Select Committee on Health Minutes of Evidence

Memorandum by British American Tobacco

THE TOBACCO INDUSTRY AND THE HEALTH RISKS OF SMOKING (TB 28)

GOVERNMENTS ROLE IN THE DEVELOPMENT OF LOWER RISK CIGARETTES

  148.  Following the publication of the Royal College of Physicians' report on smoking and health in 1971, the Government embarked on a number of public policy initiatives designed to encourage smokers to quit, to inform continuing smokers about the tar yields of brands on the UK market, and to encourage smokers of comparatively high and medium tar brands to switch to lower tar cigarettes. These initiatives included: funding a Health Education Council campaign on the reported dangers of smoking, entering into the first of a series of voluntary agreements with the UK tobacco companies on labelling and advertising; and creating the Standing Scientific Liaison Committee ("SSLC").

  149.  The SSLC was a joint committee of the UK tobacco companies and the DHSS. Its terms of reference were:

"To advise the Secretary of State on the scientific aspects of matters concerning smoking and health, specifically:

(a)  To advise on the significance to health of the tar and nicotine yield of cigarettes and how such information can best be made available to the public, and on how the determinations should be carried out for this purpose.

(b)  To receive data about the constituents of tobacco and tobacco smoke and changes in them and to release these in confidence to bona fide research workers.

(c)  To review the research into less dangerous smoking and to consider whether further such research needs to be carried out and to advise on the validity of research results and of systems of testing the health effects of cigarettes and of their predictive value to human health." (SSLC, "Background and Terms of Reference", July 1971).

Tar/Nicotine League Tables

  150.  One of the main recommendations of the SSLC's Report to the DHSS was that consumers should be informed of the tar and nicotine yields of cigarette brands. From 1973 the Department of Health and Social Security ("DHSS") began publishing tables which ranked brands by tar yield and assigned them to High, Middle or Low tar brands and from 1976 the UK companies agreed to print tar group designations (low tar, low to middle tar, middle tar, middle to high tar, high tar) on packets of cigarettes and include them in advertisements. Today, tar and nicotine yields are shown on all packs of cigarettes sold in the UK.

  151.  It is important to understand how tar and nicotine yields are measured, and what they mean. In the UK, the measurements are carried out on brand samples, supplied by the tobacco companies to the Laboratory of the Government Chemist, using an International Standards Organisation method. The method is based on machine smoking of sample cigarettes, taking a two second, 35ml puff every minute to a specified butt length.

  152.  It was understood from the outset that the measurements obtained by this method did not represent the yields which would be obtained by any individual smoker. A smoker can take a greater or a lesser number of puffs than the machine, and, within certain physical limitations, can take smaller or larger puffs. The smoker can extinguish the cigarette after taking a few puffs or can smoke it right down to the filter. No machine method can mimic all the possible varieties of human smoking behaviour:

"Critics of the machine smoking procedure have frequently failed to understand that values presented in tables published by DHSS have never been intended to be actual yields obtained by any one smoker. Rather, they enable brands to be ranked. This allows inter-brand comparison under a standard test procedure, presenting the smoker with information to enable him to choose, if he so wishes, a lower yielding brand" (ISCSH, Fourth Report, Appendix 3, 1988).

  153.  This assessment was made by the ISCSH, on the advice of the Laboratory of the Government Chemist, in 1988. It reflects British American Tobacco's understanding of the purpose and limitations of machine smoking methods, both in the UK and elsewhere. At the same time, the ISCSH observed that, although absolute yields from cigarettes could be changed substantially by variations in individual smoking parameters (for example taking longer or deeper puffs):

"the rank order of brands by yield of the main measured components is little altered" (ISCSH, Fourth Report, paragraph 21, 1988).

The Independent Scientific Committee on Smoking and Health

  154.  The ISCSH was appointed by the DHSS in 1973 (the SSLC stood down after the publication of its report in 1972). For the following fifteen years, this Committee sought to steer the search by UK tobacco companies for "lower risk" cigarettes. British American Tobacco closely monitored the recommendations of this Committee, and British American Tobacco's internal research reflects responsiveness to the issues raised by the Committee as well as by the medical and scientific community as a whole.

