METHODS

The data for this paper came from document sites on the world wide web arising from litigation in the USA.¹ Documents came from the primary sites of the manufacturers,¹³ the Council for Tobacco Research¹⁴ and the US Tobacco Institute,¹⁵ and the secondary sites of Tobacco Documents Online (Bliley, BC, Health Canada and Guildford Miscellaneous collections),¹⁶ the Canadian Council for Tobacco Control,¹⁷ the British Columbia Ministry of Health Services,¹⁸ and the Centers for Disease Control and Prevention Guildford collection.¹⁹ The search string “pagewood or moorabbin or granville or australia or sydney or melbourne or brisbane or hobart or adelaide or perth or canberra or amatil or wills” has been used throughout the project to gather documents relating to the Australian market. Of those identified and annotated as revelatory throughout the course of the project, 320 were from the 1980s and related to the issue of smoking and disease causation. This paper presents only a subset of the evidence: all 320 documents are available from http://tobacco.health.usyd.edu.au/site/gateway/docs/research.htm An overview of our approach to document searching and analysis is available elsewhere.²⁰

RESULTS

The industry’s private knowledge

The context

In March 1978, the global BAT group held its research and development conference in Sydney, Australia. The first paragraph of the minutes read:

“there has been no change in the scientific basis for the case against smoking. Additional evidence of smoke-dose related incidence of some diseases associated with smoking has been published. But generally this has long ceased to be an area for scientific controversy.”²¹

By the 1980s authoritative public health pronouncements on smoking and lung cancer were more than two decades old, and the Australian industry was aware of the international consensus. Information was distributed to the local industry by the Tobacco Institute of Australia (TIA), and by INFOTAB, the international tobacco industry’s information management organisation.⁸ Shook Hardy and Bacon, (the industry’s outside legal counsel in Kansas) and BATCo had “data banks ... capable of keeping the industry abreast of developments on smoking issues,” which the Australian industry could access.²² Australian manufacturers also attended joint industry activities such as CORESTA meetings.²³

Any contention that the local Australian industry was ignorant of the 1980s literature on smoking and disease is thus unsustainable. Further documentation of the flow of information into the Australian industry is available at http://tobacco.health.usyd.edu.au/site/gateway/docs/research.htm.

Research done by the tobacco industry

The manufacturers did not rely solely on external research. RJ Reynolds developed bioassays for testing the tumourigenicity of their products,²⁴ produced critiques of unfavourable research,²⁵ commissioned reviews of the published literature,²⁶ and cultivated links to university departments.²⁷ It is highly likely that this work was communicated to RJR in Australia, but no “smoking gun” to prove this has been found in the documents. There is, however, evidence linking PML and Wills to the disease research undertaken by their parent companies.

Philip Morris research

In the 1980s, the Philip Morris Corporation’s in-house research facilities employed eight staff to critique published research and 80 in animal laboratories and cutting edge, cellular level research facilities in Richmond (USA), Cologne (Germany), and Neuchatel (Switzerland).²⁸ Applied product development was emphasised, including work towards low biological activity products. Richmond and Neuchatel provided research support to global companies such as PML. Basic research information from external sources such as the published literature, universities, and biotech companies was organised by PM USA in a confidential database. The PMI corporate affairs department collected “up-to-date smoking and health related information” and circulated it to offices including PML in Australia.^29–31 PML undertook product development research in its Australian headquarters at Moorabbin,^32,33 liaised with PM’s overseas R&D units and received copies of their reports, including reports on biological testing,^34,35 and made visits to the central research facility in Richmond.³⁶

Wills research

The disease effects of smoking were foremost in the minds of Wills’ managers, evidenced by their labelling all corporate affairs reports “Smoking and health,” whether these addressed taxation, advertising, or social acceptability issues.³⁷ The BAT Group was connected by a GR&DC (Group Research and Development Centre) team, an international network of researchers which exchanged information via a programme of regular meetings, correspondence, and visits.^38–43 Wills was a relatively small player in the group, with a local focus.⁴⁴ Nonetheless Australia was considered one of BAT’s six “Major R&D Centres”.⁴⁴ Wills was also a “CAC company”, that is, one of the companies which sent representatives to the BAT Chairman’s Advisory Conference, and thus an active participant in decision making.

