II P ~* ~$ hJr ~1 ~~J~Q "3 ·tV~ -4l-~ll~,r~le,~,. .,·· ~~t cl~;t~ LU~ep' ~L~ (S ~ts~j i 1!1· · ,i; q- ~ ~hri; 7" e,c~zC~ ~ -'14·L~t. e-·-P-·i~J- $"~"' C ~y\' 4IIPy~· IRu i!, ii,j~~e~ue i ~en_c.·i~i-i;u ,.-..........-..... .-...-.-...~ . .·. ·. ~-~t ·-- · -·I e---"l~~i·-~- C~rreCC FJ~.~ L~1;U ...-...~--. ii L. 1. ~·~I·---I L~:l~j"~-*r_Of I 1 6 '1 // E,c_ C~~~ c~e/ ~t~l~;LVt-, ········ ·· · · -·· ----·· I AiUIC Ln.~~~~rL;L ~-li ·I~ ·- "74·1'-'·· ·--· o/Lr·?~~~r/j ' ddc~...~ ~L~I" 1... 4-~-~·Sf"'~ ~~o~C~L.BjpCyj- 6~c~t! o( IjL~J·LI I ____. . _ ~cS,..L;L~CJ~ ....._,. .._ . O 3 ._~5L -b w C-~-_.t/_~/~-3~tac~dce-_. 0 tfe BA~Co document for Legal Services : Health Canada 19 May 1999 1]J ~Y~J CI~ ~~ 4YtU~ I~ ~ CS ~LrJriCCII_-r;Lk~- i~i kC^lk ys~ ~~~~ 4ic;i1~71~__~ LI,~ VC.~CI i_I~ 4iK:f _~____~ 3 d se 4.(qq~ -i L ~Y ~I~ \i ii-i J' nhc~. b,d·-_~'· U.~A-~U~ --- BATCo document for legal tenices : Health Canada l9 May 1999 Ist Graft i GP (~ June 1992) INTERNATIONAL COMMITTEE FOR CIGAR SMOKE STUDY CIGAR SUB-GROUP OF THE CORESTA SMOKE STUDY GROUP FUTURE WORK~ Study on the influence of the parameters on the ranking of cigar yields, As for most of the cooperative tests we have used ICCSS Daranerers, and we detected a ranking of the cigar yields which, at first sight, seems i 1104 i cal in function of ueignt and shape, it seems pood to make a study of the influence of the parameters on the ranking of the cigars, 2 Study on Cigar Smoking Machines, laking into account that for Cigarette Smoking Machines major improvements have recently been made, reason indi- cates that by adopting similar changes to the Cigar Smoking Machines similar improvements in reproductabilitY and reproducibility can be expected, 3 Smoking of large, Irregularly shaped or tipped cigars, As the ICCSS now believes that the same smoking Darame· ters should be used over the full range of cigars, but there could be mechanical problems in smoking and holding some cigars, it seems wise to investigate the possible problems encountered when smoking large, irregularly shaped or tipped cigars, 4 Study on Filter Holders, S i nce studies made for filter holders for c i ga rette smok ing have indicated that the actually usee: fi Iter holder material 'Perspex' is hygroscopic, a study should be made of this problem, 5 End Cut~ Although ICCSS concluded, out of practice, that, for the opening of closed-end cigars, the 'stralpht·cut' is to be recommended the influence of the cutting parameters b still is to be determined, D -- BATCo document for Legal Services : Health Canada 19 May 1999 6 Insertion Depth, Is the insertion death of the cigars a critical parameter for obtaining reaeatahle and reproducible results? If the answer is positive, then It should be defined In function of good holding of the cigars and butt length. 7 Cigar lighting, Sole cigars are difficult to light and need pre-ignition, this Can be done in different ways that will most likely have some influence on the smoke yields, A study can clarify the mechanisms involved in lighting, B Test Atmosphere, It is known that the conditioning and testing atmosphere have an effect on the smoke yields however, for cigars, this aspect has not yet be studied, 9 Preparation of Standard Methods for Sampling, Smoking and Analyzing Cigars, When ICCSS is convinced that most theoretical and practi· cal problems have been solved, a Standard Method should be written, discussed, agreed upon and tested, 10 Orpanisation of a large collaborative test, After acceptation of a Draft Standard Method, all labora- tories interested in a collaborative test should adapt their esuipment according to this method, A list of cigars representative for all different shares present on the world market should be prepared. These cigars should be made available to all partic~pa- ting laboratories, A decision on a time table for the analysis of these cigars should be agreed, The results of the Collaborative Test should be discussed and a report prepared. - ----- BATCD document for Legal Services : Health Canada 19 May 1999 B~L~P~SIIERBIWC tlER ZIWWAE~IINC~TRIE e,V, Born 2, Jlme Ist, 1992 (sa2) Remarks to Draft 1 (II May 1992) ~ A Review of the work of the_C~gar Sub-Group 1910-1991 i. INTRODUCTION (Summary Point 1 and 2) ~n 1970 same Companies at the European and US Cigar Indllstry identifiedthe desirability to investigate the possibility of procedures for the machine smoking of cigars, To progress this aim, they formed a committee known as the International Committee for Cigar Smoke Study (ICCSS], In 1972 the committee approached CORESTA and the work of the ICCSS was incorporated under the activities of the Scien- tific Commission of COI~S'IA and became the Cigar Sub·Group of the Smoke Study Grolrp~ In the early days of its existence and for many years the purpose was to establish a better understanding of the pxo- duct and to allow technically based product development, In the time attempts have been made to achieve much knowledge of machine smoking of cigars~ This report reviews the work to date on the progress made to define a method for the machine smoking of cigars, and based on this review, CDRESTA hopes to progress further the work on cigar smoking. Since its foundation the Cigar Sub-Group has conducted 17 cooperative studies, In the course of these studies 15 diff· eren: laboratories have participated with between 4 and 12 participants in each study, The studies have covered a wide range of topics and each was designed to examine a particular aspect of the smoking process, A summary of the test results is given in the next section 3 and a more complete listing of the tests including the aims, objectives, results and ccr- elusions drawn at the time of testing are given in appendix (Table 1 - 17). It is important to note that these tests were cooperative rather than collaborative, i,e, participants were not re- quired to we identical equipment and procedures, Indeed, in many of the trials the products were not kept constant, - - BATCo document for legal Senrices : Health Canada 19 May 1999 ·2- This practice is completely valid as, at that stage, prin- ciples only were being established and indeed wide experi- ence with different products and with different procedures . was essential to establish those principles. Unlike cigarette testing, it must be remembered that cigar smoking is a very long and often difficult operation, It has not been possible therefore to undertake large scale collabo- rative or cooperative tests, BATCo document for Legal Services : Health Canada 19 May 1999 TABACOF INA·VANDER ELST ~~a~lOlE ~SCOOTICH~~ - SOCIETE ANONYME Ra. GP(HVG l~uv#1, May 21st, 1992 j66192 1Co.tht members of theln~tn~:n~alCP_mmitt~ for Ci~dtSmakE S~Y "·ll-i Mr, Aeas 3~2 ~p~II Mr, B~ckst~m ~--- II Mrs, Del Cas~lo ~~i~· ~Qnu~N· . Mr, Fomsgaard Dr. Gabriei Dr, vander S~aeten Dr, Gopzaez Mr, Vaschure X,$· M. Hdfrich Jr, Mr, Webcr Mr. Le Hellcy Mr. t~t~eatabft Mr, Horler Mt, Wright Mr, Limburg Mr, Moran Dr, Baslreritch Dr, Reif Mr, lacob Dear ICCSS·mtmbcr, Pleax f~od, annexed, the draft of the "ICCSS ~rogr#s Review Paper", which will be discussed dluing our 2Sth meeting in Paris, Yours sincmly, POPPE Chairman trm~FYmP·CIO1OIEUYP MUIYIU~UI~ mFL ·1116I IOCU;, M;I IOlpFMIZL~M.~)· CrW mlGEu · H I ·1. i. ~~mvll~H I1C O\ BL OlbC I! Y · IILa 1IPI · rlr I: 1~91a · I t w ·tv n ur vwr N BATCo document for Legal Services : Health Canada 19 May 1999 Draft 1 (1( ~ay 1992) A Review of the Work of the Cigar Sub-Group~ 1970-1991 Table of Contents, 1 BhCXGBOIRID~ Z INTRODUCTION, 3 GMEBIIl, OtfEBtfIZW OF THPI WOBX OP TIE CIGAR SIJB-GBOUP~ I DISCUSSION OP RESULTS OBTAIZlga TO DATE, J~1 Glass Ffbre Filter Trap Material Type and Loading. ~.2 Smoking Paramelers~ 1.3 End-cut 1,1 Sorting / Non-sorting of Cigars, 1,5 Statistical VarfabilltY~ 1.6 Determination of Nicotine and Water in Smoke. 5 OTH~ IT~S PDB DISCUSSION. 5,1 Atm3sphere for Conditioning and Smoking of Ci~ars~ 5~2 Cigar Holder. 5.3 Insertion Depth, 5,1 Butt leppth~ 5,5 Airflow during Smoking. 5~6 Cigar Zlghtln~~ 6 SUlgURY ARD COHCLVSIONS. ~ppend~x: SUHKABY OP TESTS 1 to II~ - --- BATCo document for Legal Services : Health Canada 19 May 1999 I~~ndn~ This report reviews the work doPe by the Cigar SPb-GroPP since its formation in 1970. in particular the 11 coopot- alive studies which have been conducted since the formation of the sub-group have been e~mari~ed, They have all been reexamined PtatisticbIly and, where appropriate, a note has been added to the original sure~nary~ I general overview has been included. ~~;J · The results obtained from tbs cooperative ltpd~8~ are dle· cussed and various conclusions have been from each teet. ~-~m~nul, Bo~resp towards rmei~rkigkl method fort the determination of tar and nicotine fsoa cigar smoke has ak been hindarad by the Iaek of ~tandardisltleP of ep~iplent and procedures. particular attention is draw to the at~o~- phere of conditioning and teeting, the smoking machine, air· flow control within the machine, the cigar holder, the smoke trap, cigar cutting, cigar selection and analysis of nic- otine and water in the k~~e~frrP~i·L~c; ~n order to n~e further progress, 111 participating COP' panies are encouraged to upgrade their laboratory faclli- ties. particular attention should be paid to labo~ator~ condltionipp and the provision of a piston operated smoking machine and chronatopraphle procedures for the determination of nicotine and water in the smoke. The Cigar sub-Group Shgpldltpyp its attention to the design of standard cigar holders, evaluation of the effects of air flow during smoking and a detailed evaluation of the smoking PtL~.~ *,Y when this ~a~8le, it a be necessary to check whether the provisiooally recommended parameters produce consistent comparisons between ~f of cigars 1. d~lVC*LL There individual developments will`be tssta by cooperative studies and, when sufficient progress has been made and sufficient laboratories are fully eguiDPed, the Spb·GrOPp will organise a detailed collaborative study, -- BATCo document for legal Services : Health Canada 19 May 1989 Draft 1 (Illlay 1992) R Ileview of the Work of the Cigar Sub-Group~ 1970-1991 1 BACI(GBOUHD~ In 1970 the cigar Industry identified the desirability to standardise procedures far the machine smoking of Cigars. Ta prepress this aim, they formed a corr~nittee known as the International Cor~nittee for Cigar Smoke Study (ICCSS). In 1972 the corrJnittee approached CORESTA and the work of the ICCSS was incorporated under the activities of the Scien- tific Commission of CORESTA and became the Cigar Sub-Group of the Smoke Study Group. In the early days of Its existence and for mary years the purpose of producing "standard' procedures was to establish a better understanding of the praduct,by allowing short term comparisons in one laboratory, thereby aiding technically based product development, In the time that attempts have been made to achieve stan- dardised procedures for cigar smoking, much knowledge has been pained~ Simultaneously, majoz improvements have b~ made to the standardfsed machine smokinp of cigarettes. 