Out-of-Specification (OOS) Investigation for Pharmaceutical Quality Control

A1.1  Defining OOS

Practice An out-of-specification (OOS) result is a test result that falls outside the specifications or acceptance criteria established in the marketing application, a compendial monograph, or an approved written procedure. The term itself is industry and FDA-guidance vocabulary rather than a phrase defined by name in 21 CFR — the regulation that actually creates the investigation obligation, 211.192, uses broader language, covered in Module A4. What makes a result OOS is its relationship to a specification, not how surprising it is, how it compares to history, or how confident the analyst feels about it.

A1.2  OOS versus OOT: two different signals

Practice An out-of-trend (OOT) result is a result that remains within specification but breaks from the product's established historical pattern — a stability timepoint drifting in a direction inconsistent with prior data for that batch or product, for example. An OOT result is not itself an OOS result and does not, by itself, trigger a 211.192 investigation in the way a specification failure does. It does, however, warrant its own evaluation, since a trend break can be an early signal of a problem that has not yet produced an outright failure.

• OOS — out-of-specification — a result outside the specification or acceptance criterion itself. Triggers the 211.192 investigation obligation covered in Module A4.
• OOT — out-of-trend — a result inside specification but inconsistent with the established historical pattern for that product or process. Not itself an OOS event; evaluated on its own terms, often as part of ongoing trend monitoring.
• Discrepancy — the broader term 211.192 actually uses, reaching any unexplained inconsistency in production and control records, of which a specification failure is one example. Covered in full in Module A4.

A1.3  The scope of FDA's OOS guidance

Requirement FDA's Guidance for Industry: Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production — originally issued in 2006 and current in its Level 2 revision, issued May 2022 — is scoped to chemistry-based laboratory testing of drugs regulated by CDER: active pharmaceutical ingredients, excipients, in-process materials, and finished drug products under 21 CFR Parts 210 and 211. It is one evolving document, not two separate guidances; "Level 2 revision" describes how substantially FDA revised its own prior version, not a newer document superseding an unrelated older one.

---

A2.1  The case

Requirement United States v. Barr Laboratories, Inc., 812 F. Supp. 458 (D.N.J. 1993), arose from FDA's action against a generic drug manufacturer over its handling of OOS test results. The court's opinion did far more than resolve that one dispute — it laid out, in detail, what a legally defensible OOS investigation actually requires, and it remains the reference point FDA's own later guidance was built on.

The court's central holding is the idea this entire course returns to again and again: an OOS result is not, by itself, proof that a batch failed. It is a trigger for an investigation. Only if that investigation cannot identify an assignable, documented cause — a genuine laboratory error, for example — does the result stand as a true, confirmed batch failure.

A2.2  The two-phase framework

Requirement The court described an informal Phase One review for a single OOS result — the analyst reports the result to a supervisor, and the two of them review the test notebook, the procedure, the calculations, and the equipment together — requiring, in the court's words, "more than a laboratory investigation consisting principally of retesting." Where Phase One does not resolve the matter, or where the failure is serious or recurring, the court described a formal Phase Two investigation: a full-scale inquiry reaching beyond the laboratory bench into the manufacturing process itself.

Practice This is not a coincidence of naming. FDA's own guidance, covered in Module A3, adopts essentially the same two-phase structure — Phase I and Phase II — and elaborates on it with specific documentation requirements. The vocabulary in the current guidance is the case law's vocabulary, formalized.

A2.3  Retesting, resampling, outlier tests, and averaging

Requirement The court held that retesting is proper "only after a failure investigation is underway," and it specifically rejected Barr's own rigid "two-out-of-three" retest procedure — the practice of retesting a fixed number of times and accepting the majority result — because it substituted a predetermined formula for the case-by-case scientific judgment an investigation actually requires.

Requirement On resampling — testing a new specimen entirely, rather than retesting the original sample — the court required "evidence, not mere suspicion" that the original sample was not representative of the batch before resampling could be justified.

“...the USP’s silence...with respect to chemical testing and outliers is prohibitory.”United States v. Barr Laboratories, Inc., 812 F. Supp. 458 (D.N.J. 1993).

