Statistical Considerations Support the Supreme Court’s Decision in Matrixx Initiatives V. Siracusano

Abstract:     
In Matrixx Initiatives v. Siracusano, 568 U.S. (2011), the Supreme Court ruled that information about the number of adverse events occurring to users of a drug need not be sufficiently numerous to reach statistically significance before this knowledge should be disclosed to potential investors. None of the opinions or briefs presented a statistical analysis of the data. This paper provides a statistical method for comparing the proportion of individuals with a particular adverse event who had used the product under investigation with the market share of the product and concludes that there was a statistically significant excess of cases of loss of smell in users of Zicam nasal spray made by Matrixx. Because adverse event reports are not based on a random sample, this conclusion is subjected to a sensitivity analysis showing that even if Zicam users were twice as likely to take the medicine as users of other cold medicines and twice as likely to report a problem statistical significance remains. It is demonstrated that the studies of effectiveness of Zicam were too small to detect a small but important risk to users. Finally, public policy concerns also support the Court’s decision since a drug manufacturer who receives complaints but does not disclose them effectively eliminates the possibility that other medical researchers would carry out an epidemiologic study.

Number of Pages in PDF File: 16

working papers series