In a verification study of a reference interval with 20 subjects, how many results should fall within the interval for acceptability?

• A. 1
• B. 2
• C. 4
• D. 6

Answer: (B)

In verification of a reference interval using 20 healthy subjects, you’re checking how well the established interval agrees with real data from a small sample. The key idea is that most of the results should fall inside the interval. Practically, no more than about two of the 20 values should lie outside the interval, which means at least 18 of the 20 results must be within the interval for it to be considered acceptable. If three or more results fall outside, the interval isn’t verified with this sample and you’d typically need more data or a new reference interval.

So, the correct standard criterion is that 18 or more results should lie within the interval (i.e., at most two outside). The options given don’t align with this, so the fundamental rule is the count inside should be at least 18 out of 20.