Too many clinical trials repeat scientific work already done, researcher says

TORONTO (CP) - Too many researchers are repeating the work of others, conducting clinical trials that are redundant and don't add to the body of scientific knowledge, an Ottawa research team argues in a new study.

The scientists, from the Ottawa Health Research Institute, said the redundant trials are ethically problematic because, among other reasons, they withhold treatment that has already been proven to be beneficial from the patients in the placebo (non-treatment) arm of the studies.

Their article was published Thursday in the journal Society for Clinical Trials.

Lead author Dean Fergusson, a clinical trials epidemiologist, said he believes the problem relates to inadequate homework on the part of the researchers. Too often researchers are not conducting thorough examinations of the scientific literature on their study topic before beginning their trials, he said.

"I don't think investigators initiated these trials with the intent to do redundant research," Fergusson said in an interview from Ottawa.

Fergusson and his colleagues used as an example the clinical trials conducted on a drug called aprotinin, used to reduce bleeding during and after operations.

He had done a meta-analysis - a systematic review and reanalysis combining all existing similar scientific studies - on the drug that was published in 1997. At that point, the combined evidence of 45 trials showed the drug clearly worked.

A few years later the team revisited the literature on the drug, finding the trials now numbered 64 - most mirroring each others findings.

Fergusson suggested university research ethics boards should require researchers to do thorough literature reviews before beginning a trial to ensure they are not setting out to repeat existing science.

Another researcher who studies clinical trials agreed that doing that type of homework is crucial. But Dr. P.J. Devereaux of McMaster University in Hamilton said he sees some important caveats to Fergusson's message.

Sometimes new studies disprove the body of evidence, changing the understanding of a disease or a treatment, he said.

A noteworthy example is the Women's Health Initiative trial, which showed that hormone replacement therapy in post-menopausal women not only didn't protect against heart disease, it could cause it.

"It's a balance," Devereaux said.

"Their point is very clear and it's true. At the same time I would not want people to walk away with the message of 'Boy, we're doing way too many trials and half these things we don't need.'

"The truth is there's a lot of times when people are making it sound like we know the answer when we don't. It's a two-sided coin."

Devereaux noted that the Canadian Institutes for Health Research routinely ask researchers seeking federal funding for their work to provide an analysis of the existing literature as part of their application.