199011991, COBESTR undertook a major revision of all stan- dards relevant to the machine smoking of cigarettes~ This review led to major changes in practice and ultimately a revision of the International Standards for cigarette tee- tin?. Although not directly relevant to machine smoking of cigars, it is probable that much of the knowledge and many of the improvements made to the machine smoking of cigarettes may have a bearing on aspects of the cigar smoking studies made in the last 20 yea:s~ Additionally, the European Co~nlselon have recently~ expressed a corrni Iment to the determination of tar and II. nicotine yields from tobacco products other than as soon as recognized standards are deflned~ This report reviews the work to date on the progress made to define a method for the machine smoking of cigars, and based on this review, CORESTA hopes to progress further the work on cigar smoking, P BATCo document for Legal Services : Health Canada 19 May 1999 2 INTRODUCTION Since its foundation the Cigar Sub-Group has conducted 11 cooperative studiee~ In the course of these studies 15 different laboratories have participated with between I and 12 participants in each study. The studies have covered a wide range of topics and each was designed to examine a particular aspect of the smoking process. I surmnaXY of the test results Given in section 3 and a more complete listing of the tests including the aims, objectives, results and conclusions drawn at the time of testing are Given in the appendix (Table 1 17). Also where applicable, new conclusions have been added which have resulted from a fresh statistical analysis of the data. In addi tied, further information from other sources La been referred to, to assist in drawing appropriate conclusions. It Is Important to note that these tests were cooperative ra ther than co i 1 abora tlve , participants were not required to use identical e~uipmeet and prpcedures~ Indeed, in many of the trials the products were not kept constant. This practice Is completely valid as, at that stage, Prln- cipies only were being established and indeed wide experi- ence with different products and with different procedures was essential to establish those principles. Un like cigarette testing, it must be remembered that cigar smoking is a laborious and often difficult operation. It has not been possible therefor to undertake large scale collaborative or cooperative tests and In manY cases labo- ratories were nor able to fellow laid down procedures due to equipment differences. BATCo document for Legal Senrices : Health Canada 19 May 1999 3 GEIOEBAJ, OVH1YTFW 01 ~HIi WOBX DF 11IX CIGAR StlB-OBWP 3,1 The first important step consisted in collecting the existing data on the smokers behaviour for om~a~d Ir~e~itt~eipfit ci~.~*hleh r,p~,ran~jthe major part of Studies from France, the United Kingdom, Germany and the U.S.A~ allowed the first parameters for ci~ar.Pmokin~ to be proposed and to start a series of cooperative studies, based on these parameters~ 3.2 A detailed report of the first results was presented to the Smoke Study Group in Nice in 1973, and in 1975 the COBESTA Bulletin ] published a CDRESTA provisional rs· commendation on parameters for cigar smoking based on the consumer data and some practical eonsiderations, namely: For studying the smoke from (untipped) cigars In the weight range 1.5 g to 5.5 g : puff volume : 20 mi puff duration : 1.5 see puff interval : 10 see (interpuff period : 38,5 seel butt length : 33 mn 3.3 In 1975 and 1976, several studies shaved that the amount of secondary allraloids and pyridlnic darivates present in cigar smoke condensate might be important; therefore the eorm~ittee made the following recomPendations: to keep the ISO spectrometrlc method 3400 for the determination of total alkalcids to propose a ~ae-chromato~aphic method for the specific determination of nicotine 3,4 A series of cooperative exercises between 1977 and 1979 allowed the Cigar Sub-group to reconrmend the glass fibre filter trap, The Increase In pressure drop of the filter assembly after smoking must not exceed 250 Pa~ From the limited number of types of cigars studied, it was suggested that the number of cigars to be smoked should be chosen so that the total particulate matter does not exceed : 100 mg per glass flbre filter of a nominal 44 mn diameter 150 mg per Glass fibre filter of a nominal 56 mm diameter 'COBESTI Bulletin 1975-1 pace 33, W BATCo document for Legal Services : Health Canada 19 May 1999 3,5 From 1980 to 1983 and later from 1987 to 1989, several smokin~ cooperative exercises were organ Ised in an attempt to achieve a better understanding of the reasons for the important within and between laboratory varia- bility~ From a cigar sample of the same production, the selec· tion by weight and draw resistance improves the within laboratory variation but ~iwC~ has no influence in the be torYvdriation, variation. ~d·-i-~·l··lht-~u;'1~;i-, 3~61In19~Y4'four laboratories presented the results of a s S~Jnt~ stud~on cigars weighing more than 5.5 g. Using the reco~unanded parameters some cigars have a tendency to extinguish between puffs and the purpose was to study If a modification of some of the parameters Would eliminate this problem. *r ~L~kL~ . with ail the other parameters rsmafninp consttg~lchang- ing the puff interval from I0 sec to 30 sec/redueduced the tendency of these cigars to extinguish between puffs. .,,, mlmn.~r nmr~ ~C woll-*r Lthe 6mnka vlbl~;fl~"Y1I However, such a 61ort puff Interval seems unrealistic and therefore is not the appropriate solution to the Dnll~p of im~)ia~ Li~an d fhji lil~l;j~l, 3.7 Another problem related to the stapd~rdisatipn of cigar smoking is the case of c~qarsyrth closed heads. They can be cut by V-cut, O~U8~Rg or piercing. For standardisation it is however essential to reconunend only one method~ Conserluently, a cooperative exercise was ~~eanised in 1986 to compare different ways of cutting under well defined conditions. ,VillrsJjA Because_ it can be_ mornLPreci selY_desetlbed Bscol- mi t tee a 4traiphl cut~' a wmai defined p~Jr~, depth (which is still to be let~s~.~ 3~8 In 1988 a cooperative test on cigars of a weight of 1~1 p was organised to investigate the affect of dlffer· ing parameters on one a~ cigar type. There was no practical problem with any of the parameters tested. So far, the conunlttee has organised 17 cooperative studies and marked improvements have been Iealised~ I BA~Co document for Legal Services : Health Canada 19 May 1999 I DISCUSSIOLI OP BF~SUI~S OBTAINgD TO MTI R The cooperative testsYmd,, attempt to understand 1 the wide range of principles invoi Where appropriate, each built on the experience fro~qrevicus work, Stat~sti· eians were not always Involved~the~ and on se-exam- ination some of the conclusions dr~ij~-~nnot be fully sub- stantiated. Where this Is the ease, a s~atfstical re·appral- sal has been added to the summary of\the test given in appendix, In the light of present day knowledge, some of the data and hence ~onclusions drawn at the time uy not be entirely valid~hs an exangle/lt is known that the type of sh6kinp machise (f~e, principle of operation) significantly affects smoke yield,' -:---" ---Z~ i" _'h"*~L~"~" pYn~d --th~ _ thP. vi I ]- _ ~~~_ ~CCOC~9d· TL M~LnC Lkf L--L-~~Y(Flb~ lr LL:- UU~L~Ot all laboratories were using the same principle of machine operation, therefore it is likely that some of the fabil- ity reported will be due to this factor~ ~iJws~ 1 Further, not all laboratories which were involved In these tests are air conditioned, but it Is now known that the ~Q i ambient conditions affect the combustion characteristics of the product~ Recent work has shown that the material of construction of the glass fibre filter holder must be non-hy~oscopic, otherwise data may be seriously affected. This new type of holder was not available in the past and all smoking was done using the older style Perspex holder. Nevertheless many of the conclusions drawn at the time are still valid and give a pood indication for the way fcrward~ A re-appraisal of the data from a statistical viewpoint has also indicated many areas where further work is required. The following paragraphs indicate the overal 1 cone 1 us ions subdivided Into subject headings, 4.1 Glass Piboe Filter Trap Katerial Type and Loadia(l. 4~~~1Test~ 9, 10 and 11 were intended to study certain performance characteristics of glass fibre filter material , 4,1,2 From previous appraisals of Tests 9 and 10, It was eor~nented that "glass fibre materials are variable; vacuum smoking machines are sensitive to change In filter assembly pressure drop i maximum loading should be 3 cigars or TPH of 200 ms per filter; Q certain filter materials are not suitable for as many as 3 cigars par pad I and further, more detailed, study Is needed"" W BAfCo document for Legal Services : Health Canada 19 May 1999 4,1l3 For Test 11 It was possible to perform a statistical re-analysis and this indicated similar mean smoking yields and associated variability for a number of different filter materials (a 'c6ntrol' material was compared with each of 9 1 abo ra t ory ' a ' own ' material). No problems of over-loading were chserved with 1 and 2 cigars per pad, as indicated by eimflax mean yields, ~~~'a~l ~IIY ~ll*L~-ZIc7~LnlrC nd, pa~wa~b~-~t Following this test the Sub·Group ~deetded to restrict the amount of TPM to approximately 150 mg per nominal 56 mm diameter Glass Flbre Fllter_Pad, 4.1.4 Focusing specifically on the effects of different numbers of cigars per pad, Test S was particularly useful since It Included i, 2 and 3 cigars per pad, albeit for lust one cigar-type with 1 Imit ed TPM Ug -.L-·.L-H.~'mr~-C~)~k~mLLL These results also suggested no problems of ovar·loadin~, with a clear indication that in terms of repeatabi 1 i ty (i~e, wi thin-laborat orY variabl ii ty during short period of testing) it would be benefi- cial to test a number of cigars per pad. 4l1.5 Given that the recoasnended type of smoking machine is piston operated and not constant vacuum, the effects noted in Tests 9 and 10 should not now be important. The variable nature of the pad material has been noted elsewhere and this my still require investigation, especially fax the determination of n i cot ine ~ Otherwi se there would appear to be no major problem due to pad stability, provided that the pad size and maximum yield of total particulate matter is defined, This may require additional eva- luation for different cigar shapes and sizes. Indeed it may be desirable to vary the number of cigars depending on the size and shape of smoked per the cigar well as yield of total particulate matter. 4.1.6 Regarding the material and construction of the glass fibre filter holder, it is clear that the observa- ti6ns reported for cigarette smoke will apply equal ly to cigar smoke ~ The effect has been reeognised by ISO and the standard for the cigarette holder includes reference to the fact that it should be constructed from materials which are chemically and non-hygroscopic. In auaY1~, preatar standardlsatfon of the holder and the glass flbre filter material is re~ulred~ :0\ ·O BATCo document for Legal Services : Health Canada 19 May 1999 4.2 Smoking Parameten, I~t.lvarious combinations of puff volume, puff duration and puff interval were considered In Tests i, 2, 3 and 17, 4.2,2 Not surprisingly, the different combinations of parameters led to differences in mean yields and, to some extent, affected the var 1 abi.litu In the results, Certain combinations also resulted in the I, cigars not staying all~ht~ (·4 4,2.3 The provisional figures suggested by COBESTA of a 20 mi puff volume, a 1,5 see puff duration and a 49 sec puff int erva 1 worked adequately for the cigars tested in the cooperative studies, In particular no reason was found to segregate cigars below 115 g from those of 1.5 g - 5~5 g. b i The chosen parameters should apply to all cigars up to at least 5,5 g, but beyond this weight it Is not yet possible to corpment~ However, the possible effects of different parameters may need to be studied further in order to establish a meaningful measurement of the full range of weight and type of cf~ars~ (~3 End-cut 11,3.1 Test 15 considered the effects of a 'V-cut' to the butt end of the cigar versus a 'straipht-cut'. The results using a 'stra~~ht-cut' were lover In mean yields, but similes to the '8-cut' in varfabilltyl 4.3.2-"~3hp+-' ~L ~kMC ~YYIL· 1L Y1·lL L~IJ~fi Il~t the !Ir.UL~'_W~U1~ ~d ~"~~~~II' .,LL~L~LL-~ -ha~~-·dift~e~LLlatln~ ~·'··---·llm"~·)~-~b ~~kh r~mn 1Ur~i~h~ II ~YC~e~C L-L~-I-LL~.·~~c~-~C~- u4u4unr I The 'straight-cut' can be justified on the basis that it Is the more usual end·cut, is easier to staodardise and provides a practical convenience in a routine laboratory testing situation, 4.3.3 Even so, a decision is still needed on the standardisation of aile of end-cut, Clearly this is not as obvious as It first seems because of the gross differences between cigar-types with respect to width and end-shape, BATCo document for Legal Services : Health Canada 19 May 1999 ~~I Sorting I Non-sdrtinl of Cigars. (,4.1 Arising from concerns about high variability between individual cigars (of one type), Tests 12 and 13 were designed to study the possible value of weight and pressure drop sorting the cigars before smaklng~ 1.4,2 The conclusion from the re-analyses Is that sorting Is beneficial in that the repeatability·l (withfn- 1 aborat cry veri abi lity) is appreciably improved~ That is, the repeatability for 1 'sorted' cigar per pad roughly equates to that for 2 'unsorted' cigars. Again, the generality of this result for all cioar- types is unknown ~just one, 3 ei~ar·type was studied), but statistical experience in many areas of laboratorylproduct testing suggests that the same principles would apply throu~hout~ 4.4.3 Despite the fact that sorting is beneficial, serious objection could be raised about this procedure on the pounds of sampling bias. Furthermore 11 would be difficult to specify a standardised procedure for cigar selection which would be appropriate for all cigar types. A sensible compromise, based on practical experience in the laboratory, would be sample rejection, rather than sa~le selection, cigars for testing should be quickly examined for defects, including natural defects such as thick mid-ribs at the butt- end and that those should be rejected. 