The search for a “safer” cigarette and Project Rio

In 1980, BAT executives still hoped that the company’s problems could be solved by the production of a “safer” cigarette. To this end, BAT was engaged in a million pound a year biological research programme. This programme had two arms. The first was concerned with developing tests of mutagenicity, biochemistry, and inhalation which would yield rapid results.⁴⁵ The second was the much larger Project Rio. Rio investigated the biological effects of smoke, and aimed to produce an “acceptable” cigarette which could be demonstrated to lower cancer risk and thus give BAT a marketing edge.^46,47 Wills contributed to Rio but also understood its potential dangers, recommending to the BAT GR&DC in 1983:

“A program has to be developed to handle Project Rio... This general area is an important one for the long term survival of the industry and as such we support it. About 12% of the research effort goes into this area. (It is the second largest area). We would feel this effort is large but reasonable.”⁴⁸

The same year CAC companies like Wills were instructed to concentrate on applied product and process research rather than basic research.⁴⁹ Wills’ responsibilities within Rio were reduced to the testing of local commercially successful cigarette brands for their “relative mutagenicity”.^46,49 Wills evaluated Hallmark, Marlboro, Benson and Hedges, and Winfield, and found that a Wills “low tar” brand, Hallmark, was the cigarette with the highest mutagenic activity—an unfortunately common finding which was discussed at BAT Group meetings and in reports.^39,46,50

Legal restructure of BAT research

Understandably, the group became anxious to distance itself from “dubious past research”, leading to a major restructure of Rio and direct lawyer involvement in “every step” of BAT’s smoking and disease activities.^49,51 By 1985, the “commercial aspects” of Project Rio were pronounced complete, and reports prepared for individual participating CAC companies such as Wills.^50,52 Southampton used Rio data to draw conclusions about correlates of mutagenicity in BAT cigarette designs worldwide, concluding that features such as filter ventilation, still common in “light” Australian cigarettes, increased mutagenicity across BAT brands.^49,52,53

For legal reasons, the drive to restructure BAT research intensified throughout 1985. The CAC laboratories were considered too “independent” and “inconsistent”, and Wills were instructed to clear any BATCo documents “dealing with scientific and other potentially controversial issues” through the groups’ lawyers before release.^44,54–57 In 1985 a new group-wide “research rationale” distinguished “product research (to be done internally)[from] smoking and health research (to be done externally) in future”. Basic research was further de-emphasised.^44,58

The scientific research group to monitor external research

“External” work to “determine the health consequences, if any, of smoking to the smoker” was monitored by a new scientific research group (SRG.) The SRG was established in April 1986, with representatives from CAC companies, to advise BAT companies on the product development and policy implications of external research work.^58,59 Although Wills objected to paying an equal share of the costs of SRG activities relative to other CAC companies, Wills contributed tens of thousands of pounds to SRG monitored work. Wills also attended some SRG meetings.^59,60

SRG members (including Wills) were conscious of their responsibilities. In 1986 they noted:“current [legal] opinion indicates a clear duty to be aware of relevant scientific literature on smoke and health issues, and, if necessary, to carry out appropriate research on the product and/or its effects on the smoker.”⁵⁹ The SRG reviewed the literature, identified gaps, commissioned and monitored external research, directed group companies on R&D issues, and met to hear presentations of the results of commissioned research. Presentations were generally on topics favourable to the industry such as the benefits of nicotine, and thus provided the SRG with “good news”.^61,62

Wills’ internal product research

From 1985 internal product research, focusing on potential regulatory challenges like measuring smoke toxicology, was reviewed to increase international coordination, quality, and reporting.^58,63,64 Southampton “maintain[ed] a knowledge base on factors influencing specific constituents” and “provide[d] background on the opportunities for removing specific constituents”. The role of operating companies such as Wills was to “develop products with reductions in specific constituents” to capture health concerned smokers in a shrinking market.^64,65

