Key result
For rare events with 0 occurrences after n trials, the upper limit of the 1-tailed 95% confidence interval can be conservatively estimated using the Rule of 3 (3/n) when n ≥ 30.
Design
Review
Authors
Loading...
Zero events do not confirm zero risk; extends statistical tools for bounding rare complications in research.
This methodological review highlights that a zero occurrence of a complication in a sample does not equate to zero risk, and provides statistical approaches to estimate the upper bound of risk for rare events.
Ho et al. (2018) conducted a review in Rare complications with zero occurrences. Rule of 3 and other statistical methods was evaluated. For rare events with 0 occurrences after n trials, the upper limit of the 1-tailed 95% confidence interval can be conservatively estimated using the Rule of 3 (3/n) when n ≥ 30.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: