6 j UL W4 BRITISH-AMERICAN TOBACCO COMPANY LIMITED BAT Corporate Affairs Managers TO: CQ D S Watterton FROM: Dr L Rudge DATE: 4 July 1994 SUBJF.CT: Australian lung cancer study Attached are comments on a study published on the I st July which claims that an increase in smoking among cchoot age girls can explain why lung cancer rates among young people in Tasmania are now higher in females than in mates. The study received considerable press coverage in Australia. a 0 -9> MIILL13ANK KNOWLE GREEN STAINES MIDDLESEX TWIS 1DY C> SMOWNG ISSUES DEPARTMENT DIRECr FAX NO: 0784 448654 Co -si BATCo document for Legal Services : Health Canada 22 October 1999 Dwyer, T. Blizz--ardL., Shugg, D., Hill.D., A 11sari.M. Z Higher lung cancer rates in young women thnn young men: Tasmania, 1983-1992. Cancer Causes and Control 5: 351-358 (1994). In Tasmania, the number of cases of lung cancer and the number of deaths from lung cancer over the last decade are reported to have been higher among women than men, in the age group 25-44 years. This trend appears unique to Tasmania and to the last decade (1983-1992). Lung cancer rates were lower in females than in males in those aged over 45 years, and in all age groups on mainland Australia. In an attempt to explain this pattern the authors investigated smoking behaviour. Using data from surveys of school age children, the authors estimate that smoking prevalence among young females passed that for males at some time between 1973 and 1984, a decade before the observed excess of female lung cancer cases. The authors warn in their conclusion, "It is likely that the change in the gender distribution for lung cancer reported here will soon be seen in other regions which have experienced similar smoking trends." Thifre are two key problems with this assumption that smoking can explain the change in the male to female ratio of lung cancer. Firstly, the change in smoking incidence occurred too late to explain the change in lung cancer incidence~ and secondly, similar changes in smoking among school age children in Australia have not resulted in a higher incidence of lung Cancer in females than waits. ù Lung cancer is commonly a disease of old age. The size of the excess lung cancer incidence in Tasmanian females aged 2544 years compared to young males was 19 cases during the 10 years of the survey. There were also I I more deaths from lung cancer in young females than males, but the difference was not significant. ù The time trends in school age smoking were estimated from five surveys of schoolchildren, but which used different questions to assess smoking incidence. Three asked whether the children had smoked in the last week, and two asked whether they considered themselves regular smokers. These questions are likely to yield different answers. and both are open to considerable inaccuracy regarding self-reported smoking. ù Nevertheless, the authors suggest that smoking peaked in schoolboys, aged between 12 and 16, in 1973 and declined thereafter, while schoolgirl smoking peaked in 1984 and is presently slightly above that of schoolboys. Assuming that the schoolgirl population increased their smoking mid-way between these two survey dates, even the eldest schoolgirls would not have been old enough to be included in the lung cancer survey until its half way point. They are therefore unlikely to explain the lung cancer rise in the 25-44 year old age group. Many scientists now accept that lung cancer is a disease with a long development time, and assume that if smoking were a cause of lung cancer, changes in smoking would not be f eflect cd in lung cancer incidence until afler a 20 to 30 year time lag, -j~:b CD L-1 01% Co BATCo document for Legal Services : Health Canada 22 October 1999 In 1973, less than 20 years before the end of the lung cancer survey, smoking incidence was still greater among schoolboys than schoolgirls. This again indicates that the changes in smoking observed are too recent to explain the changes in lung cancer rates. In Australia, the prevalence of smoking in young females overtook that of young males between 1976 and 1980, and the authors acknowledge that "[ilf this reversal of the gender distribution among young smokers had a similar timing in Tasmania, the difference in past smoking prevalence in young Tasmanians would not readily account for the higher lung cancer incidence in 25-44 year old women during the period 1983-1992, because the latent period would appear to be too short." The authors accept that * ... the excess of female cases over male cases cannot be directly linked to gender diffierences in smoking history using our data. Given the -widely accepted view that there is a latent period for effect of smoking on lung cancer, we would need smoking data on the population at risk a decade or more prior to the period for which we have data on incidence in order to