Associations between universities and the tobacco industry: what institutional policies limit these associations?

Thomson, George

Abstract

This paper examines the extent of associations between the tobacco industry and New Zealand universities, and the institutional mechanisms that have been used to limit such associations. Tobacco industry documents were searched for associations between New Zealand universities and the tobacco industry. The stratagems used by New Zealand universities, funders, professional societies and government to limit such associations were analysed, using written requests, website surveys and interviews. Philip Morris invested at least US$790,000 into research at the University of Auckland during 1988-1996, and other associations between tobacco companies and New Zealand universities have continued until at least 2004. There are still few formal policies in New Zealand to prevent such associations. In contrast, a number of prominent Australian universities formally limit their associations with the tobacco industry. If the evidence of harm to the public interest from associations with the tobacco industry is accepted, then, despite the risk to academic freedom, formal policies to address such associations may be warranted. To be most effective, policies by research institutions and funders on tobacco industry associations should be formal and explicit, and also need to be comprehensive and effectively implemented.

INTRODUCTION

This study explores the institutional mechanisms that have been used to limit associations between the tobacco industry and New Zealand universities. Our research is premised on the view that such associations are potentially unsafe. A substantial literature from 1982 has found dangers to public health and to the public interest from associations between universities and the tobacco industry ("Smoking still kills" 1982, Fischer and Richards 1986, Pierce 1986, Shaw 1986).

The potential dangers include giving respectability to the industry, the tendency to remain silent about industry behaviour and tobacco harm, the active perversion of research processes by the industry, and the diversion of public, scientific and government attention from tobacco harm ("Smoking still kills" 1982, Wolinsky 1985, Fischer and Richards 1986, Pierce 1986, Shaw 1986, Chapman 1987, Warner 1991, Cohen et al. 1999, Bitton et al. 2005, Garne et al. 2005). The documented perversions by tobacco companies--individually or collectively--of the pursuit of truth include the covert control of academic journals (Bitton et al. 2005, Garne et al. 2005), the manipulation of research processes and arenas (Barnes and Bero 1998, Ong and Glantz 2000, 2001, Tong et al. 2005), the suppression of results (Diethelm et al. 2005), and the attempted corruption of research and health organisations (Zeltner et al. 2000, Yach and Bialous 2001).

One of the most obvious potential dangers of association has been the implied support of an industry that has denied and deceived about the harm from its products. Since 1964 or before, suggestions that there was a scientific "controversy" as to whether or not this harm existed have been very largely driven by the tobacco industry (Doll 1998, Hill et al. 2003, Parascandola 2004, Proctor 2004, Talley et al. 2004).

The literature on such dangers is now supported by wider research that indicates adverse consequences from financial and other associations between researchers and the commercial funders of research whose activities are related to the research area. These consequences can include lower-quality, fewer and more biased publications (DeAngelis 2000, Lexchin et al. 2003). There appears to be no immediately obvious reason why these dangers should not apply to New Zealand. Tobacco industry associations with universities occur in the context of growing concerns about the conflicts involved in business-research links (Cho et al. 2000, Morgan et al. 2000, Cech and Leonard 2001).

Worldwide, universities have been slowly developing defences against the perceived dangers to the public interest from associations with the tobacco industry. In 1982, the University of Sydney in Australia adopted a policy refusing support from the tobacco industry (Miller 1982). However, in much of the academic world little or no action was taken until the 1990s, even in medical faculties (Blum 1992, Walsh et al. 1994, Lewison et al. 1997, Spurgeon 2002). A number of universities and research-funding agencies in North America, Britain and Australia now have policies limiting funding of research by the tobacco industry (Cohen 2001). Because of the possible dangers to the public interest, we examined the extent of such associations in New Zealand, and the institutional mechanisms that have been used to limit associations between the tobacco industry and New Zealand universities.

The consequences of associations between tobacco companies and universities can be seen as erosions of the public interest. For this paper, "public interest" has been defined as "an approach that serves society as a whole, is focussed on the longer term and is not solely in the service of special interests" (Pearson 2001). A public interest approach, in this context, would thus look at the societal implications of actions by universities.

This research has been informed by institutional theory, which argues that policy making is often shaped by the nature of the institutions involved. The theory is concerned with the formal and informal policy mechanisms and embedded ideas of institutions, such as the rules, processes and structures that frame the policy possibilities within institutions (March and Olsen 1996).

METHODS

Between May 2001 and May 2002, all available tobacco industry internal documents concerning New Zealand were collected from the United States Master Settlement Agreement websites (Master Settlement Agreement no date) by the University of Sydney tobacco document research team (University of Sydney no date). To explore aspects of the relationship between New Zealand universities and the tobacco industry, this set was searched for documents relating to universities and research. The material was supplemented by searches on the Tobacco Documents Online website, requests under the Official Information Act and searches in the secondary literature.

To understand the way such associations between the universities and the tobacco industry are handled, and to explore more widely the protections of the public interest, data were also assembled on the policies of New Zealand research funders, university and funder ethics committees, scientific and professional societies, universities, and government.

