With Congress poised to legislate new mandates for post-approval studies, it seemed like a good time to talk to Richard Gliklich, the president of Outcome and a professor of surgery at Harvard Medical School. We covered his company’s substantive conference on what happens after drugs are approved here.

Gliklich’s take-away from the meeting? That governments in Europe, Japan and the U.S. may not be writing the same regulations, but they are tossing around the same terms, and thinking in the same way. “They’re all converging toward the same concepts,” says Gliklich. They’re thinking about harmonization. The world is becoming smaller.”

Gliklich offered no predictions about what Congress will do about drug safety or post-approval requirements. But he thinks there will be similar rules across geographies. “Post approval is hot on every one’s agenda,” he says. “We’re going to see more convergence of requirements.”

Hot Off The Press

In that context, to a degree that overshadows his own company, Gliklich is eager to discuss the Agency for Healthcare Research and Quality (AHRQ) publication of “Registries for Evaluating Patient Outcomes: A User’s Guide.” The AHRQ is a 300-person division of the U.S. Dept. of Health and Human Services. The AHRQ’s role is to shepherd efforts to raise the quality of medical care and evidence-based decisions.

The new user guide, meanwhile, is an attempt to condense best practices into a single volume that academia, government and industry can agree on. “This Registries Guide is a milestone in our growing efforts to draw from medical practice and learn which treatments really work best,” HHS Secretary Michael O. Leavitt said in the press release about the guide. “If we can learn more systematically from the experience of millions of patients and clinicians in day-to-day practice, then we can discover more quickly which treatments are truly most effective, and for whom.”

Gliklich supervised the editing of the book. It has contributions from dozens of colleagues all over the world. One collaborator was especially key: Nancy Dreyer is Outcome’s chief of scientific affairs, and helped Gliklich assemble the manuscript. He says the book will help different stakeholders design registries with greater rigor and consistency. “Let’s have a common foundation for what the right quality and scientific principles are,” says Gliklich. “That stake in the ground is really important.’

Best Practices In Phase IV

The AHRQ guide offers real-world insights into how to deal with the diverse challenges facing all sponsors. We’ll just quote from one paragraph in the chapter on drug safety:

A performance-linked access system (PLAS), also known as a restricted access or limited distribution system, can be described as a special application of a registry. Unlike the less structured disease or exposure registries discussed above, a PLAS is part of a detailed risk minimization action plan that sponsors develop as a commitment to enhance the risk-benefit balance of an approved product.  The purpose of a PLAS is to mitigate a certain known drug-associated risk by ensuring that product access is tightly linked to some preventive and/or monitoring measure.

Filling The Gap

A PLAS, of course, is exactly what sponsors built for clozapine and thalidomide. It’s unclear how common they’ll be in the years ahead. But probably more common than in the past.

For Gliklich, the central point is that a well-designed registry could unite the worlds of a) observing clinical data and b) acting upon it. “If we get good observational, real world data coming in, how do we think about plugging that into decision making?” he asks. Politicians and regulators are just starting to wrestle with that, he says.

In part, he says, drug safety concerns are driving the growth at Outcome. Says Gliklich: “You could definitely construct a safety registry that would have a 90 percent power to detect if there was a significant difference in the safety signal after X thousand patients had been treated. That’s a large bulk of the kind of work we’re doing now.”

Doing Science Right

Gliklich doesn’t say it explicitly, but he appears to think that appropriately designed registries could begin to defuse the toxic political environment in which the industry finds itself.

For some politicians and more rational observers, good registries could be a first step to re-establishing the trust of the public.

Gliklich paused. “[Post approval data] can be complementary to clinical trials and in some cases fill gaps that clinical trials cannot fill,” he says. “Let’s be agnostic as to who is doing the research as long as we can have first principles. If those principles start and end with transparency and have the same quality parameters about what is good science, then it shouldn’t matter where the information comes from. That’s something everyone can get their hands around in the scientific community. Over time, that can be true in the provider community.” The new AHRQ guide, he contends, can foster the sorts of registries that will support better science and reassure the public.

Noting several major pharmaceutical companies that have slashed their sales forces, Gliklich speculates that the existence of good data could confer a marketing advantage to companies that possess it. “It’s not about pens any more,” he says flatly. “I think it’s about better evidence. [People in industry] want to know that this product is going to work better in my population of patients.”

Competitive Issues

Gliklich is not worried about competition from technology firms serving the post-approval space. His company, established in 1998, has enough experience under its belt, with a million patients enrolled in its studies and 3,000 sites logging into the Outcome system daily.

Gliklich is similarly nonplussed by the prospect of generalist technology firms encroaching on post-approval projects. “We’re seeing a lot of players come in on the technology side and the services side,” he notes. He doesn’t sound worried. “The specialists [who] are delivering the highest standards will continue to do well and differentiate themselves by being the cream of the crop.”

Project Management

“There is a big learning curve,” he says of Outcome’s niche in post-approval. “We’ve done over 100 web-based registries. That’s a lot of learning. A lot of mistakes we‘ve learned from, and can bring to the next program. It’s important to be proactive in knowing what to avoid. Anyone can go into Microsoft Project and construct a great time line. But you can’t change the solution you offer, the technology or services you offer, on a dime.”

“Prior success is what leads to our highest number of wins,” he says. “The second thing is the focused solution. We’re really integrating technology and service that exists from the ground up. Strong project management is one of our strengths. The project manager is always in the mix when the team meets with the client.”

Health Systems

Moving forward, he says Outcome is forging new ties to the world of health and hospital information systems. He lauds the harmonization work of Landen Bain in that arena. Gliklich agrees that health records are the next frontier for the clinical trial industry generally and Outcome specifically: “We are working very hard on leading the charge on EHR [electronic health records] integration. It’s very important. We’re the leader.”

The key to making more of the existing health data, Gliklich says, is to plan how to synchronize data so that it can be collected once and used by hospitals and pharmaceutical companies alike. “There are only so many times and ways that hospitals are going to collect the same data,” he notes. “It’s about harmonization of the data models and concepts up front. You need to think through these descriptions and staging of the disease and the endpoint of interest.” It will take work, he concedes, but hardly be insurmountable.

Woe From Washington

In some cases, he says, the notion of “data integration” between a hospital system and a research system may be random or even fallacious. An IT person who lacks domain knowledge may make random, clinically ignorant decisions about how to synchronize two scientific realms. Says Gliklich: “A programmer makes an arbitrary decision about how to map the data element from one system to another. That’s not what you want. It’s potentially dangerous.”

The approach at the National Cancer Institute, he says, is altogether different, and more heartening. Its effort to synchronize and harmonize terminology and data models could genuinely facilitate a bridge between a hospital and a pharmaceutical company, between a regulator and an academic investigator. Says Gliklich: “Getting all these things talking to each other is very important.”

But Gliklich is more dismissive of Congressional attempts to remedy the drug safety situation by dropping insurance claims data into the mix at the FDA. Such data, he says, are inherently dirty. They are tossed by the capricious winds of medical billing strategies and rushed coding decisions. Claims data, for Gliklich, will never form the basis for grounded scientific assessments or policy decisions about a particular drug’s risk/benefit ratio.