Interpreting Your Results
Analysis Summary — Interval Summary
The Interval Summary table provides the top-level verification outcome. It reports:
| Metric | What It Tells You |
|---|---|
| Verification Result | PASS if ≤ 10% of results are outside the proposed interval; FAIL if > 10% are outside. |
| Proposed Reference Interval | The lower and upper limits of the preestablished reference interval as entered during experiment setup. |
| Range | The minimum and maximum raw values observed across all reference individuals in the study. |
| Mean | The arithmetic mean of all reference individual results. |
| Median | The middle value of the result distribution. Less sensitive to extreme values than the mean — useful for skewed datasets. |
| Standard Deviation | The sample standard deviation of all reference individual results. Quantifies the spread of the distribution. |
| Samples Outside Proposed Reference Interval | The count and percentage of results falling below the lower limit or above the upper limit. Must be ≤ 10% for verification to pass per CLSI EP28. |
| Total Samples | The number of results included in the analysis and the number excluded (if any were removed with documented justification). |
Analysis Summary — Distribution of Data
The Distribution of Data table lists each reference individual result alongside its classification — whether it falls within or outside the proposed reference interval. Review this table to:
- Identify which specific individuals produced results outside the interval and by how much.
- Determine whether outlying results cluster at the low or high end of the distribution, which may indicate a systematic shift or a subpopulation difference.
- Review individual results for data entry errors or sample handling issues before concluding that the reference interval is inappropriate.
Histogram Analysis
The histogram displays the frequency distribution of all reference individual results, overlaid with vertical markers at the lower and upper reference interval limits. Use the histogram to:
- Assess distributional shape — A roughly bell-shaped or unimodal distribution centered within the interval is consistent with a homogeneous healthy population. A bimodal or highly skewed distribution may suggest the need for demographic partitioning (e.g., by sex or age).
- Visualize outlying results — Bars falling outside the reference interval markers represent results that contribute to the "percent outside" count. The histogram makes it easy to see whether failures are isolated extreme values or a broader distributional shift.
- Evaluate interval fit — When the majority of results are concentrated well within the interval and only a small number of tails extend beyond the limits, the interval is a good fit. When a large proportion of results cluster near or beyond the limits, the interval may not match your population.