In May-July 2004, a search was made of policy documents of the eight New Zealand universities, five New Zealand research funders, a research ethics body and two professional societies for their formal policies about the protection or enhancement of the public interest. To provide a comparison, the results were contrasted with information found in an October 2003 search of the websites of 43 Australian universities, which were searched for policies about associating with the tobacco industry, using the phrase "tobacco industry". The universities were those listed by the Australian Department of Education (Department of Education Science and Training 2003).

In examining the policies of funders, we chose the major relevant agencies in New Zealand. These were the Foundation for Research Science and Technology, Health Research Council, Heart Foundation and Cancer Society, and the research ethics committees organised by the Ministry of Health. Much of the health research funding in New Zealand requires approval by these ethics committees. The Royal Society of New Zealand is both a substantial research funder (through the Marsden Fund) and the pre-eminent scientific professional society for New Zealand.

To provide additional context, telephone interviews were conducted with two New Zealand university research administrators, and with four of the six scientists funded at the University of Auckland by the Philip Morris tobacco company. Structured interview formats were created for each of the two groups of interviewees, whom we agreed not to identify by name or current institution. The interviews took 25-70 minutes, and the transcripts and notes were thematically analysed. The research plan was approved by the University of Otago Human Ethics Committee.

The Limitations of the Research

The relative dependence on website-sourced documents means that some of the relevant formal organisation policies may not have been found. A wider and more representative selection of interviews with university and funding administrators could have provided a much richer context within which to check the documentary material.

RESULTS

Relationships between the Tobacco Industry and New Zealand Universities

The study found evidence of only one major project at a New Zealand university funded by the tobacco industry. This work, at the University of Auckland, was part of a wider Philip Morris plan during 1988-1996, called Project Cosmic (Philip Morris 1991, Mangan 1994). Two key objectives of Project Cosmic were to facilitate research and publications that might suggest that smoking has benefits, and to keep up with the "changing scientific and public policy environment" by the development of a network of experts (Philip Morris 1990).

In May 1988, the University of Auckland signed a contract with Philip Morris for a three-year research project (Ennis 1992). The project was for three experiments to further examine the theory that nicotine enhances "psychological comfort" and "mental and psychomotor performance" (Mangan 1987, Bergquist and Houghton 1988). The project was renewed (through the University of Auckland company UniServices Ltd) in 1990, 1992 and 1994, with a total funding of at least US$790,000 during 1988-1996 (Philip Morris 1991, 1996, Mangan 1994).

From 1992, the University of Auckland project workers published a number of articles about this research in peer-reviewed journals (Colrain et al. 1992, Bates et al. 1994, 1995, Stough et al. 1994, 1995). The articles included suggestions that nicotine via smoking might be beneficial for memory consolidation in some circumstances, and that this nicotine was associated with reduced decision time and inspection time, and improved intelligence test scores. The direction of the research which highlighted apparent "benefits" was at odds with work elsewhere that focused more on the increased stress for smokers created by nicotine addiction (Cohen and Lichtenstein 1990, Pomerleau and Pomerleau 1990, Tate et al. 1994, Parrott 1995a, 1995b, 1998, West and Hajek 1997).

From the interviews, there appears to have been a range of opinions within the University of Auckland about the research funded by Philip Morris. On the one hand, there were benefits to the University: "a lot of people were very happy that the research came to Auckland. It left a well established laboratory, a stack of publications, students got publications that helped them then go and get jobs elsewhere." On the other hand, there was perceived to be pressure from parts of the University of Auckland, including the medical school, not to be associated with a tobacco company.

Other relationships between New Zealand universities and the tobacco industry were also found. At the University of Auckland a statistician was employed by both the University and the Tobacco Institute of New Zealand during the late 1980s and early 1990s (Owen 1993). A University of Auckland anatomist commented for Philip Morris on the 1997 California Environmental Protection Agency paper about tobacco smoke pollution, and for the Tobacco Institute of Australia on a similar Australian paper (Tobacco Institute of Australia 1996, California Environmental Protection Agency 1997). A University of Canterbury economist spoke at a Philip Morris Seminar on Taxation, in Washington DC in 1986 (Woodside 1986). A University of Otago marketing academic in 2004 prepared a paper on health warnings for Imperial Tobacco (Todd 2004). At the Victoria University of Wellington, a very senior academic has, through his law firm, acted for the industry at least since 1995 (Curry 1995, Laugesen and Maling 2002, Victoria University of Wellington 2005). Part of his attractiveness to the tobacco industry in 1995-1996 was his academic standing (Tunstall 1995, Thompson 1996).

Institutional Mechanisms Limiting Association between Universities and the Tobacco Industry

What formal or informal institutional mechanisms might have helped avoid such associations as that between Philip Morris and the University of Auckland? The active use of ethical codes and the development of professional virtues have been suggested as solutions to the potential issues (Edwards and Bhopal 1999). Particular ways by which universities and others can influence such associations include:

* policies about the disclosure of financial or other associations with particular types of organisations

* policies that limit such associations

* supplying information to researchers and funders about the possible consequences for them, for the university and/or for society from such associations

* the encouragement of professional virtues by incentives and leadership examples.

Below we outline the more prominent of these mechanisms that we found.

