Tobacco Documents Online

A CRITIQUE OF DR AUBORT'S IN-MOUTH STUDY D G FELTON 1. IN RODUCTION 1.11 Dr J D Aubort, Cantonal Chemist of Vaud, alleges that BARCLAY cigarettes deceive the consumer in that they deliver tolthe smoker amounts of TAR and Nicotine iin excess of the declared figures of "TAR 1 mg - NIC 0.2 mg` as det'ermimed by internationally agreed'standard methods. 1.2 Md based his allegation on the results of a novel and unconventional experiment, in which human smokers were required to draw smoke from alcigarette at fiixedi iintervals of time, hold the smoke in the mouth for an unnaturallly liong period and submit to a process by which a fixed volume (35 ml) was withdrawn from the mouth cavity. The combined sample from a series of sequential puffs was then analysed for Nicotine by an approved procedure and for TAR by a colorimetrilc method'suitable only for comparative purposes. 1-3 Dr Aubort also modified BARCllAY cigarettes by prolonging the tipping paper I mm beyond the end of the filter and claimed that the modified product then behaved in this huimanismokiing test in a manner comparable to a cigarette ventil'ated by conventional methods. The reader was left to presume that cigarettes ventilated by conventional methods do not "deceive the consumer". 2. GENERAL CRITICISMS 2.1 There are several fallacies implicit in Dr Aubort's argument and thinking, which are outlined here and, then, subsequently examined in detail.

/2 2.11.1 He assumes tacitly that the figures for TAR and nicotine, determined by internationally agreed standardlmethods, are supposed toirepresent the amounts of these substances absorbed by human smokers. 2.11.2 He assumes tacitly that a1Thumain smokers react to smoke in the same way and are consistent, and!constaint, in their smoking behaviour. 2',1.3 He assumes tacitly that a single sample taken from an ill-defined region of'the mouth of'a human smoker puffing, a cigarette will be representative, in every case, of thee smoke taken by that smoker. He further confuses t'he concepts of "quantity of smoke" and "concentration of'smoke". 2.2' Each of these assumptions willl be examined i6turrr and will be shown to be incorrectly founded; therefore, any conclusions, drawn from arguments based on them; are worthless and unjiustifiable. 3. STANDARD METHODOLOGY FOR SMOKE ANALY!SIS 3.1 1 It'has long been recognised that there is a need for precise and reproducible dataifor any laboratory work involving cigarette smoke, such as ci'garette design, quality control and research purposes. To meet this need, standard equipment aindlmethods for smoking cigarettes mechanicallly and collecting and measuring the smoke quantitatively have been developediand agreed internationallly through bodies such as the International Standardisation Organisation (ISO) and CORESTA. -~•.

/3' 3'.2 Latterly, these test procedures have been adopted by official bodies such as the Federal Trades.Conmission in the USA, the Laboratory of the Government Chemist in.the UK and European Tobacco Authorities who, whilst realiising! that the test does not necessarily indicate the amount of TAR the smoker receives, appreciate that it does sensibly and reproducibly rank cigarettes in order of TAR delivery. ('J.A!.O:A:.C. 69, 598-600, 1986). 3.3' The standard methodology for smoke analyses is, in many ways, analogous to the methodology used to measure fuel consumption data for motor vehicles, which may be found im small print at the bottom of many motor car advertisements. These fueil consumptionifigures are determined under three standard conditions, viz; - urban cycle, a steady 90 kph and a steady 12D14ph. No:driver expects to match these on every occasion when he drives a particular model; nor would two drivers expect to return: ident'icali figures when driving the same car. What is achieved by fuel consumption figuires is aimeaningful: comparison of different models of car under standard conditions. 3.4 In the case of cigarettes, it has long been known that no two smokers smoke in identical ways; i~ndeed, no single smoker smokes in the same way on alll occasions. Just as ai car driiver inia hurry may accelerate toolfast for economical driving, a.smoker who is algitated or worried takes larger puffs or smokes more rapidly than maybe hiss wont when sitting relaxed in an armchair reading!a book. The amount of smoke obtained from a cigarette depends upon the volume of puff taken, the shape of the puff "profille" (the pressure-flow rellationshrip whereby the smoker may invoiuintari]y control the volume he puffs), the number and frequency of puffs and the way these are spaced throughout ./..

