Published online 15 December 2006 | Nature | doi:10.1038/news061211-19

Column

Tainted by association?

Richard Doll's links with industry are disconcerting but hardly scandalous. And they don't make him a villain, says Philip Ball.

Few things will polarize opinion like the dressing down of a recently deceased and revered figure. That's clear enough in the debate that has followed the recent media splash on accusations that Sir Richard Doll, the British epidemiologist credited with identifying the link between smoking and lung cancer in the 1950s, compromised the integrity of his research by receiving consultancy payments from the chemicals industry.

Doll's contracts with the likes of Monsanto and Dow Chemicals, splashed in The Guardian and elsewhere, have provoked howls of outrage and vague accusations that his studies into links between the companies' products and cancer were a sham.

Others have risen to defend his reputation. An editorial in the British newspaper The Times calls the charges "a cheap shot" made by "grave robbers". Several leading UK scientists have written to The Times saying that "we feel it is our duty to defend Sir Richard's reputation and to recognise his extraordinary contribution to global health".

The black-and-white character of the argument is perhaps best exemplified by a leader headline in The Observer newspaper: "Richard Doll was a hero, not a villain."

Why does he have to be either? It is a sad thing if we cannot accept the complexities of real people. And although the collaborations between academia and industry certainly create tensions and problems, there is no point pretending that they should not exist.

Doll's consultancy work is not immune to criticism even by the standards of his time, when disclosure was not mandatory. But the suggestion that his research is automatically invalidated, and his character besmirched, by such conflicts of interest (as they would now be regarded) is one that smells of piety rather than an evaluation of the facts.

The case against

Here, then, is the case for the prosecution. Doll proclaimed that Monsanto's Agent Orange posed no carcinogenic hazard, when he had been receiving consultancy fees of US$1,000-1,500 a day from the company for nearly 30 years. He compiled a review on behalf of Imperial Chemical Industries (ICI), Dow Chemicals and the Chemical Manufacturers' Association to defend them against claims that workers developed cancer after being exposed to vinyl chloride. He was paid £15,000 for that review and Monsanto was a big producer of vinyl chloride. He also argued that there was little basis for the idea that asbestos is a major health risk, while pursuing a long-term consultancy relationship with the UK's leading asbestos manufacturer Turner & Newall. Turner & Newall later donated £50,000 to set up Green College in Oxford in Britain, of which Doll was a founder and the first warden.

Doll was rather inconsistent about declaring these interests — he made no secret of some of his links to industry, but the Monsanto connection was not disclosed until a court case over vinyl chloride in 2000. All this was revealed in a document written in 2002, available online1, but only hit media headlines this November after an article in the American Journal of Industrial Medicine mentioned them again2.

To fail to disclose such connections when the work itself was conducted would be very bad indeed for any researchers working in today's society. But until the 1980s, there was no expectation that academics should make this sort of payment public, so there is no reason to expect that Doll would have been systematic about doing so. That is one of the main lines of defence for Doll's supporters: that it is absurd to judge him by today's standards — a notorious way of vilifying historical figures and events.

That's a fair point, although we have to remember that we're talking here about the 1980s, not the nineteenth century. It doesn't take a great deal of insight to see that being paid by a company while assessing their products is not ideal.

But it is notable that Doll used contract money to help set up a college, and donated other fees to charities such as the Medical Foundation for the Victims of Torture. He gives every impression of being a man going about his business in an environment that had not thought very hard about the proprieties of contracts with industry. If he did not think too hard about it either, that does not make him a villain.

Evidence

We must remember that no evidence has been presented that Doll's conclusions were biased by his contracts. (Admittedly it is hard to see how that could be established either way.) We do know that some of his results were not reproduced by subsequent studies, but researchers sometimes get it wrong.

Richard Peto, who worked with Doll on the tobacco studies, says that Doll came under pressure from the asbestos manufacturers not to publish any evidence of the harmful effects of their product. They claimed it would damage the national interest by undermining this important industry, says Peto, and even threatened legal action. But Doll resisted this, and published anyway.

ADVERTISEMENT

That speaks of a man who knew his own mind, as does the way he overcame industry's resistance to the link between smoking and cancer: hardly the act of someone in the pocket of corporations.

It is notable that the paper in the American Journal of Industrial Medicine that started this round of debate aims not to denigrate Doll, but rather to call for a tightening of policies governing disclosures of interest today. There's still plenty of work to be done in that respect (see 'Cash interests taint drug advice', 'Journals scolded for slack disclosure rules)' ). We should recognise the shortcomings of the past, and move on.

Visit our byassociation.html">newsblog to read and post comments about this story.