An OOS Result Is Not Confirmed Failure or a Field Alert

An OOS Result Is Not Confirmed Failure or a Field Alert

When an analytical laboratory generates an out-of-specification (OOS) test result during pharmaceutical quality control testing, commercial and quality teams can collapse four later identities into that first number. Treating the result as an immediate scrap write-off, or as a 3-working-day Field Alert Report (FAR) clock that has already started, mixes laboratory-record, disposition, and postmarketing-reporting questions. Under Food and Drug Administration (FDA) current good manufacturing practice (CGMP) regulations and Agency guidance, an initial out-of-specification measurement is an unconfirmed laboratory signal—not an established manufacturing defect, not an automatic lot rejection, and not a field alert.

To answer the central operational question directly: for a single laboratory test result, an initial OOS value is merely the trigger for a structured, two-phase investigation mandated by 21 CFR 211.192 and detailed in the FDA’s May 2022 guidance for industry, Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production (Revision 1). The initial result should be marked against four separate, non-interchangeable operational identities before anyone treats it as commercial disposition or a postmarketing report:

• Phase I Laboratory Investigation: An immediate technical inquiry focused strictly on the analytical measurement process. Its sole objective is to determine whether the suspect value was caused by an identifiable laboratory error, analytical artifact, equipment malfunction, or operator mistake that justifies invalidating the discrete result before the test preparations are discarded.

• Confirmed Manufacturing OOS: A formal finding reached only after Phase I rules out laboratory error and Phase II production investigations confirm that the OOS measurement reflects the true quality of the batch. This finding attributes the root cause to an aberration of the manufacturing process, storage conditions, or raw material quality.

• 21 CFR 211.165(f) Batch Rejection: A finished-drug disposition identity under 21 CFR 211.165(f) and 21 CFR 211.22. If a confirmed OOS shows that the batch fails established standards or specifications, 211.165(f) says the drug product shall be rejected. That rejection does not substitute for, terminate, or satisfy the 21 CFR 211.192 investigation, which still has to determine cause and evaluate other associated batches or products.

• 21 CFR 314.81 Field Alert Report (FAR): A postmarketing regulatory notification submitted on Form FDA 3331a to the FDA district office. 21 CFR 314.81(b)(1) applies to NDA and ANDA applicants for products that have already been distributed into commerce. An OOS on an undistributed quarantine batch does not start the 3-working-day field alert clock unless the subsequent investigation reveals that previously distributed batches are compromised.

This analytical worksheet is designed specifically to maintain the operational boundaries among these four identities on a single laboratory result. It is not an analysis of in-specification stability drift, which is examined separately in our guide to stability out-of-trend signals. It is not a guide to process performance qualification records, which is analyzed in our review of PPQ protocol versus commercial batch release. It is not an evaluation of analytical transfer criteria, explored in our examination of analytical method transfer acceptance criteria, nor is it a briefing on complete response letters, detailed in CMC-only complete response letters, or contract manufacturing agreement disputes, covered in CDMO quality agreement red flags. Furthermore, this worksheet does not invent industry-wide OOS rates, fabricated investigation-day clocks, or imaginary search volumes. Every evaluation should rest on the cited regulations and on the May 2022 and July 2021 guidances, using “should” as recommendation unless 21 CFR or the FD&C Act is the hook. Adjacent manufacturing jobs remain separate: second-site tech-transfer package gaps; PAI readiness and launch-critical facility timing; Type V facility DMF versus PAI versus CDMO CGMP; sterile fill-finish Annex 1 CCS versus FDA 2004 aseptic RFQ; FDA drug-recall lookup; FDA drug-shortage list reading; FDA warning letters by the numbers; and eCTD Module 3 lifecycle hygiene.

What Counts as an OOS Result — and What Does Not

To establish operational rigor, analytical teams must understand the precise scope of what constitutes an OOS result under federal regulations. FDA’s May 2022 guidance, Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production (issued by CDER’s Office of Pharmaceutical Quality under Docket FDA-1998-D-0019, with landing page content confirmed current as of August 1, 2024), defines the term with explicit breadth. The designation includes all test results that fall outside the specifications or acceptance criteria established in approved drug applications (NDAs or ANDAs), Drug Master Files (DMFs), official compendia (such as the United States Pharmacopeia–National Formulary), or internal manufacturer release standards.

