Reporting Bias in Drug Trials Submitted to the FDA
Author Information
Author(s): Kristin Rising, Peter Bacchetti, Lisa Bero
Primary Institution: University of California San Francisco
Hypothesis
Not all data submitted to the FDA for a new drug approval are published and certain trial characteristics may be associated with publication.
Conclusion
Many trials were still not published 5 years after FDA approval, and discrepancies between FDA reviews and published trials often favored the test drug.
Supporting Evidence
- Seventy-eight percent of efficacy trials contained in FDA reviews of NDAs were published.
- Trials with favorable primary outcomes were nearly five times as likely to be published.
- Forty-one primary outcomes from the NDAs were omitted from the papers.
Takeaway
Some drug trials don't get published, and when they do, the results can look better than they really are.
Methodology
Observational study of efficacy trials in approved NDAs for new molecular entities from 2001 to 2002, comparing publication status and outcomes.
Potential Biases
Potential for selective reporting and framing of data in published articles.
Limitations
The study could not determine why results changed from FDA reviews to publications, and data sources from the FDA were not easily accessible.
Statistical Information
P-Value
p = 0.0039
Confidence Interval
95% CI 72%–100%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website