  155.  The membership of the ISCSH consisted of scientists from a range of disciplines, acting with "complete professional independence" (ISCSH, First report, page 1, 1975), under the Chairmanship, first of Dr Robert Hunter (later Lord Hunter of Newington), and second of Dr (later Sir) Peter Froggatt. Consequently, the ISCSH was often referred to as either the "Hunter Committee" or the "Froggatt Committee". Its terms of reference were set out in the ISCSH's First Report:

"To advise on the scientific aspects of matters concerning smoking and health, in particular:(a)(i) To receive in confidence full data about the constituents of cigarettes and other smoking materials and their smoke and changes in these. (ii) To release to bona fide research workers for approved subjects such of the above as is agreed by the suppliers of it. (b) To review the research into less dangerous smoking and to consider whether further such research, including clinical trials and epidemiological studies, needs to be carried out; and (c) to advise on the validity of research results and of systems of testing the health effects of tobacco and tobacco substitutes and on their predictive value to human health" (ISCSH, First Report, Appendix 1, 1975).

  156.  The terms were expanded slightly in the Second Report "to make it clear that the Committee advises both the Government and the tobacco companies" (Second Report, paragraph 1, 1979); the first sentence was amended to read: "The Committee is appointed by the Health Ministers to advise them and, where appropriate, the tobacco companies on the scientific aspects of matters concerning smoking and health . . ." (ISCSH, Second Report, Appendix 1, 1979).

  157.  In practice, the ISCSH was concerned with two main issues; the safety assessment of additives in tobacco products on the UK market (considered in this memorandum under the heading "Ingredients"), and the promotion of a product modification strategy (ISCSH, Fourth Report, paragraph 2, 1988) aimed at the development of what it called "lower risk" cigarettes (ISCSH, Second Report, p.1, 1979). The ISCSH characterised the measurement and reporting of tar and nicotine yields of marketed brands, together with the creation of the Committee itself, as a "centrally directed, coherent `product modification' programme" (ISCSH, Fourth Report, paragraph 2, 1988). Throughout its existence, the ISCSH consistently advised that the best health option for smokers was to quit, but equally consistently acknowledged that some smokers would continue to smoke, and that many of these smokers would "have a strong desire to smoke less dangerously either by smoking fewer cigarettes or `lower risk' cigarettes" (ISCSH, Second Report, paragraph 18, 1979).

  158.  The ISCSH further recognised that the "manufacturers are engaged in a commercial enterprise which depends greatly on the acceptability of their products to the consumer", and that there had to be an "accommodation" between commercial acceptability of products and the reduction of risk to health (ISCSH, Second Report, paragraph 22, 1979). In summary, the aim of the product modification programme was to promote the development of "lower risk" cigarettes with maintained consumer acceptability.

Tobacco substitutes

  159.  The first stage of the programme to develop "lower-risk" cigarettes was to consider the value and safety of introducing tobacco substitutes to replace a proportion of the natural tobacco in cigarettes. It was thought that the replacement of tobacco with a tobacco substitute offered the prospect of an overall reduction in toxicity (ISCSH, Fourth Report, paragraph 3, 1988). In its First Report, in 1975, the ISCSH set out detailed guidelines for the toxicological assessment of tobacco substitutes. The guidelines were revised in the Second Report. The UK tobacco manufacturers agreed to follow the ISCSH's testing protocol before marketing any cigarettes containing tobacco substitutes. The protocol envisaged several stages of chemical and biological testing, encompassing smoke chemistry; acute inhalation studies using rats and monkeys; short term clinical tests on human smokers (as well as, at the manufacturers' discretion, tests of consumer acceptability); and chronic animal carcinogenicity (skin-painting), reproduction/teratological and inhalation studies. The ISCSH set out with specificity the types of data required to be submitted.

  160.  It is evident that the ISCSH believed that this testing regime might give some information as to the comparative toxicity of modified products. At the same time, the ISCSH did not believe such tests were dispositively predictive of the effects of long-term use of the products by human smokers. Consistently, the Committee emphasised that "systematic long term epidemiological studies after marketing (would be) needed" (ISCSH, First Report, paragraph 11, 1975). This approach of the ISCSH was broadly in line with British American Tobacco's long-held views of the values and limitations of biological testing.