Formal University Policies

We found no statements in New Zealand university policies about associations with the tobacco industry, and a general lack of specific statements recognising a duty to protect the public interest. The focus in research policy statements was on individual human rights rather than on the societal impacts of research. A major exception was in the charter developed by the University of Auckland in 2003. This recognises that the "research undertaken in the University and disseminated within the wider community has important cultural and social outcomes." One of the duties of the university council is to be concerned "for the public interest" (University of Auckland 2003).

In contrast to the New Zealand situation, a number of Australian universities have specific statements about aspects of associations with tobacco companies. Policies about tobacco industry funding were found at 14 of the 43 university websites searched. These include the Australian National University and the Universities of Western Australia, South Australia, Adelaide, Melbourne, Sydney and Newcastle. The most comprehensive statement found was that in the academic policies of Edith Cowan University, West Australia. The policy was to not "do or allow anything which may directly or indirectly encourage or support the use of tobacco products" or "enter into any association or other arrangement with any person known to be involved in or connected with the tobacco industry" (Edith Cowan University 2001).

Ethics Committees

The Health Research Council guidelines for ethics in health research are focused on the individuals or populations directly involved in the research process, rather than the effects of the research process on society. The types of research needing ethical approval are stated as those "using animal or human participants, animal or human materials, personal information, or involving clinical trials" (Health Research Council of New Zealand 2002). Likewise, university ethics committees tend to focus on individual research participants rather than societal consequences. Where groups and communities are considered, it is as sites or subjects for research.

The ethics committee process of one university does appear to recognise the need to be aware of the possible societal implications of research processes and outcomes, but only in a very general way. The Auckland University of Technology requires researchers to "be cognisant of potential implications or interest that the process and outcomes of the research might have for other cultures or groups" (Auckland University of Technology Ethics Committee 2004).

The Formal Policies of Funders and Professional Societies

Except for the Cancer Society, we found no policies by the New Zealand research funders that had specific reference to tobacco industry associations. Of the policies studied, only that of the Royal Society demonstrated a move to include explicit consideration of the public interest in their code of ethics (Royal Society of New Zealand 2003). Since the early 1990s, the Cancer Society has had a policy to "not knowingly employ any outside agency that works in association with the Tobacco Industry" (Cancer Society of New Zealand 2002).

Of the relevant professional societies, the New Zealand Psychological Society also moved its stance over the last two decades. Its 1986 code of ethics focused on responsibility to individuals (clients, students, research subjects) (New Zealand Psychological Society 1986). A new code formed in 2002 by the Psychological Society and the New Zealand Psychologists Board has a new section on "social justice and responsibility to society" (Code of Ethics Review Group 2002). However, the implementation of any such code depends on the researchers concerned being members of the professional body.

Informal Mechanisms

From the interviews, we found that there were a number of effective informal methods of reducing the risks to the reputations of universities, and to the public interest, from associations with tobacco companies. These included the establishment of perceptions that tobacco industry funding would not be acceptable, the processes involved in research proposal approval, and collegial pressures.

The perception of some of the nicotine research team who had been at the University of Auckland was that New Zealand and Australian universities would now refuse a research proposal funded by a tobacco company. One university administrator who was interviewed recounted the belief of senior colleagues that a policy of not accepting tobacco industry funding was in place, despite the lack of a written policy.

Institutional processes may lend themselves to informal safeguards. One administrator suggested that the multiple levels of approval needed for each research proposal, from university department to university senate, helped to ensure that through the process considerations about the public interest would be applied. This "safeguard by process" appears to require active scrutiny at each stage, rather than "rubber stamping."

Other safeguards suggested in the interviews included the collegial pressures within universities for agreed behaviour. These pressures could prevent proposals that might be controversial from getting to the formal proposal stage, because of the risk of wasted unpaid time by funding applicants. Some interviewees mentioned the perceived pressure on research institutions, from health-focused non-government organisations that funded research, not to accept any funding from the tobacco industry. It was suggested by an ex-University of Auckland scientist that "people in the university that receive a lot of medical research funding" would "put pressure on the people above them not to let the university jeopardise that".

A further form of pressure that was suggested was the formal and informal ties to overseas academic groups. These academic groups include universities, university associations, and international professional and academic disciplinary societies. For instance, an association with the tobacco industry by a New Zealand university could influence their recruitment of international and local staff and students.

The Ideas Underpinning the Debates about Tobacco Industry Associations

Two sets of conflicting ideas underpin discussion about institutional mechanisms to limit association between universities and the tobacco industry: those relating to "the public interest" and those about "academic freedom". Three interviewees commented on this conflict, seeing it as inherent in any policies that restricted funding from particular sources. A university administrator argued that the principle of academic freedom meant that a university cannot say what particular sort of funders should be used--it can only decide on each research proposal as it is presented. A number of interviewees were concerned that rules about association with outside groups would affect the nature of the research done. One theme from the scientists interviewed was the need for "some level of divorce in the research debate [between the] research that gets done, and societal pressures on that research". If universities did not allow that space "then science becomes subject to the whimsy of the different societal views."

Universities are naturally wary of any suggestion that there be limits on the subjects or methods of research. The University of Otago policy states that "the University will jealously protect an environment in which members of the academic community can question and test received wisdom, put forward new ideas and state controversial or unpopular opinions" (University of Otago 2004).