/4 the cigarette. (Creighton and Lewis, p. 301 in 'Smokiing Behaviour', edited by R'E Thornton, Published by Churchill-Livingstone, Edinburg,1'9Z8).. 3.5 Toisuggest that figures for TAR'and nicotine, determn'nedl under standard conditions, should be interpreted as showing the intake every smoker may be expected tolobtaiin under every kiindi of condition is to be simplistic in the extreme. There is ample evidence (e.g. Creighton and Lewis, p. 289, op. cit.; Rawbone, Thorax, 1984 39, 657) to show that smokers' behaviour patterns are not constant. 3.6 Dr Aubort's first implicit assumption that TAR and nicotine,,d'etermiined!by i~nternationally agreed standard methods, are supposed to represent the amounts of these substances absorbediby human smokers, is therefore untenabie. 4. COMPENSATION 4.1 "Compensation" is the term used to describe the change in smoking behaviour often shown by a, smoker when presented wiithi a cigarette product diffieringi in standard deliivery from that of his habitual brand. It would seem that many smokers have a level of smoke iintake with which they feel comfortable and to whichi they have become habiltuatedi, when smoking! under "normal" (i.e. unstressful) circumstances. If tfiey are then given a product with a different standard delliivery, they adjust their smoking behaviour in a number of ways iniorder, ifpossible„ to,maintainithecustomarylevel of smoke intake. If the difference between the new and their usual'product is too great, they may find'the adjustment in, smokiing behaviour to be too extreme for comfort and will reject the new product. Evidpnce for "compensation" can be found in a number of the papers in the book "'Smoking Behaviour" (ow. ci t. ) and in papers cited by Ra:rbone (loc_c.iti_)'.

/5. 4.2 Thus the second of Dr Aubort's implicit assumptioms is incorrect. Smokers are not constant, or consistent, in their smoking behaviour. Given products of differing standard delivery, they try to adjiust the way they smoke to obtain the smoking response they 1'ike to experience. It iis consumers, rather than cigarettes, which "cheat the League Tables". 5;. VENTILATED CIGARETTES 5.1 A number of inedicallauthorities have counselled smokers, who wiishito continue to smoke, to switch to "'lowiTAR" products i.e: cigarettes yielding less than 10 mg TAR when smoked under standard conditions. Government authorities, acting on this advice, have urged manufacturers to reduce TAR levels of brands generally and have adviisedlsmokers to buy low TAR products. There has even been discriminatory taxation against higher TAR'brands. 5.2 2 Of the ways avai'lable for reduciing, the TAR yielded under standard smoking procedures„ the one most generally adbpted has involved ventilated filters. These are filter tipped cigarettes bearing one or more lines of minute perforations in the tipping band, allowing air to be sucked in duri'ngi the puff, thereby reducingi the volume of air drawn in tfiirough the cigarette coal (the burning tip) and diluting the smoke produced in the puff. This design has the added advantage of diluting the gas phase of thee smoke and reducing the intake of carbon monoxid'e -andlother gases, which might be considered deleterious. /..

/6 5.3 When smokers of mo.re,conventional products, i'.e. those yi'elding TAR leyels of 10-15 mg, switch to low TAR cigarettes, they frequently report them to~be unsatisfying. Studies, suchi as that by Rawbone (loc.. cit.),show that they attempt to compensate by altering their smoking behaviour. Only rarely do they increase the number of cigarett'es smoked~. 5.4 One theory to explain this compensatory behaviour suggests that smokers are seeking a particular level of nicotine intake. Acting on this, Russell (p. 336 in 'Smoking Behaviour' op. cit.) proposed that cigarettes should be designed to offer low levels of'TAR, but normal levels of nicotine. However, in tests with consumers, such cigarettes with altered ratios of TAR to nicotine were rejected as being "too strongP. This filnding lent support to an alternative theory, namely that smokers of low TAR cigarettes are seeking some minimal level of mouth stimulation, ai sensationi described as a"Imouthfuli of smoke"'. This led to the development of the ACTRON filter. 6. THE' "ACTR'ONI FILTER 6.1 It was found that delivering the diluti'ngiair through ai small number of channels (optimally 4)'„ slightly offset with respect to the longi'tudinall axis of the filter, induced turbuilence i'n the smoke issuring at the mouth end of the ACTR'0Nlfilter. This is in distinct contrast to the well-defined, narrow pencil of smo~ke issuing from the usual type of'conventionally ventilatedifiilter. The aerodynamic principles underlying the production of smoke turbulence have been descriibedlin a mathematical paper. (A J Baker, J A 0rzechowski', and G'E Stungis. Technical, Raper A1AA83-0'289 A1AA Aerospace Sciences Meeting, Reno, NV, January 1983). .

[7 6.2' The consequence of the smoke turbulence produced by the ACTRO'N filter is that human smokers report greater mouth sensation, due to spatial'summatiion of stimuli, and an enhanced feeling of'satisfaction, which they equate to that given by a convent'iionalllly ventilated cigarette of a nominally higher TAR delivery. 6.3 3 Detailed pharma: codynamilc studies, reported by Gori andl Lynch, Rqg. Toxicoli, and Pharmacol. 3,, 110-120 (1983) demonstrate definitivelly that smokers of cigarettes with ACTRON filters do.not receive more smoke than when they are smoking conventionally ventilated cigarettes of the same TAR levell„ as measured by standardimethods.. 7. MOUTH' SANIPLING 7.1 Dr Aubort's allegation that smokers of BARCLAY, a cigarette bearing the ACTRON filter, are deceived is based on his so-callledl"mouth sampling" analysis. 7.2' The mouth sampling procedure requires that a constant volume (35 ml )lbe withdrawn from the mouthieavity of all smokers in the test. However, it is knownithat the puffing habits of different individuals cover a wide spectrum of'puff volumes (on average between 101and 700 ml). Accordingly, the constant volume withdrawn from the mouth.cavity will compriise a variable mixture of smoke and of dil uti ng a~i ral ready i ni the mouth or comi ng from~ elsewhere in the naso-respiratory system. It would appear that the mouthi sampllingi procedure is based on the assumption that human smokers willl take alconstant volume puff (35 ml?) from each andlevery cigarette, irrespective of the varying deqrees of smoking pleasure