validly infer this." Despite this obvious fault in their analyses, they argue that * ... this is not sufficient reason for dismissing changes in smoking prevalence as the primary cause of the reversal in the gender ratio of lung cancer incidence. It could be that females in the relevant cohort who have smoked have done so in a different way than the males." This is a desperate attempt to make the data fit the hypothesis. Other regions seem to have experienced similar smoking. trends. The surveys also showed that more Australian schoolgirls than schoolboys now smoke, as in Tasmania~ but the increase in female lung cancer relative to males was not seen. The authors fail to mention this. They also state that "...data from the US indicate that the prevalence of smoking in that nation by girls of school age has exceeded that of schoolboys since 1977". Again, no shift in the male / female ratio of lung cancer is mentioned. No explanation is given as to why only Tasmania should have experienced this. Finally, the authors suggest that "...the need to search for other explanations of the higher female lung cancer incidence other than change in the pattern of smoking, could be determined.". This is the only indication that the authors have even considered that something other than smoking could be important. They do not consider any other factors in their analvses. 9 BATCo document for Legal Services: Health Canada 22 October 1999 :~El% T B~ a. -- : I- T-91 : ~Ijtlj ISAL-,3 DEN. - -, II 5Lr,/ BRITISIT-AMERICAN TOBACCO COMPANY LIMITED Te: BAT Corparate Affairs Managers Total Pages: Mr J.H Ram Daitugut Arg‡n—nq B. omisore mr M. Riordan Australia P. Ade‹besalt Nigeria Nb. D. Stm=n TIA, Austrarla b& 0. Oddaker Norway Nir R-A- Mazurn…r Bangladesh (Tiedenmu) I& A.C. Stmte Barbados j& Cr. Mani paidsm ?& F. de Vr‡cY B‚lgium Mr A. Tejada panama Dr L.T. Caruso Brazil b& T. Faite Papua New Guinea Mr 1. de Baffas Franco Bra2il Mr S.J. Abery Rusia Mr 1. matiiis Camercon Mr B. A. 0. King Sieffi Leone Mr M. Dessoetcaux Canada Nir F. Combe Sinfflre Dr. S. MaswY canada (BATUKE) Mr B. R=890 Chat Mr D. Quek Situppore mg B. Chow China Ms il Thomson South Afiica Mr E. Cordeio Costa Rica Mr 1. Aranaz Spain Mr P. Neophytou Cyptus Mr. M. chacon Spain (CI) Dr P. Nladm D,¯uitafk - . . I& Vp. Majalagokam Sri Lanka Mr D. Vep FI Salvador Nb 0. Renten Suriname Nin S. Ka=noia FŒnbm Mr C Martift swbedam Mr C. floudard France Mr T.R. Ivey Titvm Dr D. Pangritz Germany Mr k CWI Tbailand Mr L idun-08de Ghana N—s D. Demming Trinidad Mr 0. Scbmstedt Guatemala Mr E. Pcpplcs U.SA ,vir C.JL Quinfin Guyana Dr. S. Appleton U.S.A. flonduras P. C" U.S.A. Dr A. SantRfflda (BATUKE) Mr B. Brady Rang Kong Mr M. Gwoke Uganda Mr L. LaWY Hungary MrA. Sarkar India (TI) Mr E. Antich V‚nezuela Dr Charadec Indita (ITC) Ndr. M. Fry Vietnam india (ITC) B. Mavambu. zaire Nin A SYam Zimbabwe j~— B. Bbcxttacha~ee Indit (VST) Mr PL B"Son Mr G. St.,anto indonesia Mr E. -k LUMO Kenya b& M. Prideaux %Windsor Heuse Nir S Kenya W D. Bama Corp Affairs - OPulah ~&. G. Read 'R & D, Soton Mr ~u Nightio Malawi le Mr D. Rabian. Malaysia Nt K. Abduƒtif mainitius Nir P. de Nbre Netherlands 1& 1. 1-L J:Ioppenbrouwers Netherlands CHW) 1,– ri. Robertson 14ew zealand, Mr. M. Thompson TINZ mr C. Amidar Nicaragua SMoIKI14t; ISSU ES DEW. DUOECT ]FA-X- 0-, $4 443654 ®1>. MujjŒL~tNK l(NOWLE rREEIi STA WS MIDDLESLX TW18 ID'y CD L4 Co CD BATCo document for Legal Services: Health Canada 22 October 1999 0784LI18654 SLY BY: S. A. T. ;v10- 6-94 : 13:31 :SMOKING 15SUFS DEPT.- 09-03 "779715--- Given that the EPA report on ETS led to a huge number of scare stories on the issue, these are very important quotes rrorn the man who spearheaded the organisation. Reilly then goes on to say thaf-- 'one in a million is a very remote risk .... The anmtal, not lifetime. risk of death by homicide in Washington, D.C. is now more than 900 in a millionl' Reilly also makes a number of quotes about active smoking, I personally doret believe that smoking a cigarette or two really raises your chance of dying, or takes ffteen minutes off your life, as the models show! - although he goes on to suggest that there is no indication that there is a Tully safe exposure to tobacco smoke! We would strongly urge all operating companies to use these articles with their media contacts in our increasingly important efforts to encouragc balanced reporting in the media on these issues. Both articles are significant enough to warrant meeting with media contacts to brief them on these latest developments on ETS. At a minimum the quotes can be used as the basis for letters to the editor in response to articles printing negative stories about ETS. Please contact the Smoking Issues department if necessary for further assistance in using t~ese articles. With kind regards ~~L ;~ - . - 0 (J-4 Co BATCo document for Legal Services : Health Canada 22 October 1999