Why is multiple testing a concern, and how is it addressed in outbreak analyses?

• A. It increases false positives; address via corrections like Bonferroni or false discovery rate control.
• B. It decreases false positives; corrected by increasing alpha.
• C. It has no effect; ignore.
• D. It guarantees finding true associations.

Answer: (D)

Testing many exposures or time frames increases the chance of false positives. In outbreak analyses, investigators often explore a large number of hypotheses—different foods, vendors, time windows, or geographic sources—which can produce apparent associations purely by chance even when no real link exists. To prevent chasing noise, analysts adjust for multiple testing. Common approaches include controlling the family-wise error rate with methods like Bonferroni or Holm-Bonferroni, which tighten the significance threshold as the number of tests grows. Another approach is controlling the false discovery rate, such as with the Benjamini-Hochberg procedure, which limits the expected proportion of false positives among the declared findings and often preserves more statistical power. In practice, teams may pre-specify key hypotheses, use sequential or hierarchical testing, or validate findings with independent data or permutation-based methods. The goal is to reduce the chance that a detected signal is just random noise while still allowing genuine associations to be identified.