SBA — Single Best Answer · Review
Question 700
Statistics → Epidemiology
Classification
Question
SSBAQuestion Header
Statistical significance
Question Stem
In a very large study, an exposure increases the risk of an outcome from 0.0001% to 0.0002%, and the difference is statistically significant. Which conclusion is most appropriate?
Options
A
Statistical significance necessarily means the effect is clinically important
B
Statistical significance and clinical significance are distinct
C
The difference cannot be statistically significant because the absolute risk is small
D
A statistically significant result must have a narrow confidence interval
E
The null hypothesis should be retained whenever absolute risk is low
Explanation
A very small absolute difference can achieve statistical significance, particularly in a large study. Statistical significance therefore does not necessarily imply that the magnitude of an effect is clinically important.
Option Validity
A) statistical significance does not establish clinical importance.
C) a small absolute effect may still be statistically significant.
D) statistical significance does not itself guarantee a narrow confidence interval.
E) the decision concerning the null hypothesis is not determined simply by the absolute risk.
Further Reading
Review the distinction between statistical significance and clinical significance.