Out-of-Specification Results in Raw Material Testing: How to Investigate, Document, and Prevent Recurrence

Every QC manager has been there. A raw material shipment arrives, your analytical testing laboratory returns results, and one value falls outside the specification limit you agreed on with the supplier. Now what?

The instinct — especially under production pressure — is to either reject the lot outright or, worse, quietly push through retesting until a passing result appears. Neither is correct, and under 21 CFR Part 111 and FDA’s 2006 Guidance for Industry on OOS investigations, neither is compliant. A proper out-of-specification investigation isn’t bureaucratic overhead. It’s the mechanism that tells you whether you have a supplier problem, a sampling problem, or a testing problem — and that distinction carries real financial and regulatory consequences.

What Actually Qualifies as an OOS Result

Before starting any investigation, be precise about definitions. An out-of-specification result is any test result that falls outside the established acceptance criteria written in your specification — whether that’s a pharmacopeial monograph limit (USP, Ph.Eur., or BP), an internal specification agreed upon with your supplier, or a third-party standard you’ve committed to meeting.

This matters because “OOS” is sometimes used interchangeably with “out-of-trend” or “atypical result,” and those categories call for different responses. A value that’s technically within spec but drifting in the wrong direction across multiple lots is an out-of-trend situation. A value that crosses the specification line is OOS — and it triggers a formal investigation under GMP, full stop.

Common raw material parameters that generate OOS findings include heavy metal levels (particularly lead, arsenic, cadmium, and mercury under USP <232>/<233>), microbial limits under USP <61>/<62>, identity confirmation failures by HPTLC or FTIR, and potency or active constituent assays for botanical extracts. In our experience across multiple raw material categories, roughly 3–5% of incoming shipment tests generate an initial OOS result before any investigation begins — a useful calibration benchmark if you’re sizing your QC team’s workload.

Phase 1: Start With the Lab, Not the Supplier

FDA’s 2006 guidance — and the parallel ICH Q10 framework for pharmaceutical quality systems — prescribes a two-phase OOS investigation structure. Phase 1 is laboratory error investigation. Phase 2 is the field or manufacturing investigation. You must complete Phase 1 before escalating to Phase 2. Not because lab error is the most likely explanation, but because it’s the most controllable one and the most documentable.

A Phase 1 investigation asks whether the OOS result could be attributable to analyst error, instrument malfunction, a reagent problem, or a sample preparation failure. This is a documented review, not a casual conversation. It should cover:

• Analyst training records and demonstrated competency for the specific test method
• Instrument calibration and maintenance logs from the time of analysis
• System suitability results from the analytical run in question
• Reagent lot numbers and expiration dates used during analysis
• Sample chain-of-custody and preparation notes
• A complete review of the raw data — chromatograms, spectra, integration records — not just the reported summary value

If a confirmed laboratory error is identified and documented, the OOS result may be invalidated and a retest initiated from the original sample. This is the only condition under which retesting to obtain a passing result is GMP-compliant. Retesting without a documented assignable cause is explicitly cited in FDA warning letters as a data integrity and potential falsification risk. If Phase 1 finds no laboratory error, the result stands and you move to Phase 2.

Phase 2: Field Investigation and the Supplier Conversation

A Phase 2 investigation shifts focus to the raw material lot itself and the manufacturing or handling conditions on the supplier side. This is where things get operationally complex — because now you’re potentially initiating a supplier corrective action request (SCAR) while also deciding what to do with a lot sitting on hold in your warehouse.

Phase 2 documentation should work through four areas in sequence:

1. Reserve sample retest. A retest of the retained sample from the original shipment, conducted under full analytical controls. This is a retest from the field investigation perspective — distinct from a Phase 1 retest — and it serves to confirm the original result and eliminate sampling variability as an explanation.

2. Expanded sampling. If the original test used a single composite sample, expanded sampling across multiple containers in the lot can reveal whether non-conformance is isolated or uniform throughout the shipment. USP <1> and standard statistical sampling plans — ANSI/ASQ Z1.4 for attribute sampling or Z1.9 for variable sampling — provide established frameworks for determining appropriate sample sizes.

3. Supplier data review. Request the supplier’s raw test data for the failing lot, not just the summary COA table. Compare the analytical method they used against yours. Method differences — different versions of a USP chapter, different extraction protocols, different reference standard lots — can create real inter-laboratory discrepancies that don’t indicate fraud or negligence on either side. Identifying a method mismatch here can redirect the entire investigation.

