1.Scientists are only human
2.Big academic guns roll out (again) to protect Richard Doll's reputation
GM WATCH COMMENT: The revelations about Sir Richard Doll's consultancy work for Monsanto and other corporations has prompted an important commentary on conflicts of interest from Richard Smith, former editor of the British Medical Journal.
Here's just one example from the catalogue of evidence he produces showing not only that conflicts of interest are critically important but that they often remain undisclosed. This article really is worth reading in full.
EXTRACT: ...in 2001 a medical student and I looked at 3,642 articles in the five leading general medical journals and found that only 52 (1.4%) declared authors' conflicts of interest.
1.Scientists are only human
The Guatrdian, December 8 2006
The revelations about Sir Richard Doll's undisclosed income reflect our increased, and justified, anxiety about conflicts on interest.
Richard Doll, whom I knew, admired, and liked, has been accused of failing to disclose payments he received from Monsanto. The story is presented almost as if all of his outstanding work must be doubted. In fact he never hid his conflict of interest: it simply wasn't usual to declare conflicts at that time. But the scientific community has become steadily more anxious about conflicts of interest - and rightly so.
Scientists used to like to think that they were above conflicts of interest. What mattered was the quality of the science, not any conflicts. They enjoyed a fantasy that science was an objective discipline based on evidence and data and so immune to human failings. This is, of course, nonsense. Science is a human activity and so prone to abuse, fraud, bias, misjudgements, incompetence, greed, and the full rainbow of human frailty.
None of us likes to think that we are influenced by conflicts of interest, and it's important to understand that conflict of interest is a condition not a behaviour. It often operates unconsciously, and there is substantial and growing evidence of its influence on how doctors prescribe and treat patients, what research is undertaken, and how research is interpreted.
A systematic review of the effect of financial sponsorship looked at 11 studies that compared the outcome of those sponsored by the drug industry and those that were not sponsored - and in every one of the included studies those that were sponsored were more likely to have findings favourable to industry.
Another study looked at 69 randomised trials of non-steroidal anti-inflammatory drugs, which are prescribed on a huge scale for arthritis. All of these trials were sponsored by the drug industry, and the sponsor's drug was deemed to be as good as the comparative treatment in three quarters of the studies and better in a quarter. In not a single case was the drug being investigated worse than the comparative treatment.
The main determinant of whether scientific reviews find passive smoking to be harmful or safe is whether the authors have ties with the tobacco industry. In the intense debate over whether third generation contraceptive pills increased women's chance of developing thromboembolic disease, all industry sponsored studies concluded that the pills were safe while all publicly funded studies found that they increased thromboembolic disease.
Despite the evidence that financial conflicts of interest have a powerful effect on the results and interpretation of studies, authors did not until recently declare them. The International Committee of Medical Journal Editors stated that authors should do so as long ago as 1993, but in 2001 a medical student and I looked at 3,642 articles in the five leading general medical journals and found that only 52 (1.4%) declared authors' conflicts of interest. Yet most authors have a conflict of interest.
Another study looked at the financial conflicts of interest of authors of 75 pieces in prominent medical journals on a particular category of drugs. The investigators asked the 89 authors of the articles whether pharmaceutical companies had provided them with reimbursement for attending a symposium, fees for speaking, fees for organising education, funds for research, funds for a member of staff, or fees for consulting. They also asked about the ownership of stocks and shares in companies. Sixty-nine (80%) of the authors responded, and 45 (63%) revealed that they had financial conflicts of interest. Yet in only two of the 75 articles had conflicts of interest been exposed.
Medical journals have become steadily stricter in requiring authors to declare conflicts of interest, but newspaper stories appear regularly showing that authors have failed to declare conflicts of interest. Just two weeks ago it was disclosed after publication that an editorialist in the BMJ advocating the use of blood products had failed to declare that he had received funding from manufacturers of the products. Such stories damage the credibility of journals, and there is increasing anxiety - particularly in the United States - that, in the words of an editor of the New England Journal of Medicine, academic medicine is "for sale."
