Tobacco Documents Online

Brief com=ents on letter and pgpers b~ Richard Peto Author : P.N. Lee Date : 20.1.83 2£ i71 New Orleans talk This is heavily based on the famous 1981JNCI paper (666, 1191) and contains little new. The major points made are as follows: (i) (ii) (ill) smoking is a c~use of a great proportion of lung cancer, i~s effects considerably outweighing those of air pollution, radiation, asbestos and other occupational /actors, duration of smoking is particularly important and failure to take this into ~ccoun¢ (e.g. by the recent National Research Council Report) can easily lead to misassessment of its effects and those of other causes of lung cancer, ~ar reductions not only have been responsible for a substantial reduction in the lung cancer risk associated with smoking so far, but will be responsible for further reductions, partly because of the further reduc¢ions in ~ar levels that are occurrlnE but also because of ~he greater propor¢ion of thelr smoking lifetime %hat smokers have smoked lower tar cigarettes. The Zhird poln¢ obviously overlaps tO some extent with what I plan tO ~alk about in ~ew Orleans. I sm not sure thls matters particularly. ~y ideas, which I will assemble into a draf~ in the next week or two, are ~o talk alonE the following lines: (a) (b) (c) (d> brie¢!y outline how cigarettes have changed, summarize evidence on compensation in terms of number of cigarettes Smoked, which I will show is vlrtually non-existent, summarize evidence on compensation in terms of of inhalation, which I ~ill show exists, hut is by no means complete, summarize :he epidemiological evldence, as in Lee and Garflnkel (1981~ but updated somewhat showln~ smokers of lower Car products have considerably bess rise of lun[ cancer than smokers of hi~her ¢ar products CD C:D t_,4 LJq CO

(e) show how the favourable trends are also reflected in UK national figures, (f) crltlclse the US NRC report suggesting the USA trends in lung cancer reflect an adverse effect of lower tar. WHO chapter Much of the first part of this chapter (up to page 14) gives messages similar to those given in the Mew Orleans ~alk, although it discusses other aspects of the smoklng/lung cancer association - dose/response, effects of stopplnE smoking, importance of cigarettes vs. pipes, interaction with other factors such as asbestos, problems of diagnosis. IZ is interesting (and gratifying) that Pezo (on page 12) quotes the conclusions of Lee and Garfinkel as support for his view that the switch to lower tar cigarettes has resulted in a substantial reduction in risk to the smoker. I essentially agree with all the conclusions Pe%o reaches from the epidemiologlcal data. In the last par~ of the chapter (pages 14-19) PeZo goes on to "practical action: discouraging sales and decreasing tar levels", and treads on ground which is more outside his (and my) main area of expertise. He makes a number of points: (1) tax increases may not be sufficiently emphaslsed. In support for %his he notes that an increase of 20% in clgaret~e prices produced less than a 20~ decrease in cigarette sales so that "the tobacco manufacturers complained of unemployment in the industry while the government collected more t~x". This seems somewhat simplistic to me, firstly because it does not consider the possibility that increasing tax would always continue to increase revenue (the recent decrease in ~ revenue from increasing alcohol revenue might have been mentioned) and secondly because the reference zo unemployment comes over only as an indicator of success of the strategy (the industry's 0 0 O0

(ii) (iii) - 3 - pips squeaking) and makes no deference to the fact that un- employment is not only undesirable but costs the government money. more attention should be given to qunntitatlvely emphasising the dangers of tobacco, pointing out evidence of unawareness of the facts in the general public (e.g. people in Britain erroneously ~hink traffic causes more deaths than tobacco, whilst in the US they think background radiation from nuclear power plants is a greater health risk than tobacco whereas tobacco is several thousand times more important. The methods of quantitation he advocates, e.g. "about a quarter of all regular cigarette smokers will be killed before their time by the habit" are not new. more attention should be given to tar reduction. He points out, correctly in my view, tha~ ~YHO and UICC concentrate too much on smoking avoidance and that though ':the aim is to produce circumstances in which very few people choose to smoke, but in a world where cigarette sales are still increasing rather than decreasing it is not wise to let the perfect be the enemy of the possible". Whether he overesrimaues the ease of tar reduction is arguable, however, the unsupported sSatement that "smokers hardly notice gradual changes in tar deliveries" is questionable. Draft RCP chapter on chronic airflow obstruction Much of what is contained in this chapter is a reiteration of the conclusions of ~he detailed study of Fletcher and Peto (1977), with emphasis being placed on loss of FEV (forced expiratory volume) as an indicator of disease status. As noted there, non-smokers and the ¢ majority of smokers lose FEV at a slow rate and have no problems whilst a "susceptlble minority" of smokers lose FEV at a greater rRte and do have problems. Giving up smoking, for susceptible smokers, does not recover lost FEV but leads ~o further loss occurring at the slow rate. The fac:ors causing susceptibility are not known. C~ C~