4,5 Statistical Varlability~ I.5.1Host of the tests provided some information on the variability of smoke yields; see appendix for details of the Individual tests. However It must be noted that this information was obtained from vasf- ous laboratories without standardlsation of the equipment, procedures, etc~ 4.5.2 Looking across the separate tests, ~tssrem~ttr~en~ ~b~C~tbti~-~e~ I d ~kr, there was no clear Indication that the variability (neither within or between labarat- ories) was correlated with mean yield~ Note, however, that this could be misleading since an underlyln~ correlation may have been concealed by changes in smoking method details that occurred across tests. ilji /ru BATCo document for Legal Services : Health Canada 19 May 1999 4.5.2,2 Secon/dIy, we note that oysrall average SD's for /TPH, IIFDPH and Smok~ tine are: TPM NFDPM~iS_m,Nic, Hean yield mg 50 /50 3 SD bet`bean laboratories / 4~~ 4.6 0,3 SD P~tween channels (uiihin laboratories) ~1 3.2 O~i ~N~B. Average n cigars per chan~1 was approl[Imately 2) 4.5.3 Fpxr NFDPH, for/Bx~ample, tgSb~~e woul4·Qit~ a~pro- imate 95% confidence ~C~inits on thepiean Meld fpr\ one eipar-~3a tested/by any one~oratory Fbdsen " at randop/of tthe oFdar of ~10PJP.4 ~~6' + 3' 2 !nY" I where,//n Is the n)t~ber of ~els of smalCMlp over which the mean)Y~eld is obtained). ~.5.rA number of other aspects of statistical varlahilltY should be noted. 4.5.1.1 The effect of pre-sorting the cigars for smoking was mentioned earlier. This is valuable from the point of view of economy of testing (i.e, a required repeatability achieved by smoking fever cigars i. However, it is relatively unimportant in the context of estimating an 'absolute' mean yield since the associated confidence limits would be governed largely by other components of variance (e,8. between laboratories), ~~5.~,2 The repeatability effects of smoking a number of cigars per pad was also noted earlfes~ As for 1,5~4~1, testing efficiency can be Increased by smoking a number of Cigars per pad but its in- fluence in the wider contexts of variability is limited. 9.5.(,3 There are other, very important components of variance which describe method andlor product variation over longer periods of time; these are relevant in the context of cluantffyinp the con- fidence limits in mean yields, laboratory c6mp- arfsons, rank-ordering, ete~ These effects can- not be determined from the Individual tests that were re-analyzed (i.e~ method and cigar-type changes occurred between tests)~ BATCo document for Legal Services : Health Canada 19 May 1999 4.5.5 Proper standardisatlon of equipment, procedures, etc, wi 1 1 lead to improved be tween- 1 abora t cry comparisons and the repeatability within a lab- oratory may also be improved. It Is therefore essential that much more empha- sis is placed on these aspects when further between laboratory trials are considered. 1.6 Determination of Nicotine and Water in Smo~e.~~, Tests 1, 5, '1 and II included attempts to study the effects of certain aspects of Water and Nicotine p~ea- suremen t. Unf ortuna lel y this was not very successful since the results were either fneoncluslve or did not permi t a meaningful statistical re-analysis. However, experience in the area of cigarette testing strongly suggests that there would be no problem If modern cc methods were used as the basis for standardlaatlon~ Nevertheless, even if this seems the appropriate route, special attention must be given to the highly variable Water levels Indicated by many of the Individual test~lG results (~F~.~6~ttn the variability for HFDPX or DPM wasdj5c~~ as hlp~h aslo~ hiCher than,f~r TPM). .·L-~L4""11 ·03 BATCo document for Legal Services : Health Canada 19 May 1999 5 DTHgB ITMS FOE DDCUSSION. As well as the topics lifted in section I, there are a number of other factors, which have not so far been standardised, which need to be given careful consideration before a final decision Is taken to perform a full collaborative study, Some of these are listed bel6v~ In addition, consideration should be given to the $eat variety of cigars available e.~~ there are more than a thou- sand cigar shapes, length varies from CS mm to 250 nml weight varies from less than I g to Mrs than 15 g, diameter varies from 6 mm to 20 nun, the shape may not be regularly cylindrical especially at the mouth-end~ 5.1 Atmosphere for Conditioning and Smoklq of Cigars. Hhlrst most laboratories have used a condltionln~ pro- cedure for products prior to testing, no systematic ap- praisal has been made on the effect of the length of conditioning time on e~utlibrium, or the effect of product conditioning on yields. Those two effects must be e~tab- lished before a meaningful standard can be produced to define the cigar conditlonin~ procedure. Of greater Importance and relevance to the variability of snake yields, is that not all laboratories that partici- pated in the cooperative tests, have air conditioned laboratories. It is known that this is very important and yields are affected. This Is therefore another aspect which needs to be taken into account before conducting a collaborative experiment~ 5.2 Cigar Holder, At present there Is no definition of a cigar holder. All laboratories have been free to devise their own method of holding the cigars during smokin~~ Integral with holding is the question of sealing and If the cigar is not firmly held and sealed It will obviously not provide reliable data. More work is required in this area, 5 ~ 3 Insert ion Depth ~ Linked with the puestfon of the holder is that of Inser· tion depth. Unlike the papas wrapper of cigarettes, the cigar wrapper is not porous. The depth of Insertion is likely therefore to be much less Important. This needs to be determined and if it Is so then this fact can be used to advantage and a holder designed to give optimum depth of insertion ~ This would assist wi th the holding and sealing referred to earlier and also remove another uncon· trolled varfable~ I a BATCo document for Legal Services : Health Canada 19 May 1999 5~1 Butt Length, The COXESTA provisional recorrmmendation for butt length is 33 mm for untlpped cigars, This was based on observations of discarded butts~ Bearing In mind that the brtt may depend on the cigar geometry, and will affect the holding of the cigar, It may be prudent to reconsider this aspect. 5,5 Airflow during Smoking. Currently, no attempt has been made to measure or control airflow over the cigars during smoking. Since airflow is bound to affect the combus t i on oharaeleristlcs of the product it is likely that It will affect the tar and possibly nicotine yields, However, there Is very little .inter-puff smoulder and because of this there Is no pre- dicting what effect airflow will have on yleld~ Further- more it is possible that the magnitude and direction of any effect may vary with cigar size and shape. Certainly lack of central of air flow could contribute to the vari- ability of the data and clearly this needs investipatlon~ 5,6 Cigar Lighting, Special attention may need to be directed to the problems of lighting cigars, especially these normally reouirlng pre-ignition, BATCo document for Legal Services : Health Canada 19 May 1999 Appendix Page 1 SUH~ABY OP TESTS 1 to 171 This Appendix contains statements of the Objeetfves and Pro- cedures for each of the 17 teats, along with a surrunary of the main results an the conclusions drawn. Inmost cases, these results and conclusions have been au(l- mented by a statistical reappraisal of the data. An outline of the basis of these statistical analysis is given below with a brief explanation of some of the terminology used. Outline of Statistical nethodology, The statistical analyses were carried out with a view to achieving one or more of the following objectives: 1 Comparison of mean yields (e~~. 'stralght' versus 'Y' style of end cut) removing as far as possible the effects of other factors. t Comparison of variances (e~q~ variance of the data from selected versus unselected cigars). 3 Estimation of individual components of variance, especially between and within laboratories, with a view to providing some information on the reprodueibi~lty and repeatability of the method. In general the statistical technl~ue used was that known as the Analysis of Variance, h mathematical model lunderlying the Analysis of Variance) was formulated to represent the factors being studied in each of the tests~ The data were then used to estimate the means and variances associated with the factors characterised by the model. The following is an explanation of some of the terminology used: 1 Statf stieal significance in a statistical context, an observed difference (e,p, between two means) is said to be significant if It 16 ~eater than can reasonably be attributed to chance varia- tions, wether it is of practical consepuence or not. For this reason, a test of siqnfficanee Is not the only point to be considered when drawing conclusions from the results of an experlment~ Experiments are usually designed not only to establish whether or not a factor has a 'signlficant" effect but also to establish whether or not this effect has pra~cti- cal consequences. BATCo document for Legal Services : Health Canada 19 May 1999 Appendix Page 2 2 Variance and Standard Deviation A Standard deviation (often abbreviated to SD) is the square root of its associated variance, and is a measure of the "spread" tar variability) of data~ If the mean and SD of the measurements of a parameter are known, then approximately 958 of all these measurements will be in the range (mean - Z SD's) to (mean + Z SO's)~ 3 Within Laboratory and Between Laboratory variance It is a well-known fact that when a test Is repeated time after time In any one laboratory, the results of these repeated tests are rarely idcntical~ The within laboratory variance Is a measure of the amount of variability observed in these repeated measurements. It Is also a well-known fact that if identical samples are tested in a number of laboratories, then the results from these laboratories are rarely identical, The between labora- tory variance is a measure of the amount of variability seen in the results from different laboratories testing identical samp las. 'Per Channel" analysis of the data During the course of the 17 tests described In this appendix, the number of cigars smoked through a ~ln~le channel varied from 1 to 3. Thus the smoke yields reported are effectively the mean of 1 or 2 or 3 cigars, depending on the number of cigars smoked, Tn peneral the statistical analyses described in this appendix have been based upon the observed data, without any adjustments being ~nade for the number of cigars smoked on each channel. Such adJustments would assume that all of the within-channel variability Is due to the variabllltl of the cigars themselves. A limited analysts (not included in this document) suggests that this is not the ease 1 For example, in the measurement of TPM It seams probable that the cigars themselves account for less than 75% of the w~thin- channel variance. In addition to the statistical reappraisal carried out for each test, some analyses were done on results pooled from some or all of the tests, Details of all analyses, including these additional analyses, can be provided, BATCo document for Legal Services : Health Canada 19 May 1999 )gp~tiX : table 1 table 1 lo~erati~e Test r : 1 ~CGSS-inetipgP 1 bate cb ~etiP~ to a~ae test : DecePtPr 1970 Sm~b$g tests to be d~Foe m stall d~arr Idii~eter 5 9 re) a~eepdia~ to 2 different sets of pa~eters : 1) Conrts cigantte pa~g~eters 2) Parsaeters rep~a~ed in F~in~ce to be typical for cigar s~w - : OBJECTIPE - Ib Qte~iae Imet~u~ the 'TreDeh" p~ters e~ be used for oachiae s~idDJ of d~J, - To st~ the inauare of both sets of pa~petw on the aprmts of tar ad nicotine free e~gar st~te. PROCfDURC - Each labc~ata~i to spoke : · their cwo ~all dpars, std · 100 sa~e cigars (&9 ~ bian~ter, 1l0 n leo3th, 2,2 g weight), nei~t 3 drbr resistance selected. fach latspa~eo~y to ~p~be aeee~di~ to 2 sets of ga~ters : ~e~ta cigarette "frencb" pudmeters ira~i~etw PdfmaatieP tsec 2u Puff Pol~Pe 35ml 20~ ~eral 60# Usee - ~a~ labnatcay to a its own e~iplent and FPae~Pe~. -~lat~at~ie~ I~tidpated. ORIGIIIII BESU~TS and COlleLUSIONS i, mese d~aR an be s~hed m a sa3kiag machine udap 'T~m~ Fax;m~ters wit~wt practical diffic~, 2, me yields of tar ad nirotine are affected by the ~nb~p pa~i~aters. 