How To Disseminate Complete Clinical Trial Data
October 14th, 2011 // 3:02 pm @ jmpickett
The growing controversy over releasing clinical trial data – and the extent to which underlying patient-level data should be disclosed – has yet to be resolved. In the interim, debates have raged about patient safety, the utility of meta-analyses for deciphering safety issues and obligations of drug and device makers to physicians and patients. And so a pair of academics is offering a proposal in hopes of rectifying the situation. Here is how they frame the problem:
“There are no uniform international standards requiring that study protocols, statistical analysis plans, and study results be made available, and there are no requirements that completed clinical trial data be publicly available or posted for independent analysis. Even data submitted to the US Food and Drug Administration are not made publicly available,†write Harlan Krumholz and Joseph Ross of the Yale University School of Medicine in the Journal of the American Medical Association.
So what do they suggest? They propose an “external coordinating organization†that is independent and has legal resources could contract with drug and device makers. These would provide access to patient-level data, documents and case reports from all trials – before and after FDA approval – and safety programs. Data, they say, will be “deidentified.†The source materials, they write, are “essential†for determining study veracity. Reviews would assess adequacy and quality of the data.
From there, this coordinating organization would have two other qualified groups of scientists run independent data reviews, so it would be possible to “to check the reproducibility of the findings†and offer confidence that no one is being “manipulated by the funding agency.†Afterwards, other researchers could have access to run their own analysis or confirm results. They also envision a steering committee comprised of “subject-matter experts,†patients, researchers and policy makers.
Of course, this begs a question – why would drug and device makers agree to participate? Krumholz and Ross write that industry gains by improving a reputation damaged by various safety scandals and benefits from reviews that can mitigate unnecessary speculation. They also argue this would field into a “new model (that) would support competition.†[It is worth noting that Krumholz is also a board member of the Center for Outcomes Research and Evaluation.] Finally, they add there is “a mounting recognition that it is untenable to withhold critical information about safety and effectiveness†(here is the JAMA abstract).
They note that one example already exists. In the wake of a scandal over the safety of the Infuse spinal device, which contains a genetically engineered version of a naturally occurring protein, Medtronic has agreed to such an arrangement with Yale. For those who may not recall, a Senate probe began earlier this year after revelations that doctors with financial ties to the device maker were aware of serious problems with the product, but never disclosed potential health complications in articles in medical journals. The probe is also initiated after revelations that some doctors received millions in payments. Many studies were repudiated (back story here and here).
Of course, one might argue this is a cynical attempt by a pair of Yalies to generate more work and enhance their own reputations. But, in fact, they do raise some important – and unresolved – issues. Naturally, implementing such a proposal can be complicated, because aligning interests is never easy. Nonetheless, in an era when safety remains paramount and there is an increasing expectation that comparative data will eventually be regularly sought by insurers, such an approach may offer some valuable and worthwhile insights.
Source: Pharmalot