Controls on associations between research and the tobacco industry were seen by interviewees as part of the wider problem of "who should control research--society or researchers?" One solution offered to this question by a scientist was that "if society was going to have more of a say about what sort of research was going to be done, [universities] need to have better mechanisms to inform them [society] of exactly what we're doing and why we're doing it."

DISCUSSION

The Advantages to Philip Morris of Association with the University of Auckland

The research at the University of Auckland can be seen to have helped Philip Morris achieve two of the specific objectives of Project Cosmic--to get scientific research on smoking "benefits" published, and to maintain contact with the science and policy environment. Because of the differences of opinion among scientists about the cognitive effects of nicotine (Turner and Spilich 1997), funding research in this area also fitted with the tobacco industry strategy of maintaining "controversies" about tobacco and health (Warner 1991, Saloojee and Dagli 2000, Yach and Bialous 2001). These benefits are in addition to the general benefits to tobacco companies from associations with universities--which include access to expert networks and the chance of an increased public perception of companies' legitimacy.

Institutional Mechanisms

There is a range of institutional mechanisms available to protect the public interest from the dangers of the tobacco industry. The institutional mechanisms used in New Zealand have changed over time. Between the 1980s and 2003, the formal policies of one relevant organisation, the Cancer Society, have moved to include explicit mention of tobacco industry associations. The policies of the Royal Society, Psychological Society and the University of Auckland have moved to include explicit consideration of the public interest.

However, the evidence indicates that policy changes within New Zealand since the 1980s on associations with the tobacco industry have largely occurred at the informal level. There seem to be no formal policies that would ensure that universities would not be associated with the tobacco industry: neither specific policies about such associations nor general policies that are intended to protect the public interest. The absence of such formal policies may allow some associations with the industry to persist.

Some universities and funders have general statements about "ethical practice" or "social viability". The problems with such statements are that they are vague and allow scope for interpretation. For instance, how much argument and evidence would be needed to make a link between apparently innocent research funded by a tobacco company, and damage to "social viability"?

The reasons for the lack of formal policies may include:

* the risk to academic freedom created by such explicit policies protecting the public interest

* the conceptual leaps required to directly link associations with the tobacco industry with damage to the public interest

* the blinkering consequences of the specific functions of organisations, where any effects of associations with the tobacco industry are considered to be outside the organisation's mandate

* a perception that the effects of these associations are so nebulous that they do not require action

* a perception that the chance of such an association occurring is so low as to not require action

* the lack of distinction by policy makers between the tobacco industry and other businesses.

Because the costs from the association of organisations with the tobacco industry are often widely diffused, it is difficult for those outside the health sector (and for many within it) to see those associations as a significant issue.

Academic Freedom and Policies about Tobacco Industry Associations

Academic freedom consists of freedoms of speech, publication, inquiry and study (Gillin 2002). It is tempered in New Zealand and other countries by stipulations on students about entry qualifications and courses of study, and on staff to present opposing viewpoints in class, to keep to a prescribed syllabus and to carry out academic duties. An allied concept is institutional autonomy, where universities have freedoms to decide on staff, student entry, courses and standards. However, there are matters "necessary to the good functioning of a university as an institution, that should not be decided by each individual academic at his [or her] own discretion" (Kemp 2000).

Thus staff commitment to the purposes of their university may be at odds with their desire to obtain funding or to do private work. This is a well-established tension in the university decision-making process.

 Any regulation of a faculty member's professional life thus
 runs counter to the general thrust of academic freedom ... The
 precise location of the dividing line between freedom and
 obligation in various situations is established by prevailing
 practice, and can be expected to change with time ... Since a
 conflicts policy will of necessity forbid or limit some
 professional activity, a tension between such a policy and
 the principle of academic freedom can be anticipated. (Stein 2002)

One may argue, however, that policies restricting association with the tobacco industry are no attack on academic freedom. Because tobacco companies, individually and collectively, have actively perverted the pursuit of truth in universities and research-based organisations in a number of ways (Bitton et al. 2005, Garne et al. 2005, Barnes and Bero 1998, Ong and Glantz 2000, 2001, Tong et al. 2005, Diethelm et al. 2005, Zeltner et al. 2000, Yach and Bialous 2001), associations with them would appear to offer the graver risks to the quest for knowledge and understanding. Researchers should be able to publish results independently of their funders' interests, but commercial funding of academic researchers may create bias (DeAngelis 2000, Lexchin et al. 2003). In particular, funding by tobacco companies has been associated with apparent consequent bias by systematic review writers (Barnes and Bero 1998).

Policy Implications for Associations with the Tobacco Industry

This research suggests that health and equity advocates should consider arguing for the adoption by universities of policies against associations with the tobacco industry that are formal, explicit, comprehensive and effectively implemented. Advocates could also argue that research funders, professional societies and governments should adopt similar policies. Formal policies may have more permanence, visibility and enforceability than informal ones. They also avoid questions of covert and/or unjust processes.

Funding agencies can require that the organisations they fund should not be associated with the tobacco industry. Governments can require that those who take government business should not also work for the tobacco industry. Universities can require that their staff not be associated with the industry. At a more fundamental level, the development of an awareness of the societal impacts from such associations could have greater priority in the education of researchers and policy makers. Advocates of such policies need to be mindful of the counterargument about the limits to academic freedom from policies that protect the public interest.

