Define the charted quantity and population first
Relative potency, reference EC50, control recovery, slope, asymptotes, and replicate CV answer different questions. Select a result that is stable enough to trend and define which runs belong to the baseline population.
Do not combine incompatible method versions, matrix conditions, instruments, or control lots merely to increase the number of points. Stratify or annotate known changes so a shift can be interpreted.
Distinguish statistical limits from specifications
Control limits describe expected process behavior under a defined baseline. Specification or acceptance limits define what the method or product requires. They can coexist, but substituting one for the other changes the question.
Estimate limits from an appropriate stable period, preserve the baseline version, and define what happens when a rule signals. A flag should initiate review under a procedure, not silently invalidate an assay.
Preserve the context behind every trend point
A plotted point should link to its run, method and settings version, sample or control lot, instrument, analyst, result, suitability state, exclusions, and report. Filters and annotations should not alter source records.
When a point signals, reviewers should be able to compare relevant runs, see known changes, record the investigation, and retain the conclusion without changing the original result.
- Versioned baselines and control limits.
- Method-, lot-, instrument-, and analyst-aware filters.
- Informational rule flags with visible definitions.
- Annotations for planned and investigated events.
- Direct navigation from a point to its source analysis.
Frequently asked questions
Are Westgard rules automatically appropriate for potency assays?
No. They were developed in a particular laboratory QC context. Any transferred rule needs a justified charted quantity, error model, false-signal assessment, and procedural response.
Is every point outside a control limit an invalid assay?
No. A statistical signal indicates behavior that merits evaluation. Validity and release decisions follow the approved method and quality procedure.