skip navigation nih record
Vol. LXIV, No. 8
April 13, 2012
cover

next story


Baby Steps Not Enough
Clinical Trials Need Major Remodeling, Califf Says

On the front page...

Duke’s Dr. Robert Califf calls for a major overhaul of clinical trials in the U.S.

Duke’s Dr. Robert Califf calls for a major overhaul of clinical trials in the U.S.

We need to rethink the clinical trials system—from the ground up. That’s what Dr. Robert Califf proposed at a recent Mind the Gap seminar, “Innovative Approaches to Clinical Trials.”

“I hope to convince you that incremental changes in our clinical trials system will not be sufficient,” he said. “We really have to prepare ourselves for a major transformational change.”

Vice chancellor for clinical and translational research and director of the Duke Translational Medicine Institute, Califf has studied the business of conducting clinical studies for decades. Outlining some of the problems with the current system, he said the nation’s biggest challenge might be overcoming a mindset about the way we’ve always done clinical trials until now.

“You’ll see that in the context of defining the problem, it’s technically a non-problem,” he explained. “[The hurdle] turns out to be a social and cultural set of issues that we’re going have to get over if we want to make the next set of advances in health and health care.”

Continued...

‘Toss of a Coin’

Califf said the medical research community should develop a system that continuously records well-defined, coded data in practice, and in the midst of quality improvement and observational studies, it should not lose “focus on interventional trials that are randomized…The only difference between clinical practice and a pragmatic randomized trial should be the toss of a coin. That’s the way we need to change the fundamental way we’re thinking about medicine and clinical practice, as we evolve into a true learning health system.”

Califf said it costs an estimated “$400 million to do a global clinical trial with industry for a chronic indication.” We could do a more efficient version for less than half that amount, he surmised.

Califf said it costs an estimated “$400 million to do a global clinical trial with industry for a chronic indication.” We could do a more efficient version for less than half that amount, he surmised.

Photos: Bill Branson

Much of what we assume about our current medical practices may be based on myth anyway, he pointed out. Using national clinical practice guidelines to examine the evidence base for medical decisions, for example, he and some of his colleagues in heart disease research asked themselves a question.

“How often,” he asked, “when we make a significant recommendation do we have high-level evidence from multiple randomized trials, and how often is it ‘level C,’ which means, ‘We have no earthly idea, but there are experts who think they know’?”

The answer surprised them.

“About 15 percent of the time we have high-level evidence to back up a recommendation,” Califf reported. “So far no other specialty has been willing to stand up and say they base their decisions on more than 13 percent. We’re making decisions every day that are life and death—or chronically important—knowing that empirically it’s possible to answer the question, but [we’re] just not doing it…because the way we currently do [clinical trials] is so labor-intensive and so expensive that we could only answer a very limited number of questions in the universe.”

The same team of researchers took the query a step further: “What percentage of the time are we completely flying without instruments?” they wondered. The answer: “It’s about half the time.”

What’s Stopping Innovation?

That’s not to say the data is not available to inform decisions, Califf noted. Truth is, we now have so much information that we’ve run out of ways to harness it for our purposes.

“The problems I’m talking about are a good thing,” he said. “We have these problems because we’ve made so many advances that we’ve exhausted the old methods and the old way of thinking…The bottom line is, our knowledge has accelerated beyond the ability to provide empirical support for decision-making.”

Califf suggested a “3-compartment model, with innovation in every compartment.” The components would be proof-of-concept and biological mechanism trials, studies to identify risk/benefit balance (late phase 2, early phase 3 trials) and studies to gauge effectiveness at societal and individual levels.

“Types 2 and 3 are merged,” he proposed. “We’re going to increasingly see less of a distinction between risk/benefit balance of individual products and effectiveness in the community.”

Another big problem, Califf suggested, is too many needless rules. Overregulation defeats innovation, he said.

“In such a regulated industry—where there’s so much at stake if an error is made—it leads to a situation where people are so afraid to take risks that they rarely do anything new,” he said. “We need an ‘innovation envelope’ to try new things until the system is transformed. Experiment with agreement ahead of time, [so researchers know] it’s okay to do the experiment and you’re not going to be punished for trying, even if things go bad.”

Streamline Trial Design

Califf (r) greets well-wishers after his lecture.

Califf (r) greets well-wishers after his lecture.

In closing, Califf addressed two other issues that prevent advancement: health systems, driven by the need for efficiency, are not supporting clinical research, which in its current form is prohibitively expensive; and protocol design, which has become too complex and burdensome.

“I call it the ‘Christmas tree effect,’” he said. “You start out with a beautiful green tree that should be admired and then everybody in the family wants to put an ornament on it…and no one will take grandma’s ornament off the tree. So you end up with a protocol that is impossible to do and is very distracted from answering the question you originally had.”

Califf said it costs an estimated “$400 million to do a global clinical trial with industry for a chronic indication.” We could do a more efficient version for less than half that amount, he surmised.

“It’s our belief,” he concluded, “that you could markedly reduce the costs of clinical trials if you are willing and able to take the streamlined approaches that we’ve all talked about for years now.”

Watch Califf’s entire talk online at http://videocast. nih.gov/ under Past Events. NIH’s Office of Disease Prevention, NHLBI, NIA and NIAAA presented the lecture.


back to top of page