Designing and Implementing Policies

Universities have a general choice between policies that cover some particular points about their relationships with the tobacco industry, or statements that endeavour to be more comprehensive, such as the aforementioned policy laid out by Edith Cowan University. Examples of "particular" policies are those that ban the acceptance of funding from the industry. For instance, the University of Adelaide "will not accept research funding offered to the institution or to individual researchers by the tobacco industry" (University of Adelaide 1997).

Comprehensive policies can help to deal with wider issues that arise, beyond direct funding. Three such issues are:

* association via the sharing of staff or governance figures who work for the industry in their private time

* association through the use of contractors, consultants, law firms and other service providers who also work for the industry

* unpaid cooperation between universities and the industry (such as allowing the industry to recruit on campus).

There are a number of potential implementation issues for such policies. For both particular and general policies, a degree of proactive screening by a university may be necessary, because funders and others may not declare their tobacco industry associations without direct questions (and even then might not). Another issue is the definition of "tobacco industry". A number of Australian universities define a tobacco company in a similar way to the Australian National University, as one "engaged directly in the production, manufacture, distribution, promotion or marketing of tobacco or tobacco products as its primary business" (Australian National University 1998). Australian National University gives discretion to their Vice-Chancellor on the decision of what is "primary". Several also include organisations funded by the industry within their definitions (Australian National University 1998, University of South Australia 1999). Again, a comprehensive approach could help avoid the need to widen definitions, when indirect funding or association is discovered.

Wider Implications

The conflict between the public interest and academic freedom can be seen in a wide range of research activities. The spectrum of relationships between universities and commercial organisations ranges from collaboration within fixed terms, the sharing of staff, contracts for services and training, payments for patent use, through to joint ownership of companies (Howard Partners Limited 2001). Common to many of the relationships are the possible tensions between private profit and public information (Cho et al. 2000, DeAngelis 2000, Cech and Leonard 2001). There are potential conflicts when research funders or others associated with a university have objectives that are in opposition to those of the wider society. There are also potential conflicts when research funders or associates have objectives that are contrary to those of universities (DeAngelis 2000, Lexchin et al. 2003).

The conflict between research values and business needs is also, of course, seen in a wide range of other research organisations besides universities, and it is because of this that scientific and professional groups worldwide have considered avenues to address these conflicts. An analysis of proposed scientific or engineering oaths found that a common theme of recognition of the "responsibility for the societal consequences of research", which was included in over half the oaths (Iverson 2001).

Conclusions

New Zealand universities do not appear to have formal policies to limit associations with the tobacco industry. Informal mechanisms, where present, have at times been ineffective. What institutional policies will best limit associations between the tobacco industry and universities? Informal mechanisms such as research approval processes and collegial pressures are important. However, despite the risk to academic freedom, we suggest that if the evidence of harm to the public interest in the case of the tobacco industry is accepted, then formal policies to address such associations are also warranted. We suggest that to be most effective, policies on institutional associations with the tobacco industry need to be not only formal and explicit, but also comprehensive and well implemented. The policies of Edith Cowan University (2001)--to not "do or allow anything which may directly or indirectly encourage or support the use of tobacco products" or "enter into any association or other arrangement with any person known to be involved in or connected with the tobacco industry"--could serve as a model for many institutions.

(1) Acknowledgements

This study was assisted by funding from the New Zealand Heart Foundation, the Wellington Division of the Cancer Society, a University of Otago Bridging Grant to George Thomson, and the National Institutes of Health (United States) grant #1 R01 CA87110-01A1 to the University of Sydney. The comments of Alistair Woodward, Philippa Howden-Chapman and Louise Delany on a draft are much appreciated. The work on the document references by Fiona Byrne of the University of Sydney was of great help.

REFERENCES

Auckland University of Technology Ethics Committee (2004) Ethics Committee Guidelines, Auckland University of Technology, Auckland, www.aut.ac.nz/research_showcase/pdf/guidelines.pdf [accessed 14 July 2004].

Australian National University (1998) Tobacco Industry, Funding from, January, Australian National University, Canberra, http://info.anu.edu.au/policies/Policies/Research/Other/ Tobacco_Industry_Funding_from.asp [accessed 14 March 2005].

Barnes, D. and L. Bero (1998) "Why review articles on the health effects of passive smoking reach different conclusions" Journal of the American Medical Association, 279(19):1566-1570.

Bates, T., G. Mangan, C. Stough et al. (1995) "Smoking, processing speed and attention in a choice reaction time task" Psychopharmacology [Berl], 120(2):209-212.

Bates, T., O. Pellett, C. Stough et al. (1994) "Effects of smoking on simple and choice reaction time" Psychopharmacology, 114(2):365-368.

Bergquist, P. and K. Houghton (1988) Research Funding Agreement Between University of Auckland and Philip Morris USA, 31 May, Bates Nos. 2022884812-4815, http://legacy.library.ucsf.edu/tid/mfc78e00 [accessed 23 July 2004].