  161.  Two submissions were made to the ISCSH under the guidelines for testing tobacco substitutes, one jointly by Gallaher and Rothmans for the product known as "Cytrel" and one by Imperial Tobacco for the product known as "NSM" (or "new smoking material") (ISCSH, Second Report, paragraph 3, 1979). Internally, British American Tobacco had developed its own tobacco substitute, known by the code-name "BATFLAKE", and had begun to assess it against the ISCSH's testing protocol. Brands containing Cytrel and NSM were placed on the UK market in 1977, but were, as the ISCSH acknowledged, "a commercial failure and . . . were eventually withdrawn" (ISCSH, Fourth Report, paragraph 5, 1988). For obvious reasons, it was not possible to carry out the long-term epidemiological observations envisaged by the Committee, so it remains an unanswered question whether the availability of such products would have correlated with a reduction in smoking-related diseases among their users. For equally obvious reasons, British American Tobacco, knowing that its product faced the same consumer acceptability problems, did not move forward with the BATFLAKE development, either in the UK or elsewhere.

  162.  The second phase of the ISCSH's programme was to seek "further reductions in the tar yields of conventional cigarettes" (ISCSH, Fourth report, paragraph 6, 1988). This was against the background of "notable progress" by the tobacco manufacturers in reducing tar on their own initiative (ISCSH, Second Report, paragraph 20, 1979). Indeed, the ISCSH recognised that there had been a reduction of approximately 45 per cent in the sales-weighted tar yield of UK cigarettes in the years 1965 to 1977, and a reduction of approximately 35 per cent in sales-weighted nicotine yields, in filter brands, over the same period (ibid). Further research showed that an overall downward trend in sales-weighted tar yields could be traced back to the 1930s (ISCSH, Fourth Report, paragraph 9, 1988). Nevertheless, it was clear that the ISCSH wished to work together with the UK tobacco companies to make further, systematic progress towards a lower tar, and indeed nicotine, market.

  163.  The basis for this strategy was set out explicitly: "We have reviewed the epidemiological data which associate lung cancer and chronic obstructive lung disease with cigarette smoking. Human data, acute and chronic in vivo studies in animals and short-term in vitro toxicity studies lead us to concur that tar is the major tobacco component in lung cancer causation" (ISCSH, Third Report, paragraph 7, 1983). Indeed, the ISCSH noted epidemiological evidence already in existence which was suggestive of an association between the consumption of lower tar cigarettes and a decline (in men) or levelling off (in younger women) of lung cancer death rates. The view that tar reduction was a reasonable strategy was consistent both with the published research of the TRC and with British American Tobacco's own internal research.

  164.  On this basis, the ISCSH, in 1979, recommended that "the Secretaries of State obtain the manufacturers' co-operation for the achievement of further substantial reductions in tar yields" (ISCSH, Second Report, paragraph 23, 1979). In the following year, a voluntary agreement was concluded on product modification, between the UK companies, represented by the Tobacco Advisory Council and the Government. The Agreement set a sales-weighted average target ("SWAT") for tar of 15mg by the end of 1983. It also set "tar ceiling" constraints on the introduction of new brands, and on the reformulation of existing brands. A further voluntary agreement on product modification was concluded in 1984 which set a target of about 13mg by the end of 1987. Although no further tar ceiling agreements were formally concluded, the ISCSH proposed in its Fourth Report a tar SWAT of 12mg by the end of 1991. These targets were met. Reductions in tar and nicotine yields have continued in the UK market; the most recent figures available to us (1997) show a sales-weighted tar average of 10.3mg per cigarette and sales-weighted nicotine average of 0.8mg per cigarette.

Tobacco Products Research Trust

  165.  Just as it was recognised that long-term epidemiological studies would be required to establish whether any actual health benefits accrued from the smoking of products containing tobacco substitutes, the ISCSH recognised that similar studies would be needed to monitor the long-term effects of reducing tar yields on human populations. The necessary projects were funded by the Tobacco Advisory Council through the Tobacco Products Research Trust ("TPRT"), which was established in May 1982. Over a period of 14 years, 37 projects were funded, covering a wide range of issues. The total funding amounted to £8 million. (Swann C and Froggatt P, "The Tobacco Products Research Trust", Royal Society of Medicine Press Limited, 1996).

