Start with the decision, not another test
A surprising laboratory result creates two questions: does it fail an established acceptance criterion, and what evidence is needed to understand it? A passing repeat test does not, by itself, explain the first result. This article offers a starting point for a QA/QC discussion, rather than a substitute for an approved investigation procedure.
OOS and OOT: different signals
OOS means a result falls outside an established specification or acceptance criterion. OOT describes an unexpected pattern or departure from an expected trend; a result can remain within specification and still raise a concern. Define trend rules for the product, method and data set in your approved procedures. Do not invent one universal percentage change for every test.
For example, a hypothetical assay specification of 95.0–105.0% makes a result of 94.0% OOS. A sequence moving from 100.2% to 99.1% to 98.0% remains within that example specification; whether it triggers an OOT review depends on the approved trending approach and context. These numbers illustrate classification, not release criteria.
Make the initial review traceable
Begin with the original record: sample identity, method version, raw outputs, calculations and the sequence of events. Section 211.194 requires complete laboratory records, including testing data, calculations, results and review. A useful review note tells the next person what was examined, where the evidence is stored and what remains uncertain. Avoid replacing a documented observation with a broad label such as “analyst error.”
When should the investigation expand?
FDA’s OOS guidance describes an initial laboratory assessment followed by a full-scale investigation when a clearly causative laboratory error is not established. Its stated scope is chemistry-based testing of CDER-regulated drugs; it is not a universal rulebook for every medical-device test or biological assay.
Separately, 21 CFR 211.192 requires thorough investigation of unexplained discrepancies and specification failures, including consideration of other associated batches or products. Rejecting a batch does not answer whether a wider problem exists. Write down the question that requires escalation, the affected scope and the accountable quality-unit owner.
Keep retesting separate from explanation
FDA’s guidance discourages repeated testing aimed at obtaining a pass. Additional testing needs a scientific purpose and predefined controls. Ask what hypothesis the next test addresses and how every result will be evaluated. A result is evidence; it is not a shortcut around the investigation.
Turn the review into a useful handover
Before the next review meeting, use the free checklist to organize questions and evidence references. Identify the person responsible for each open item. Keep confidential batch records in your controlled quality system, not in a public download or a website form.
For further study, EduCraft’s on-demand OOS and OOT Investigations Masterclass covers investigation phases, root cause analysis, trending, retesting, CAPA and documentation. Review the full curriculum and instructor details to decide whether it fits your role.
Go deeper into OOS and OOT investigations.
9 hours · On-demand · Meredith L. Crabtree
Explore the curriculum, intended audience and faculty background. Confirm current price and access terms on educraft.us.
Explore the masterclass →Sources and scope
- FDA: Investigating OOS Test Results for Pharmaceutical Production — May 2022, sections III–V.
- 21 CFR 211.192: Production record review.
- 21 CFR 211.194: Laboratory records.
Educational overview for pharmaceutical QA/QC, reviewed against these sources on 18 November 2026. FDA guidance contains nonbinding recommendations; applicable regulations and approved procedures govern actual decisions. The OOT example and review checklist are EduCraft educational tools. This article does not establish a release decision or imply FDA endorsement.