Bitton, A., M.D. Neuman, J. Barnoya et al. (2005) "The p53 tumour suppressor gene and the tobacco industry: Research, debate, and conflict of interest" Lancet, 365(9458):531-540.

Blum, A. (1992) "Ethics of tobacco-funded research in US medical schools" Tobacco Control, 1(4):244-245.

California Environmental Protection Agency (1997) Health Effects of Exposure to Environmental Tobacco Smoke. Appendix B: Summary of Public Comments and Responses on the 970200, Final Report, Environmental Protection Agency, Office of Environmental Health Hazard Assessment, Sacramento, California, Bates Nos. 2062614445-4705, http://tobaccodocuments.org/ness/50546.01.html [accessed 15 July 2004].

Cancer Society of New Zealand (2002) Cancer Information: Tobacco Control [position statement, December], Cancer Society of New Zealand, Wellington www.cancernz.org.nz/[accessed 23 July 2004].

Cech, T. and J. Leonard (2001) "Science and business: Conflicts of interest--moving beyond disclosure" Science, 291(5506):989.

Chapman, S. (1987) "Tobacco industry health research 'blood money': The British Health Promotion Research Trust" Community Health Studies, 11(2):139-142.

Cho, M., R. Shohara, A. Schissel et al. (2000) "Policies on faculty conflicts of interest at US universities" Journal of the American Medical Association, 284(17):2203-2208.

Code of Ethics Review Group (2002) Code of Ethics for Psychologists Working in Aotearoa /New Zealand, New Zealand Psychological Society, www.psychology.org.nz/about/Code of Ethics_2002.html [accessed 28 July 2004].

Cohen, J., M. Ashley, R. Ferrence, et al. (1999) "Institutional addiction to tobacco" Tobacco Control, 8:70-74.

Cohen, J.E. (2001) "Universities and tobacco money" British Medical Journal, 323(7303):1-2.

Cohen, S. and E. Lichtenstein (1990) "Perceived stress, quitting smoking, and smoking relapse" Health Psychology, 9:466-478.

Colrain, I., G. Mangan, O. Pellett et al. (1992) "Effects of post-learning smoking on memory consolidation" Psychopharmacology [Berl], 108(4):448-51.

Curry D. (1995) Letter from the Manager, Public Health Regulatory Services, to the Secretary of Commerce, 27 October, Ministry of Health, released under the Official Information Act, Wellington.

DeAngelis CD. (2000) "Conflicts of Interest and the Public Trust" Journal of the American Medical Association, 284:2237-2238.

Department of Education Science and Training (2003) Higher Educational Institutions, Department of Education, Science and Training, Canberra, www.dest.gov.au/highered/ausunisalpha.htm#alpha [accessed 16 October 2003].

Diethelm, P.A., J.-C. Rielle and M. McKee (2005) "The whole truth and nothing but the truth? The research that Philip Morris did not want you to see" Lancet, 365(9461):86-92.

Doll, R. (1998) "Uncovering the effects of smoking: Historical perspective" Statistical Methodology in Medical Research, 7:87-117.

Edith Cowan University (2001) Tobacco Sponsorship and Promotion, 27 April, Edith Cowan University, www.ecu.edu.au/GPPS/policies_db/tmp/ac017.pdf [accessed 23 August 2005].

Edwards, R. and R. Bhopal (1999) "The covert influence of the tobacco industry on research and publication: A call to arms" Journal of Epidemiology and Community Health, 53:261-262.

Ennis, D. (1992) Review of Smoking Behavior Research Project Conducted at the University of Auckland, New Zealand, from January 1991 to Present, Philip Morris, Richmond, Virginia, tobaccodocuments.org/pm/2023773761.html [accessed 23 July 2004].

Fischer, P. and J.W. Richards (1986) "The pathogenesis of atherosclerosis--and an exchange on research support by the tobacco industry" New England Journal of Medicine, 315(10):644.

Garne, D., M. Watson, S. Chapman et al. (2005) "Environmental tobacco smoke research published in the journal Indoor and Built Environment and associations with the tobacco industry" Lancet, 365(9461):804-809.

Gillin, C.T. (2002) "The bog-like ground on which we tread: Arbitrating academic freedom in Canada" Canadian Review of Sociology & Anthropology, 39(3):301-323.

Health Research Council of New Zealand (2002) Guidelines on Ethics in Health Research, Health Research Council of New Zealand, Auckland, www.hrc.govt.nz/assets/pdfs/ethgdlns.pdf [accessed 23 July 2004].

Hill, G., W. Millar and J. Connelly (2003) "'The great debate': Smoking, lung cancer, and cancer epidemiology" Canadian Bulletin of Medical History, 20(2):367-386.

Howard Partners Limited (2001) Mapping the Nature and Extent of Business-University Interaction in Australia, Australian Research Council, Canberra, www.arc.gov.au/pdf/01_03.pdf [accessed 18 August 2005].

Iverson, M. (2001) Should There Be an Oath for Scientists and Engineers? Committee on Scientific Freedom and Responsibility, American Association for the Advancement of Science, Washington, DC, http://www.aaas.org/spp/sfrl/projects/oath/oathsummary.htm [accessed 26 April 2004].

