Failure to thoroughly investigate an unexplained discrepancy or a batch failing to meet its specifications has been the number one or number two cited FDA 483 observation for four consecutive years. 21 CFR 211.192 is not an obscure requirement. It is the single most commercially consequential regulation in the entire Code of Federal Regulations for a working QA or QC professional, because it is the one inspectors cite most often when a laboratory result falls outside its specification.
Two 2026 FDA warning letters confirm the stakes have not declined. This article covers the complete two-phase OOS investigation framework FDA expects, the legal foundation behind it and the four citation patterns that consistently appear in 483 observations and warning letters.
21 CFR 211.192 requires that any unexplained discrepancy or the failure of a batch to meet any of its specifications be thoroughly investigated, whether or not the batch has already been distributed. The investigation must extend to other batches that may have been associated with the specific failure and must include a written record with conclusions and follow-up.
The legal foundation - United States v. Barr Laboratories
The modern OOS investigation framework traces directly to the 1993 federal court decision in United States v. Barr Laboratories. That case established the legal principle that an OOS result cannot simply be discarded or retested until an acceptable result appears. FDA's 2006 guidance on Investigating Out-of-Specification Test Results formalised the two-phase investigation model that remains the industry standard today.
The Barr decision is worth understanding because it explains why FDA's expectations are structured the way they are. The court found that a manufacturer cannot presume a laboratory error caused an OOS result without conducting and documenting an investigation that supports that conclusion. Retesting alone, without a documented and scientifically justified basis, is not an acceptable response to an OOS result.
The two-phase investigation framework
Laboratory investigation
The first phase determines whether the OOS result is attributable to a clear, documented laboratory error. This requires a full review of the analyst's technique, the equipment used, the reagents and standards, calculation checks and a review of the raw data. Retesting can only proceed once a specific, assignable cause has been identified and documented - not simply because the analyst suspects an error occurred. If Phase I identifies a clear laboratory error, the investigation may conclude at this stage with the original result invalidated.
Manufacturing investigation
If Phase I does not identify a clear laboratory cause, the investigation expands into the manufacturing process itself. This covers batch record review, process parameter review, equipment qualification status, raw material and component history and a review of any other batches manufactured using the same equipment, process or materials. Phase II is where the investigation determines whether the OOS result reflects a genuine product quality issue rather than an analytical error - and where the scope assessment for other potentially affected batches occurs.
The four citation patterns FDA consistently documents
Citation 1 - Retesting without an assignable cause
The single most common OOS-related citation is retesting conducted without first identifying and documenting a specific, assignable laboratory cause. FDA's position since the Barr decision has been unambiguous: an analyst cannot retest simply because the first result was unexpected or inconvenient. A documented investigation must precede any retest, and the investigation must identify a specific reason to suspect the original result does not reflect true product quality.
Citation 2 - Undefined or inconsistent retesting strategy
A related and frequently cited gap is the absence of a pre-defined, SOP-based retesting strategy. How many retests are permitted. What statistical or scientific basis determines which result is reported. Whether an outlier test is used and whether its application is scientifically justified for the specific test method. Inspectors look for a documented, consistently applied approach - not case-by-case decisions made after seeing the results, which creates the appearance and often the reality of testing into compliance.
Citation 3 - Scope not extended to other batches
21 CFR 211.192 explicitly requires that an investigation extend to other batches that may have been associated with the specific failure. A narrow investigation that examines only the failing batch, without assessing whether the same root cause could affect batches manufactured on the same line, with the same materials, or around the same time, is a frequent citation. FDA inspectors specifically ask what other batches were assessed and what the documented rationale was for including or excluding them.
Citation 4 - Root cause not identified or investigation closed prematurely
An investigation that concludes with no root cause identified, or with a vague conclusion such as probable analyst error without supporting evidence, does not satisfy the thoroughness requirement of 21 CFR 211.192. FDA expects investigations to reach a documented, evidence-supported conclusion. When a genuine root cause cannot be determined despite a thorough investigation, that must be documented explicitly along with the rationale for the batch disposition decision - not left as an open question.
When FDA investigators review an OOS investigation file they typically request the full laboratory notebook and raw data, the retest authorisation and rationale, the scope assessment for other batches and the final disposition decision with its supporting justification. A file missing any of these four elements is likely to generate an observation regardless of what the ultimate batch disposition was.
Building a defensible OOS investigation programme
A compliant OOS investigation programme requires several elements working together, not just a procedure document.
- A written SOP defining the two-phase investigation process, retesting criteria, statistical methods for outlier evaluation where applicable and timelines for investigation completion
- Trained analysts and investigators who understand both the mechanics of the investigation and the regulatory rationale behind it - not just a checklist to complete
- A documented decision framework for when retesting is appropriate, how many retests are permitted and how conflicting results are resolved
- A scope assessment template that prompts investigators to systematically consider other batches, equipment, materials and time periods that could be affected
- Quality unit oversight of every OOS investigation before final batch disposition, consistent with the quality unit's authority under 21 CFR 211.22
OOS investigations sit at the intersection of laboratory science and regulatory defensibility. A scientifically sound investigation that is poorly documented will still generate a citation. A well-documented investigation that lacks scientific rigour will not survive scrutiny either. The Barr Laboratories decision and three decades of subsequent FDA guidance and enforcement have made clear exactly what a compliant investigation requires - the gap between knowing the requirement and consistently meeting it is where most 483 observations in this area originate.
Key regulatory references
- 21 CFR 211.192 - Production record review
- United States v. Barr Laboratories, Inc., 812 F. Supp. 458 (D.N.J. 1993)
- FDA Guidance for Industry - Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production, 2006
- 21 CFR 211.22 - Responsibilities of quality control unit
Our OOS Investigations course covers the full framework
The complete Phase I and Phase II investigation model, retesting strategy design, scope assessment methodology and the Barr Laboratories legal foundation. Certificate on completion.