The lab is where inspections slow down
Investigators spend time in QC laboratories because lab data decide whether a batch is released. FDA’s FY2025 drug 483 data show it: 211.160(b), laboratory controls that are not scientifically sound, was the fourth most-cited provision with 121 citations, and 211.192, inadequate investigations, was second with 164; testing and release for distribution (211.165(a)) also made the top ten (GMP Insiders).
This article splits the work between the QC manager, who owns the system, and the QC supervisor, who owns what happens on each shift.
What the regulations ask of the lab
Laboratory controls (211.160)
Lab control mechanisms, and changes to them, must be reviewed and approved by the quality control unit, followed and documented; deviations must be recorded and justified. Controls must include scientifically sound and appropriate specifications, standards, sampling plans and test procedures, and instruments must be calibrated under a written program; equipment that fails its specifications must not be used (21 CFR 211.160).
Laboratory records (211.194)
Each laboratory record must include (21 CFR 211.194):
- Sample description, source, quantity, lot and dates sampled and received
- Each test method used, and where its accuracy and reliability data can be found
- Weight or measure of sample used
- All data, including instrument graphs, charts and spectra
- All calculations
- Results compared with specifications
- Analyst initials or signature and test dates
- Second-person review confirming accuracy, completeness and compliance
Records are also required for method changes, reference standards and reagents, instrument calibration and stability testing.
Investigations (211.192)
Any unexplained discrepancy or failure to meet specifications must be thoroughly investigated, extended to other potentially affected batches, and documented with conclusions and follow-up (21 CFR 211.192). FDA’s guidance Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production (May 2022, Level 2 revision) sets out how it expects OOS results to be evaluated, and applies to in-process tests too (FDA).
QC manager vs QC supervisor: who does what
| Task | QC manager | QC supervisor |
|---|---|---|
| Specifications, methods, sampling plans | Approves and maintains | Applies; flags problems |
| Calibration program | Owns | Checks status before use |
| OOS investigations | Approves conclusions, trends results | Starts Phase I lab investigation promptly |
| Record review | Sets review standard; audits a sample | Second-person review every day |
| Training | Defines curriculum | Coaches and observes analysts |
| Inspection | Presents the lab system | Shows real-time practice and records |
A daily review routine for supervisors
- Check that every test started has a complete record trail, including instrument data.
- Review OOS and atypical results the same day; make sure the investigation starts before retesting.
- Confirm instruments used were within calibration.
- Spot-check calculations and units.
- Make sure second-person review happens against the original data, not a printout summary.
Monthly checks for managers
- Trend OOS results by product, method and analyst.
- Review investigations closed as “lab error” for evidence.
- Audit a sample of records against all eight 211.194(a) elements.
- Review method changes for supporting data.
- Check overdue calibrations and stability pulls.
Who you work with
- QA: release decisions and investigation approval.
- Production: in-process testing and batch timing.
- Analytical development / method validation: method changes.
- Metrology / engineering: calibration and instrument qualification.
- IT / CSV: laboratory computerized systems.
Common weaknesses
- Investigations concluding “analyst error” without evidence.
- Retesting before an investigation is opened.
- Second-person review that checks the report but not the raw data.
- Missing instrument data or unlabelled chromatograms.
- Methods changed without recorded justification.
Checklist
- Sample of records checked against 211.194(a)
- OOS procedure aligned with FDA’s May 2022 guidance
- OOS and invalidated results trended
- Calibration status current for all instruments in use
- Second-person review covers raw data
- Supervisors and analysts rehearsed for investigator questions
How AskDona helps
Lab records are long and detailed, and missing elements are easy to overlook. AskDona reads laboratory records, OOS investigations and methods against 211.194, your procedures and the specifications you provide, and flags missing elements or unsupported conclusions with page citations. Run a sample laboratory record review on one month of OOS investigations. AskDona shows the evidence; your QC team decides.
Editor’s note: CFR text is paraphrased from LII. The OOS guidance’s detailed content is summarized only at the level shown on FDA’s page. The manager/supervisor split is a typical model, not a regulatory requirement. Not legal advice. LinkedIn counts were supplied by the client.
FAQ
What must a GMP laboratory record contain? The eight elements of 21 CFR 211.194(a), from sample description to second-person review.
What is FDA’s current OOS guidance? Investigating OOS Test Results for Pharmaceutical Production, revised May 2022.
Was laboratory control a common FY2025 observation? Yes. 211.160(b) was fourth, with 121 citations.
What is the QC supervisor’s most important daily task? Thorough second-person review against original data.
References
All pages opened and checked by October 23, 2026.
| Source | Type | Used for |
|---|---|---|
| GMP Insiders: FY2025 FDA inspection observations | Industry analysis | Citation counts |
| 21 CFR 211.160 (LII) | Regulation | Laboratory controls |
| 21 CFR 211.194 (LII) | Regulation | Laboratory records |
| 21 CFR 211.192 (LII) | Regulation | Investigations |
| FDA: Investigating OOS test results (May 2022) | FDA guidance | OOS scope |