Aspects of the design protocol and the statistical methods for analysis of tar, nicotine and carbon monoxide yields in cigarette smoke that can affect the measurement variability within collaborative studies.

Statistical principles described in ISO 5725-1 (1994) are a robust basis for evaluating cigarette smoke data from collaborative studies under the ISO 3308 machine smoking and for specifying the criteria for the removal of outlier data and determination of mean yields and their variability. However, the standard only provides recommendations on outlier removal that should be taken into account by experts who undertake data interpretation. The potential for over-interpretation of data from small numbers of laboratories is highlighted and recommendations made to deal with this possibility. Key variables to the statistical analysis, the number of cigarettes per replicate and replicates performed in each laboratory, the number of participating laboratories and the use of linear and rotary smoking machines in smoke collection, are identified and their relevance to obtaining robust data are considered. The statistical methods routinely used for data analysis from the ISO regime, are re-assessed for their suitability to analyse data obtained under the Canadian intense (CI) regime, where yield differences between linear and rotary smoking machines are found. This machine effect can lead to more outliers being detected and difficulties in outlier detection which may affect the provision of robust estimates of mean yields, repeatability and reproducibility.

Effect of puffing intensity on cigarette smoke yields.

Two US blend style cigarette products, one ventilated, were smoked under 16 smoking regimes. 'Tar', nicotine, carbon monoxide (TNCO) and water smoke yields determined with these regimes, are shown to form part of continuous functions linked with puffing intensity (the product of puff volume and puff frequency) and total puff volume (the product of puff volume and puff number). This allows the prediction of yields for any regime and leads to the conclusion that the characterisation of cigarette products with these analytes is achievable from using a single smoking regime. The rate of increase of TNCO yields decreases as the puffing intensity increases, due to the more rapid burning of the tobacco available for smoking, although (particulate phase) water yield, relative to TNCO, increases considerably with intensity. Total puff volume is linearly related to TNCO machine yields from a range of regimes, to duplicated human yields and to the nicotine and solanesol retained in spent filters. The concentration of these smoke components is essentially independent of the regime used to generate them. This is not the case with water for which the yield in smoke increases exponentially with the total puff volume and its concentration increases rapidly with intensity.

The fate of ingredients in and impact on cigarette smoke.

A series of experiments are described, undertaken on both volatile and non-volatile ingredients either during cigarette smoking or under pyrolysis conditions that try to simulate cigarette smoking. In particular, the fate of a series of deuterium and (13)C labelled volatiles was studied which demonstrated that, in a similar way to unlabelled volatiles, a large proportion of each was transferred intact into mainstream smoke. The unaccounted material, which was not transferred intact, in the studies of both volatile and non-volatile ingredients was primarily transformed into products of complete combustion such as carbon monoxide or carbon dioxide with only very minor amounts transformed into products of incomplete combustion. In addition, the studies on both unlabelled and deuterium labelled compounds demonstrated that the utility of pyrolysis studies lies mainly in distinguishing between those compounds that transfer intact into mainstream smoke from those that might be liable to degrade. Pyrolysis does not provide a robust prediction of the compounds that are formed from ingredients during cigarette smoking studies.