By Ray Dunne
BBC News Online health staff
One of the world's most respected medical journals says it should never have published a controversial paper on MMR.
The Lancet now has tougher rules on conflicts of interest
What steps do journals take to ensure studies are robust and trustworthy?
Richard Horton sees thousands of research papers pass his desk every year. As editor of The Lancet, he decides what is published in one of the world's most prestigious journals.
In 1997, he received a paper from Andrew Wakefield, a doctor at the Royal Free Hospital in Hampstead, north London.
Dr Wakefield and colleagues had carried out tests on 12 children. They claimed to have found a possible link between the three-in-one MMR vaccine and autism and bowel disease.
The study was published in 1998 and its findings sparked a media furore. Many parents subsequently decided to shun the three-in-one jab.
Last week, Dr Horton was told Dr Wakefield may have had a potential conflict of interest.
Within two days, The Lancet editor had issued a statement acknowledging the potential conflict of interest. He told journalists the study had "fatal flaws" and should never have been published.
The Lancet maintains it should have been told that Dr Wakefield was being paid to carry out another similar study.
It says Dr Wakefield should have been aware of the potential conflict of interest after reading the journal's guidelines on the issue.
In 1998, these stated: "The conflict of interest test is a simple one. Is there anything...that would embarrass you if it were to emerge after publication that you had not declared it?"
Today, those guidelines are slightly more detailed. The Lancet now demands that contributors declare all potential financial conflicts of interest.
It no longer accepts articles from anyone who has been employed by or held shares in a relevant company or its competitors during the previous years.
It will accept articles from people who have received money towards research, travel or accommodation from relevant companies but only if these are declared in the published paper.
Possible sanctions against doctors
A letter of explanation pointing out genuine misunderstandings
A letter of reprimand and warning as to future contact
A formal letter to the relevant head of institution or funding body
Publication of a notice of redundant publication or plagiarism
An editorial giving full details of the misconduct
Refusal to accept future submissions from those involve for a stated period
Formal withdrawal or retraction of the paper
Reporting the case to the General Medical Council
The new rules reflect a growing trend by medical journals to be seen to more transparent.
"People have really only started looking at this seriously in the past few years," says Dr Richard Nicholson, editor of the Bulletin of Medical Ethics.
"The pressure has come from the American journals. Some journals have had big problems with undeclared conflicts of interest."
Last year, the Nature Publishing Group announced that it would require all authors to declare if they had financial ties to products.
It followed the news that the author of one paper on experimental treatments for depression held a patent, stock options and was being paid consultancy fees by a company named in the article.
In 2002, the New England Journal of Medicine introduced new rules banning articles by people with "significant" financial interests in relevant companies - namely those who have received $10,000 or more from these companies.
"Around three out of four authors in medical journals have some sort of conflict of interest," says Dr Richard Smith, editor of the British Medical Journal.
"For some, this is being paid to go to a meeting or receiving a research grant.
"A few years ago, very few of these conflicts of interests or competing interests were declared. Things have changed but we still have a way to go."
In 1997, some of the editors of the leading medical journals got together to form the Committee on Publication Ethics (COPE).
It aimed to provide editors with a sounding board and help them "to deal with possible breaches in research and publication ethics".
In 1999, they issued guidelines aimed at stamping out research fraud and potential conflict of interests.
They state that editors and contributors must always declare potential conflicts of interests. "If in doubt, disclose," they advise.
The committee has also drawn up sanctions to be taken against doctors who fail to stick to the rules.
They range from a simple letter pointing out errors or misunderstandings to a decision to report the doctor to the General Medical Council, which has the power to ban them from practising medicine.
This latest controversy has led to calls for these guidelines to be tightened up.
"Guidelines need to be sufficiently clear so that people have absolutely no excuse," says Dr Evan Harris, a Liberal Democrat MP and a member of the British Medical Association's medical ethics committee.
"It is vital for the credibility of science that all possible steps are taken to ensure that everything is above board and everything is seen to above board."
Dr Smith, who is also vice-chair of COPE, acknowledges that more needs to be done.
"The problem has been that people have believed the myth that science is a pure objective activity.
"It's not. It's a human activity and it's prone to all of the joys and downsides of being a human activity. We've fooled ourselves."