- 4 - Peto points out that t~e mechanism by which cigaret:e smoke produces chronic airflow obstruction is not known nor which component of smoke is involved. He notes that the epldemlology gives some indication of a benefit from lower tar but that there is a'problem in interpretation (relevant to the Alderson study and well known to us) in that smokers with low FEV may switch to low-tar brands because of the disease, just as they may give up smoking. InterestlnEly he feels that national trends in mortality from COPD (chronic obstructive pulmonary disease) which show substantial decreases beginning to emerge among middle aged males and among early middle aged females are quite likely to be due to an important extent to tar-reduction. This contrasts with the often held view that the decreases are a response to the Clean Air Act and the consequent reduc~lons In air pollution. ReQuest for smokln~ data I believe Peto's request to be supplied wi~h data on smoking statistics should be taken seriously. While his work on quantification of the hazards of smoking is obviously damaging to the indusZry, his, surprisingly in some ways, s~rong advocacy of low-tar cigarettes is ~o their advantage, and it h~rdly need be said he is a bona fide research worker. The Research Committee are well aware of my views regarding suppression of publication of smoking statistics but it should be rea!ised that there is nothing in the data since the last edition of RPI which would add fuel to his views on the damage done by smoking but there are in contrast some li~tle known fac~s that support the switch to low-tar cigarettes. Thus, inspection of the appropriate talbes in TD 1628 (finalized version includ!ng 1982 data is shortly to be prepared) reveals that low-~ar smokers smoke less cigarettes per head and inhale less than smokers of middle tar cigarettes, that consumption per smoker has been steadily falling for more than 5 years and that there has been no evidence of an increase in tlme of the proportion of smokers who say they inhale a lot. Coupled with the fact thaz the percentage of sales of cigarettes that are low tar has increased sub- s~antia!!y over the period since 1975, this is Imporzant evidence that the UK smoker is not compensating for tar reduction as so many researchers su~es~. C-D t.,'~ (.,'0

- 5 - Dr. Wald already has a request in for smoking data and Imperial have agreed in principle to supply him with information when the precise tables required are known. Peto's request could perhaps be linked in with %his. CD C_~ t~4 C30 C~

CLINICAL TRIAL SERVICE UNIT RADCLIFFE INFIR.~IARY, OXFORD OX2 oIIE. Pustal =,.t.J, rc~= 6or |,¢t,t.~;$ ~4%tttt no ~t.:'t¢'op~,: C.T3.U.. FREEPOST. OXFORD OX2 6BE. CNIVERSITY OF OXFORD T¢l~'l)lh~llC nUlllO~; :'or Dlr,gl:t ;~IILJ ~ fill I'~.I I I0 II ,4;VI*%:. line O~i;)rd (10865)40972 724567 Peter Lee 25 Park Road Cheam Sutton Surrey SM3 SPY 17 January 1983 Dear Peter I enclose (i) an outline of what I'll be saying fn New Orleans - i hope you agree; (2) a paper on lung cancer that i recently drafted for WHO, which emphasises the likely benefits of low-tar cigarettes in ear_a~ middle age; and (3) a paper on COPD that does likewise (see especially Table 2) and that emphaslses how misleading Alderson-type case/control studies of COPD may be. Other than your own re-analysis of the 1968-70 Whitehall FEV data, do you know of any decent studies of FEV in relation to tar deliveries? Also, do you agree with the general perspectives of the COPD chapter? Yours sincerely Encl: Richard Peto New Orleans MSS x 2 COPD RCP chapter WHO lung cancer chapter PS: In exchange fcr this lot, can you arrange for me to be on the mailing list for the TAC blue books that you used to produce? I need statistics on smoking, especially in the UK. :~=t Ftl¢.:+ld IJuil. I R3. I R+ I'. l :nc:,lll~ I':,,1¢~+,: ,,I \l~.th+in¢. llu+l, l'hr~.',.'t¢lL lulf,,~.'tlal ('+llL~'t J4.~.~q..jr+l| I "llt~J C:lll++%'r+ UlIIIP R++h;=rd Pcto, %IA. ~IS,, i(.'Ri P..'J~.-r ,+| ( .=+1,¢r .~+tltll¢~ l$.+r+,.;rd ll~l.~-r+ +++r.~!l~:tvlll ~I., I. t '11++%'J'+II% .,I {,+~.l,++t| Cb C> Co c~ 0-,,