Kemp, K. (2000) What is Academic Freedom? University of St Thomas. St Paul, Minnesota, courseweb.stthomas.edu/kwkemp/Papers/AF.pdf [accessed 14 March 2005].

Laugesen, R. and N. Maling (2002) "Law on order" Sunday Star Times, 5 May 2001, C1-2.

Lewison, G., G. Dawson and J. Anderson (1997) "Support for UK biomedical research from tobacco industry" Lancet, 349(9054):778.

Lexchin, J., L. Bero, B. Djulbegovic et al. (2003) "Pharmaceutical industry sponsorship and research outcome and quality: systematic review" British Medical Journal, 326(7400):1167-1170.

Mangan, G. (1987) Research Proposal to Philip Morris, University of Auckland, Auckland, Bates Nos. 2001260144-0150, legacy.library.ucsf.edu/tid/kfa88e00 [accessed 23 July 2003].

Mangan, G. (1994) Nicotine, Smoking Behavior and Psychological Variables, University of Auckland, Bates Nos: 2051517571-7587, legacy.library.ucsf.edu/tid/win52e00 [accessed 23 July 2004].

March, J. and J. Olsen (1996) "Institutional perspectives on political institutions" Governance, 9(3):247-264.

Master Settlement Agreement (no date) Information About Archives: Master Settlement Agreement Provisions Regarding Internet Documents, TobaccoArchives.com, www.tobaccoarchives.com/maininfosettle.html [accessed 30 August 2005].

Miller, C. (1982) "'Public Enemy Number One' gets a broadside salvo from University Senate" Medical Journal of Australia, 2:391-392.

Morgan, S., M. Barer and R. Evans (2000) "Health economists meet the fourth tempter: Drug dependency and scientific discourse" Health Economics, 9(8):659-667.

New Zealand Psychological Society (1986) Code of Ethics 1986, New Zealand Psychological Society, www.psychology.org.nz/about/code_1986.html [accessed 28 July 2004].

Ong, E. and S. Glantz (2000) "Tobacco industry efforts subverting International Agency for Research on Cancer's second-hand smoke study" Lancet, 355:1253-1259.

Ong, E. and S. Glantz (2001) "Constructing 'sound science' and 'good epidemiology': Tobacco, lawyers, and public relations firms" American Journal of Public Health, 91(11):1749-1757.

Owen, J. (1993) Letter to Dr Sharon Boyse, Corporate Affairs Dept., BAT Staines, from Wills New Zealand, 26 July, British American Tobacco Guildford archive, Bates Nos. 50086961-6963.

Parascandola, M. (2004) "Scepticism, statistical methods and the cigarette: A historical analysis of a methodological debate" Perspectives on Biological Medicine, 47:244-261.

Parrott, A. (1995a) "Smoking cessation leads to reduced stress, but why?" International Journal of Addiction, 30:1509-1516.

Parrott, A. (1995b) "Stress modulation over the day in cigarette smokers" Addiction, 90(2):233-244.

Parrott, A. (1998) "Nesbitt's paradox resolved? Stress and arousal modulation during cigarette smoking" Addiction, 93(1):27-39.

Pearson, D. (2001) "Defining the public interest" Australasian Council of Public Accounts Committee Conference, Canberra, February, Office of the Auditor General of Western Australia, www.audit.wa.gov.au/pubs/publicinterestpaper.html [accessed 10 June 2003].

Philip Morris (1990) Project Cosmic: Proposals, Contracts and Progress Reports, 20 November, Philip Morris, Bates No. 2022884747-4751, legacy.library.ucsf.edu/tid/mgc78e00 [accessed 15 July 2004].

Philip Morris (1991) Project Cosmic: Budget/Spending Status, February, Philip Morris, Bates No. 2023160927, legacy.library.ucsf.edu/tid/jgk48e00 [accessed 15 July 2004].

Philip Morris (1996) R&D 1995 Research Grants, Philip Morris, Bates No. 2057992643-2644, legacy.library.ucsf.edu/tid/gan13e00 [accessed 23 July 2004].

Pierce, J.P. (1986) "Tobacco sponsorship of medical research" Medical Journal of Australia; 145:658.

Pomerleau, O. and C. Pomerleau (1990) "Behavioural studies in humans: Anxiety, stress and smoking" Ciba Foundation Symposium, 152:22-39.

Proctor RN. (2004) "The global smoking epidemic: A history and status report" Clinical Lung Cancer, 5(6):371-376.

Royal Society of New Zealand (2003) Code of Professional Standards and Ethics, Royal Society of New Zealand, Wellington, www.rsnz.org/directory /code_ethics.php [accessed 23 July 2004].

Saloojee, Y. and E. Dagli (2000) "Tobacco industry tactics for resisting public policy on health" Bulletin of the World Health Organization, 78(7):902-910.

Shaw, J. (1986) "He who sups with the devil ..." Medical Journal of Australia, 144:617.

"Smoking still kills" (1982) British Medical Journal, 285(6351):1295-1296.

Spurgeon, D. (2002) "Canadian universities' links to tobacco industry shown" British Medical Journal, 325:734.

