Atlanta, GA—A plenary session billed as a “candid discussion with medical and pharmacy directors” on the value of health economics research in real-world decision-making proved to be all that and more, with several of the managed care medical directors on the panel taking issue with the type and quality of health outcomes research supplied by researchers and the pharmaceutical industry.
Even as he acknowledged the health services research expertise of the audience members at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) meeting, Albert J. Rizzoli, MD, a medical director at Presbyterian Health Plan in Albuquerque, NM, also expressed skepticism regarding their work, because decision makers such as himself only see a very small part of what those researchers do. “Largely, what we see in managed care are pharmacoeconomic models that are brought to us by the pharma industry,” he said. “That’s almost entirely what we see.”
Models Lack Real-World Connection There is a disconnect between models and real-world experience that often leaves the real world not meeting best-case modeling assumptions, Dr Rizzoli argued. For a model to actually work, “the best result has to happen from…linked studies,” he said. Models also often consider indirect costs and tout the importance of the employer’s perspective, said Dr Rizzoli, but these factors are difficult for managed care decision makers to digest. Studies frequently emphasize the importance of the treatment on the patient’s quality of life, he said, and seem to say, “we know this costs a lot of money, but think about the quality of life,” with the implication that if the quality of life is improved, the medical care costs will be lessened.
Another difficulty is presented by the current US Food and Drug Administration approach of comparing drugs with placebo, which makes it necessary for a decision maker to compare one placebo study with another, Dr Rizzoli said. There is very little comparison of new therapies to existing products, so there is no way of knowing whether the new product provides any advantage over an existing product.
The research time frame is also an issue, Dr Rizzoli said. Companies may offer an argument for long-term return on investment—usually many years— but “in managed care, we’re looking for return on investment over a short period of time,” he said. Despite arguments about the positive impact of efficacious treatments to the broader budget, pharmacy costs are still contained in a silo. Dr Rizzoli knows “better than anybody that everything we do has an effect on the entire system,” but that’s still not the way the world looks at it. “You don’t get much credit for medical cost-savings,” he lamented. In addition, Dr Rizzoli worries that widespread off-label use distorts costcontrol efforts, as do coupons and copays offered by manufacturers, which allow patients to get around drug tiering systems.
Concern for Economic End Points From the perspective of Kenneth L. Schaecher, MD, a medical director with SelectHealth in Salt Lake City, UT, health outcomes research represents an attempt to add an economic end point to studies that are traditionally clinically focused, thereby creating economic value. But in his eyes, those end points are rarely pertinent. “If you don’t give us anything pertinent, we’re going to make the best decisions we can. We’re not trying to ration care,” he said, adding that when controlling cost is part of the mission, “you’re going to make a conservative decision.”
“Gleevec was a game-changer,” Dr Schaecher pointed out, but “the fifth ARB [angiotensin receptor blocker] in the class was not a game changer.” Nevertheless, health services research was used to support both drugs, despite only 1 truly being innovative, he said. “We are a cynical, skeptical lot in the insurance industry,” he acknowledged.
John Watkins, RPh, MPH, a pharmacy manager in formulary development at Premera BlueCross, a network plan in Seattle, WA, also worries about slanted studies. “It’s a no-brainer— funding bias is something forefront in our mind,” he said, pointing out that study funders get to control the frame around that study. Nevertheless, 2- way communication and some level of trust must be developed. He emphasized that when his group evaluates studies, they try not to be cynical, but will give a fair hearing to the evidence presented. At Premera, Mr Watkins shows a model’s decision-analytic tree to his clinical experts and asks if it resembles how patients are treated. If it bears no real-world resemblance, it goes no further.
A final panelist, Harold “Hank” L. Gardner, MD, at HCMS group in Cheyenne, WY, argued for less complexity and the primacy of the consumer. How do outcomes get translated into verbiage that patients and business managers at employers can understand, he asked.
Putting Meaning to Models Overall, Mr Watkins took a more charitable view of relationships between pharma, healthcare researchers, and decision makers, and offered several suggestions about what might be done to improve them. First, he suggested, manufacturers should keep several pragmatic questions in mind.
“What is the decision that [decision makers] need to make, what data do they need to properly inform that decision, and how do I get from there to studies that will provide that information?” he asked. He suggested it is essential to keep in mind who the decision maker is and who the end user is, because this may influence the model used.
Mr Watkins said he would like to raise the bar in cost-effectiveness modeling studies by having clinical and economic end points included in the trial design, and making sure the studies are adequately powered for their end points. The outcomes research developers should be at the table when the trial registration process is being planned, he argued.
Changing Models’ Focus Dr Schaecher had several suggestions. First, health outcomes research should include appropriate financial outcomes related to the therapy and the intended population. Conducting studies that will get a drug covered will also be increasingly important, and when those studies are conducted, they will be considered and may change his firm’s policies, Dr Schaecher said. “Evidence that shows quality will lead to an improvement in cost of care over time,” he said. In addition, it is important not to settle for surrogate end points when it is possible to measure meaningful end points.
For Dr Rizzoli, the minimum requirement is that models should be transparent, understandable, and focus only on cost and utilization. “We have a hard time quantifying quality of life and translating that into financial data,” he emphasized. And although the National Institute for Health and Clinical Excel lence in the United Kingdom has engendered much controversy, Dr Rizzoli said that this may be an approach worth trying. “We desperately need that kind of model in the United States. We need somebody objective,” he argued.
During the question session, Milt Weinstein, PhD, of Harvard University, indicated disbelief at what he was hearing from Drs Rizzoli and Schaecher, saying it sounded like they did not care about outcomes, but only about costs. Marc Berger, MD, of Eli Lilly, pointed out that insurers have only recently started to ask for better evidence, and he echoed the importance of a skeptical, not cynical, approach to evidence by decision makers and “the need for good dialogue. If there’s cynicism on either side, we’re not going to get where we need to get to.”