Critically, the guidance clarifies that OOS principles apply equally to in-process laboratory testing that is conducted against established in-process specifications. If an in-process laboratory test is outside established specifications, the May 2022 guidance treats that event as an OOS result. 21 CFR 211.192 still requires investigation of specification failure of a batch or any of its components, whether or not the batch has already been distributed.

However, the May 2022 guidance also defines explicit technical boundaries regarding what does not fall within its scope:

• Biological Assays: The guidance applies to chemistry-based laboratory testing of drugs regulated by CDER, including chemistry-based testing of biotechnology products under CDER’s jurisdiction, performed on APIs, excipients and other components, in-process materials, and finished drug products to the extent 21 CFR parts 210 and 211 and FD&C Act section 501(a)(2)(B) apply. It is not intended to address biological assays (for example in vivo assays or immunoassays).

• Process Analytical Technology (PAT): The guidance does not cover PAT operating models where continuous real-time process controls, multivariate statistical models, or automated feedback loops serve as the primary release mechanism in lieu of traditional end-product testing.

• Real-Time Equipment Adjustments: In-process testing performed strictly to trigger immediate operational adjustments to manufacturing equipment (such as compression weight adjustments within a verified operational range) does not constitute an OOS investigation if established product acceptance criteria are not breached.

• Out-of-Trend (OOT) Signals: An analytical result that falls comfortably within registered specifications but deviates noticeably from historical batch averages or statistical process trends is an out-of-trend observation, not an OOS result. Footnote 6 of the May 2022 guidance notes that while the general investigative principles of root-cause analysis can be helpful for evaluating OOT trends, an in-specification OOT signal does not trigger mandatory 21 CFR 211.165(f) product rejection or 21 CFR 314.81 field alerts.

21 CFR 211.192 requires that any unexplained discrepancy or the failure of a batch or any of its components to meet any of its specifications be thoroughly investigated, whether or not the batch has already been distributed. FD&C Act section 501(a)(2)(B) is the CGMP statute; 21 CFR 211.192, 211.165, and 314.81 are regulations. The May 2022 and July 2021 guidances are nonbinding except where they cite statute or regulation. Write “should” as recommendation unless a cited regulation is the hook.

The May 2022 guidance states that the responsibility of a contract testing laboratory in meeting these requirements is equivalent to that of a manufacturing firm. The laboratory should convey its data, findings, and supporting documentation to the manufacturing firm’s quality unit, which should then initiate Phase II whenever no clearly causative laboratory error was identified. A quality agreement does not transfer 21 CFR 211.22 approve-or-reject authority away from the owner’s quality unit, and it does not supersede the contract laboratory’s own CGMP investigation duty.

ICH Q7, as FDA’s September 2016 Q7 Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients, is the API overlay, not a second finished-product worksheet. Section 11.15 states that any OOS result obtained should be investigated and documented according to a procedure, and that any resampling and/or retesting after OOS results should be performed according to a documented procedure. Section 6.72 states that all deviation, investigation, and OOS reports should be reviewed as part of the batch-record review before the API batch is released. The May 2022 OOS guidance says part 211 citations pertain to finished pharmaceuticals but are also consistent with Agency guidance on CGMP for APIs with respect to laboratory controls, including OOS investigations. API OOS still requires an investigation identity; 21 CFR 211.165(f) is the finished-drug rejection rule, and 21 CFR 314.81 is an NDA/ANDA applicant report, not an API-only caption.

Phase I: A Measurement-Process Investigation That May Invalidate

Phase I represents the initial laboratory investigation. Its essential purpose is to evaluate the technical accuracy of the laboratory's data before the manufacturing process is implicated or disturbed. The first phase should include an initial assessment of the accuracy of the laboratory’s data, whenever possible before test preparations (including the composite or the homogeneous source of the aliquot tested) are discarded. The purpose is to identify the source as an aberration of the measurement process or an aberration of the manufacturing process.

The technical integrity of Phase I depends heavily upon operational discipline at the analyst bench. The guidance emphasizes that, whenever possible, the laboratory should keep the original test preparations—including the composite sample, sample stock solutions, chromatographic dilutions, and the homogeneous aliquot source—before they are discarded. Preserving these preparations allows hypotheses regarding instrument malfunction, pipette inaccuracies, volumetric dilution errors, or column degradation to be tested directly on the identical solutions that yielded the suspect result. Once an analyst recognizes that an unexpected measurement has occurred, they should not knowingly continue an analytical sequence that they expect to invalidate later for an assignable cause.

