) parameters and indications from the literature verified the selection of a lognormal distribution as the overall best model. When individual data were not reported in the literature, statistical summaries of results were used to model sets of lognormally distributed data that are intended to mimic the original data distribution. Grouping the data into various categories led to 31 frequency distributions that were further interpreted. The median values in nonsmoking environments are about half of the median values in smoking sections. When different continents are compared, Asian, European, and North American median values in restaurants are about a factor of three below levels encountered in other hospitality facilities. On a comparison of nicotine concentrations in North American smoking sections and nonsmoking sections, median values are about one-third of the European levels. The results obtained may be used to address issues related to exposure to ETS in the hospitality sector.

Urinary cotinine in children and adults during and after semiexperimental exposure to environmental tobacco smoke

Urinary cotinine (U-cotinine) as a biomarker of environmental tobacco smoke exposure was evaluated in 14 children (age 4-11 y) and in 7 adults who were exposed to environmental tobacco smoke at an air nicotine level of 110 mg/m3 for 2 h in a bus. Nicotine in air and U-cotinine were measured by gas chromatography/mass spectrometry before, during, and after the experiment. U-cotinine rose rapidly to a maximum after a median of 6 h following the end of exposure; remained at an apparent plateau for half a day; and then decreased exponentially, with a mean half-time of 19 h (95% confidence interval 18-20 h; no significant difference between children and adults). The maximum U-cotinine was higher in the children (mean = 22 mg/l) than in the adults (13 mg/l; p = .005); decreased with age among the children (r = -.74; p = .002); and increased as the estimated inhaled nicotine dose increased. Therefore, the findings of the present study showed that young children had higher U-cotinine than adults at the same experimental environmental tobacco smoke exposure, probably because they had a higher relative nicotine dose because of a higher relative ventilation rate, and possibly also because of metabolic differences; the elimination rate did not differ. The long half-time makes U-cotinine a good biomarker of environmental tobacco smoke exposure; the time of sampling is not very critical. Dilution-adjusted concentrations should be employed, and in children, preferably by density correction. A certain urinary cotinine level indicates a lower environmental tobacco smoke exposure in a small child than in an adult.