Wills remained a relatively minor player in the global research team⁶⁶ but continued to study Australian smokers under the internal research programme. In 1986 “a relationship was established” between Wills and the Southampton GR&DC, “in which equipment for monitoring human smoking patterns was supplied [to Wills] to support the research programme on smoke quality. Data collected [were] transferred instantaneously to R&D [Southampton, from Sydney] for duplication of smoking behaviour and determination of smoke component deliveries.”⁶³ Wills also contributed financially to the centralised GR&DC research programme, although their contribution was smaller due to their relatively lower volume of cigarette sales.^67,68

The BAT research programme winds down

As the 1980s wound down so did the BAT research effort: in 1989 the Southampton laboratory staff was slashed from 500 to 150²⁷ and in 1987 BAT’s lawyers reviewed BAT’s central R&D files containing decades of work, a confidential exercise intended to be protected by legal professional privilege and not disclosed to third parties by R&D staff.⁶⁸ Allegations that BAT engaged in systematic document destruction in Australia to protect themselves from legal action have been made both in the courts and by whistleblowers in the media.^69,70

The industry’s communications

Internal strategies on smoking and disease

The early 1980s was a time of “definite change in the attitudes of the [international] industry, [from]...a generally passive stance to an active approach on smoking issues”.⁷¹ This was prompted by the realisation that their two decades of relative silence on smoking and disease were eroding industry credibility.^72,73 Australian manufacturers were no exception to this change. PML’s corporate affairs department set out to “minimise the impact of unfavourable events arising from the smoking and health ...controversy” via a range of PR strategies including the use of political, advertising, and journalism contacts.³⁰ Wills characterised the Australian industry as having a “policy”, as follows:

“advocat[ing] strongly its right to do business in the market place, that the question of whether people smoke or not is a function of individual adult choice from a range of lifestyle options (of which smoking is merely one) and that further research is needed to help resolve the unresolved questions surrounding smoking and its possible relationship to human health.”³⁷

The 1980s through to the early 1990s was a unique period of intense, aggressive, obdurate denial, in the face of the industry’s aforementioned private agreement that smoking caused illness.

In the 1980s, employees and affiliates at all levels were indoctrinated with the official policy: that disease causation was still an unresolved question.^74,75 There are many examples of internal communications defending smoking. A 1985 document from PML Legal Department files dismissed specific disease claims as unscientific.⁷⁶ At a 1986 Annual General Meeting PML exhorted shareholders:

“the company does not accept that there is overwhelming scientific and medical evidence indicting cigarette smoking as the largest cause of preventable death and disease in Australia... a significant body of eminent medical and scientific opinion does not hold that view... the true causes of the illnesses remain unknown... the tobacco industry in Australia has, through the Australian Tobacco Research Foundation, supported over 100 independently conducted medical research projects in Australian universities and teaching hospitals.”⁷⁷

At a 1984 meeting of the board of BAT Industries, major shareholder in Wills’ parent company AMATIL, directors were instructed:

“[this year] more anomalies have come to light to support the controversy on causation... despite 30 years of intensive research, no compound or group of compounds in cigarette smoke have been identified as having a causal link... As in the past, the only recourse for the industry is to undertake or support more research... possibly during the next five years and almost certainly within 15 years... smoking and health issues will be resolved, one way or the other.”⁷⁸

Into the late 1980s, BAT’s meetings of its regional and national managers featured “scientific” presentations that denigrated “statistics”, highlighted apparent epidemiological anomalies, and rehearsed analogies that undermined the evidence against smoking.^79–81 BAT’s publications for staff, distributed to Wills, included a booklet promoting low tar cigarettes and the company’s research investment, a massive “Issues guide” for corporate affairs management, and the detailed “Compendium of epidemiological studies,” a three volume work for scientists highlighting the “controversy”.^81–85

Internal conflict on smoking and disease

There is extensive documentation of turmoil within the BAT group on the nature of their public stance. In 1980 LCF Blackman, BAT’s head of research in the UK, felt that:

“other companies or industry associations...are...tending to over-state the causation is not proven argument which we believe will not convince any audience—and will probably further reduce rather than raise the credibility of the Industry.”^86,87

BAT’s public affairs, legal department, and R&D personnel worked for a whole year to find “ways of breaking out of the dilemma without prejudicing the legal position [of the industry]”.⁸⁶ This resulted in a proposal for a “new approach of open communication”. The proposal was that BAT should “acknowledge (but...not accept) the strong body of opinion that smoking is associated with risk” and should promote their development of reduced risk products.⁸⁶