The laboratory supervisor plays a pivotal gatekeeping role in Phase I. The supervisor's assessment should be objective and timely, with no preconceived assumptions as to cause. A Phase I supervisory review should evaluate and document these elements:

1. Analyst Interview & Method Verification: A formal discussion with the testing analyst to confirm that approved, validated analytical procedures were executed without deviation, and that the analyst understood their procedural responsibilities.

2. Raw Data & Instrument Audit: A rigorous audit of original chromatographic integration baselines, absorbance profiles, spectral matches, baseline drift, retention time stability, and raw instrumental readouts.

3. Calculation & Conversion Verification: A line-by-line verification of all mathematical formulas, volumetric dilution factors, tare weights, moisture correction algorithms, and computerized data conversions.

4. System Suitability & Instrument Qualification: Verification that the analytical instrument satisfied all system suitability criteria (including theoretical plates, tailing factors, and injection precision) and was in a fully calibrated, qualified operational state.

5. Reagents, Standards & Solvents: An inspection of reference standard lot numbers, potency assignments, preparation timestamps, expiration dates, mobile phase formulations, and solvent purities.

6. Historical Method Performance: An evaluation of the analytical method's historical performance, verifying whether the method has exhibited similar anomalies, baseline artifacts, or instability during prior batch testing.

Under FDA CGMP standards, true laboratory error should be a relatively rare occurrence. Frequent laboratory errors within an analytical facility indicate systemic quality failures, including inadequate analyst training, poorly maintained or uncalibrated instrumentation, or careless laboratory technique. Consequently, the regulatory threshold for invalidating an OOS result is exceptionally high: invalidation of a discrete test result may be done only upon the observation and documentation of a test event that can reasonably be determined to have caused the OOS result.

If Phase I investigation establishes clear, documented evidence of a causative laboratory error (for example, documented spill, confirmed misidentification of a chromatographic peak, or verified use of an expired reference standard), the original laboratory result is formally invalidated. However, if the evidence of laboratory error remains unclear or unproven, the OOS result cannot be attributed to analytical error. In such cases, the result stands as an uninvalidated OOS, and the investigation should proceed into Phase II.

Furthermore, scientific invalidation must never be conflated with record expungement. As established in the FDA’s December 2018 guidance, Data Integrity and Compliance With Drug CGMP: Questions and Answers (Docket FDA-2018-D-3984), and codified in 21 CFR 211.180 and 211.194, invalidating a test result changes how that result is utilized in batch release calculations, but it does not erase the original data from the CGMP record. Complete raw data, audit trails, and the formal Phase I investigation justification must be permanently retained in the batch file and presented to the quality unit.

When testing is performed by a contract testing organization, the laboratory should convey its Phase I data, findings, and supporting documentation to the manufacturing firm’s quality unit. When investigation by a contract laboratory does not determine an assignable cause, all test results should be reported to the customer on the certificate of analysis; FDA also recommends that OOS investigation reports be provided to the customer. Under FDA’s November 2016 guidance, Contract Manufacturing Arrangements for Drugs: Quality Agreements, a quality agreement should designate responsibility for investigating laboratory OOS and out-of-trend results and for sharing those investigation reports. The agreement does not strip 21 CFR 211.22 approve-or-reject authority from the owner’s quality unit.

Retests, Resamples, and Why a Passing Average Is Not the Original Result

Once an OOS result enters Phase II, additional laboratory testing is frequently performed to probe the root cause. Regulatory compliance during this phase hinges upon distinguishing two fundamentally different sampling operations: retesting versus resampling.

• Retesting: Involves analyzing additional aliquots drawn from the same original homogeneous sample preparation that was evaluated during the initial test. Retesting is executed to assess analytical precision, chromatographic consistency, or instrument repeatability on the identical material that produced the suspect reading.

• Resampling: Involves collecting an entirely new physical sample from additional units collected as part of the original sampling procedure or from a new sample collected from the batch, if that is necessary. The original sample should be large enough to accommodate additional testing. In some situations it may be appropriate to collect a new sample. An investigation might later conclude that the original sample was prepared improperly and was therefore not representative of batch quality. Resampling is not a license to collect a new sample merely because the original representative sample failed.

