Yesterday, Bill Gardner weighed in on the conflict of interest firestorm set-off by Lisa Rosenbaum. On the surface, his New Republic article seemed to offer a middle-of-the-road and nuanced counterpoint to the vigorous—and at times spiteful—counterattacks to Rosenbaum’s NEJM series. But despite his efforts to achieve a balanced perspective, Gardner failed to resolve the question with clarity.
While he conceded that Rosenbaum made valid points, Gardner advanced the standard and seemingly indisputable 3-part argument in favor of COI disclosure: 1) Big pharma is hugely powerful, 2) big pharma is guilty of having committed great mischief and, most important, 3) big pharma can only be dealt with by demanding more, not less, transparency.
Now, transparency seems like one of these wonderful ideas that no one in his right mind could ever be against. It’s the first word on everyone’s lips when some shenanigan is revealed: “We need more transparency!” If something is transparent, then anyone can see through it. And if anyone can see through it, then no shenanigan will take place. A no brainer!
But what exactly is the “it” that should be more transparent? And by what means is transparency to be commanded? According to Gardner,
The remedy for bad science is to make science more objective by making it more transparent.
So we can make science more transparent by making data more visible. Gardner wants drug companies to “store the data from published clinical trials online forever” for anyone to re-analyze and check the results. What data does Gardner want to have disclosed and put online? He doesn’t exactly say, but by thinking through his proposal, we may begin to get an idea.
The execution of any large clinical trial involves myriad decisions every step of the way. The endpoints obtained and the information typically released to the public are only the very tip of an iceberg of data that determine the intermediate choices made about the how, when, what, where, and why of trial design and conduct. Everyday, employees and executives of pharmaceutical companies make decisions—based on certain data points—to decide the best course of action for a given clinical trial in the immediate, short and long term. And obviously, at each point in the process, undue influence may take place.
What is published in medical journals, therefore, is obviously insufficient to ascertain the integrity of the science. This entire iceberg of information, then, must be exposed if transparency should stand a chance to have its hoped-for effect. So long as information is kept out of view, the opportunity for mischief remains. The only effective solution, then, is a system of totalitarian surveillance directed at the pharmaceutical company. Is Gardner seeking an alternative activity for the NSA now that the Patriot Act has finally expired?
Ironically, with all his attention focused on transparency, Gardner misses the elephant in the room. When he brings up the infamous VIOXX case as an example of industry misconduct that could allegedly have been averted by more disclosure of COI, it does not occur to him to bring up a very pertinent point: how is it that a mere 3 mis-classified cases, out of 8,000 trial enrollees, should tip the scales for the trial and have such a momentous effect on countless patients?
The cult of “evidence-based medicine” has given rise to the worship of the mega clinical trial, meant to detect minute clinical effects through the enrollment of thousands of patients. The results of these mega trials are fodder for population medicine, and influence clinical decisions for millions of Americans and people worldwide. Cui bono from such a misplaced focus? Precisely the giant pharmaceutical companies which are supposed to be kept in check!
Meanwhile, alternative ways to think about enhancing therapeutic knowledge are woefully ignored. But unless we deal with our warped framework of clinical analysis, unless we adopt a perspective suitable for the individual patient, calls for transparency will only obscure the issues at hand.