  166.  A number of the studies supported by TPRT funding looked at aspects of the health effects of reduced tar yields in human populations. The projects and their results were described by Cheryl Swann and Sir Peter Froggatt in a book on the TPRT programme (Swann C and Froggatt P, "The Tobacco Products Research Trust", Royal Society of Medicine Press Limited, 1996). A "tar pooling" project, directed by Professor Wald, combined data from four prospective studies conducted in the UK. It found "a consistent trend of increasing mortality from smoking-related diseases with increasing level of tar yield", statistically significant for a range of diseases, including lung cancer. Calculating the mortality risks in smokers of low tar (15mg) to high tar (30mg) cigarettes, the study reported that "about one quarter of deaths from lung cancer, coronary heart disease and possibly other smoking-related diseases would have been avoided by lowering tar yield from 30 mg to 15mg/cigarette" (ibid, p.45-46).

  167.  Swann and Froggatt drew the following overall conclusions from these and the other studies reported:

"Over much of its range, tar yield of cigarettes was positively and independently associated with the incidence of a number of smoking-related conditions including lung cancer and chronic obstructive lung disease and less strongly with fatal and non-fatal ischaemic heart disease and stroke; These associations may be less marked, or in some instances even become undemonstrable, with lower yielding cigarettes (c12mg tar); Generally tar yield was less strongly associated with (relevant) disease incidence than amount of cigarette tobacco smoked"(ibid, p.3-4).

Nicotine

  168.  The ISCSH and nicotine's role in smoking: When the ISCSH turned to the question of the acceptability of lower tar products, the Committee was in no doubt that nicotine, as a pharmacological agent, held the key:

"There are many reasons why people start to smoke but dependence on nicotine is probably the most important single reason for their continuing to smoke" (ISCSH, Third Report, paragraph 16, 1983).

  169.  There was nothing new in the claim that nicotine was an important constituent of tobacco products. It has been known in the scientific community since the "classic work" by Langley and Dickinson in 1889 that nicotine has a mild pharmacological action (US Surgeon General, "The Health Consequences of Smoking: Nicotine Addiction" p.10, 1988). Over the years, researchers extensively described nicotine's effects on the central nervous system, including on neurotransmission; its complex properties as a stimulant and relaxant; its anti-diuretic effects; and its action as an appetite suppressant. A valuable and massive resource on the published nicotine literature was created by Professors Larson, Haag and Silvette in 1961 (Larson PS, Haag HB and Silvette H, "Tobacco: Experimental and Clinical Studies: a Comprehensive Account of the World Literature", The Williams &Williams Company, Baltimore, 1961).

  170.  Throughout the twentieth century, researchers in this field have hypothesised that smokers smoke for nicotine. In the late 1950s, British American Tobacco funded research at Battelle, an independent contract laboratory in Geneva, to investigate nicotine's pharmacological effects and its metabolism in the human body. Sir Charles Ellis hoped that such research would provide a scientific under-pinning to the development of Ariel. This work resulted in a series of reports by Battelle on what was known as "Project Hippo", and a report on metabolism, "The fate of nicotine in the body". Peer review of this work by Professor JH Burns (who then held the Chair in Pharmacology at the University of Oxford), and Dr AK Armitage of the TRC, (who subsequently became one of the most widely published and cited workers in this field) demonstrated, regrettably, that the work was unsound in its methods and conclusions. British American Tobacco, however, through its membership of the TRC, continued to support work in this area, not least by Dr Armitage, which resulted in the publication of more than 20 nicotine pharmacology studies in such journals as "Nature" and "The British Journal of Pharmacology".

  171.  The ISCSH and compensation: It has been suggested recently that lowering tar yields is an ineffective strategy for lowering risk because of the phenomenon known as "compensation".