The main response to conflict of interest has been disclosure, but is that always adequate? There must come a point where the conflict is so substantial that more is required. No journal, for example, would have an editorial on a new drug written by an employee of the company that manufactured the drug - except unknowingly as has just happened to the BMJ. Few journals, however, are transparent on what the cut off point is for excluding authors. Nor do journals tend to disclose the scale of conflicts of interest. Yet a financial conflict worth tens of thousands of pounds may have a much greater impact than having lunch bought by a pharmaceutical company.
Finally, I can't help reflecting that although science has some way to go with solving the problem of what to do about conflicts of interest, it is way ahead of the media. Newspapers virtually never disclose conflicts of interest.
2.Big academic guns roll out (again) to protect Richard Doll's reputation
by Murdo Maguire Injurywatch, 9 December 2006
It was no surprise to see the big scientific and academic guns being wheeled out in defence of Sir Richard Doll's reputation. Despite clear and obvious flaws in his studies and massive undeclared payments made to him from interested parties at the time, Sir Richard was always an establishment and industry favourite. Here injurywatch poses the questions his supporters must answer...
Under his mentor Austin (Tony) Bradford Hill, who discovered the link between smoking and lung cancer, Richard Doll justifiably built a reputation as one of the world's great epidemiologists.
But in later life the beknighted professor secretly took payments from the leading UK asbestos polluter Turner and Newall for more than thirty years.
Injurywatch can document a GBP50,000 payment from Turner and Newall to Green College, where Sir Richard Doll was founding warden, and which was led by his wife, and has film available of Sir Richard touring Turner and Newall sites in 1982 following a TV exposure warning of the dangers of asbestos.
The purpose of Doll's trip round the T&N plants was to tell workers their industrial exposure to asbestos was "largely safe".
Sir Richard Doll later expressed his gratitude for the T&N payment to Green College in gratitude "for the work I have undertaken on their behalf."
Currently in excess of 4,000 people a year die from asbestos-related cancers in the UK and more that 1900 people die from the cancer of the thorax, mesothelioma, where exposure to asbestos is the only known cause.
Turner and Newall, with plants across the UK, went into chapter 11 bankruptcy in 2001 and UK victims agreed to settle earlier this year. Many claimants are dead, but their descendants are only likely to receive between 10 and 20p in the pound of the compensation due to them.
In the chemical sector, injurywatch has been able to verify that secret payments were made to Sir Richard Doll by Monsanto of at least $1000 a day (later increased to $1500 a day) between 1976 and 2002.
During the time he was secretly being paid, Doll consistently produced reports which would be deemed to support the Monsanto position. He testified in favour of the Monsanto product Agent Orange before an Australian Royal Commission in 1985 without ever declaring a financial interest of being paid by the company which produced the chemical.
Injurywatch believes there were flaws in both the Doll-Peto study: "The Causes of Cancer: Quantitative Estimates of Avoidable Risks of Cancer in the United States Today," (Journal of the National Cancer Institute 66 (1981).
Specifically we would like to know why the parameters of what purported to be a neutral study of environmental illness would be drawn to exclude African Americans and those aged over sixty when cancer is known to be a disease of the poor and the old.
Doll also produced a 1998 study into vinyl chloride, Effects of exposure to vinyl chloride. An assessment of the evidence. Scand J Work Environ Health 14(2):61-78. Doll R. 1988. The study was not declared at the time to have been funded by the Chemical Manufacturers' Association and Doll's contract with Monsanto, a major vinyl chloride manufacturer was again undeclared.
The study, which largely exonerated the vinyl chloride industry of any blame, seems to have obvious perversities. High risk/exposure individuals seem to have been deliberately excluded from the study while low risk/unexposed workers were drafted in.
Respected scientists have long called into question Sir Richard Doll's findings and motivation and yet 35 years after Doll Peto when its core figure of 6,000 annual environmental cancer deaths is demonstrably wrong, HSE continues to promote the research as the best available.
It is regrettable that Blakemore, Walport, Rees, Bell, Markham and Peto have not put their efforts into conducting an immediate review rather than viewing the exposure of Sir Richard Doll's financial links as an attack on his otherwise emeritus legacy.
[Injurywatch are compensation and injury claims specialists]