3. 'Ihe'Tre~" p~Ptws giPe sli~tly larer between la~adtcpy biffeteo~es for an ad nicotine. I, me la~est between la~P~ttny differeoces ~ae in the ~mt of water, rasaed in Iffl~ STATISTICAL REIPPRIISAL i. c~clusico ng 3 is not ~pFe~tel by statistical aaatpl, 2, Cenclusiea o0 I qk inferred ~ the hetkeen la~patcoy S,D,'s fcP an and DAI. 3~ Pith 2 cigars per pad the within labePatoty S.D,'s are loner fcP the TrPm~ pa~te~s, 12th Iku 1992 ~i I - - BATCo document for Legal Senrices : Health Canada 19 May 1999 table 2 ~pe~dix: table 2 e~erative Test g 2 ICCSS - Peetin~ NO : 2 Date of PeetitlO to agree test : E~rcbes t9n ~akin~ tcr'3 te b? h m s~e ~arJ acccdiP~ te 2 different sets od pa~i~t~s : 1) "Ren~n p~a~tw ~ed in t~ ~e~iac; test 21 ~ao pa~i~ters, reppted fron shdies in Cerreny, TrmwP i~ the 0.if., and ~utt length studies in t~u! V,Sa. I~ie~r pa~t~s), 083ECTiVE -'h,dete~ne wt~tber the ~ie~ pa~ten an be ~ed for nacbine gD5~ Of dgars, - TO sAdY the influence ed ~tb sets of para~eters op the an~Mts of tar and nicotine frw cigar smP]Pe. PROCEDURE - ~cb ~d~cpateay to ~P~ke : · 100 s~t~e cigars D-9 on biapeter, 1l0 En length, f2 I tRi~mt, Irmtiyped : i.e. Ji~dlar to those nsed in o~epetative test nO i), weight (mtan ~1-5~)6 draw~istaow(m~8an ~l-lt~ seleaed. · 2 cigars to be smbzd per Class rilap filter T~ (~T) of 56 a - u latrmtmy 0 ~ amab~L~ to the 1 rets d paa~aten : 'Trench" 'IjRT Puff ~rati~ 2see 1,5 sec Puff ~Pl~me 20ml 20~ Fre~Pcy 41 sec D see htt length Mtipgad u~ 33 n tiFF~d tip+l'lnm -5latc~atcais ~artidpa~ed. ORf GlNAL RESULTS and e6NCLUSI6NS i, Ibese cigars can be srmted on a ~io0 ma~iae ~io~ liew" ywameters witbwt paeticl difficulty. 2. The yields of tar ani nimtitu! are affectai by the sanldug pa~ters. 3, R high variation of results for water in IPlr xas fo~d for birth sets of Fa~i~oetets. STATTSfIClf RE1PPIlITSAl The ~n la~atc~y ~aria~ity derived fr~ the ~ecr pa~eters is FPobabiy lower than tb3se derived ft~n the "Freneh" p~am~ten. 12th $ 1991 W BATCo document for Legal Services : Health Canada 19 May 1999 ~Peabix : table~ table ~ Cwpe~ative Test F ICCSS - lleetipg ND : 3 Date of ~eetia? to aae~ test : Bil 19~n ~dpg tests to b~ dole m larg~ dgars 8eccading to the "1W' pa~i~tcrs used in the poePicus test. oeJEcTlvE - To rtu~ the li~tinj of larger ci~s when ~aldn3 to the new pati~eters. - To establish the Iwel d bctweep la~raatorl agce~meat when S1POJ~D~ la~er a~ars, PROCEDURE - 69 s~ple dgats (1Cd mp length, U no d$peter, 4.6 g wei~t, ~ti~83)~ - Flater and nie~tipe to be ~s~ed ~ 2 E~Pled e~aFs. Coly ~ dpar to be ~oked per CI~ ed: 56 rn. -Ila~atraies partidpated. ORIGINAL RESULTS and CDNCLBSIONS Ho diffiolltie~ vere e~erieneed in tbe lightirg ed tbese la~e dgias, ~irp tbe'l9er)' parapetw, 2. l~e~r pat;npeters 43d be adopted in fuhpp 1 nf#SS - pa~Ii~te~Sn, 3. Altb~ there Kas rea#oable a~ee~nt on IPW, there var, r~nriderable ~et~a la~a~oty discrepanrl for dry partic~ate ~tter mRI), m~e to the big var$bility in ate~ results, STATISTICAL REIPPRAISAI Ho ~e~nts, 12th ~ky 1992 P BATCo document for legal Services : Health Canada 19 May 1999 E~E~E~E~E~E~E~E~E~E~E~E~E~E~~ : table I Cwperative Test k 4 Ccaesta Sub·Crolp - Lting r : 4 me d rsetjpF to agree test : Dec~ter 1912 Smskiw s~rlies to ~e F~f~rmd to iwestigate the water ~asure~nt, oaJtcrIvE -m~~baun,h~vawnt~~crE~rraenuu~~ otsetPed. i -'lo meYure the water cmteot cb s~ples of extrachfn scent with )mc~midentical water aoteat. PROCEDURI - 24 ~ple dgars distribhed (lCO re lea5th, 14 re di~eler, 4.6 g weight, MtipFed : i,e. similar to e~se wd in aopeatiR test o0 3). - Chly 1 cigar t6 be smsked per (a~ on 56 am. - Each labca~t~y to carrp wt 2 separate Ltcrm~nati~, i.e. : · by am method (Xarl ~is~er or CdSL~IDatCQSapb) Ed · by a gra~in~~ic ~et~d (silicagel) - ~ li~nae the water aat~nt of the bisbibrtel solwnt sample by am n~tt~d. - 7 labsrstmies partidpatd. ORlCIBAI RLSUbTS and COHCIIIS~OIS i. Ditficylties were experienced in the use ed the gra~imetrie rt~, leading to aot~d~ictepp~ results. i. Ib~g their on netr8 Ilb~a0mier abt~i~uri ec~p~ablr ren~le m the solvent samples with JPtwD water CeOt~t~ ?. It was therefore agreed to persevere with the Ihrl Fisfher or GasEhrrmatcgr~hic met~l. I, Ihe hiph variatim in rwlts fa water in particulate matter repcoted in Feevia~s tests remained ~me~ained. 5, Further invertigatico eo sm3]drq p~b~edute ~ req~tired. STATISTICAL REAPPRAISAL No Boen~aents, 12tb ~i 1992 0, BATCo document for Legal Services : Health Canada 19 May 1999 table 5 ~I~gdiX : table 5 CmperatiPe Test HO 5 ewesta 9JM;r~ - ~etiag NP : 5 Drte of ~eti~ to agree test : ~hy 1913 Sm3~i~ st~dies to be perfolmzd to epdluate the s~o~in~ pro~e~ae. OBJECTIYE - To determine the ~ti~m p~e~ of dears which cap be ~nked m a I1I~f. - m deterndw the p~ec~sirn Of results. - To determine the water #oteat of II~ - To aeanrre the water omteat d a s~e d ~x~x~x~x~x~x~x~x~x~x~~ sol~nt c~ntainin~ a ~k~wn water fmtent, PROCEDVRE - 3 pm br~ds of Cipk7 to be s~ked, eac~ to be within tim limits of 1,5 - I,O p, - For each cigar t~p~ I 2 and 3 dears per QTT of 56 m to be ~ked. · AU la~c~atraies to use ICeSS - pa~i~t~s. · ppopaml to be Iwd as extracti~ wleent. · The water rmtePt of the distrihrted solvent to be dete~m~oated by cwn rethol, - 6 labaatrcie pafildpted. ORJ~C~NI~ REaUfTS and COHC1USIOHS i. For the ~an~e of dears st3bied sP3tine 3 dears per QIT d 56 ~x~s preferred, as this ~cm~ced an improved Feedsiea of refl~ltr, 2. It was agreed, cm water Qterm~oati~n, that further distrituticp of solvent samples was rurt rbpJired, md that iso~l~aml was the preferred solvent, STATISTICAL REAPPRI~SXt No ~g~ts. ~th hay 1992 P BMCo document for Legal Services : Health Canada 19 May 1999 ~P~sdix : ~I CEdperat~ kt r ~t~ S~bG~o~p - Xeetin~ r : Ye d r4 b p ~ : Cct~he: ~1 SmkiP~ strditr ~ L ~d ~ further c~ultl ~ ~oidcq pP#ma~. OIJ~CTIYE - b waluate ~e di~ d la~at~y ~~ d & 4 0 4 PPOC~DBIB - Im s~s Ltrib~td (I a 1~, P.5 · diatetu, 13(~f ~m~ti~ed), ~L d tt r~~ xlettld, - 3 cigan ~ chnnel. ~ ~n~p~s ~ ~ e J r. - m - pa~eta ~ h ~. - ~ Ib~ 0 recad Ihl~ aoidap ~~, ~a d ~dn~, ~ m, ~~ ~ ~~ resllt, -Ikb3~atories ~i~~~ OR~CIN1L RESULTS Ind CONCIUS~ONS 1. 4 iabrrataies rep~td ~ C ~ ~ari~ticn, ~ ~ labcratais asso~d ~ ~ lafcra~ mditims. 1. BchRen la~at~xia ~f#a~s i ~ re~ults ~ ~ ~, ~ fo dm~ ~ q ~ Isso~i~t~ dB oct~i, ~ Jbculd h fur~ber kt~~. ST1T~STIC1L RE1PPR1ISI1 ~ an~dr~dts, BCWCPLI ~ L ~ ~ ii ~ ~E~I(~Udbilit~repaWI~ilj~~ ~ % P ~ ~ Itt~t~ b separab b ~' ~~t ~ L · IPlfOi~ cla?Pt d ~ arcr~ll wilb~ ~ b h arisirg ~ ~ ~ latontc~y, m~a a - --c~--~- B~Co dacument kr Log~ Senkls : HPdth CtaldP 18 r~y ~II table 3 )i~E~: tablel ccoFeratiw Test tP 1 Cceesta S~C~olp - btiag N" : 1 IBte ed ~ctirq to agree test : September 19~ ~kiDJ st~dies to investigate nio3tine deteminatim in ~oke. OBJECTIVE ~63 yl "p·P'S -Tolwestipate 34b[P (ste~ distillatim ~d spect~try) . - To in~9estipate the use of acetic add in plaa of sulph~ic add. PROCEDURE - Ckn dgars to be wed, · 10033 - pa~im~Bte~s to h used. · At least 10 e~aets, xitb 4 diJtillatimr Fer ext~st ~ each of the folloninF : · TSO - n~etbo~ xitb sulp~sie add · 193 - netbri with acetic add · ~kn ~tbal Kit~ sJlFbndc acid · Ccn oetho~ xith acetic acid - ? labaatd~ies Pdltid~ated ORICTNAl RESULTS and eONeZUSIdgS i, Al! la~atc~ies effept eoe repeated Iws nieotioe oal~s ard irproped Foe~sim w~ usirp acetic add. a. r~rthet in~esti~atirp of the add method is re~ired 3. R ~c~E~arisra skuld h made against Gas Liquid Ibtc~lcgra~l, 4. h the ~antime therefore, tbe C~Psta S~wp will ccatirme Kith the determiPatieo of total alkaloid~ ~sicq the sul~h~cic add ~t~J, t DIS-~403 : Deft ~ntefiatiea~ Standard, plblisbed as I93 3~e~316, ~21h nin~ 1992 p -- BATCo document for Legal Services : Health Canada 19 May 1999 REwrdix: table8 table B ColFerative Tea N" 6 ~esta S~C~o~p - ~leetin~ N" : 6 ~ate of Yleetipg to agree test : Ihy 1915 $nbing st~dies to b~ Ferfc~msl to evaluate f~th~ t~f smo~jp~ Fcceedure. OBJCCTIVE - B evaluate the eftect d la~ea~crp· caditim anl the dsi to y ~iatim of results, PROCEDURE -120 ciga~ di~trib0e~ (1,9 g vei~t, Mtipled), weight and ~in t~istdnce selected. - lCCSS - pa~tus to be used. M latcaatorl to gp~ke ad cigars m ea~ of 2 dsF of coe week, ad this to be repeated the fallorirg week : i,e, 96 cigars in all, - 3 cigars ~ 56 re ~IT to be apked. - tdbyrinth seals ca artifidal ~ to be used I*J~ dental das]. - ~8ta to be entered cc a standard f~ - B labxatepies participated, ORICINAI RESUbTS and CONCbUSIOIS 2 lab~tepies repc~t~d;nableP when sm3~b3 3 dpars per ~T Icallapse) . 2. For the tim~ beirg the F~aocemae to ~Pke 3 cigars per rsET w~ld taaain, but kme~tipated further, i. Yitbin lab~at4li~ : .'IP~ : d l~platc~ies re~ted no significant diffe~oc~s, 2 la~pratories did, · Yater : 3 l~atoria reFeted si~ifi~nt differ`e~e~ bet~ days, partly cwiD~ to lab3~atcPy ccaditimr. · Nicotine : 1 lateaatcpies re~ted si~ificant differeaws ~twe~ days. I. hhreeP latcwties : · TR] : pePerally close a;lre~I~nt, apart frcm 1 labaatary. · Yatp : variable results. · RWII~ : ~eDlf~ly gocd a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~a~t, apart fm 1 labxatety, · ~mtine : t~ data fell into 2 gmups, 5. Data for nicotine in ~ke of dgk'S is n~tbol specific, and a ace reliable metbrd a re~ire~. 6. h the meantime the fI~3 · rmt~Dd feo total alkaloidr is to be retained. ?. lilternative Pethcbs fca aicPtipe in ~gar smoke to be ep~~ted. STATISTICAL REIPPRRISAL I~e to tbe absence of raw data detailed cmrlusi~eP~ carmot b anti~m~d, ~er Imal~sis of the dat~ to hard eeafirn~d what night be ~a~e~Eted, namely that there is ~aria~hlity ht~ labwatraier, and between days Kitrin la~a~e~ies, 12tb hay 1992 W .N 1~0 jO\ BATCo document for legal Services : Health Canada 19 May 1999 table 9 Aprpodix : table 9 Cwperatiw Test NO Ccn~Jtd ~i~cup - Ltin~ r : 9 Date of ~eti~ to agree test : A~ 1916 sm3~d~ st~die~ to investigate glass files filters. OBJECrIYE - To impstigdte whetber tbe cpllsse ed tbe filter is a ~artica ed tbe filter n~terial, of the s~bg para~etetj, or eb t~b cigars tbe~sel~es, PROCEDURI - a s~es ed different sized dgars (untiFped) bistribrted, - C~ss fibre filter ~aterial from 2 saoeer to be sent to la~atceies : · rpe sbnriPg mllapse in~epio~ tes's, · cne not s~awipg collapse in PFepiw tests. - IOSE - pat~m~ters to ~ used. -~~atc~ies ~ar~icipatel, ORIGINIII RESULTS and COACIUSIONS i. Glass fi~w filter material van variable. 