I’m losing my faith in clinical trials. As a medical journalist I review quite a few scientific studies every week and it doesn’t take long before you start to smell a rat. They may be couched in scientific jargon but reading between the lines you detect a definite sense of long bows being drawn and creaky hypotheses being propped up by very wobbly numbers.
I’m not alone in suspecting that a lot of research is rubbish. One of the most widely downloaded articles from the Public Library of Science is by epidemiologist John Ioannidis entitled “Why Most Published Research Findings Are False”. He says researchers are too intent on finding relationships between different variables - risk factor X and disease Y - and scientific journals are too eager to publish articles that show these relationships.
His argument is that there are scores of unrelated variables that may falsely appear to be in a causal relationship, and that more repetitive studies rather than large ‘landmark’ studies are needed to show the true picture.
Then there are vested interests. A few years ago a blog called Pharma Watch published a “Cheat’s Guide to Clinical Trials” describing the many ways in which industry sponsors could skew the findings or derive favourable interpretations for clinical studies they have a hand in designing. Using unequal comparators, for example, or finding 'benefits' based on surrogate endpoints, composite endpoints and secondary outcome measures.
Is this a genuine concern? Well, only this month the American Journal of Psychiatry has published a trial which appears to show that suicide rates increased when warning labels were added to SSRI antidepressants. But experts have since pointed out that the modest increase does not tally with prescribing rates nor with the introduction of warnings.
So what are clinical researchers doing to address the perception of bias and lack of integrity in clinical trials? The adoption of clinical trial registers is long overdue, although this is an initiative being forced upon researchers by editors of medical journals.
There may well be many other efforts underway to bolster and demonstrate integrity in research, but you won’t read about them in 6minutes. Those involved in clinical trials are a reticent if not secretive bunch, and this week they refused us access to their national conference. I was hoping to speak researchers at the NHMRC-sponsored International Clinical Trials Symposium in Sydney on Wednesday, but we were not allowed to attend.
If researchers want to restore trust in their trials they should be striving for openness and transparency, not meeting behind closed doors.
Comments
I congratulate Michael Woodhead on his excellent column "Clinical trials must come of the closet". We can't provide the best care until the flawed foundations of evidence based medicine are repaired. It is great to see that Michael and 6minutes are willing and able to debunk misleading claims.
Posted by Dr Peter R Mansfield on Tuesday, 2 October 2007
Come and see some clinical trials in progress and talk to researchers, clinical trials staff, monitors, and research subjects and patients at our clinical trials centre at St Vincent's Hospital. It is an important enterprise but should be transparent to all stakeholders including the media. It is increasingly difficult to undertake clinical trials that are poorly designed, potentially biased, and incompetently performed. We subscribe to the view that there are not enough clinical trials - there are a lot more important questions than answers in clinical medicine. Posted by Prof Ric Day on Tuesday, 2 October 2007
