Why Adverse Event Data Gets Misread¶
FDA’s own documentation is clear on each of these points. These are the misreadings that show up most often, with the wrong and the accurate way to state each one.
A Report Isn’t Proof Of Cause¶
FDA states that “a causal relationship cannot be established” between a product and the reactions listed in a report.
"Drug X caused 500 cases of liver damage."
"500 reports mention liver damage in patients taking Drug X."
Counts Aren’t Incidence Rates¶
FDA doesn’t receive a report for every adverse event, so the data cannot be used to estimate how often an event happens.
"Liver damage affects 2% of Drug X users."
"The data can't tell us how common liver damage is with Drug X."
A Drug In A Report Isn’t Always The Suspect¶
Reports also list concomitant and interacting drugs. Check drugcharacterization before attributing anything. See the annotated example.
Who Reported It Matters¶
Nearly half of reports originate with a consumer rather than a health professional
All openFDA drug adverse event reports with a known reporter type (19.9 million), by primarysource.qualification, which records who originally reported the event. Many of these reports reach FDA through drug manufacturers. Captured October 29, 2026; data as of July 30, 2026.
A Spike Isn’t Automatically A Signal¶
News coverage, lawsuits and how long a drug has been on the market all change how many reports get filed.
More Reports Can Mean More Use¶
A widely prescribed drug will collect more reports than a rarely used one, whatever its safety.