2. vaanm oarhires rae nrae sensitive to ~atge in filter ~ss~bly ~es~ae dmp, reP3tiaJ in ~mJes in g~f ~al~e. 3. It xar agreed that the lorerta - rea~Ppded ~Pg F~aoee~w fc~ testing dgars s~uld be to Iln F~tcn s~r~irg ma~biPes, sm3kin~ up to three cigars per 56 nm ~IT, or up to a mzdt~ 1RI of 200 rig per Wter~ Uth ~ky 1992 O BATCo document for Legal Services : Health Canada 19 May 1999 table 10 Anoedix: table 10 CcaperatiPr yest k 10 Ccwsta SuMltolp - htiPj b : I~te of ~etinl to agree test : Ibsrh 19n $Itiag s~dies to investigate f~ther glass fi~e filter material. OBJLCTrVE - TD investigate the swrce ed d the reasras feR variability in glass fihre filter s~ply, PROC EDUBE - IPdivim~al Iz~atraie~ to ~p~e glass fi~oe filter mateial frco different scn~rce~, ORIGINAL RESULTS and CONCLVS~ONS 1, Rpbable saaee of all glass fitoe filter mate~s xas the B3llit~iw3h C Vese Ccmpany in the USA, 2~ Several kb~atoris reFc~d filter ~terials that wre ~modtabk fw srmldrq 3 dgars Fer filter aecarding to the I[SSS - p~i~P~ters. 3. A firrtber, mpe detailed, st~dy xas repJked, ~tn $ 1999 Lcl BATCo document for legal Services : Health Canada 19 May 1999 ApDerdix : tab~e_U table It Cc~e~dtiq Test ~ U Ccwsta Su~Qwp · )(eetieg NO : Date of ~eet~ to a@e test : April 19f8 Smidm ShdieE to ~stber in~pestigdte glass fibre filter ~teria. OB:iCTIVE - Tb determine the pesape drcp of the Glass fitae filter trap after ~n~ 1 rmd a cigars per ~. PROCED~RE · 1l0 s~ple cigars distributed (132 re l~ng~, 12 ~ diameter, 3.2 O wei~t, ~ti~ced), wei~t ard ~aw resistance selected. - Cats fitae filter n~ter~al ~ i~U121 I~plied) imd w filter ~te~ial to bused, - Ca~ lab to snJle · 24 cigars at i dgar per QIT, - 2P ci~s at 2 cigars per Gm, · ~ Idb to smlke op a filters : - sc 4on - aa~ filter - recss- parameters to be wed. - Results to indrde tlpe of M1 Class filPe Filter ~hterial (QMI d p·e~sure ~op of the QIT after aetiPg. - U la~ratories participated, ORlGINAI RESULTS and eONC1USIOHS 1, H, cases of bloc~age ~ mllaF~ of the Qi~ ww rap~ated in this st~, 2. lhe increase in IPe~ure d~ cb the QIT did mt exceed WI Pa (2,5 mbat) at a a01 tate ed 17,5 nUsec whPn mSBPed after ~j~D~~ ~x this ~s cmbjlmed as ~ipg a re~ke~t. 3, It was apridered that the mrdnam increase in ~a~s~pe drog of the QIT after w3king is ~mlikelg· to b;~ce~ded if the total cmdepsate frc~ the smolda3 of cigars is less than abxB 103 q fw the 44 ~ dia~ater QIT or W ng tar the 56 ~ ~T, ii. Impmped ~aedsi~ of s~nkiog rcn~lt~ was obtaiaed fw ~ more than me cigar per QR. 5, ~ ~alr~s feP B Were ~para~ently mt affected by n~t~ of dQ~q s~ked, w by the type of a. STATISTICAL RE1PPR1ISlb Imdsam~lbls~t~ll ~h);(i;lli *rllJI;~· Iztb bay i~a w BATCo document for Legal Services : Health Canada 19 May 1999 ~p~di~ 1~P Cmp~tilz Yr P Clresta 4~W~p ·~lecti~qr : P nb d ~~ ~ p ~t : s~ldao ~t~i~s b Isscss ~ ~t ~ b rc~i~tapa ~eaicn ~ dan, alJref~9g b ~ ~ t~~ d miFbt ~ ~ ndstaa decticP a ~ ~ ~wtiae ~~. IIQCIDI11 · I stple dp (~ · lealth, ~5 · diwtr, i O ~t, Imtippdl, ~i~t ~ b ~~ ~~, J Y ~~ ~I dp Dd ~ 1~, n~ · ~imtcr, i ~1Ri~t, ~fiPFed) ~ btritut~d, · Yeight seleetica : ~ ~~ 5 SI a resistlrm decti~ : ~ ~- D · ~ labaa~ ~ ~ t Irt ~ dgars ~ & ~~ dp, ~ ~ dgus ~ blmxlcad dp. - ~ · paran~ta 0 L ~, - t la~aatds ~dpltcd. al~F~N1~ RISUltS Ind CDNCLUS~OIS ~ L ~ecticn ~ dp in~pdoes & ~~ ~ a~ 1~~ da~ity d ~ ~a~ine s~dng d~~ r L ~ mlues ~ ~ ~ ~~ a d apgarently ~~a ~ ~ selecticp. STXTIST~CII REI~PRIISI~ c~ntlusicn ~ 1 ~ ipmma 1 rt~pet ~ L bet~ labnto~y ~hbility, ~e : ~ jectico pr~n~ss i~ ~ b plspd ~ ~ rmclu~a~ ~ ~.) B6mter i~ Ihbmato~ d~~ ~ ~ ~d $ ~ redured b ~ ddi~ I~ress, ~ Lh NPP3ft ~ ~~ij~ B~Co document for ~g~ Sen~es : He~th ClnldP ~9 Yq 1999 ~oe~x : table U tableU Ca~peratiPe Tot r U CcPesRS~C~arp-~leetingWO: 13 1Bte of ktip~ to agree test : ~ 1980 S~DkiD~ sbdies to repeat e~eperaeiPe stedy nO 12. OBJECtlVE - To sh~ the effect e~ xei~t and draw resistance seleeticP Ea to d nicotine yield. PROCEDURE - M) sample d~ (TM a leaJt~, 12.5 ~ dian~ter, 3 g wei~t, ~ntipped), weight aDS cfrslr resistance selected, and 60 ~selected sa~e dgars (105 ~ leapb, 12.5 m biameter, 3 g weight, rmti~ped) woe dis~ihit~d. - Xeight se3ectieo : lmae,l- 5 t ~aw rrJistance selectiep : aean +I- - E~b lataratco~ to ~e UI eipdts frem th selected dgdrs, ad I dgars ~ the ~nselected cigars, - Detailed I~dolm~ns ~d.na~im plans Inre s~itted to partidpants. IL~SS · puameters to k Ilsed. -12 latc~st~ils participated. ORIGINAL RESULTS and CONCLUSIONS i. Nan ~alues for tar ad PimtiPe ~e not affected $ dOilr selectim. 2. he within labaatrPy Dliabilit~ is i~pmoed by selertim. 3~ his i~opeea~t ~ to seleetim deer P3t e~ain EP rem~ce the difference ia renilts hh~een la~orat~ia, STAT~STleA~ )~1PPRA~SII b ~ents, 12tb ~j 19(n 6\ - BATCo document for Legal Services : Health Canada 19 May 1999 table Id Egperdix : table id CcaF~ative Test HO II Cotesta S~C~cup - ~etiog NO : 11 Date of heating to agree test : April 1981 Xeasurement d nicotine in reference and c~adensaee s~utic~. cBJECTIYE - To ~E~ine the betweep la~at~y differences in nicotine detumdPatim, PROCEDURE - Reference and Eeod~ePsate wlutim to be dltriEuted by 1 lahPatcPp. Nio3tine in the sblutiens to te measured by 193 - atd own ~tbsl, - 12 la~aatwies participated, DRICIIIAL RESU1TS and CONC·IUSIONS i, AU atbcds gave reasmable agr~P~nt fep the refeten# solution, 2. Ta, few lab~ratcaia used natb3ls other than ISO far valid mrclusicns to be drawn, 3. )leaeae~Pt of nio3tine in smJre solutim ming the lSO ~th~ gave ~ to ~me~hiwd variable results. 12tb ~ 1992 p BATCo document for legal Services : Health Canada 19 May 1999 :table Appendix : table.l5 tcoperatio~ Test NO 15 CcP·esta SU~C~ - htinj NO : 18 ~ate of ~etirg to i~ee test : April 1955 Sm3ldPp ~w~ to investigate elmed n~tbp~ dgan, OBJECTIYE - To assess the effect of different ~tt~ds of oPenin3 closed n~utb~d cigars m draw tPSLtaPc~ Md sm3te ~ie~ds, PROC EDURE -70 s~eple dgars (U2 an length, 1'1.5 rc di~naeter, 5,19 wei~t, closed autheai), were bistrihrted, - A cigar cutter capable cE p~cin~ a 9~ - olt and stt~~t olt a distribaed. - a~9~ to h cmditimed fcp at least Idays (60 it R.B. aal 22 OC) - bcimne aPt~Pt to be determined before sDs~, - 20 dgars to t~ cut by each oetboi plus w method of altting, - Cig~s to be wei~ed befwe d after attti~, - IcCSS - pa~i~eters to be ~re~. - 2 dpars per 55 am diameter QIT to be apked, - 5 ~aPPets to he ~3~Rd on each of 2 days. ~a Petbais fa nicotine ad water to h used. -10 lab3~ator~es participated, ORIGINAL RESULTS and CONtLUSIONS i, wean Palue~ f(o Strai~e Olt were ~fodstently ~nRf ·tb~ for V - cut, 2. V - cut ard s~ai~t att s~t~wed similar levels of pmiabllity ~ath ~etweeo ard within laboratories, 3. Fut~ae tests wwld be perfotaed using strai~t cut, since this a nore o~pa~ with factory olt d9i~s· Prpdse details d the ~Jt depth to ~ determdpld. STATISTICIL R~APPRAISal No a~P~eP~f. 12th $ 1952 BATCo document for Legal Services : Health Canada 19 May 1999 Ippendix : table 16 Cc~e~iltim, Test F 16 ecPePta Su~a~p · Lti31IP : 29 Fate ed ketia~ to agree test : ~4~ril ls'l ~P~bO shdies to fi~ther iwestigate s~kiP~ pmce~aes~ OBJECTIVE - To e~aluate the s3lrces of Paria~dlity of cip~ ~king re~lt~. PIOCBDURE - 60 ss~e d~s 1165 nm lergtb, 12.5 w ditaetet, 3 g weight, unSgli, selected co wei~t (Ee~ +1- 5 U, dm redrttarp (wan ~I· lo S1 anl sutherd di~m~~er. · Cigars to h aaiiticaed to ISO - standards for 'I dsjs, - I~dst~Pe cdntenr to be determ~Iu~ befLw ~kiag. - t~n DP~al s~o~in~ P~hiw to be used. - 24 x 2 ci~sl~, O B6 ~ ca~T. · ~Didas o7er 2-3 days. - IC3S - I~te~ to be use~. -12 lat~nt~aies Famdpated· ORf eIHAt RESULTS and CONCfUS~ONS 1, between and within la~co-atcay differences were Do ~ettet in this st~y tb~ in prwi~s st~di~, ~iD1 similar dp~s~ 2. Cven selectim of dgars · by 1Ri~t, draw redstara and n~xrt~nd dia~eter s~t~g - did D31 ppoPide a Imristeat set cd ~gar ~ke yields. 3, la~caatcry n~t&dolegy bad wt been shfdardised and this might have e~lained ~any of the differences repxted, STATISTICI~ RI)APPRAISAL Clacl~ic~ nP 2 is Mf~nded, since ~mseiected cigars were wt awlysed in this test, rzth $ 1992 4 --- BATCo document for legal Services : Health Canada 19 May 1999 ~ppe~ix: table 11 table 11 Cmperatipe Test H" 1'1 Ccaesta O~Csc~ - htit~ HO : 21 Lte of ~eetiot to agree test : Fay 1988 ~dD~ shdigs la ~all ci~s, OBJECtTVE To assess within ard between la~Pat~y variability wbsl loi~ 2 pltf duratims and 2 Fuff intwals. PROCEOURE - 250 sample cigars (18 a length, 8.4 ~ diameter, r,l gwei~t, Imtipped), xei~t se~t~ (aan ~I- 40 ngJ, were distrib~tal. -Cigars to~eaditimed, - ~a~e~toru cmliti[ns to be Foo~ded. - C~n rrom~ ~D~jD~ P~c~ine to k used, - a x 3 cigars per channel a~ a 56 mp QFI to be ~oked for ea~ of the 4 pa~eter ~patieps, to be re~ated on 2 different Q~s, - 10 lab~itories Participated. ORIC~H)IL BESUlfS and CDFCLUSIO1S i, Puff droatim was cbww~ to ha~e ~ y~tical effect on plff mm~e~ cc ~ke yield at ccestant pdf i9te~al, 2, Puff ~betr and s~ke yields were lower for the 60 see iPte~al than for the 40 sec interval. 3. Hew of the oigars exti~g~dsbs~ for the a and 60 see ipte~als. 4. ~a~ labaatcay ms able to obtain c~erpnt sPo~dng re~lts. S. Between lat~atcay oariabilit~ Kuld be j~p~soved by a better stao~arbisaticn cb the apparat~ and rethcds, ind by a better respect ed the twt a~aspb~e. STATISTICAL REAPPR1TSAI ~mclusico PO 4 waJd be ~tt~ writtep as fPUrms ; tbe within la~eatc~y parid~ity in this test a Icmr tb~m F~a~ic~sly ford. 12th $ 1992 Irl BATCo document for Legal Services : Health Canada 19 May 1999 k. a~cA a~48 cF ~ ~~(~ g~l~ ~C ~,~P. rF P~6L)hl 6c(ilR ~~H ~ ~XPI~ pl~ G qC ~ ~i D 3 - ~76~ Vi - 4~0)g ~ /gi Fir~i (O ~c~--~-~-, ~L~3 ~~ HRelrY. C1 B~Co document for Legal Seniees : Health Canada 19 May 1999 k17 ~ ~c~b~ I - - ]~ "5" ,,, Q ~ k 5 t q, ~,,i ~II· ~ L1 Uj~ct~ ~ I ~~ ~ i; J 6'~ L ~,i~ ~ 5 L4 k ~~~~P*w L L~ ~ 1;~ L~ L~L ~ Z ~ II~ Stu I~-· CI~~ ;L tt~t~ e ~ ·i- t ~, t rr ~, t I rL t a ~,,, 1 ,a~ ~(1L ~· ~, t J,~ 6 e~e~t, ~~. LI,~ r ~~ 3~ ~ ~ Auy-CIC ~ BATCo document for legal Services : Health Canada 19 May 1999 Q " i'- aiul eRc~~ ,~ , auid~j L~a~a~2L, I ~1, Lk·dJ L~i~ *Y·L W~1_ ,,p~~ ~w- -~ ~-~" "7" ~-/ k "r! "~"1""' ~tLeU/ ~e~~-t~. ~ P~i h~: O" (I-- j J" "~"2 ,,,~~ a O IrJ 0\ -- ------- --- --- BATCo document for legal Services : Health Canada 19 May 1999 G4PZ~P~L~ ~UMI·1ARY OF ;~STS 1 to 17. ihis Appendix contains statements of the Objectives and procedures for each of the 17 tests, along with a summary of 'nla mair~ results ant the conclusions drawn. in most cases these results and conclusion have been augmented jy a statistical re-appraisal of the data. An outline of the 325iS of these sta~isiical analyses is given below, with a~~: ~riei expianztion of some of the terminology used. ~3utline of ~t~tis;ical ~ethodolasv ;he statistical analyses were carried out with a view to achieving one or more of the following oDjectives: iai ioml~;ison ~i mehn yields ie,g, "s~raighl" versus "Vee" style of end cutj removing as far as possible the effects of other factors. tji Comparison of variances (e.g, variance of the data from selected versus unselected cigars). icl Estimation oi individual Components of variance, a~peci~!·y nctueen and within laboratories, witn~a view to providing some information on tne reproducibilty and repeatability of the method, in general the statistical technique used was tnat known as ;rl; Analysis of Variance. A mathematical model (~nderlying tne Analysis of Variance) was formulated to represent the ia~tors being studied in each of the tests. ihe data were then used to estimate The means and variances associated with the factors cnaracterised oy the model. i~e following is an explanation of some of the terminology uSe3; (i) Sts~iscical significance In a statistical context, an observed difference [e.g. between two mdh~lS) is said to be significant if it is greater than can reasonably be attributed to chance variations, whether it is of practical consequence or not. For this reason, a test of significance is not the only point to 38 CD~SiO~TeO when drawing conclusions from ~kr results of an experiment. Experiments are usually dCSi~ileO 110t only to establish whether or not a factor' haj a 'S19niiiChntu ~~~bCt but also to establish w~le~iler ~I lot tills effect has al)Y practical Cons~qu~iii~~, BATCo document for Legal Services : Health Canada 19 May 1999 iiii Variance ano Standard Deviation x Standard Deviation (ofte~ aboreviated to SO) is the sauare root of its associated variance, and is a measure of the "~pte~d" ((jr variability) of data, If the mean and SG of t~lb measurements of a parameter are known, then ~ppr6~imalely 95X of all these measurements will be in the range iFroan - 2 SO's) to (Mean ~ 2 SD'sl, (iii) #ithin laboratory ano Between laboratory variance It is a well-known fact that when a test is repeated time after time in any one laboratory the results of these repehleo tests are rarely identical, The Within laboratory variance is a measure of the amount of variability olserveo in these repeated measurements, It is also a well-known fact that if identical samples are tested in a number of lasoratories then the results from Ihese Itinoracories are lately identical, The aetween laboratory variance is a measure of the amount of varia~ility seen in the results from different iabora~oies testing identical Samples. !;V] "Per channel" analysis of the data During the course of the 17 tests described in this appendix the number of cigars smoked through a single channel varied from r to 3, Shus the smoke yields reporiec tire effectively the mean of 1 or 2 or 3 cigars, deoenoing on the number of cigars smoked, In general the stacis~ichl analyses oescribed in this appendix have been based upon the observed data, without any adjustment being made for the number of cigars smoked on each c~lannel, Such adjustments assume that all of the rithin-cR~nnal variability is due to the variability of the ClgbTS themselves, A limited analysis (noc included In this document) suggests that this is not thecase, For example, in the measurement of TPM it seems possible that the cigars themselves account for less than 75r of the wit~i~-channei variance, In addition to the statistical re-appraisal carried out for each test, some analyses were done on results pooled from some or ail of the tests, Details of all analyses, including these additional analyses, call be provided and may be included in a later draft of this rda0rt, BATCo document for Legal Services : Health Canada 19 Pay 1999 MP~BL TOBACCO ~I~ C;OAP SMORPtG PBO~URES in ~e course of de preparation of ~e Odt Repor: of de IC~ to te CORE~i ?noie Study es~, aor Fuestions be been raised vhicb have eause;, r concen, mre haw ha investigated as far a possi5le ~~g ~-ho~e ~onation, and ~ pplY~ remain unresolved. Ms paper therefore sets our ~C cmea~, and presents senple Monti?n whici we believe shows ~at ~e cmca are ~C and teed O be addressed, h pa~~r we a e~nvinc~ ~~ t5c report being prepared needs :D address ~ase csncc~, and will not be valid alas they are d~te!p dealt vid. ne jFrten were o:!