However, the presence of an American company (Brown & Williamson) in the BAT group raised the spectre of US legal liabilities and Blackman’s advice was quashed.^86,88 Wills opposed Blackman’s position, at least in part because “parts of Dr Blackman’s paper would appear to conflict with the propositions Widdup [Managing Director and Chairman of AMATIL] put to the Senate Committee [Inquiry into tobacco] five years [before]”.⁸⁹ Questions of product liability were not to become a serious threat in Australia until the end of the decade.⁹⁰

Communications beyond the industry

Australian manufacturers, individually and through the TIA, implemented the policy of misleading the Australian public on smoking and disease. This was done via publicly distributed newsletters, advertising and interviews in news media, self published books and pamphlets, medical conferences, and sympathetic publications such as the Australian Retail Tobacconist. The misrepresentation of the state of medical research was deliberate. The TIA, in its information–management role, sought out the tiny body of work that dissented from medical authority (for example, a TIA staffer’s 1983 request to the US Tobacco Institute for “articles refuting the US Surgeon General’s report or UK Royal College of Physicians report”)⁹¹ and then concentrated these ideas in user friendly form to promote to the public.

The Australian industry had scant regard for science when formulating these publications: in fact, twice during the 1980s even BAT was outraged by the inaccuracy of the Australians’ claims. The TIA’s 1983 pamphlet “The smoking and health controversy...why more research is needed”, sent to parliamentarians, libraries, and tobacco employees,^92,93 was strongly criticised by BAT’s researchers in Southampton. They called it unbalanced, simplistic, superficial, and unconvincing, noting that it conflated evidence and opinion and ignored the published literature.⁹⁴

In 1985, the BAT Group were concerned about a Wills position paper which proposed that “people dying from smoking-associated disease do so at a later average age than those from non-smoking-associated diseases”.^95,96 Their statistical consultant reviewed the paper and advised:“l am afraid that this paper is very naive statistically and the conclusions and implicit inferences to be drawn from them are in many cases unsound”.⁹⁷ He called it “deliberately antagonistic”, and noted that the writer did not seem to understand the points that he or she was making.⁹⁷

There are many examples of the Australian industry publicly contradicting the scientific consensus. The TIA published a press advertisement in the Australian Financial Review entitled “Smoking. Let’s be sensible about it”,⁹⁸ which insisted that tobacco control advocates were “imply[ing] scientific certainty where none exist[ed]...[and] simplistically...blam[ing] cigarette smoking for various diseases and deaths”.⁹⁹ John Dollisson (TIA) stated in 1986:

“It is misleading to attribute even one death to smoking... You are talking about poorly constructed studies conducted in America 20 odd years ago. They have been extensively criticised by scientists the world over as not being representative of America at the time. How could they possible accurately represent Australia today? The smoking and health controversy should stay with the scientific community until resolved one way or another. The only appropriate role for the Government is to inform people of the medical and scientific facts.”⁷⁵

Geoff Bible, then managing director of PML, promoted the policy in a speech to journalists where he encouraged smokers in the room to “take heart [because] your lungs are as pink and pretty as those of the nonsmoker sitting next to you”. Bible claimed that “thirty years ago, even twenty years, it was possible for an honest man to believe that smoking was the cause of all the ills that flesh is heir to”. However, he argued, after spending $100 million on research, the tobacco industry had not been able to find the “one ingredient” to blame that could be removed.¹⁰⁰

AMATIL (Wills) were not only true to the policy of claiming more research was needed in the 1980s, they were happy to publicly concede that the industry had jointly agreed to that position. Duncan Fairweather argued on Australian radio in 1984:

“our position on health is quite clear... we think there are some questions to be raised about the statements that are made by the medical authorities, in the meantime there’s a lot more research needed. That’s a standard position that the industry has adopted and everyone is well aware of it.”¹⁰¹

Many other examples of the Australian industry’s consistent and aggressive support of the policy are available from http://tobacco.health.usyd.edu.au/site/gateway/docs/research.htm, including a 30 minute radio sound file featuring John Dollisson, the head of the TIA, responding to radio listeners’ questions in Sydney in 1984.

