Evidence based medicine is moving from relatively simple questions about a drug's safety and effectivenesss to thornier questions of relative clinical effectiveness and cost effectiveness, international experts agreed at a meeting last week. The experts, from the United States, Canada, the United Kingdom, and Germany, had assembled to discuss these issues with congressional and policy staff in Washington, DC.
"While there is a lot of data out there, unfortunately it is often very poor," said Dr Peter Littlejohns, director of the UK's National Institute for Health and Clinical Excellence. This leads to an evaluation that is based more on cost than on effectiveness, within a framework of ethical values, he said.
Given that policy decisions will be based on imperfect information, what is the tolerance of error, and what kind of error can be tolerated (false positives and false negatives), asked Dr Marc Berger, a vice president for research with the drug company Merck.
He argued that the answers are likely to vary, depending on the severity of the medical condition and what treatment options were available. Involving all stakeholders in setting priorities and in determining the methods and the degrees of certainty needed to reach conclusions is integral to building trust in final recommendations, he said.
For Dr Andreas Laupacis, chair of the Canadian Expert Drug Advisory Committee, the critical element is the effectiveness of the drug. The Canadian Common Drug Review has examined 22 drugs and recommended that provincial health agencies add only nine of them to their formularies for coverage.
The Common Drug Review tends to be conservative, because it is hard to remove a drug once it has been approved, but it is relatively easy to conduct a re-evaluation, should important new data become available, Dr Laupacis said.
He pointed to two examples that supported this approach. Although the US FDA used preliminary data to approve the anticancer drug gefitinib, the Common Drug Review hesitated, and subsequent data showed the drug to be ineffective. And even though the Canadian body approved rofecoxib, the provincial government of British Columbia was not convinced of its effectiveness and chose not to add it to its formulary. That decision saved patients in the province from unnecessary exposure to risk.
Mark Gibson, deputy director of the Drug Effectiveness Review Project at the Oregon Health and Science University, said that within the US—perhaps due in part to a heavy campaign of advertising directly to consumers—educating doctors to prescribe first the most cost effective drug for a medical condition was sufficient to shift patterns of usage and market share by 30%. Administrative or formulary changes were necessary to achieve greater levels of change in clinical practice.