4. Manufacturing history. If the supplier is cooperative and GMP-registered, request deviation records, environmental monitoring data, or in-process failure logs for the specific batch. A supplier who can produce this without hesitation is one worth keeping.

The outcome of Phase 2 is a disposition decision: reject the lot, accept with a documented risk assessment and conditional justification, or request a replacement and initiate a SCAR. If the lot is rejected, quarantine the lot number in your materials management system with a permanent non-conformance record — one that’s reviewed during every future supplier qualification assessment.

CAPA: Where the Investigation Becomes Value

Too many teams treat CAPA — Corrective and Preventive Action — as paperwork bolted onto the end of an OOS investigation. That’s a missed opportunity, and FDA inspectors know how to spot a CAPA that was written to close a record rather than fix a problem.

A meaningful CAPA addresses the root cause identified during the investigation, not just the immediate nonconformance. It specifies preventive actions with ownership, due dates, and verification criteria. Under 21 CFR Part 111, all QC-related records including CAPA documentation must be retained and available for FDA inspection at any time.

If Phase 1 identified a reagent management gap, CAPA might require revising your reagent SOP to add a pre-use verification step and updating analyst training records. If Phase 2 identified that a botanical supplier shifted growing regions without notifying you — and the new origin material has systematically different marker compound levels — CAPA might involve revising your acceptance specification, adding supplier change notification requirements to your supplier agreement, or introducing an origin verification step to your incoming inspection panel.

The preventive action component is where strong QC programs distinguish themselves. Fixing the immediate problem is table stakes. Asking “what would prevent a similar OOS result from occurring across our entire raw material portfolio?” — that’s the question that drives program-level improvements and compounds value over time.

When to Bring In a Third-Party Analytical Testing Laboratory

There are specific situations where involving an external contract analytical testing laboratory isn’t optional — it’s the operationally sound call.

The clearest case is an inter-laboratory dispute. If your in-house testing returns an OOS result and the supplier insists their COA shows passing results, the investigation cannot be resolved internally. A third-party analytical testing laboratory accredited to ISO 17025 for the relevant test methods provides an independent, defensible data point that carries weight in supplier negotiations and, if necessary, in regulatory proceedings.

A second case is capacity or capability constraints. Phase 1 and Phase 2 investigations are supposed to be completed within your batch disposition timeline. If your lab is running at capacity, or lacks in-house capability for a confirmatory method — ICP-MS for elemental impurities under USP <232>/<233>, or PCR-based identity testing for botanical materials — outsourcing the confirmatory analysis to a contract lab is not a compliance shortcut. It’s a reasonable and documentable decision.

The third case is audit readiness. If a raw material with a prior OOS history is under heightened internal scrutiny, or you’re preparing for a facility inspection, independent third-party data behind your disposition decisions adds significant defensibility. Inspectors routinely ask for the data supporting a “lot accepted” decision when there’s a nonconformance history on that material.

When selecting a contract analytical testing laboratory for OOS confirmatory work, verify ISO 17025 accreditation specifically for the test method in question — not just general lab accreditation, which doesn’t guarantee method-specific scope coverage. Ask for the scope-of-accreditation document, not just the certificate.

A Practical Note on Documentation Timing

One pattern that consistently appears in underdeveloped QC programs: the investigation is conducted informally, the right conclusions are reached, the right actions are taken — but the documentation is written retrospectively, days or weeks after the fact.

This is a compliance liability regardless of how thorough the actual investigation was. Under GMP, records must be contemporaneous. “Contemporaneous” means created at the time the activity occurs, not reconstructed from memory. If an FDA inspector finds a 14-day gap between your OOS result date and the start of your investigation record, that gap becomes the observation — irrespective of the quality of your eventual documentation.

Build your OOS investigation form into your LIMS or document management system so it opens automatically when an OOS result is flagged. Even a one-line entry — “Phase 1 review initiated, analyst [name], [date]” — is enough to establish contemporaneous documentation of the investigation’s start.

The OOS result you catch at incoming inspection is a data point. The investigation you conduct afterward is the process that converts that data point into organizational intelligence. Build the process properly, and it gets faster, more consistent, and more defensible with every iteration.

---

Written by Nour Abochama, VP Operations, Qalitex | Quality Consultant, Ayah Labs. Learn more about our team

Talk to our team about raw material testing — from incoming OOS investigation support to full supplier qualification programs. Contact us

Related from our network

• ISO 17025-Accredited Supplement and Raw Material Testing — Qalitex Laboratories provides contract analytical testing services for dietary supplement manufacturers, including OOS confirmatory analysis and supplier qualification testing.