Stein, P.C. (2002) Conflict of Interest and Conflict of Commitment, Cornell University, Ithaca, New York, web.cornell.edu/UniversityFaculty/docs/2002ConflictsProcess /CONFLICTSBOOKLET.pdf [accessed 14 March 2004].

Stough, C., G. Mangan, T. Bates et al. (1994) "Smoking and Raven IQ" Psychopharmacology [Berl], 116(3):382-384.

Stough, C., G. Mangan, T. Bates et al. (1995) "Effects of nicotine on perceptual speed" Psychopharmacology [Berl], 119(3):305-310.

Talley, C., H.I. Kushner and C.E. Sterk (2004) "Lung cancer, chronic disease epidemiology, and medicine, 1948-1964" Journal of the History of Medicine and Allied Sciences, 59(3):329-374.

Tate, J., C. Pomerleau and O. Pomerleau (1994) "Pharmacological and nonpharmacological smoking motives: A replication and extension" Addiction, 89:321-330.

Thompson, M. (1996) Letter from the Director of the Tobacco Institute of New Zealand to the President of the Tobacco Institute of the USA, 24 September, Tobacco Institute of New Zealand, Auckland, Bates Nos: 10991543-1545, tobaccodocuments.org/nysa_ti_m2/TI10991543.html [accessed 27 August 2005].

Tobacco Institute of Australia (1996) Submission on the NHMRC Working Paper on the Health Effects of Passive Smoking, May (estimated), Tobacco Institute of Australia, Bates Nos. 2502216744-6841, legacy.library.ucsf.edu/tid/lvx22d00 [accessed 23 July 2004].

Todd, S. (2004) Review of Research Undertaken into the Placement of Health Warnings on Cigarette Packages, September, Dunedin, report for Imperial Tobacco obtained by Official Information Act request.

Tong, E.K., L. England and S.A. Glantz (2005) "Changing conclusions on secondhand smoke in a sudden infant death syndrome review funded by the tobacco industry" Pediatrics, 115(3):e356-366.

Tunstall, J. (1995) Legal Rights to Advertise Liquor in New Zealand, Letter to P. Viner of BatMark, 19 September, 1995, Wills New Zealand, Petone, Bates Nos. 503062267-2297, tobaccodocuments.org/guildford_misc/503062267-2297.html [accessed 27 August 2005].

Turner, C. and G.J. Spilich (1997) "Research into smoking or nicotine and human cognitive performance: Does the source of funding make a difference?" Addiction, 92(11):1423-1426.

University of Adelaide (1997) Funding from the Tobacco Industry, April, University of Adelaide, Adelaide.

University of Auckland (2003) Charter, University of Auckland, Auckland, www.auckland.ac.nz/docs/sa/University_of_Auckland_Charter_2003.pdf [accessed 23 July 2004].

University of Otago (2004) Policy on Ethical Practices in Research and Teaching Involving Human Participants, University of Otago, Dunedin, telperion.otago.ac.nz/acadcomm/policyethicalpractices.html [accessed 24 June 2004].

University of South Australia (1999) Research Funded by Tobacco Companies, July,

University of South Australia, Adelaide, www.unisa.edu.au/policies/policies/resrch/res14.asp [accessed 14 March 2005].

University of Sydney (no date) Tobacco Industry Document Gateway, University of Sydney, Sydney, tobacco.health.usyd.edu.au/site/gateway/docs/index.htm [accessed 30 August 2005].

Victoria University of Wellington (2005) Laws 505: Public Law-Exploring the Edge,

Victoria University of Wellington, Wellington, www.lawschool.vuw.ac.nz/vuw/content/course.cfm?id=473.

Walsh, R.A. and R.W. Sanson-Fisher (1994) "What universities do about tobacco industry research funding" Tobacco Control, 3:308-315.

Warner, K. (1991) "Tobacco industry scientific advisors: Serving society or selling cigarettes?" American Journal of Public Health, 81(7):839-842.

West, R. and P. Hajek (1997) "What happens to anxiety levels on giving up smoking?" American Journal of Psychiatry, 154(11):1589-1592.

Wolinsky, H. (1985) "When researchers accept funding from the tobacco industry, do ethics go up in smoke?" New York State Journal of Medicine, 85(7):451-454.

Woodside, A. (1986) "Social costs and the political economy of tobacco taxation" address to Philip Morris International Seminar on Taxation, September 10-12, 1986,

Washington, DC, Tobacco Documents Online, tobaccodocuments.org/pm/2025823917-3936.html [accessed 31 January 2003].

Yach, D. and S. Bialous (2001) "Junking science to promote tobacco" American Journal of Public Health, 91(11):1745-1748.

Zeltner, T., D. Kessler, A. Martiny et al. (2000) Tobacco Company Strategies to Undermine Tobacco Control Activities at the World Health Organization, July, World Health Organization, Geneva, www.who.int/tobacco/en/who_inquiry.pdf [accessed 14 March 2005].

George Thomson (1)

Louise Signal

Department of Public Health

Wellington School of Medicine and Health Sciences

University of Otago

Correspondence

George Thomson, Department of Public Health, Wellington School of Medicine and Health Sciences, Box 7343, Wellington South, New Zealand, Ph: +64 4 918 6054, fax: +64 4 489 5319, email: gthomson@wnmeds.ac.nz.