  172.  It has even been suggested, in the face of plain evidence to the contrary, that tobacco companies possessed covert knowledge that compensation was undermining the low tar programme (Jarvis M and Bates C, "Why Low Tar Cigarettes Don't Work and How the Tobacco Industry has Fooled the Smoking Public", Action on Smoking and Health Papers, (www.ash.org.uk/papers), 1999). In fact, it was the tobacco companies who supported early research on compensation (eg Ashton H and Watson DW, "Puffing Frequency and Nicotine Intake in Cigarette Smokers", BMJ 3:679, 1970); research supported by the Tobacco Research Council, the implications of which were explained to the ISCSH:

"The industry has told us, however, that the level of nicotine in some cigarettes has already reached a point where consumers are trying to maintain their nicotine intake by methods of `compensation'"(ISCSH, Third Report, paragraph 17, 1983).

  173.  The ISCSH were concerned throughout this period that reductions in tar yields would lead to consumer rejection of what they hoped might be lower-risk products. It was recognised that the main techniques used by manufacturers to reduce tar, filtration and ventilation, also had the effect, although somewhat less marked, of reducing nicotine. The members of the ISCSH, like many other members of the scientific community, and indeed like many scientists within the tobacco industry, apparently believed that nicotine, and more specifically the pharmacological effect of nicotine, was the key to cigarette acceptability.

  174.  Indeed, these very issues were the focus of an ISCSH conference held in 1986, the proceedings of which were later published (Wald, N and Froggatt, Sir Peter, "Nicotine, Smoking and the Low Tar Programme", Oxford University Press, Oxford, 1989). At the conference, Professor Nicholas Wald, and co-workers, reported that, because of compensatory behaviour "smokers who reduce the tar yield of their cigarettes by half will, on average, reduce their intake of tar by 24 per cent" (ibid, p 100). Dr Martin Jarvis argued that "nicotine is the primary reinforcer underlying tobacco use" (ibid, p.171). Dr Michael Russell at the same conference, also identified nicotine as the major controlling factor in smoking behaviour (ibid, p.151, p.175). Dr Frank Fairweather, however, observed in his contribution to the symposium, that research on factors other than nicotine, such as taste, flavour and the common chemical sense, had been piecemeal, but that it was clear that nicotine was not the only determinant of smoker compensation (ibid, p.212).

  175.  In the face of accumulating evidence of compensation, both from the tobacco companies' own research and from independent scientists, the ISCSH remained convinced that the low tar programme was valid and was showing benefits. As Professor Wald and Sir Peter Froggatt reported:

"Nicotine, while not the only factor in controlling overall smoking behaviour, is recognised as being an important factor in regulating compensatory smoking. This can be used to advantage by reducing nicotine yields less than tar yields and thereby reducing the intake of tar as yields are gradually reduced. Reduction in tar yields will reduce the incidence of lung cancer and will probably also reduce the incidence of chronic obstructive lung disease . . . Trend data on smoking and its related diseases support these conclusions"(ibid, p 229).

  176.  The ISCSH and product development: While maintaining that nicotine yields, like tar yields, should generally be reduced, the ISCSH made a number of recommendations in relation to nicotine and product acceptability:

—  "In the longer term, since it is nicotine that the majority of dependent smokers appear to require, it may be necessary for manufacturers to modify the nicotine delivery of cigarettes or alter factors which could influence its rate of absorption from inhaled smoke into the tissues of the body" (ISCSH, Second Report, paragraph 24, 1979).

—  "Some companies have suggested that the addition of natural nicotine or nicotine salts to ultra low tar and nicotine products would produce a more acceptable smoke for dependent smokers. If this practice resulted in an increased dependence among smokers, then it would be difficult to approve it. The Committee will continue to review the health implications of the addition of nicotine to tobacco and it seems likely that it will advocate toxicity testing in animals and other studies in man before it can recommend the addition of exogenous nicotine either in the form of natural nicotine or its salts to the smoking product" (ISCSH, Second Report, paragraph 25, 1979).

—  "We . . . recommend that, in general, nicotine levels should fall. We also believe that there should be available to the public some brands with tar yields below those of the present principal Low Tar brands (ie below about 8mg/cigarette), but with proportionately higher nicotine yields (up to about 1mg)" (ISCSH, Third Report, paragraph 20, 1983).