~~y sdcetcd a being me bat compromise be~~ the ways that ~t ~ diPferent oigar types a o~bd~SIB an 9FnpC was also used :pr ~~tttes, was adopted, and is still used. W" believe however ~a: ~e use of an rj!~~ proc~ may not be as gp:cp:iate for oigars, The ~ce of any suoh proeed~ is that it allows comparisor, bttrta pm~c~, aud thet the consumer ona ~ meaningful ~o!ces. ~ is valid if the way the ca~cr is likely to smoke ~~a~ ~e same :~ni~ng betu~ PrCduCts as ~e oe!ec:d arbitrary ProCednre, vM~ is de case ~os; vi~ol~ aclpt!pl for ~~t:cs, The data appended (I) sugges-,s to us this is not ~e case for ~~, and that de ~ order ~ 8n~c~y ~~d by de sdac:ion of parameters. CPnlf~ed by a sccoad ~~ earlier study, whiob also shove the ~terl~ora:ov differences I~e~d from r~d d~ in I can:~o~d cooperative lady. jt) ~ additional eooesn is that wherever me industry puts ~~ard has to be tl~hlc, In the case of ~~e# this is olear, ~e design features nerd~ to give a lover tar re tc~c~p supportable, end a ansctd ~s ranking. 16 cocpcndve test ha b~ car?:ed our m date bg the lC5S to da~ ~t ~ applies. Efforts have rightly b# devoted to ao~l some of ~ basic pr$~el#, Ecvaer we believe that this has to be c~ec~ed a a r~er of priority, We have analysed in I orude way ~ data from the cooperative tests mdcrfa by tbe IC~. While ~ oorparisons are not I~e~l valid (caqacing ~l!~t ~~c~ at different times, d~l~s0 of lls "d C3~~11, and not CIIC~r I Idccd range appmpr-atcry) mel give " implausible n~t - de ~ ~e~Pe~ed is roughly inversely proportional to oiger weight! (IWe~diu 3) Ye helieve ht without adecuete d~tm~ad~n, Jld a fin2iag is so $l~~le ~~ it vonid ~~e d" c~b~~ of de organisl~~ ~ugges~ it. hle points ~e 6 ledw eman, adw need be to be dgdl dealt v~~ in ~t IC~ raper:. BATCo document for Legal Services : Health Canada 19 ray 1999 bppa~ Z c~~ ~~ p~ (1983) NFDP~ RPJ ·~ End, Clss, wt ciat:e aaak Iaass a~ P~nten Paraneters ~, Irp, 4,0g 113 1 a, I, ` 3·5 11! 2 gj 5 C, S, 2.0 10; 3 n, i, 4,1 to; 4 54 i, S, 1,4 88 5 5i 3 S, 1,8 81 6 61 1 a, s, 1.4 86 7 60 z a, s, 1,8 84 8 52 6 I, 1, 3,7 64 9 25 14 J, s, 1,8 64 10 42 g K, M, 1.0 53 11 49 s, 1.? 12 37 12 M, M, 0,9 56 13 4j 8 N, S, I,? gj Is 34 1S a, M, 0,4 50 15 41 11 P~ M, 0,9 4; 16 42 4 ~Sa~a~ cigsrs n~guishd duriag $~I eptt 331 cKeept W, ~t un 20mn BATCo document for Legal Services : Health Canada 19 May 1999 r 1··*~·*lrlll rl uiEri~ ILPlillnslnl+··Ol Ii~cP~rl·IIDA·**n* ~l*~r I JiIl a X ~l~n·16··1*1 I ,rr IL J( Iv II Irl~C··ll~ill,*~r ILCII·lcrllPlo*nr ILI*l ~210~ c *rP,roPccolr D I * ri I PIZlfl··3aic*a~~ni I~ei~·l*laanotnna IIP~e~ ~l~lt ~Iri a ri I It N 1 Y 14 Z **ll··Pllle* IL 4 11 P C I I) E IL Y- II Irl * ~1 L)~ u I) 1, I IZ1LlOol·l*LO.~~a * lbell·I·~0011d**A 011 IP31L *P i~l~l Lu~ I IZlUL JII OL1 · IIe a rO ·I* I X ~·11*~616131 01 p I c r* L - Z I IL I n I r r 1~1 r I *J~ P ~IElllnblt··*rOllF y ILelllb~~PI~O·~on Ou ~10*o~ I0I 1 ~I~I: I I P I a ~~tl a LLO a I 011 O r a I *. i I Y 1·*···~··8~* *11 i r r a a I * nIl C n a IL e o * rl *~ I P vll Ir ~l·s·~rly~rlpo iul O I IZrLIPIIFr6·061*A lIL iar~~*l OIJ IP~I~ ~O r ILIII P C U I~~OC IDI Y *IL e Lp L. so~ar~i**ooa 0)01 )U( r I·A,*,A,~III II I I LS1~ J a N 1I1) I * Or D ie ~ r ~~~P~I~P~rZZ IIIL 1 ILLI a P I * O )~~ I a ·-·.·.~.·.·· Irei D ~OPPYL~rr7rZ Pbil X P Q P d BATCo document for LIgal Isntr : Haalth Canada 1) May 1981 Rppmdix 3 TPn Dlliu~rin ax :3333 Tnt~ [1978-91) Ci~~r Wlipht v TP~ X 5 XY 4 P ~ X 'p 3~ X XX o X X X X I 1 31 (1 18 CO Trm (~yeiglr) O L~J Cr 6 trl ;r, BATCo document for Legal Services : Health Canada 19 May 1999 F Ct~F rF ~ ~fF~ ~F ~ n~ ~dp~ I. P~~blPdJP,~ : Plb ~, ~ ~~ tLcU4 ~ _ ~~~jY~I FS~-~t~4~i~s·X_- I~. i L~~s .3,~Ll~,r-?ICt;l_~_~d~~dg ~i~B~~e~ii'~~~UI~ s ·~II_ __~ ~4~ '~·~-~r~J ~_ P - ip~-~W I_S~IYIL~J3~ · ~I~YrCe~e ~ ~ ~__~i·-"iwlL?,i ~rb~ ~cl~_u~ilPl~y,e~sS t N B~Co document for Legal Seniceo : Hea~h Canada ~9 Yay ~999 Z ~~ Li~t·~# ' ~-·------------- il ~~~ rb~4~ ~.t~~_ CCxr*urx_d t~~~-L··L·-L;i~.~-;l- --~1.-~-1-.-- k ~~c_Clul~2 ------ -'--I -~--~ ~--~---~ - ~-~- ~r.38H ' YJ I*LCc~ 4c9L ~ ~t,M ire hSla~ t; 1 o w a w N BATCo document for legal Services : Health Canlda 19 Yey 1999 4)-b~f~M71QZt~8~.flhk~i~ · j~ r-L~C~- ~~ ~ ~ ,~ea.8__ ___~ _____.____~ _~Y~Y IC~~~~~P~~L~-JIICD_ sLib~~·lt C4( C~-*1~I~-C~I--- - ~.sr_k j~C. L*~J~~;~PLl_l~c~L(_a~_~_ A 14~~ u~t;; Y~ ;~_t,, il~~ n~.ULi~eY-~. r~d t L~hlt_l~d~Ct_~·_l~!y~d ~~ Q~-- BATCo document for Legal Sen[ces : Healh Canada ~9 Yay ~999 'I' ,iil c~.~Jr~r-~ ~i · ~c~;a ~t _hC~_~ Ils b ~ ~XI·i~je gMI~~ 7 L.-. _________ __ ··~·~~~ - -~ ~ r ' ~ ~1------ --11~1 -- :1---· - ------------~-' -·-·------ ---- ____ g j~4~L~ r' ~?~tl·~·~-i~- -c- ~d;y"j~ I~ku/-rY; ;i~hce?4~~. ~C ---~-~ ._a _k~C~rp~ -I-e; ~eA_ O ._lane_.l~ i~i~~'4i-~L--------- : W BAfCo document for Legal Senlees : Hea~h Canada ~9 May 1999 1 n, kS~c t F·"TBb^14;·4~ a, YCS~$,,·tx·P_L~-~tl-··"7 "~~"'~ t I .._. ~~I._..~._.. ~~----------- r ·r- Y" .. ~L~ DC*LeS4~:I;U - ··-rl~-·-·---·f·--· -- -~~I- t ···- I··----· t·· -· 1···I-···-·· ~~- ~I~UQ 4~Xi~l·~6~tw~t c"""~-~ -- - Pm ~2~ i b~ ~Jn ~t~l·! ~pJc3 ;---ili Porer, --------- - ·Ct.~t"t~;w LLI~-~L ~-----------~-- ~eLL·tN ...._...., ~~ . ·--···-··-----·-· -r--~~-- L~:U*_~?aL~'dl~pll ILJ 1~Y·~4~Yt~ : /14c3 _ti~t$4 -··--~-- ----···-··- -· ----~~-~-~' ~L,,tl lit - -·~-'~~"""'" Jyr~ w~R, hen ~t.dt-.uti~-l-- lc~ ---- "- c-~" Ci ihj~i~uR Ic,;rl1I~GC 7u~~1*~M~ Ir-J ~2~g:Yl;il~r~lUL~Llzi~ PW 0\ N P BAfCc document for Legal Services : Health Canada 19 May 1999 I·. '.·~·.' ···-··· ·- · · laq~Y likiJ~d ~ *J 4 r.l. ------.----.-- --.-------------- -- - -- - - ------ ------~~ c- ~-'~'-- ~a~h'C'~'itYI-~b~ ____~ ~ __ _,,~._ _,.---------------·---· I~ ,p;i ~ilxl;tc=, ~bt~~t-~- ___ _~_~___~ _~___ ,. __,~,.--. -.------------------------ cc-~ ~___ _ _~~ ~~ _ ~_~____~___ ~ _ ~ ~ _I-I--·~------ --~------ ~flt~;yP~ _~ ~____~~_ ~~~~_~ ~~_____~_~_,_.,., ..-.--------- ·--- ~~R~-~ _~_ ~ ~~~______~___ ~_~,~~-~---- ~---------·--------- --·--·----- -------c~'--.-~-~-.'" I--~--~- __,,,...c---------- --·--------~c- I ____ --- · ,,dJu~t' : id C~*t~ BATCo d6cument for legal Senices : Hea~h Canada ~9 Ysy ~999 Mw·e I ci,~c~Q ~ceuYt~ --·· ~tf·.~d--r "-~"` ~~--- _-··1~1--111--~-- ·- "--'4-1 I~%].;~ . ~_L~-~_a~L~y~_~;~jolt ,-~.··U·i~,,~Eqo·i--._~cl_~_~t `~'I~·tL-- "F~qL· I i ~ ~t~Ja~l,Le. ~ ~L·~ .UI_UL --;1 I.L·? --- ·----I-·--·--··- · 9~- --·~r-·r·-~l~·- · 1·+- L~L,_ ~ldd / -·I·I ·-- ~~Ln·~l) o~ BATC6 document for Legal Services : Health Canada ~9 May 1999 b-' 4-2· ~ ~ aisP u: · bl ;;jj cl i ~ c L-~ LI L ~k rH LL~~ ~~~ ~ tr, cti It o*C~ ,Y-P r~lk 9YI 9 41 6~,.-~~Lu.~td i i ~- w~r c, 1·cl~l(' i f ~"~ /~"~ ? ~' t L ~jit BATCo document for legal Senices : Health Canada 19 Try 1999 ~199~8 !1(6 r~10 irp a;::p or~zssasa7 1:-I-- 10105~ 0212!:~101;; ( El From 1910 to 19E1 Ind Ilt~ from !911 to 1989, IIvIrhl Il~fP( oooPorAtiVo WlfCII(I Hero fllljl9d for L bitter UPdItlfMdlPg 01 thl rP1IOI9 for Ih( imPortIllt *b~~l·~~ jrrcm I tlFlr Ianpl( oi the 1~ prodl~t~~, m~ lal~c- itIon 5g u~lPt rd drlw ~ol~~tmo~ ip~cvs~·th~ within I~er~torr VUiItion bit ~I PO IPtllOPoI i) the ClfYOQ IIboratarP rulltloll tpd~ It ii vlry dl!8~lt tJ Ilnd 1 O~~lllly III· ~~iliP( IwDIP~rlcn to: rbl bid ~spblukR of a~ia( ~llulls dc:ing I eoaparItlvI oxltelasl the main tlllon I(PI to b0 tbl orIIt EI~Lr VUllblll. tVI nrll thI l:rl~l~ritp of tbO 21661, of the WP1Pnf, O! thl dllmlfbrl Of the drIw flllltlPCI II wI11 II ~~I proolnoa of applrlnt valnl induclno Ilak·Irtioo~ I thI lovsl of the ElgIr holdor durfno ns0han!ell .e 11 ~n lP8~, it the roFuIIt of tbe oomnlttII, ioUt l~ort· tori~l pr~~pntld thI:saults ot I ItudY on blp ~iP~~, it II to aIy iJIIohinO ~FI thin 5~J o, I Owl ~l~arr h~~ in~l~d I ',ondInoy to axtlnelllh · bstwIIn two pr!!! I·' to Itudu i! 1 Podiileatlop of rom~ part~t~ did not Illlfsltl 1PeDm(tflpe~, all the other perI#lItIrI II·IPIP( ~n~hap(~d~ thl PIleeOI at 13~~f fnt~~al from 10 IIa to 30 Iee Itro· ~,~aPll rldaeoI thit ~lnd~ncy of bit oiollrI to t~tlPailh blfuIen the PP!lll~hl PJ~r of POiil II well II Ikl j H~vl~, Iceordino to the lnolDbIrI of tbo oommitteIl 1POb I t~lt Puff l:IqllPel II unroolIItio end d~vl~to~Ik too moon from fhl aa~l c~t~r ImakIr bl~la~, c Cpq(gb(ptlt it I1II'IVldllt th(r Ill 1PPTIII( of (hl puif frIpueney II not nIdslIIlily thI bIIt ~olPtloo to the prPbll of big elCr IRokingl t~thl~?ra b?:iC:; ~s uil~bt, otb~r crlllt~oP~ II tL M1~1, th( di~lt9~, Ib( Ilnpfn Ind pl~h~p tb dIPIity IhoPld be aonlidlrld In rblohPlsl of pu~· for blp oiplt ~~in0~ lUlathIr ~tabla rllltld ~D thb It·dltdl~LtlPI Of bll oiOlr ImoklnO !I thl clIl of ~I~~ with eloIId b~~d~~ ~nd~od, ii the Imokor oIn outlthIm by Y·eul, ~llla· tilla or plltPlal, IZ 1I houtvl: 1IIoPtlIl to rIooavlInd only ooe ~lthad In thI flIld of ItlndlrdlIat P ,K"te~~ LL-Y~ 13 ~ ~1 jc~.b BATCo document for Legal Services : Health Canada 19 May 1999 ., ·,··· II · I ··II d tb i, n pi''- ~-3 Tlj ~,~ rYjaW ~ YRJlff I ~I~. ..~. k~"; , 1~5~~~LO ~hqhkLLW~~~_~_~ --···.·-·· ----- ··-··-·-·--·---·- r -·--; -· iy.n~L-~J1 :P~i~fae~ c .I..~- · ··..-.p, K~*rl~Z·jl~C~' 3C~ I1C-,11,1 9 ~~ 3n J~: f;_cb~tfk .~.. ~~~... I~ _··__-ll·--C- ·---·~-·· t ~i4 1*1 , ·I C, L ~~4''··r~'~'t·-a~·-i~~ ·Rj·'L- .--.. ~..-.~.-~I~-~--rc-' __ ,,;le~~t,~-,i-~..~ _..____.~__~_-- _____~J, ·i. - -L~d-t~ it ,,~ ilt~t j~lu,t,.~~c~ i~c~ i,LIL-"··.Llr-~17. ·.·. -.-----~S~"-;e L, ~ulp~ II, E~~di·~..l·····---l~·i-e~$- ~~~_~~~?_~3S_UY(~~IL* i~j~_~. ~~~-'---~--'~~~~~~~~~ o~,· o" BATCo document for Legal Services : Health Canada 19 May 1999 II~ · ·1· - -· ····--· S,~iui~ ~pB~ Uwy ·-:~d~kiZ-Oau~ ·In~w~cha, rL~. ji gI4~zctj: s~d pl· ii 7 /"~ it UuvljlJ7t -~j. "-~-~--~~' ~I .~ . .-C.------. X~-C ~t I - I~- -I-·~-CC I~.--I-- ~' -- '-1-·11~-----" .-.--~-·I-1` -~.~ --.c~-··r~- --- ,-- ~-·- ·--· ) i~. C~-~._......~...._ ... __~.___~.___.._.._.. ~--- ~"~-3~e _. .__ .I____.._ ,_.,,,..,_,~-.. ·---~--- 1. -I- 3, ---~- ~' 11-1b ___-_, ... -- --i~-~·t-";-~·Y;i·~3-1-~-;4~~·- I- -~ -. osil. · ~i~L~l~birc~~ ~e~?~Li_ ~~tY~ ~ ~G·C~A n*A L, itt~tt~tt~tt~~e~2~-~iP-4~01·~-- FL~L~( ·------- ~ ~~Sir~---- CLT II ~vP ~_~___ b·eg" ~3P ,.I ··-·---·, - --- I~ IUy ----21-*-7--"-y IJ - ---- O ~eiJ~·~-I·- 1~ P 0\ n Ctl W O BA~Co document for legal Services : Health Canada 19 May 1999 ~i~-i·L~~ " ~t ,~~ ~LI~ K b,~c,/~i·~ 4Cii-"-~' t 41 ~~ C~P- r; ;r ly7~ 6rn~,cg ,A BATCo document for Legal Services : Health Canada 19 May 1999 d.S.P tl~ r'DP" I~ W C~ P7a3id6~ lot Draft I 8P ~EBNarrP~ CO~T~EE I~ CI~ ~~ ~r ElfGER Il·~P dP 3 e38g5~ SKQIG OtDDY ~1P Proprlaa raport rl Z~tdootlor~ ln 1970 the elplr. lndultry tn th~ U,l~h~, T~ Linltod Kln0d~ hnd 811~1~ raPo~iPod tho n~~d to etndy PltbOdl to Ipsl~(~ 0111~0t(l In daoaabcr 1970 dllsgat~~ of rhl ct(u Indl~ttf~~ of thpcl eorptr~as mat In PLrfl and oat OP tb8 fPtl~ltlOP11 Corittle for eioa: b~~ Itrdl, Pc ebjlotiva for IeC88 n~ dailaad aoI thl ~t·~t~~tion of ~~inl preoaduraa and ralptiosl ~stbPd~ for tar and nlcotlba dsts~natloa in al~ar ~~s, ~P aXt(Bd itl knowlcd9a in thl ~i~ld of Ei(ar loko ra)pnll, tha PQfttl( H~PI 1P 1971 Lhl cPPporf 91 EO~lfl and In 1972 thb ~~3~ 80~a Sludy Group Ptlatad tbl Cioat Btb·ptaPy' !ar parllatlp~ I ·o~niall Ir~lat Pathed Ier clolr~~ DEal rappttio( to tha ~~9~ 8molla StPdt OrouD ws~ doas dUrlno :hl tearly ~oo~iopl of tha droUD, ii tha roclipg 1)91 Ip J:r~c~t, tho Sloka SIPbY OrDu3 propolod the el~~ I~·~taap ta P2~D~I I proorooa rapQrf on iti ~otiYilli~l 1E~OSn Irll~tlP 10tl·2 DIIa 11 _ -------~---~e BATCo document for Legal Services : Health Canada 19 May 1999 d^ih/itrW;ryC~ f~ GbJj a I1YclR~'~I-lit".