Additional testing during an OOS investigation should be governed by a predefined written SOP. The maximum number of retests should be specified in advance and should not be adjusted depending on the results obtained. FDA inspections have revealed a strategy of repeated testing until a passing result is obtained, then disregarding the OOS results without scientific justification; the May 2022 guidance calls this practice of testing into compliance unscientific and objectionable under CGMP.

The May 2022 guidance addresses averaging during an OOS investigation. It states that averaging the result(s) of the original test that prompted the investigation with additional retest or resample results obtained during the OOS investigation is not appropriate. Averaging individual numbers hides sample variability, masks batch heterogeneity, and creates a fictitious appearance of compliance. To illustrate the severe risk of this practice, the FDA provides two mathematical examples in the guidance text:

• FDA Example 1 (Confirming Marginal Quality): Consider a drug product with an assay release specification of 90.0% to 110.0% of label claim. The initial laboratory test produces an OOS result of 89.0%. During the subsequent investigation, two retests yield 90.0% and 91.0%. If an analyst averages these three numbers (89.0% + 90.0% + 91.0% divided by 3), the resulting average is exactly 90.0%, which appears to meet the specification. However, evaluating the individual results clearly indicates that the batch is sitting right at the lower limit of acceptance, and the retests actually tend to confirm the validity of the initial OOS finding rather than refute it.

• FDA Example 2 (Masking Extreme Dispersion): In a second scenario, the initial assay result is 80.0% (substantially OOS). Two subsequent retests produce results of 85.0% and 105.0%. Averaging these three values (80.0% + 85.0% + 105.0% divided by 3) again produces an arithmetic mean of 90.0%. Yet relying on this average would be disastrous: the individual values exhibit massive variance (a 25% spread across individual measurements), demonstrating either complete analytical instability or profound physical heterogeneity within the product batch.

In both scenarios, the quality unit should evaluate the individual analytical determinations, not the obscured average. The only context in which averaging is scientifically legitimate is when an approved, validated analytical method explicitly specifies that multiple replicate determinations (such as duplicate injections from the same chromatographic vial or multiple titrations) are averaged to constitute a single reportable result. However, such method-specified averaging is permissible during an investigation only if it was the exact protocol utilized during the original test that produced the OOS result. Averaging is inappropriate for dosage-form content uniformity or in-process blend uniformity, where the fundamental purpose of the test is to measure unit-to-unit variability.

Confirmed Manufacturing OOS Is Not the Same Identity as 211.165(f) Rejection

When Phase I does not determine that laboratory error caused the OOS result and the testing results appear to be accurate, a full-scale Phase II OOS investigation should be conducted using a predefined procedure. Conducted under the leadership of the quality control unit, Phase II is a comprehensive, multidisciplinary investigation that engages manufacturing operations, process engineering, validation specialists, maintenance personnel, and contract manufacturing partners. Its core objective is to identify the manufacturing root cause and evaluate the broader operational impact of the defect.

Under the framework established by the May 2022 guidance and 21 CFR 211.192, every completed OOS investigation reaches one of three mutually exclusive concluding identities:

1. Scientifically Invalidated Result: Phase I investigation uncovered clear, documented proof of laboratory error (e.g., equipment malfunction, documented sample contamination, or verified calculation mistake). The suspect OOS value is formally invalidated and excluded from batch release considerations. If a laboratory error is clearly identified, the retest results would substitute for the original test result, but all original data must be retained under 21 CFR 211.180 and an explanation recorded. Even a legitimately invalidated result remains in the quality-unit packet with the investigation report.

2. Confirmed Manufacturing OOS: The investigation confirms that the measurement process was sound and that the OOS result reflects the true quality of the drug product. The defect is attributed to a manufacturing process failure, raw material degradation, or environmental lapse. Under 21 CFR 211.165(f), a batch failing to meet established specifications shall be rejected. Crucially, batch rejection is a commercial disposition identity—it does not terminate the 21 CFR 211.192 investigation. The inquiry must expand into a full batch-failure investigation across other batches of the same product and other products sharing common raw materials, equipment trains, or process steps.

3. Inconclusive Investigation: The investigation does not reveal a cause and does not confirm the OOS. The original OOS result should be given full consideration in the disposition decision. The quality unit might still ultimately decide to release, but only after a full investigation has shown that the OOS result does not reflect the quality of the batch, always erring on the side of caution. FDA’s own labeled illustration is an 89.5 percent composite assay, then seven passing retests clustered near 99 percent, with robust process history and other tests consistent. That path is not Phase I invalidation and not a confirmed manufacturing OOS. An inconclusive file is a third concluding state.

