Describe › Reference intervals

Non-parametric reference interval

A non-parametric reference interval takes percentiles directly from the data by the CLSI EP28-A3 method, with 90% bootstrap CIs on each limit.

What is Non-parametric reference interval?

The percentile method takes the 2.5th and 97.5th percentiles of the reference sample directly as the lower and upper limits. No distributional assumption — works on skewed, heavy-tailed, or mixed-distribution data where the Gaussian method fails.

CLSI EP28-A3 specifies n ≥ 120 for the standard percentile method (so the 0.025·n = 3rd value is the lower limit, with reasonable bootstrap precision). Smaller samples give wider bootstrap CIs and less stable limits.

When should I use Non-parametric reference interval?

Most clinical lab analytes (skewed, non-normal).
When you don't want to commit to a parametric distribution.
When the Gaussian assumption is hard to defend.

What data does it need?

One numeric column (≥ 120 strongly recommended by CLSI).

What does it report?

Lower and upper percentile-based limits + 90% bootstrap CIs on each (B = 1000).

What does it assume?

Reference sample representative.
Independent observations.

How do I interpret the result?

Wide bootstrap CIs on the limits ⇒ insufficient n; recruit more reference subjects rather than over-trusting the point estimates.

References

CLSI EP28-A3 (2010). Defining, establishing, and verifying reference intervals in the clinical laboratory.