Requirement The court held that outlier tests — a statistical method for excluding an extreme result from a data set — are not appropriate for chemical assays, reserving them for biological or antibiotic assays given their "substantial innate variability." On averaging, the court warned that relying on an average figure "without examining and explaining" the individual OOS results within it is "highly misleading and unacceptable," because averaging can "hide[]" the variability among individual test results that the investigation is supposed to be examining in the first place.

---

A3.1  Phase I: the laboratory investigation

Requirement A single OOS result triggers an informal Phase I laboratory investigation, conducted by the analyst together with a supervisor. FDA's guidance describes a specific checklist of steps this investigation should cover:

• Discuss the test method with the analyst — confirm the analyst's knowledge of, and performance against, the correct procedure. Aimed at identifying analyst-level or procedural errors first.
• Examine the raw data — review chromatograms, spectra, and other instrumental output directly for anomalies, rather than relying only on the reported summary result.
• Verify the calculations — confirm that the arithmetic and any data transformations applied to the raw data are scientifically sound.
• Confirm instrument performance — check that the equipment used was functioning correctly at the time of testing.
• Confirm reference standards, solvents, and reagents — verify that these met their own quality control specifications before the test was run.
• Evaluate method performance — compare the result against the method's validation data and the product's own testing history.
• Document and preserve the assessment — fully record the Phase I review itself, regardless of outcome. This record is what later distinguishes a genuine investigation from an after-the-fact justification.

Requirement Where Phase I identifies a genuine, documented laboratory error, FDA's guidance is direct about the consequence: "In the case of a clearly identified laboratory error, the retest results would substitute for the original test result." Where Phase I does not identify an assignable cause, the matter is not resolved — it escalates.

A3.2  Phase II: the full-scale investigation

Requirement A formal Phase II investigation is triggered when Phase I does not identify an assignable cause, or when the failure is serious enough (or has recurred often enough) to warrant it. FDA's guidance specifies what a Phase II investigation must document:

Requirement Specifically, FDA's guidance requires a Phase II investigation to include: a clear statement of the reason for the investigation; a summary of the aspects of the manufacturing process that may have caused the problem; the results of a documentation review, with an assignment of actual or probable cause; the results of a review for prior occurrence of the same problem; and a description of the corrective actions taken.

---

A4.1  The text, in full

Requirement 21 CFR 211.192, titled "Production record review," reads as follows, in its entirety:

“All drug product production and control records, including those for packaging and labeling, shall be reviewed and approved by the quality control unit to determine compliance with all established, approved written procedures before a batch is released or distributed. Any unexplained discrepancy (including a percentage of theoretical yield exceeding the maximum or minimum percentages established in master production and control records) or the failure of a batch or any of its components to meet any of its specifications shall be thoroughly investigated, whether or not the batch has already been distributed. The investigation shall extend to other batches of the same drug product and other drug products that may have been associated with the specific failure or discrepancy. A written record of the investigation shall be made and shall include the conclusions and followup.”21 CFR § 211.192, Production record review.

A4.2  What each clause actually requires

• Quality control unit review before release — every production and control record, including packaging and labeling, must be reviewed and approved by the quality control unit against established written procedures before a batch is released or distributed. This is the gatekeeping function an OOS investigation ultimately feeds into.
• Two independent triggers — an "unexplained discrepancy" (the regulation's own example is a theoretical-yield percentage outside its established range) or a specification failure. An OOS test result is the most common trigger encountered in practice, but it is only one of two named categories.
• "Thoroughly investigated," regardless of distribution status — the obligation to investigate does not depend on whether the batch has already shipped. A batch already in commerce does not excuse or diminish the investigation requirement.
• Investigation must extend outward — to other batches of the same product, and to other products that may share the same cause. This is the regulatory basis for the "look-back" scope FDA warning letters routinely cite when a firm's investigation stopped at a single lot.
• A written record, with conclusions and followup — not merely a record that an investigation occurred, but one that states what was concluded and what was done as a result. This is the regulatory seed of the CAPA requirement Arc D builds on directly.

Practice Every idea covered so far in this arc is, in a real sense, an elaboration of this one paragraph. Barr Laboratories interpreted what "thoroughly investigated" has to mean in practice. FDA's Phase I/Phase II structure operationalizes it into a specific procedure. The distinction between OOS and OOT exists to identify, precisely, which results this regulation's investigation duty actually reaches.