21 CFR 211.22(a) assigns the quality control unit responsibility and authority to approve or reject drug products, including products manufactured, processed, packed, or held under contract by another company. The owner’s quality unit retains final release of contract-manufactured product. That assignment does not transfer the contract facility’s own CGMP investigation duty, and a quality agreement cannot waive it.

Recent FDA regulatory enforcement underscores the strict separation between these investigative identities. In a Warning Letter issued to Unipack LLC (MARCS-CMS 716621, December 19, 2025), the Agency cited the facility for repeatedly invalidating original out-of-specification assay results without identifying an assignable laboratory error. The firm executed multiple retests until passing numbers were obtained and subsequently released commercial lots for distribution without identifying root causes. Furthermore, the FDA issued a separate violation under 21 CFR 211.194(a) for failing to maintain complete laboratory records, The 211.194(a) complete-data citation is a separate inspection ask from the 211.192 investigation citation. Unipack is enforcement evidence, not this worksheet’s fictional result.

A Field-Alert Question Exists Only if the Product Is Already Distributed

A field-alert question is a different identity from the laboratory result. An OOS does not start 21 CFR 314.81 merely because the result is out of specification.

The regulatory requirement for Field Alert Reports is codified at 21 CFR 314.81(b)(1). The regulation mandates that holders of an approved New Drug Application (NDA) or Abbreviated New Drug Application (ANDA) submit a formal report to the FDA district office responsible for the facility within 3 working days of receiving information concerning:

• Any incident that causes the drug product or its labeling to be mistaken for, or applied to, another article (mix-up); or

• Any bacteriological contamination, or any significant chemical, physical, or other change or deterioration in a distributed drug product, or any failure of one or more distributed batches to meet application specifications.

The critical operational boundary is distribution. Under the FDA’s July 2021 guidance, Field Alert Report Submission: Questions and Answers (issued by CDER, CBER, and ORA under Docket FDA-2018-D-2326, with landing page confirmed current as of November 5, 2024), this boundary is analyzed through definitive regulatory questions:

• Question 1g (Undistributed Lots): If a drug product has not been distributed and an OOS result is discovered during release testing, a Field Alert Report is not required. The FAR requirement exists specifically to alert the Agency to quality threats present in the commercial marketplace. However, Question 1g establishes a vital caveat: if the investigation of an undistributed batch failure reveals that previously distributed batches are compromised or may fail specifications, a FAR must be submitted for those distributed lots within 3 working days of receipt of the 314.81 information.

• Question 1h (Distributed Lots & 3-Day Invalidation): If an OOS result occurs on a distributed drug product (such as during ongoing annual stability testing or reserve sample re-evaluation), a FAR is required unless the result is scientifically invalidated within 3 working days. If laboratory error is clearly identified and documented within 3 working days, no FAR is required. If an OOS result is not scientifically invalidated, a FAR must be submitted within 3 working days of initial receipt of the OOS information. The May 2022 OOS guidance recommends a follow-up FAR when the investigation is completed; follow-up and final FARs are recommended, not a second independent legal clock.

• Question 3a (The 3-Working-Day Clock): Working days are Monday through Friday excluding U.S. Federal holidays. The FAR Q&A’s timing example treats the day information meeting 314.81(b)(1) is identified as day 0; the day a contract laboratory learns of a sterility failure is day 0 for the applicant. The time frame applies regardless of where the information is identified.

• Question 2 (Applicant Accountability): The NDA or ANDA applicant remains legally responsible for submitting the FAR, even when the OOS is first identified by an off-site contract manufacturer or third-party laboratory.

To maintain compliance hygiene, quality leaders must also keep related reporting frameworks separate from the 314.81 FAR identity:

• Media-Fill Failures: Question 1i clarifies that a media-fill failure is an aseptic process simulation failure, not an automatic FAR. A FAR is required only if that later investigation discovers 314.81(b)(1) information about distributed product.

• Recalls & Shortages: Question 1k notes that initiating a product recall under 21 CFR Part 7 or notifying FDA’s drug shortage staff under FD&C Act Section 506C does not replace or satisfy a mandatory FAR filing.