~d~,I, ~:"'c;iL '"'"- b~tr·- ,iG, .I~ I ebi;G~-~y &-l,wYj, L. ~idd . ~'~~ ""'-- -~"' "` "c~ ""- '-'~~ ..... ~~.......----- I- : ~L, ~,glLli;, ue~_~~·l·n,*~84~L^-~1~;1;~.C~. nll-~J · -...-I~-~-··--··· ·-· · ··-··--·-· -· ··-··------ -~--·----------------- ---· dblu_1_ · ' -~111--·-·-1---1-1- ·11·- · 1··I*I ·-··- ·1···---··-1-1·11-··-I-~-I· .~ii~*~E4*~-L /V9~,~1u~"~3tk~L·~e~e~_ · ""` ' ""'"~'~'''' ~'-'~"''~' ~"~'-~"'~ ''~-~~I--r-~~' '' ~~~., i~sn; / I P(I~IL~LV~ g,lacr; -· ~~ ~;t"~7.-·-.--C~~~(,1;~e~·M~~ ~fLb/ ~b ?~ Efl~BTIOB BATCo document for Legal Services : Health Canada 19 May 1999 ,~ i L~ -1P I~L L ~~L ,~k 41~' II~ - -- ---- ---~I·~ ·-·-- ---- ~------------------ · .k ;A -i~rl~~ ~~b~~~tt ~14=·~?51~1rltl-~;-j~ot~Lc~ __ _~·~.~J,~lu~iL~~'~~4i$c~lt;_ -~Ln: t- -I~-- I I c~ A Y~~r~A~C~l_~~illi~~ t ~.U C~~~ILI I~ ~~LXa~ -~~-c---·---·------· C ~4,--"- --- BATCo document for legal Senices : Hee~h Canadl ~9 Y1J 1999 ~-c·- · -· -. -· n · -~ ·-1 -· ·----·~'C Ik I i~lJ a~~ ~ ii~cd ~18 & i· I,.~. Adl;k u ~" ~ -~------·--------~~ -'-- -.Dr~.BI~~ d~ _ __ ____~ _~_______ _~ _ _ _ _ _____ b~a~~. _L RIF L,~, ~1~4 k~l?r\ ,I 1 ~_ ~--~~--~-- -- -"1 L ~ct~~~-~7~·~~c~i_~cL~_ O P B~Co document for legal Senlces : Hea~h Canada ~9 Yay ~999 13 ~, ,s, I~ B ·' ~~ ~ ~ B~i~i~t __ li~c~)~l~, L~G~IC S~QI ____~___,_. ~" -· ~--·------------- ·- ------~------~---- i,cu~, _____~ ~tW~ L.i_._. ~~~~I-~-~---~- ,c~ ,SL -~-.- ·---·-------~ --~-----~~--~----- ---~~ '-~'--~- P ik___ -I---~-~-c- ~ ~--~ -~ ~t~ ~~-~-- lr~ _I-·----LIIC-----~ ·_ _ ~ _ __ ___ _~ ~--1~ ---L-~----~ ~1··1·-·---·~·-·-~^ I-----I ---L--III~-LI _ ___ ______ -- ·~II~----LCI- _ -Ir-·C- I-C~I-~--~ -----------~-~--- _··__·-·· -_~~l-~·-CI-- ~llr---- ---- I __-· ·1--·-··r--I-~--- ---- I--C1I ·_I~ ·--r ----r·--·-- t --~ -~--~·--I1CI~-·l~·-~IIICIII~C- -~·_~_·-II --~--- -II· I __I .^-· ~-·I··~ --~·-~I· I·~I----·- ~ I~ _· _____t~ -rlC ·II· · - --r-- ·----·-C ~-- ~-C~- - -----~ ---~ 1-·11I1-----·----· -- ---- -·· ---~rC--·~-·~--- -LI- -~'-I' -~~--~--~'-~ 'a o -- w ---------·--·--·- · ---------------·- a __ w w BMCo dacumelt for Lsg~ ten~ls : HollB Clnldl II Yly ~999 ICCSS Tests 1 · 1'1 : S~arl of Statistical Findings Summary 'ihis document notes the main findings from statistical re·ana!yses of the ICCSS's past inter-laboratory studies (Tests 1 - 17). These re-analyses were requested at the last meeting of the Sub-Group (16117 March, leuven) as part or its requirements in producing a progress report of the work done towards the development of a standardised method for the machine smoking of cigs , In summary, it seems clear from the statistical re·analyses that we have reached a stage when many aspects of the methodology need not be studied further, However, there remain a lew areas in which further work seems essential before it would be feasible to decide upon the basis of 8 meaningful standardised method, In particular: the maximum leeding per pad needs to be established in order to cover very high yielding cigars; some thought, perhaps requiring minor experimentation, needs to be given to the site (or siresj of end·cut; and it seems crucial to carry out studies into the possible interactive effects of puffing parameters and butt lengths across a wide range of cigar types in order that a relevant and reproducible rank order can be achieved, In addition to these thoughts on future work, please note that numerous comments in nDraft No. I" of the progress report are either ststisticaily unsupported by our re-analyses or irreleva~tl~isleading; see summary details appended, Aspects of standardisation of the smoking machine itself have not been considered, ttl." j/L" B4 BATCo document for Legal Services : Health Canada 19 May 1999 -2- Introduction R.is document attempts to co-ordinate I interpret the main findings from our recent statistical re-analyses of the ICCSS Tests 1 - 17; individual summaries of the re·analysel tests are appended. Its main purpose is seen as providing a 'check-list' in order to avoid stating things in the proposed "ICCSS progress report" which are technically not supportable, Where appropriate, this is supplemented by reference to tests ?, 9, 10 & lil which, in the absence of accessible data, we were unable to include in our re·anelyses; any such comments, therefore, have been taken from earlier appraisals as noted in nDreft No, 1" of the progress report. Oomnents are also included where further study and clarification seems necessary before it might be possible to progress towards a meaningful standardised method for the machine smoiting of cigars, A.~ Cambridge Filter - Material. Type 6 Loading 1 Tests 9, 10 and 11 were intended to study certain perfarinar,ee, characteristics of glass-fibre ii!ar ~9Pf ~` TU~~h~ ' From previous appraisals of Tests 9 1 10, it was commented thatrC (I gl~s-fibre materials are variable; vacuum smokings are sensitive to cha?ge in filter assembly pressure drop: maximum leeding should be 3 cigars 2` or TP~ of 200 ng per filter; certain filter materials are nat suitable for as many as 3 eigarslpad; and further more detailed study is needed~ For TeS: 11 it was possible to perform 8 statistical re·anelysis and this indicated similar mean smoking yields and associated variability for a Ij n~mber of different filter materials (B 'cont:ol' material was ccwp~are~_~: witi~ each of g laboretor~s~ Ipwn' filters). No problems ofover·loading were observed vit~ t and~2 cigars/pad, as indica~d·by simi,l~mea? ~g~~~i"t~~x~~~ ~iJ i~a~d ""_d~OU~~B~ ~t,te I; ds was io~S~~"";-t~7.t~L~l~o~ Slu hehe~e~H~S spetlricaly on the effects of different numbers of cigars per pad, Test 5 vas particularly useful since it included i, 2 and 3 cigars per pad. albeit for just one cigar·type with limited TPM tie, about 150 mg total, from 3 cigars). These results also suggested no problems of over-loading, with a clear indication that in terms of repeatability tie. vithin·laboratory variability during short period of testing) it would be beneficial to test a number of cigars per paE~ C~BrkiPikec T~c~~~ *3~ Itr~ (56 mm) and loading would not seemf$pose ton ~ueh cbss, r, ~Y~m:"tF~ further study seem to be on the verii highest cigars bef a final on how many can be per pad. ,~~g Sarlnetm Various combinations of puff volume, puff duration and interval between I puffs uere considered in Tests i, 2, 3,qand 4~ O Not surprisingly, the different combinations of parameters led to W differences in mean yields end, to some extent, affected the variability in ~P the results. Certain combinations also resulted in the cigars not staying O~ alight. , W - BATCo document for Legal Services : Health Canada 19 May 1999 ·3· The IOlt rtCtPt~nbF~ I cabination seas to be a 20 11 pull voluae, a 1.5 sec, ddtion ad sec iatcrvg~~ia is so~std bp co·ePrt~ i·d · if. I~~d ~ ' L.l~d. ---X 3.r ~ don't 2 to the-";8X tsret~nicotineFeP ------- ~ UPP understand this its t! mis la~r point prolgts us to c~phsiJe th~a~ws should be clear on t~ f ly finaWlOiCB of paracetefsPrrd mi reasons given p~ort~ble , to this, It essential that ~e effects B,~f difl parraters are a to establish a ed robust across all C. Batt teoOth In Test if the butt length to which the cigars were sloked vs 30 11. We prcsut that this arose fron the present lOOSS definition of one-third tora cigar length up to a I·fla of 33 r; velre not sore that this was the basis of the 30 11 because us didn't know the lengthof the cigar-type under study, In.s, mis ststelent of the ICCSS definition nay be incorrect ed needs to he cla~licd)~ Regarding the lCCSS definition of butt length, there we no basis within the re-analysed tests for assessing any possible shore-co~~g~ exanple, it seens reasonable to suppose that this particpisr definition :0P~d lis-FlpCCIPnt certain eiglr-~nes. Poue~er, to our Lvlsdbe, this h* Pat been investigated vithin ICCSS. In view of this re think there is a crucial need to carry oat detailed studies of butt lengths versus the full range of cigar-types, perhaps incorporating different colbinations of puffing paraeters. It seens essential to do this before laking any fin tacolmeadatiom towards a would·be standardised saoking Pethod, D, ~d·Clt lest 15 considered the effects of a 'v-cat' to the bltt ad of the citer versus a 'straight-cut', Zhe results using a 1BtldBt.Clt' were lower in leg yields, bat siailar to the 'o·at' in variability. ~ the basis of these results it could be argued that the 'v-cut' would be preferable on the basis of better differentiation between sp~a yields (i.e. the same variability as the 'straight-cut: but supetilpased upon LY~arlca~la4. Hovwer, ~s~E~ups the 'atreight-cut' eab justified on the basis that it is the lore usual end-at, wat~ki-be~s~f to Jtrdardiac ed lod6pt~ida,a practical convenience in a rou laboratory testing situation. ~cn so, I decision is still needed a the shdatdisation of sire of e~d·cut. Clearly this is not as obvieus es it first sems becsuse of the gross differences between cilr·tbe~ with aspect to width end end-shape, E, Sortiagllon-sotting of Cigars brisiog fron concerns about high variability between individual cigars (of ons type), 'fests 12 and 13 were designed to study the possible value of veight ed pressure drop sorting the cigars for snoking, P BATCo document for legal Services : Heath Canada 19 May 1999 -4· n The cbaclPsfea tmi de n·ePpes lsPnat softbf is beneficial in thet the repeatebili 9~ appreciably crdtdbA ~at is, the repeatability lot 1 'sorted' cigar pr pad roughly equates to that for Z 'rartad' cigars, Igdn, the ~ncr~itJ of this result for all cigar-types is mbcP (I·t one, 3 gm cigar-type was studied), but statistical experience in Iry areas of labprae~qlprodaet testing suggests that the sae principles would apply throughout, perhaps ironically, a lore l~sly objtetim to the routine snokial of pn·sortd cigars could be Iwl external sources on grounds 6f suspicion r about biases in aaplin(. in addition, if it ran dcidd to dot a sorting procedure it would be necessary to specify a standardised nethod of cigar selection which would have to be appropriate for all types of cigar, P. Statist:ica]. VariabilitJ Idft of the tests peovldedsom iPloaatioP On ~~~ibllitr of Ilb~e j yf elds; see appendixlfor :ests, n~tt~7 individual ~yf~/~~~J~d~I;~ r~ULi~e~w(cr~ Loting scmsp the separate tests, two results seel 91 particular interest, ;irstlg, there was no clear indication that the variability (neither 3 within or between laboratories) was correlated with wean yield, ~e~ca~4 ~ote, correlation may have been concealed by changes in spokinl lathed details that occurred across taste, Secondly, we note that overall average SD's for hi, PYF ad Sloke Aicot~e re: TpM pMWNF Sl,fic Mean ~ ~. 50 50 3 SD between labon~~rics 4,1 0,39 SD between ports (vithin laboratories) Z~ 5.1 3,2 0,31 Average ~Pber of cigars par port was approxilately 2)~ Far EMWIP; for example, these SD' a wauld give approxiDate 951 confidence !.imits on the pe· yield for one cigar-type tested by pi ma laboratory ~ca~ ordt: l101f U.c. ~2~.6' ~flP)4' where n is the number o~lPo~as~l saOking overwhich the Ic· ie;d is ~n~Llaopealr it au be polpil~ to achieve a rcduetfpl in these X SE' a (particularly that bt2~ standardisation. 