The FDA clinical trials used to evaluate drugs between 2005 and 2012 varied widely in their thoroughness, researchers report.
Published in JAMA, the study is the first systematic analysis of the standard used by the FDA in making drug approval decisions.
“We found that during the study period, more than one-third of the drugs were approved on the basis of a single trial, without replication, and many other trials were small, short, and focused on lab values, or some other surrogate metric of effect, rather than clinical endpoints like death,” says first author and Yale School of Medicine student Nicholas S. Downing.
Downing and the team evaluated the strength of clinical trial evidence supporting FDA approval decisions for new drugs by characterizing key features of efficacy trials, such as trial size, design duration, and end points.
They used publicly available FDA documents to identify 188 novel therapeutic agents for seven years. These medical review documents summarized in great detail the rationale behind FDA approvals.
“Based on our analysis, some drugs are approved on the basis of large, high-quality clinical trials, while others are approved based on results of smaller trials,” says senior author Joseph Ross, assistant professor of internal medicine at Yale School of Medicine. “There was a lack of uniformity in the level of evidence the FDA used.”
He adds: “We also found that only 40 percent of drug approvals involved a clinical trial that compared a new drug to existing treatment offerings. This is an important step for determining whether the new drug is a better option than existing, older drugs.”
Downing says survey data show that patients expect drugs approved by the FDA to be both safe and effective. “Based on our study of the data, we can’t be certain that this expectation is necessarily justified, given the quantity and quality of the variability we saw in the drug approval process,” he says.
The Pew Charitable Trusts partially funded the study, which Nicholas and colleagues conducted at the Yale Center for Outcomes Research & Evaluation.