An out-of-specification result falls outside the acceptance criteria applicable to the test. It identifies a discrepancy requiring assessment. The result alone does not tell you whether the cause lies in the material, sampling, preparation, measurement or reporting.
Confirm the comparison that produced the label
FDA’s 2022 guidance defines out-of-specification results by reference to established specifications or acceptance criteria. Its pharmaceutical production context is a useful quality-management example; it is not a statement that Novum products have regulatory approval.FDA — Investigating Out-of-Specification Test Results for Pharmaceutical Production (opens in a new tab)
For an original simple case, a hypothetical assay has an inclusive acceptance interval of 9.0–11.0 mg per vial. A valid result of 8.6 mg lies outside that interval. This is a comparison with the stated criterion, not a recommended peptide specification.
| Item | Why it matters |
|---|---|
| Measured property | Content and chromatographic area are different attributes |
| Unit and basis | mg per vial differs from mg per mL |
| Applicable criterion | A limit from another product may not apply |
| Decision rule | Rounding and uncertainty may affect boundary decisions |
If the criterion or basis is absent, describe that missing information first. You cannot reliably classify a result by guessing the intended specification.
An observation is not yet a root cause
A low assay value could be a genuine low amount or could arise from a problem in its measurement. Neither explanation should be adopted solely because it is convenient.
FDA describes an initial laboratory assessment and, where no causative laboratory error is established, a broader investigation. The aim is to explain the result rather than presume laboratory error from the outset.FDA — Investigating Out-of-Specification Test Results for Pharmaceutical Production (opens in a new tab)
In an original investigation example, a recorded calculation uses a dilution factor of 5 when the documented preparation requires 10. That is a specific, checkable discrepancy. Saying that the analyst probably diluted it incorrectly supplies no comparable evidence.
Keep additional findings in the same record
Suppose the original 8.6 mg result is followed by a 10.1 mg result. Their disagreement deserves explanation. The later number does not tell the reader whether a preparation error was corrected, a different vial was selected or the material varies.
A comparison is most informative when it identifies what changed and why. A new analyst, fresh standard, different original unit and changed calculation are different interventions; changing them all together can make a cause harder to isolate.
A confirmed error can justify revising the interpretation, but the evidence connecting that error to the original result should remain available. An unexplained result should remain unexplained rather than acquire an invented cause.
Likewise, averaging a failing and passing value does not by itself investigate the failure. Whether averaging is appropriate depends on the method and the original decision framework, not the appeal of the new average.
Read the investigation outcome separately from disposition
The investigation outcome explains what was found about the discrepancy. The disposition records the decision made about the affected material. They are related but not interchangeable: rejecting material, for example, does not establish the mechanism that caused its result.
A useful account identifies the affected sample and attribute, the applicable criterion, evidence examined, supported cause if any, and remaining uncertainty. It should also make clear whether other results depend on the same identified problem.
For a reader outside the laboratory, the appropriate conclusion may be that the available report does not resolve the discrepancy. That is more accurate than declaring either the laboratory or the material correct without the investigation evidence.
Sources and further detail
- FDA — Investigating Out-of-Specification Test Results for Pharmaceutical Production (opens in a new tab)
May 2022, revision 1; definition and phase I/II discussion read. Used for investigative principles, not legal advice or a Novum procedure. All numerical examples are original.
Sources checked 19 September 2026. Worked examples are illustrative unless a supplied report is explicitly identified. This article has not undergone independent scientific peer review.