—  "While the overall aim should be towards reductions in the tar/nicotine ratio, this should not be through the enhancement nor solely through the maintenance of present-day middle range nicotine levels (around 1.3mg/cigarette). In general the sales-weighted average nicotine yields should fall, and on the lines of the suggestion made in our Third Report (para 20) there should continue to be some brands available to the public with nicotine yields below 1mg and with tar yields reduced to a proportionately greater extent (below 8mg)" (ISCSH, Fourth Report, paragraph 34, 1988).

  177.  The implications of these proposals can be spelt out as follows. At the very least, the ISCSH were recommending that, if only for some brands, the tar/nicotine ratio should be reduced. This might mean, of course, that nicotine would rise only in relation to tar, with both tar and nicotine yields overall being reduced. The ISCSH were also expressing an interest in ways in which the rate of nicotine absorption could be influenced. Finally, it was not ruling out the addition of nicotine or nicotine salts to the product, although it expressed concerns about such a strategy.

  178.  Scientists outside the ISCSH were making similar recommendations at this time. For example, Dr Michael Russell of the Addiction Research Unit at the Maudsley Hospital argued, in a series of papers, that "(p)eople smoke for nicotine but they die from the tar" (Russell MAH, "Low-tar Medium-nicotine Cigarettes: A New Approach to Safer Smoking", BMJ, 1:1430, 1976), and that therefore nicotine yields should be maintained while tar was being reduced:

  "(T)he safest cigarette is likely to be one with a low tar yield and a low CO yield but a high, rather than low, nicotine yield. Such a cigarette would minimise the amount of tar and CO it is necessary to inhale to obtain a given amount of nicotine" (Russell MAH, "Realistic Goals for Smoking and Health: A Case for Safer Smoking", Lancet, 1:254, 1974).

  179.  With this impetus from the scientific community, British American Tobacco conducted research in the 1970s and 1980s into the properties of nicotine, the design of products with both reduced and enhanced tar/nicotine ratios, the prospects for augmenting the nicotine yield of products or increasing the transfer of nicotine from tobacco to smoke, and factors affecting the rate and site of nicotine absorption.

  180.  Not least, British American Tobacco collaborated with a number of external researchers in the field. For example, British American Tobacco in the UK supplied Dr Russell both with funding and with experimental cigarettes to test his low tar/maintained nicotine hypothesis. Similarly, British American Tobacco in the US funded the development of a special breed of tobacco, originally developed at the US Department of Agriculture, which came to be called Y-1. The plant was developed by traditional plant breeding techniques.

  181.  In practice, British American Tobacco has found that lowering the tar/nicotine ratio, by whatever means, does not necessarily support product acceptability. Smokers often do not like cigarettes with a higher nicotine to tar ratio than that to which they are accustomed, and reject cigarettes with absolute higher nicotine yields. The results of one collaborative study with Dr Russell showed that the subjects tended to compensate (that is, smoke more intensively) when confronted with low tar cigarettes, whether or not the nicotine yield was maintained or enhanced (Stepney R, "Would a Medium-nicotine, Low-tar Cigarette be Less Hazardous to Health?" BMJ, 283 (6302) p 1292, 1981). Where cigarettes with the higher nicotine yield were smoked, this resulted in unacceptable levels of nicotine uptake, poor sensory characteristics and a rejection of the product by consumers. The study also suggested that smokers, when they compensated, were compensating for something other than nicotine's pharmacological effects. As Dr Fairweather had reported in 1986, further research on factors other than nicotine pharmacology would be necessary in order to fully understand the phenomenon of compensation. Swann and Froggatt, in their report on the work funded through the Tobacco Products Research Trust, concluded:

"Compensatory smoking" was common in the product modification programme but was usually appreciably less than 100 per cent. It resulted in a smaller reduction in the incidence of smoking-related diseases studied than expected or extrapolated from the cigarette (machine) yields; and reducing the tar/nicotine ratio has an as unyet (sic) fully unresolved role in product modification programmes" (Swann C and Froggatt P, "The Tobacco Products Research Trust", Royal Society of Medicine Press Limited, p 4, 1996).

---

---

Prepared 28 February 2000