• Licensed Biologics (BLA Products): Products licensed under a Biologics License Application (BLA) are not subject to 21 CFR 314.81. Distributed biologic quality defects are governed by 21 CFR 600.14 Biological Product Deviation Reporting (BPDR), which requires submission of Form FDA 3486 within 45 calendar days. A 3-working-day Form 3331a FAR and a 45-day Form 3486 BPDR belong to separate legal regimes.

flowchart TD
    A["One laboratory result outside specification"] --> B["Phase I laboratory investigation 21 CFR 211.192"]
    B --> C{"Documented measurement-process test event?"}
    C -->|"Yes"| D["Invalidate that discrete result; do not use it for batch quality"]
    D --> E["Retain original data in the CGMP record 211.180 and 211.194"]
    C -->|"Unclear or no laboratory error"| H["Phase II quality-unit production investigation"]
    H --> I{"Already distributed under an NDA or ANDA?"}
    I -->|"Undistributed"| M["FAR field missing for this OOS unless a distributed batch is later implicated"]
    I -->|"Distributed"| J{"Scientifically invalidated within 3 working days?"}
    J -->|"Yes"| D
    J -->|"No"| K["314.81 Form FDA 3331a is an applicant report identity, not the investigation"]
    H --> O{"Concluding identity"}
    O -->|"Invalidated"| D
    O -->|"Confirmed manufacturing OOS"| P["Use the result; 211.165(f) rejection is disposition; extend 211.192"]
    O -->|"Inconclusive"| R["Original OOS still considered; any later release only after a full investigation shows it does not reflect batch quality"]

One Fictional Laboratory Result: Present, Missing, or Unknown

Work FDA’s published numeric illustrations as a labeled hypothetical, not a real batch. The May 2022 guidance’s first averaging example is a finished-drug assay specification of 90 to 110 percent, an initial OOS of 89 percent, and later retests of 90 percent and 91 percent. No interior laboratory packet, batch record, distribution history, or Form FDA 3331a was reviewed for this page. Mark four fields as present, missing, or unknown on that labeled hypothetical:

At the opening snapshot, none of the four identities has been established. Distribution status, NDA/ANDA versus licensed-biologic reporting, and contract-laboratory notification are adjacent unknowns, not extra confirmed fields. Averaging 89, 90, and 91 percent to 90 percent is not a passing result.

Keep the four fields from collapsing if later documents arrive. These are identity tests, not a disposition or FAR-filing instruction for the hypothetical:

• Phase I invalidation would be Present only after observation and documentation of a specific test event. Invalidation changes how the result is used in the disposition decision; it does not erase the original data from the CGMP record. This page does not approve a named lot.

• Confirmed manufacturing OOS would be Present if Phase I does not invalidate and the investigation indicates a factor affecting batch quality. That result should be used in evaluating the batch and should result in rejection in accordance with 21 CFR 211.165(f). Rejection then changes the file into a batch-failure investigation that must be extended to associated lots under 211.192. Confirmation is not itself the FAR.

• A 314.81 field-alert question would be Present only for an approved NDA or ANDA product that is already distributed, unless the distributed OOS is scientifically invalidated within 3 working days. An undistributed OOS is not a FAR; the question can reopen if the investigation then indicates failure of one or more distributed batches. The NDA/ANDA applicant owns Form FDA 3331a even when a contract laboratory first sees the result. A licensed biologic’s distributed deviation is a 21 CFR 600.14 Biological Product Deviation Report (Form FDA 3486, 45 calendar days), not this FAR field.

• Adjacent unknowns stay labeled as such: whether a contract laboratory sent complete Phase I data and the OOS to the manufacturing firm’s quality unit; whether the product is an NDA/ANDA or a licensed biologic; and whether the file is inconclusive rather than invalidated. An in-specification out-of-trend observation remains the published stability-OOT encyclopedia, not this worksheet.