Attacks on accurate information

In addition to actively misleading Australians on disease, the industry opposed what it termed “anti-smoking propaganda”.¹⁰² The industry instituted a “policy of vigilance against attacks by its opponents, while taking further constructive initiatives to place them on the defensive”.¹⁰³ Generally the industry’s attacks demanded that messages be removed or demanded statistical or clinical substantiation of claims.

These actions were designed to intimidate health authorities, advocates, researchers, and editors and ultimately to deny Australian consumers access to accurate information. They were conducted by the TIA or its lawyers, Clayton Utz, were sometimes supported by Don Hoel of Shook Hardy and Bacon or by PML, and sometimes used third party strategy via advertising industry contacts. Public statements by individuals as well as government health promotion campaigns were targeted, directly and through regulatory bodies including the Advertising Standards Council, Federation of Australian Commercial Television Stations, Media Council of Australia, and the Australian equivalent of the US Federal Trade Commission, the Trade Practices Commission, which received a series of complaints alleging that consumers were being misled by statements on smoking and disease.

In some instances the industry’s objectives were achieved: the offending messages were withdrawn by the self regulatory agency in question.¹⁰⁴ The industry even considered taking action against advertisements for nicotine chewing gum¹⁰⁵ and in 1987 attacked the Medical Journal of Australia regarding the content of published research articles,¹⁰⁶ a paradox considering the industry’s continuing protestations about the need for continuing medical research. Further evidence of industry’s monitoring of tobacco control and counter-attacks is available at http://tobacco.health.usyd.edu.au/site/gateway/docs/research.htm.

Impact on public opinion

Throughout the 1980s the TIA continually monitored changes in public opinion about smoking.¹⁰⁷ Despite their efforts, opinion from 1983 to 1990 showed changes unfavourable to the industry:

• “smoking is bad for you even if you smoke only a few a day”: increase in agreement from 75% to 83%

• “the medical profession is correct in everything it says about smoking being a cause of lung cancer, heart disease and other health problems”: 75% to 80% agreement

• “the medical profession has exaggerated the risks of smoking”: decrease in agreement from 28% to 21%

• “smoking helps some people handle situations in which they experience stress”: decrease from 72% to 63%

• “people are sufficiently aware of health concerns and further government sponsored anti-smoking and health advertising would be a waste of public money”: 40% to 45% agreement

• “people should be constantly reminded that smoking is harmful”: consistent agreement around 80%

These results demonstrate that the industry’s aggressive public communications did not change majority public opinion. They also show that the industry enjoyed continued support of a diminishing minority of Australians on most issues, a minority which may in fact have consisted of smokers seeking reassurance.

DISCUSSION

We have provided evidence that, in the 1980s, Australian cigarette manufacturers were continuously made aware of the international consensus on smoking and disease via both the TIA and their parent companies. Australian executives attended meetings where the diseases caused by tobacco were openly discussed, and had the research literature digested and delivered to them. PML was informed on PM’s international biological research programme and visited PM’s Richmond research facility. Wills helped fund and manage the BAT biological research programme, tested the mutagenicity of Australian cigarettes before such activity stopped in 1985, and continued to study human smoking patterns and smoke component deliveries in Australian smokers beyond 1985.

Despite this knowledge, the Australian manufacturers’ policy throughout the 1980s was to argue that the question of a relation between smoking and disease was unresolved and to advocate their “right” to do business and smokers’“right” to smoke. Internal and external communications conformed to this policy. Considerable resources were devoted to cataloguing the research literature, which overwhelmingly implicated tobacco as disease causing. Further resources were then committed to designing formal responses which sought to deny the case that had built against tobacco. So eager was the Australian industry to deny the disease effects of tobacco that at times BAT contended that they did not understand the science of their product. The industry also attacked public health advocates making statements on the relationship between smoking and disease.

It is clear to any layperson that these are morally unjustifiable acts. They are also unjustifiable even by the standards of the tobacco industry’s own culture: business. The Australian Chamber of Commerce and Industry (ACCI) is the peak council of Australian business associations, and the conservative voice of business in Australia. The industry’s actions clearly defy the ACCI policy on consumer protection, which states that “consumers ...have the right to expect...safety standards from goods and services purchased [and] protection from misleading and deceiving practices...”.¹⁰⁸ Given the extent and deliberateness of the industry’s activities, it seems appropriate that the executives, the directors, and the lawyers employed by Wills and PML from the 1980s and early 1990s should be investigated with respect to their possible liability under Australian law.