they experience. Observation and common sense tell us that this assumption is unjustified; smokers take a smaller puff from a cigarette they consider unpleasant or difficult to draw and they also vary the way they puff a cigarette according to their psychological mood. Consequently, the amount of smoke heldiiin the mouth cavity together with air already present will be variable and need bear no relation to the amount of smoke C"smoke delivery") collected inithe laboratory when a!cigarette is smoked under carefully controlled standard conditions. It follows that the.amount of smoke found in a sample of constant volume (35 mli) withdrawn from the variable mixture held in the mouth cavity will bear even less relation to the standard.delivery. 7'.3 The procedure ignores the principles underllying the special design of BARCLAY. Indeed, it may well be considered to be framed to frustrate the objectives for which the ACTRON fiilter was intended. 7.3.1 Fromiwhat has already been written (Para 6) about the ACTRON filter, imparting a swirling action to the smoke as it enters the mouthi- inicontrast to the narrow pencil of smoke produced by other ventilated cigarettes of the same. nominal ddllivery - it will be readily apparent that sampling at an ill-defined and arbitrary point in the mouth cavity will prodlace different results in the two cases, with the smoke-swirliiing,action of the ACTRON filter ensuring that more smoke will reach the sampling point. 7.3.2 A simple analogy may make matters more understandable. Consider two identical cups of coffee to each of which iss addedia single sugar lump. One cup is stirred viigorously,

/9 the other is not. Then, identical volumes of coffee aree withdrawn by syringe from each cup andiare analysedifor sugar content. Unless the sampling point is at the bottom of the cup, where the sugar will be concentrated in the unstirred cup, the unstirred cup will generally reveal a lower sugar concentration than the stirred,,everr though identical amounts of sugar were added to:the same volume of'coffee. It is clear that the degree of mixing and the position of the sampling point are important factors inithe interpretation of'the result. 7.4 Justification of'this argument has been provided in ani experiment described in~the report by Dr M Hissink of the TNO,laboratory, Apeldoorrr, Netherlands, (File No. 8712-13154,, Ref No,8'5'-03'13, dated January 1985). 7.4.1 In this study, alglass model of the human mouth iis,used, with provision for sampling at the front ("mouth-sampling") and at the rear ("t'hroat'-sampling"). The A'ubort protocol was folTowed entirely, except that samples of 35 ml voJume were withdrawn simultaneously from "throat" and "mouthP and, for greater precision,, spectrofluorescence was used, rather than optical absorbence, to;quantify TAR content. A number of Swiss brands, iincliuding BARCLAY, which yield fiigures of 1mg TAR under standardlsmoki'ng conditions, were used~ as welli as more cpnventiioinal' products. 7.4.2 The results show that BARCLAY differs from the other products in the way in which TAR andinicbt'i'ne are distributed between the two sampiing points. While ./..

/10 BARCLAY gives the~highest values for the "tnouth sample",, comparediwith the other ventilatedlcigarettes, these latter give higher results for the "throat sample". When the results for both sampli'ngipoints are combined, theree is a general correlation with the labeliled deliveries for all the cigarettes, including BARCL'AY, despite the noteworthy differences in the ratios of mouth to throat. 7.5 It should be noted that, in this model procedure, the cigarette i's~ smoked mechanically, usiing a fixed puff volume 351m1. In the case~of human smokers, however, the individual puffing habi'ts are unknown and uncontroTled'(as has beenipointed out above - Para 7.2) and all that is measured!is the concentration of smoke in the 35m1' sample withdrawn from the mouth,. 7.6 Quite clearly, the report by the TNO shows that a single sample of smoke withdrawn from an ill-defined region of the mouth cavity is uinrepresentati've of the smoke taken from all kinds of cigarette. Moreover, the confusion between concentration of smoke and quantity of smoke is unscientific,. Dr Aubort's third'assumption is therefore totally untenable and the methodological basis for mouth sampling must be rejectedias meaningless. 7.7 Finally, the procedure ignores what may happen to the smoke once it is in the smoker's mouth. Most smokers inhale the smoke rather than hold'it in the mouth for a prolonged period; indeed, it is an unnatural practice for most smokers to hold smoke in the mouth onliy. ../..