Publish or perish � or lie, cheat and steal trying

Medical publishing is a minefield of ethical dilemmas. The body that enforces the code names and shames the culprits in their annual report

By Owen Dyer

There are few sacrifices some people won't make in the name of science, including ethics it seems. For some, lying, cheating and stealing are just part of the game � anything to get published.

The pressure to get research published is intense, and not everybody is willing to let their work be judged on its merits alone. The Committee on Publishing Ethics (CPE) is a group of medical journal editors who meet annually to discuss ethical issues in the industry. Their recently released annual report lists some of the misdemeanours that these editors grappled with in 2003.

The overall number of detected cases of dishonesty � 29 � is vanishingly small compared to the number of articles published. But they run the gamut from plagiarism to malpractice to outright bribery.

The most common offence is submitting the same research to multiple journals. This accounted for seven of the 29 misdemeanours listed by the CPE. A survey of ophthalmology journals published in Nature last year suggested that over 1.5% of articles were duplicates. Just last month, a study in the Journal of the American Medical Association mentioned two articles that had been published five times each.

Copying others' work might seem unprofitable in an age of internet search engines, but there are still a few brave souls willing to try. In the two such cases reported to the CPE, articles were submitted that were almost verbatim copies of previously published work. The authors of one paper admitted that their article overlapped by 5% with the original. The author of the other paper simply hung up when contacted. Neither article, needless to say, got published.

More alarming than plagiarism are shortcuts taken with patients. The sad fact is that when ethics committees refuse to sanction research, not every researcher listens. Journal editors don't normally ask to see proof of ethics committee clearance, unless they have reason to believe medical ethics were breached in the research, and even then it's not a universal practice. One suspicious case involved taking blood samples from healthy babies. Sure enough, when enquiries were made, the issue of ethics approval was "never cleared up."

Conflicts of interest are a pervasive problem in medical research, and only the most serious cases attract any notice. One as-yet-unresolved case involves an author who wrote an article for an unnamed journal that seemed to minimize the dangers of passive smoking. A bit of digging revealed that the author had taken money from tobacco companies throughout his career and even lobbied actively for the industry.

Bribery is another tool occasionally used by desperate authors, though they are rarely so overt as to offer a wad of bills stuffed in an envelope. In one case, an anonymous caller offered to buy 1,000 reprints if a paper was published. To sweeten the deal, she promised to take the editor to dinner "at any restaurant you choose."

A woman representing a drug company made the same restaurant offer to Dr Richard Smith, editor of the British Medical Journal, if he would accept a paper sponsored by her employer. She was clearly unaware that Dr Smith is one of the most active members of the CPE.

"She stopped short of offering to go to bed with me," says Dr Smith, "but I was surprised by her crassness. 'Are you trying to bribe me?' I asked. 'No,' she answered brightly, 'just being nice.'"

Dr Smith and his counterpart Dr Richard Horton, editor of The Lancet, have been driving forces behind the tightening of ethics in the medical publishing community. But they don't lack critics in the US, who say their whiter-than-white attitude is all very well in Britain, but completely unfeasible in the aggressively commercial American environment.

So there may have been a little schadenfreude in some quarters when Dr Horton and The Lancet were bitten rather badly by an ethical scandal of their own last month, involving one of the most famous papers ever published in the journal. Dr Andrew Wakefield's 1998 study linking the measles-mumps-rubella vaccine to autism was seized on by the media and led to a catastrophic drop in immunizations as nervous parents refused shots. Experts condemned Dr Wakefield's findings almost unanimously, but the autism link was firmly implanted in the public mind.

Recently, a journalist's investigation revealed that Dr Wakefield had been paid £55,000 ($134,000) to investigate allegedly vaccine-damaged children for possible legal action by their parents, but had failed to mention this clear conflict of interest. Dr Horton, who had been regretting the article for years, was quick to repudiate it, as were Dr Wakefield's fellow authors.

"There were fatal conflicts of interest in this paper. In my view, if we had known... it would have been rejected," says Dr Horton. But he added the authors had a duty to disclose such conflicts. "The whole system depends on trust and honour."