It may be that the industry’s efforts were not highly effective in the general community. Nigel Gray, Australia’s leading tobacco control advocate in the 1980s, has characterised the industry as largely ineffectual in public debates,* and the lacklustre results of the industry’s public opinion polls support his views. Newspaper editorials advocating for continued tobacco advertising at the time frequently qualified their arguments by stating that smoking certainly caused disease. The Advertising Age editorialised in 1983 that “the industry’s credibility began its nose dive when Phil Scanlon, a spokesman for the major Australian tobacco company AMATIL Ltd., said on national TV that the tobacco industry believed smoking did not constitute a health risk” and noted that “the reaction against the ad [Smoking: Lets be sensible about it] was enormous. Federal and State health ministers condemned the industry’s attempts to deny that 16,000 Australian die every year due to smoking-related diseases and questioned the misleading presentation.”¹⁰⁹ According to an advertising industry spokesperson “Smoking: Lets be sensible about it” was a “poorly conceived and executed non-event” which did the industry more harm than good.¹⁰⁹

However, when the TIA’s Brian Gapes defended the production of the “Let’s be sensible about it” advertisement, he claimed “a lot of smokers applauded our action”.¹⁰⁹ In 1984, Andrew Whist of PMI wrote with regard to Australia that “antismoking propaganda” was starting to persuade “even smokers” that it was an “undesirable habit”.¹¹⁰ This is an important distinction, particularly given that the industry knew that committed smokers were more likely to support them on crucial policy issues. John Dollisson (TIA) argued in 1986 at an international workshop:“Smokers and non-smokers are not homogeneous groups when responding to industry issues. The stronger the smoker’s commitment to smoking the stronger their support on issues.”¹¹¹

The main objective of the industry’s aggressive communication on smoking and disease may have been to maintain the trust of smokers already predisposed towards the industry’s position, thus providing their consumers with reasons to doubt the “anti-smoking propaganda”. Australian smokers are more likely than non-smokers to agree with statements like “the medical evidence that smoking causes cancer is not convincing” or “many people who smoke live to a ripe old age, so smoking is not all that bad for you”,^112,113 positions constantly reiterated by the industry throughout the 1980s. The industry’s opinion poll data suggested that a slowly decreasing minority continued to agree with their views in a decade when Australian male smoking prevalence reduced from 40% to 30%, and women’s from 30% to 25%.

It seems probable that a diminishing cohort of committed smokers were responsible for the diminishing 20–30% of agreement with the industry in their population based opinion polls. Industry advocacy may have allayed smoker’s latent concerns, provided seemingly plausible rationalising life rafts, and ultimately, reassured them in their decisions to keep smoking. Today’s sick and dying smokers, without any advanced epidemiological knowledge or skills in the critical appraisal of evidence, who now say “I heard lots of doctors and professors on the radio saying that there was actually no proof that smoking caused cancer” cannot be deemed to be simply ignorant or ill informed. A natural policy outcome of this work, then, is that smokers who were the recipients of the industry’s policy of obdurate denial deserve to know the truth, not only about the disease effects of smoking, but about the efforts of the industry to deny these effects in the 1980s. Smokers have a right to access the information presented here, ideally via a mass media public information campaign.

The industry’s cynicism towards smokers became apparent in the 1990s, as product liability litigation in Australia became a serious cause for concern. Overnight, the industry stopped arguing that there was no proof that smokers were at risk, and instead advised smokers that they should have listened to the public health authorities, and the responsibility for their illness was theirs. The industry’s primary motivation in Australia in the 1980s, as ever, was not the pursuit of truth or science, or the wellbeing of its customers, but its own financial survival. Despite privately accepting that the case for smoking causing disease had long been established, the industry vigorously denied this via its public affairs activities. The revelations we have summarised provide important evidence for litigation by smokers exposed to such industry campaigns, and for advocacy against industry mendacity.