1 I nalber of other aspects of statistical variability should be noted. lhe ~ffectj of prc-aortinl the cigars for IlokiPI was I(PtiOPed earlier Ido~P·E), This is valuable Irol the point of view of eeonolg 3 of testing (i,e, a required npeatability achieved by saklng fewer cigars), However, i: is relatively Pflport·t in t~ context of eltiaatinl an Q 'absolute' lean yield since the associated confidence lilits would be 6 governed largely by other colponents II variance (e·l· between W laboratories), BATCo document for legol Senlcet : Health Canada 19 May 1999 ·5- (ii) Toe repeatability effects of smoking a number of cigars per pad was also noted earlier laPte~), As for (i), testing efficiency can be q increased by smoking a number of cigars par pad but its influence in the Wider contexts of va~isbility is limited. One partieolar pointle-edd on the manipulation of results which have been~a~b~ained wlftt~flferen~ number~afrcigafs per p~d 19 that the, (S the bests of ~igar per port in the 3P~i Ho, 1" report is It ass~eSthat all of t~krepeatability ~gPriation is due to the cigars. / A IfmifQd~nelysis~not included in this document) suggests j ( ispbt the for Ti~M, ~xample the may ndt account more than ~SiT Of th~M~·t~. We t~ii~e it vould~b~more correct and less misleading, therefore, not to oresedE the result~in the praposed~Fogess report op~the of a number cigars perpadJinstc~d, we"· ~I, 11 that the refer to ~hl vac~ability betue gual~fi~C8tiPn provid~d by statement bf the number o~figars per port, (iii) There are other, very important components of variance which describe method andlor product variation over longer periods of time; these 13, are relevant in the context or gilantffying the confidence limits in mean yields, laboratory comparisons, rank-ordering etc. ecak ~, individual tests that were re-analysed tie, method and cigar-type changes occurred between tests). Finally please note that numerous comments made in earli Ho. 1" o(the progress report, Tre misl and / per pa~i,e, over- will tend reduce ty, i cal to suodose that suEn factors ubuld aff Jes we included tS in appended r~na~ysis sllah~srises fo~individual tests, rests 4, 5, 1 and 14 included attempts to study the effects of certain / aspects ol~ Wale: end Nicotine Pleasurement. Unlorbnatellr this veS not very successful since the results were either inconclusive or did not permit a meaningful statistical re~sis,~g~ever, experiencel s~eniecb in the area of cigarette testlng~bbc~3jlQ~ia~*that there would be no problem il$r Icy· ~c~ GC methods were used as the basis far standatdisation, Nevertheless, even if this seems the appropriate route, special attention rw~Le Sbwld be given to the highly variable Water levels indicated by many of the individual test results e~ed tie, often the variability for ~lkWT~pp~ was as high as or higher than TPM), andll-re$er to ~pe~possihle erfetts of drop In filter y. ARaln no co~ents can be dede on the of statiseical analysds but fcomm t~e 1latest ~xperience of Ecigarette this 16 not thoug~ to be an a~ which re~ee furthcrWpdy. MW Whittaker (TT1) O E B Hirkes (8-AT) 6 nay 1992 6 ·- BATCo document for Legal Services : Health Canada 19 May 1999 APP~nlDI x N~d. Yn;nS aVJYL~ ~Ure weal i BATCo document for Legal Services : Health Canada 19 May 1999 SENT BY:TXB VE LEUI'EH 129- 9-32 ; g;jl I 201Cj04010J 180332 112 ist Draft I GP (21 April 149tJ IWEBNRTZDHRt COMnfTTEg FOE CIGAR SMOKE STUDY CIGAR SUD-GROUP bP TME COBs9T~ SMOKE STUDY GROUP Propreea report Table of ccntant~~ fntroductlonl II~ General overview of the work done, III, Problems of Machine Smoking of Cigars, Al Product-specific problems, 8~ Determination of relevant smoking perameters~ CI Procedural and analytical problems, D. Cooperative Smoking Tests E. Relevance of #ecLnicai Smoking Results IP~ DieeueeiPn of Cooperative Teet Beaulte. V, Future work. VI, Concluaibns, Rppendlx 1 Objective, Procedure, Results and Ccnclueidne cf Cooperative Teet Nr~ II Appendix 2 ObJeetive, Procedure, Results and Conclusions of Cooperative Teet Nr~ 2~ Appendix 3 Objective, Procedural Results and Cor,clueione of Cooperative Teet Ns~ 3~ Appendix 4 Objective, Procedure, Results and Conclusions of Cooperative Teet Nr, ~, Appendix 5 Objective, Procedure, Results and Conclusfons ct Cooperative Test Nt. 5. Appendix 6 Objective, Procedure, Beiults and Conclusions ct Cooperative Teet Nr~ 6~ a Appendix 7 Objective, Procedure, Resulta and Conclusions of Cooperative Teet Nrl 7. I~J BATCb document for Legal Services : Health Canada 19 May 1999 SENT BY:TS1B VE LEUVEN ;29- 0-92 ; 9:38 t 207050~0'10319C331 re 3 Appendix B ObJectlvel Procedure, Results and eoneluslons of Cooperative Test Nr~ 8, Appendix 9 objective, procedure, Results and eoneluelans of Cooperative Test Nr, 9~ Appendix 10 objoetlvb, Procedure, Results and conclusions of Cooperative Teet Hr. IO~ APPsndlx 11 Objective, Procedlrel Results and Conclusions of Cooperative Teet Nr~ ii, Appendix 12 Ob)ectlve, Procedure, Results and Cancluslone of Cooperative Teet Nr. 12. bppendlx 13 objective, Procedure, Results and Conelusicoe of Cooperative Teet Nr. 13~ Appendix 14 Objective, Procedure, Results and conclusions of Cooperative Teet Nr~ Appendix 15 objective, Procedure, Results and epnclualons of Cooperative Teet Nrl 15~ Appendix id Objective, Procedure, Results end Concluolcns of Co6perative Teet Nr, 16. RDpendix 17 Objective, Procedure, flesultp and conclusions of Cooperative Teet Nr~ ii. BATCo document for Legal Services : Health Canada 19 May 1999 SENT BY:TAB VE 1EUVEN ;29- 4-92 t 9:39 1 2010594C103 180332 ;J 1 Ist Draft I GP ~IPT~B~UI~IOBUI CDMnITTgg FOR CIGRB SHIIKI STUDY CIGEJI SUB·CBOUP OF THE CORESTA SHDKE STUDY GBDUP Progress report II Introduction: In 1910 the cigar Industry In the O~S,A,, France, The United Kingdom and Belpium recopnieed the need to study methods to analyze ctghremoke, In december 1970 delegates of the cigar industries of those countries met in Paris and set Up the International Co~nittee for Cigar Smoke Study~ The objective for ICCSS was defined as: the standilrdlsat~on of smoking procedures and analytical methods for tar and nicotine determination In cigar amoke~ To extend its knowledge in the field of cigar smoke analysis, the eorrsnittee asked in 1971 the support of CORESTA and in 1972 the CORESQ Smoke study Group created the Cigar Sub-group ' fPr perfecting a mechanical smoking method for cigara~ Oral reporting to the CORESTA Smoke Study Group was done during the Yearly meetings of the Groug~ Rt the masting 1991 in Otrecht, the smoke Study Group proposed the Cigar Sub-group to prepare a progress report on its sctivitie. 'COBESTA Bulletin 1912-2 page i, I~ BATCo document for Legal Services : Health Canada 19 May 1999 SENT BYI1AB V~ IEUVEN ;25· 0-99 ; 9;39 2010TO·0103 180332 ;R 5 General eventlew of the work donor A, The first important step consisted in collecting the existing data on the smokers behaviour for small and medium weight clparst which represent the major part of the market, Studies from Francejpe~nlted ~lnpdom, Germany and the U,S~A, allowed~P~the first paramatarl for Ituil~s2~p~er~p~:·i~Ps results was presented clear smoking to start a series of cooperative B1 A detailed report of the to the Smoke Study Group in Nice In 1973, and In 1975 the COREWA Bulletin ' published a CCRESTA provisional recommendation on parameters for cigar smoking, namelyl For studying the amoke from (untfpped) cigars In the weight range l,S B to 5~5 g I puff volume : 20 mi puff duration 1~1 sec puff interval 1 40 sec butt length 1 33 mm c, rn 1915 and 1976, several studies showed /Cnat the amount of secondary alkaloids and pyrldlnidderivates present in cigar smoke condensate 13io/ ICPOftffnt ' ~9refpre the cana~ittpe ~adceh~tlFF~rf~ts~ ~c~ l.~i. S~z; ~I TCi~c~cuyl~ 1, to keep the 190 spect'rometrle method 3400 for the determination of total alkaiolda 2, to propose pas-ehran~atogxaphic method for t specific determination of nicotine D, A series of ooopera',ive exercises between 1917 1979 allowed the Cigar Sub-group to reco8unend the ~k· glass flbxe fl I ter trap ~e~f~'-9C-rmA=r~I~d~am~ ; The ~ ~ increase in pressure drop of the filter assembly after dG*sL smoking ~i~ not exceed 250 Pa, Fr~tl·1~I~I r~' X~u/ ~1/R'rt~ttieaf~Y, the nwtlber of clp.ars to he smoked muft be ~i?njjb ""'L' '" '"L '' '"'"' ""'"!"' "':'. ""' "" exceed I ~rrr~$ 100 mg per glass libre filter of'44 mm 150 mg per glass flbre filter of 55 a Ydikm~t~l. a 1COXESTA Bulletin 1915·1 pa~e 33, BATCo document for legal Services : Health Canada 19 May 1999 SENT BY:R9 VE 1EuL,EN ;29- 4-92 ; 9:CO 2C1050·0103 180332 ;1 6 ·~·~·" a~e~ ~ E, From 1980 to 1983 and later from 1987 to 1999, several smoking cooperative eaetcisee were ~a~rr~B~a better understanding of the reasonb: for the ~impqrtaat within and between laboratory di~perP~anlwrs~·a~ ~ From a cigar sample of the same production, the aelec- tion by welOht and draw resistance improves the within laboratory variation but ~eZ.no influence In the between laboratory variation oday It Is very difficult to find a completely sat· efulng explanation for the h~k$C~t~eR of smoking ul~s durirs 1 c~ops:a~ivs.lIEiOl· ~j~j:4 The main reason seems to be the preat clQar varlablll~ tYI namely the irrepularl ty of the blend, of the we I Oh t, of the diameter, of the draw resistance se well as the presence of apparent veins inducing leak- ape at the level of the cigar holder during mechanical smoking, ,", in 1991, "'" ---··--UCM- ---m't~~b four labora- :o:iel prarsatd rB ~rultr or a atuly on ~Lg ol~all,p ~ ~isfZF~aap weiphlnp more than 5,5 1~0L~q ~QCy~Wh~YSr(P~I2~yb ,I ./p~i~ J Scne cl~are have i~doo~a tenp~ency to eKtinquleh between 0~ tuffs and the ~wet·i·en was to study If a,I·C modlticatian of sone~arametera ~ellmtnata this ~IPBB~P~BRCQC;Ek II w;.cU 8 j·J~ All the other parameters I~;~ild P~tttf~k, Puff inttrvnl fromllD sec to 30 ace ears ~F"S ' II ~~i~aBi~redueod thdq tendency of ~i~ cigars to extinguish ha~wean ~b Duffa~~t number of puffs an well an the smoke yields Increas ~jB~i~ '·-i-?30 Sc~ (; tt;P~:y ~ However, accer~glo the members of the committee, ~3~1YI· sue~f~t~ffr;oqpb~y Is unreallstie and deviates 1 .~d~mn~t~trom the normal ciop BPoker behavloc:, ~Gd·c~c. 4~u~~-· CPF~e~iteP~y It L(QOb O~ORt tk8t an Ij;t~j~of ',he ,_ pu~fre~er~ey is not tb~ooosi~the ~fion to ~ax; 7he problem of ~cr~zr smoklnp~lun ~~t~r)c. ~_2Tk Purthermara ]3eeidee the weight, other erfterians/aa the volume, the diameter, the length and perhaps the density should he considered in the choice of paras meters for blO clOar smoking, Cb Another Problem related to the standardieatlon of ~9~ ciosr smoking is the case of cigars with closed heads, I~eh~kil c~nairiRrScan cut ~SteRr by P-cutr ~uillo- a tine or pierclqt Is hovevar essential to reeorrJnend .O only one method a a BATCo document for Legal Selyices : Health Canada 19 May 1999 SEMI BY:TAB VE LEUVB 129- C·92 9:11 201050·0103 180332 ;: 1 ~Rekr~etFts,*llhf~·method pivipg(thd mp~et~ repFo~uc~ble results should/he retain Consepuently, a cooperative exercise was arganleed in 1986 to compare different ways of cutting under well defined condltiane~ --- --- -- Ali participants used the same type of ei~ar·cutter, capable td produce a straight cut and a V-cut~ ;7 The raeulte ~(P~skO the following conclusions: 3 i. The 8traipht cut gives lees puffs and a slightly lower smoke yield than the V·cut, due to the dlr- ference in length of the cigar after cutting, 2 ~ 511 thn~l~l~l t tl p sa tlfi fv Inn , the within and between laboratory variations of the smoking results axe of the same magnitude for both cut types, I Because It can be more precisely described, the cam- mlttee racoP~u~-nd~a straight cut with a well defined depth, ~at~h~ch is still to ha Dtefiwd, ·i~easeRt -n ~cF~RPlnrdCP;~F~I~BR, ~dbl~ji). as small gays" sug; ed t a s d be ~n ttl whi I he tar n 1998~ 9 t ive~iis~o~i~B# let anrd df~t~we(p~ht YIL~ ~ I*CR~CT. thb Sidbd~idl 14Ulilll 17 Eoe~nrna studies and ~Lt~S 1410YBIBGtl havB bean re~ilee4. witSrin and' tb9\between !B~Boratoru~iati s than ~Ba: ile f se t BA~Co document for Legal Services : Health Canada 19 May 1999 SENT 8Y:TAB VE IEUVEN 129- 1-92 ; 9:J1 201050t3103 180332 tt B IITI problems of Hner~...a SAKlkinl of ClOatsI There are a series of problems with reliable (repeatable and reproducible) smoking of cigars, which can be claa- eified into the following ea;egorise~ Al Product-specific problems 8, Determination of relevant smoking parameters C, Procedural and analytical problems Al Pr6duet·opeciEie Problems i, great variety of cigars : a, lore than a thousand different shapes~ b~ Different len9thl from 65 mm to appr, 250 mm, e~ Different weight, from less than 1 P up to more than 15 O per unit, d, Different diameter, from 6 r~n up to more than ZO a , Fur Ihermsra, the d i ame t er of c i gars in many cases is not regularly cyl~ndrical~ The diameter is Influenced by pressing Into various forma~ f~ Different shaping at the mouth end, 2, dreat range of tobacooe used for filler, wrapper and binder a, The filler tobaccos for cigars are thrashed and not cut~ the consequence is a.different sine of the particles and therefore an inhomogeneity of the tiller within a cigar. This makes the pressure drop and the smoking characteristics of cig