The regulatory system for reporting side effects caused by prescription drugs is producing its own kind of side effect — incomplete information about injuries that patients may have suffered. And the reason for the lack of data is that many drug makers are filing patchy reports, according to a new analysis published in the Pharmacoepidemiology and Drug Safety.
At issue is the Adverse Event Reporting System maintained by the Food and Drug Administration. The database is the key method for collecting side effect data. Both doctors and consumers can voluntarily report problems to the agency or a drug maker. But drug companies are also required to investigate and report side effects that may be attributable to their products.
Drug makers typically file the vast majority of reports of a serious or fatal outcome to the FDA. Yet compared with the side effect reports turned in by physicians and consumers, drug makers generally fail to include key data that the FDA might use to assess future warnings. For this reason, one of the study authors contends there are still wider implications.
“With increasing pressure for the FDA to approve drugs fast but with less clinical testing, it is a major concern that postmarket surveillance has major problems that are not being addressed,” said Thomas Moore, a senior scientist the Institute for Safe Medicine Practices, a nonprofit that tracks drug safety issues. “It is time for the FDA, the medical community, and industry to start work on a badly needed modernization” of this “critical tool” for monitoring safety.
The analysis found that in 2014, the FDA received 528,192 new reports of a serious or fatal side effect, of which 4.7 percent were filed with the agency directly by doctors and consumers. Of those, 86 percent included complete information about four important data points – patient age and sex, the date the side effect occurred, and a specific medical term to describe the problem.
By contrast, drug makers filed 95.3 percent of side effect reports, but most were incomplete. For instance, 40 percent of so-called expedited reports — those that must be filed within 15 days of learning of a serious side effect — were incomplete. And 51 percent of periodic reports, which drug companies generally file up to 90 days later, were similarly incomplete.
For instance, among the reports of serious side effects submitted by drug makers to the FDA, nearly 38 percent lacked the patient’s sex and age, and 47 percent did not have the date when the problem occurred. Overall, reports involving patient deaths offered the least amount of complete information for all of the key data points.
Moore suggested the reporting system has not kept pace with changes in pharmaceutical marketing. Since the last upgrade in 2001, interactions between patients and physicians, and patients and drug companies, have evolved, partly thanks to the Internet. This means a report may not be based on spontaneous information.
We asked the FDA for comment and will update you accordingly.
As for the drug makers, the analysis reviewed 25 companies that submitted 5,000 or more reports in 2014, and the overall performance was not very good. Less than 25 percent of the reports filed by two companies — GlaxoSmithKline and Gilead Sciences — contained all four key data points. The best performance was turned in by Biogen, which filed complete reports 67 percent of the time.
A Gilead spokeswoman wrote us that “patient safety is of the utmost importance to Gilead. Gilead monitors the safety of all its medications, reports adverse events in a timely manner, and is compliant with FDA and other health authority safety reporting requirements.”
As for Glaxo, the analysis noted that one episode likely accounted for the poor showing. In 2014, the company submitted more than 20,000 side effects reports related to a settlement of legal claims involving its Avandia diabetes pill. Although the FDA required Glaxo to file the reports lickety split, the company ”apparently had no further information immediately” about patient age, sex, event date, or a medical term.
[UPDATE: An FDA spokesman later sent us this: The FDA is “aware that the quality of adverse event reports may vary,” although only requires four data elements in a report for it to be filed: “an identifiable reporter,” “an identifiable patient,” “an identifiable adverse event” and a “suspect product.”
“The system does not reject electronically submitted cases on the basis of content quality beyond those elements… Improving the quality of information in (the reporting system) is of great interest to FDA, but is challenging because of the voluntary nature of reporting.”
He also acknowledged that a majority of FDA’s safety-related labeling changes come from these reports as the sole source.”]