Interpreting Your Results
Summary Statistics Table
The results table shows five rows for both PT and INR side by side:
| Row | PT Column | INR Column |
|---|---|---|
| Geometric Mean (Normal Patient Mean) | The NPM — the geometric mean of all PT values in the study. This is the value that will be used as the denominator in the INR formula for patient testing. | The geometric mean of all computed INR values. For a typical normal population this will be very close to 1.0. |
| 2SD Range | Arithmetic mean ± 2 sample standard deviations of the observed PT values. Describes the spread of the study distribution. | Arithmetic mean ± 2 sample standard deviations of the computed INR values. |
| Proposed Limits | The PT Lower and Upper Limits as configured in the experiment setup. | The INR limits — either configured directly or derived from the PT limits via INR = (PT / NPM)^ISI. |
| % Out of Limits | Percentage of normal subjects with PT outside the proposed PT limits. | Percentage of normal subjects with computed INR outside the proposed INR limits. |
| Status | Pass if ≤10% of subjects are out of PT limits; Fail otherwise. | Pass if ≤10% of subjects are out of INR limits; Fail otherwise. |
Histogram Analysis
Separate histograms are displayed for PT and INR. Each histogram shows:
- Y-axis: Percentage frequency — the proportion of subjects falling within each bin, expressed as a percentage of the total sample count
- X-axis: PT values (seconds) or INR values
- Vertical dotted lines: The lower and upper reference interval limits. Bars that extend beyond these lines represent subjects who fall outside the reference interval
A well-characterized normal population will produce a roughly bell-shaped distribution centered near the NPM (for PT) or 1.0 (for INR), with the reference interval lines capturing the central 90% or more of the distribution. If the distribution is skewed, bimodal, or has heavy tails extending beyond the reference interval lines, this may indicate that the subject population includes individuals who do not meet the normal subject criteria.
Interpreting Pass vs. Fail
A Pass result means the reference interval is appropriately wide for the local patient population — no more than 10% of normal subjects fall outside it. The reagent lot may be placed into service using the established NPM and the verified reference limits.
A Fail result means the reference interval is too narrow. More than 10% of normal subjects fall outside the limits, which would result in an unacceptably high false-positive rate in clinical practice. Options when a study fails:
- Widen the PT reference limits and re-evaluate. The 2SD Range shown in the summary table is a useful starting point — limits set at the observed mean ± 2SD will, by definition, capture approximately 95% of a normally distributed population.
- Review the subject pool for inclusion criterion violations — subjects with undiagnosed coagulation disorders, liver disease, vitamin K deficiency, or anticoagulant use will shift PT values higher and inflate the out-of-limits percentage.
- Verify the ISI value entered matches the package insert for this specific reagent lot. An incorrect ISI shifts all INR values systematically.