Keep this one-result worksheet separate from adjacent jobs. The table below marks those boundaries; none of them is this four-field laboratory-result split:

Enforcement history reinforces that regulatory authorities scrutinize these distinctions rigorously. In a Warning Letter issued to contract testing laboratory Tentamus India Private Limited (MARCS-CMS 720463, March 3, 2026), the FDA cited the facility for re-issuing test reports that replaced an original 'Too Numerous To Count' (TNTC) microbiological result with a passing count without conducting an investigation. The Agency warned that private quality agreements with clients do not supersede CGMP duties to investigate all OOS findings and notify sponsors. Similarly, in a Warning Letter issued to Auriga Research Pvt. Ltd. (MARCS-CMS 730694, August 12, 2026), the FDA cited repeated testing into compliance and the failure to transmit suspect analytical data to customers. Quality agreements and commercial contracts can allocate responsibilities, but they cannot rewrite CGMP investigation or customer-notification duties. Tentamus and Auriga are enforcement evidence, not the worksheet result.

The operator job is to mark the four fields on one laboratory result and leave adjacent programs to their own pages. This page does not give lot-specific release, rejection, or FAR-filing advice, and it does not invent typical OOS rates or investigation-day clocks beyond the 3-working-day FAR rule.

Different jobs, already published: stability out-of-trend launch-supply encyclopedia; PPQ protocol versus commercial batch release; analytical method-transfer acceptance criteria; CDMO quality-agreement red flags; recall-notice triage; and shortage-list reading. None of those pages is this four-field laboratory-result worksheet.

Sources

• U.S. Food and Drug Administration (CDER), Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production (May 2022, Revision 1), Guidance for Industry, Docket FDA-1998-D-0019.

• U.S. Food and Drug Administration, Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production — Level 2 Revision Landing Page, content current as of August 1, 2024.

• U.S. Food and Drug Administration (CDER, CBER, ORA), Field Alert Report Submission: Questions and Answers (July 2021), Guidance for Industry, Docket FDA-2018-D-2326.

• U.S. Food and Drug Administration, Field Alert Report Submission: Questions and Answers — FDA Guidance Landing Page, content current as of November 5, 2024.

• U.S. Food and Drug Administration (CDER), Field Alert Reports, regulatory overview and instructions for Form FDA 3331a, content current as of December 12, 2025.

• Office of the Federal Register / FDA, 21 CFR 211.192 — Production record review, requirements for thorough investigation of any unexplained discrepancy or specification failure.

• Office of the Federal Register / FDA, 21 CFR 211.165 — Testing and release for distribution, finished pharmaceutical testing requirements and mandatory rejection under 211.165(f).

• Office of the Federal Register / FDA, 21 CFR 211.22 — Responsibilities of quality control unit, quality-unit authority and responsibility to approve or reject drug products, including those manufactured under contract.

• Office of the Federal Register / FDA, 21 CFR 211.160 — General requirements (laboratory controls), laboratory control documentation and representative sampling standards.

• Office of the Federal Register / FDA, 21 CFR 211.194 — Laboratory records, requirements for complete analytical data and raw instrument records.

• Office of the Federal Register / FDA, 21 CFR 211.180 — General requirements (records), retention of original analytical records and periodic review of investigations.

• Office of the Federal Register / FDA, 21 CFR 314.81 — Other postmarketing reports (field alert report), 3-working-day reporting requirement for distributed NDA and ANDA products.

• Office of the Federal Register / FDA, 21 CFR 600.14 — Reporting of biological product deviations by licensed manufacturers, 45-day postmarketing deviation reporting for licensed biologics (Form FDA 3486).

• U.S. Food and Drug Administration, Data Integrity and Compliance With Drug CGMP: Questions and Answers (December 2018), Guidance for Industry, Docket FDA-2018-D-3984.

• U.S. Food and Drug Administration, Contract Manufacturing Arrangements for Drugs: Quality Agreements (November 2016), Guidance for Industry, recommendations on cross-facility investigation duties.

• U.S. Food and Drug Administration (ICH Q7), Q7 Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients (September 2016), Section 11.15 on investigating API OOS results and section 6.72 on reviewing deviation, investigation, and OOS reports before API release.

• U.S. Food and Drug Administration, Warning Letter: Unipack LLC (MARCS-CMS 716621, December 19, 2025), enforcement regarding improper invalidation of OOS assay results and release without root cause.

• U.S. Food and Drug Administration, Warning Letter: Tentamus India Private Limited (MARCS-CMS 720463, March 3, 2026), contract laboratory enforcement regarding duty to investigate OOS and notify customers.

• U.S. Food and Drug Administration, Warning Letter: Auriga Research Pvt. Ltd. (MARCS-CMS 730694, August 12, 2026), enforcement regarding testing into compliance and customer communication of suspect results.