Kratom OOS and Atypical Lab Results: How a Responsible Investigation Should Work

Suggested page / content hub: Learning Center → Lab Testing & Product Quality
Secondary phrases: kratom OOS result, kratom failed lab test, kratom retesting, kratom out-of-trend result, 7-OH test investigation

Suggested SEO title: Kratom OOS Results: Failed Tests, Retesting and Investigations
Suggested meta description: Learn what should happen after a kratom result fails a specification, conflicts with history or falls near a 7-OH threshold.
Suggested URL slug: /learn/kratom-oos-result-investigations/

The short answer

When a kratom test result does not meet an approved specification, the correct response is not to discard it, average it with a better number or keep testing until the lot passes. The material should remain controlled while qualified personnel determine what happened, evaluate the full record, decide whether the result is valid and document the lot’s disposition.

A strong investigation asks two different questions:

1. Did the laboratory produce an invalid result because of a demonstrated analytical error?
2. If not, what does the result say about the tested material, manufacturing process and potentially related lots?

Those questions should be answered with evidence. “The number looks wrong,” “this supplier is usually good” and “the repeat passed” are not root causes.

The basic workflow is:

1. Secure the material and preserve the original data.
2. Notify the responsible quality reviewer.
3. Perform a prompt laboratory assessment.
4. Expand into a material and manufacturing review when laboratory error is not conclusively demonstrated.
5. Use predefined, scientifically justified retesting or resampling only when appropriate.
6. Evaluate every result, affected product and related lot.
7. Make and document a disposition decision.
8. Correct the cause and verify that the correction worked.

This article uses federal dietary-supplement current good manufacturing practice requirements as its regulatory foundation. It also borrows investigation principles from FDA’s drug-specific out-of-specification guidance, but clearly identifies those principles as a technical framework rather than a kratom-specific rule.

OOS, OOT, atypical and invalid are not synonyms

Good investigations begin with accurate terms.

Out of specification, or OOS

An OOS result falls outside an approved acceptance criterion. Examples include:

• Salmonella detected when the specification requires not detected;
• Lead above an established maximum;
• Mitragynine below an approved finished-product range;
• 7-OH above a defined product or jurisdictional limit;
• Capsule net fill outside an approved range; or
• A package seal that fails a specified integrity examination.

The specification must exist before the result is evaluated. A laboratory reporting limit is not automatically a product specification, and a result is not OOS merely because it differs from a preferred marketing number.

Out of trend, or OOT

An OOT result may still meet its specification but differs meaningfully from established history or an expected pattern.

Suppose ten botanical leaf lots historically report moisture between 6.8% and 8.1%. A new lot reports 10.9%, while the approved maximum is 12%. It passes the numerical limit, yet the shift may justify review because it could reflect a supplier, season, drying, storage or sampling change.

OOT limits need an evidence-based trend rule. A reviewer should not label every normal botanical variation a failure.

Atypical or unexpected result

An atypical result conflicts with other evidence even when no formal statistical trend exists. Examples include:

• A white-variety leaf powder has a chromatographic profile unlike the supplier’s other leaf lots;
• Two results from the same composite disagree beyond expected method precision;
• A pure-leaf capsule result resembles a concentrated extract;
• A microbiology control behaves unusually; or
• A sample reports a compound that the formulation and supplier records do not explain.

Suspect result

A suspect result is under assessment because there is a specific reason to question it. Suspicion is a status, not a conclusion. The original result remains part of the record.

Invalid result

An invalid result is one that a documented, scientifically sound investigation demonstrates does not represent the sample because of an assignable cause. Examples can include a confirmed calculation error, documented sample spill, instrument malfunction, incorrect standard or proven preparation mistake.

“A repeat passed” is not by itself evidence that the original failed result was invalid.

Inconclusive investigation

An investigation is inconclusive when no clear laboratory or manufacturing cause is established. Inconclusive does not mean the initial failure disappears. The disposition should account for the unresolved risk and all available data.

The dietary-supplement quality-control baseline

FDA states that kratom has not been lawfully marketed in the United States as a conventional food or dietary supplement. Referencing Part 111 here does not imply that FDA has approved kratom or any Kiody product. FDA and Kratom.

Part 111 nevertheless provides a useful and directly relevant quality-system framework for businesses manufacturing, packaging, labeling or holding products represented as dietary supplements.

Under 21 CFR §111.113, quality-control personnel must conduct a material review and make a disposition decision when, among other triggers:

• A specification established under §111.70 is not met;
• A batch deviates from its master manufacturing record;
• An unanticipated occurrence may lead to adulteration;
• Instrument calibration suggests a possible quality problem; or
• A dietary supplement is returned.

When a specification is not met, quality control must reject the component, supplement, package or label unless it approves permitted treatment, in-process adjustment or reprocessing. The reviewer must document the material review and disposition at the time it occurs.

Part 111 also requires:

• Specifications for applicable components, processes, finished products, packaging and labels under §111.70;
• Appropriate, scientifically valid tests and examinations under §111.75 and §111.320;
• Written laboratory procedures and contemporaneous documentation of test performance and results under §111.325; and
• Quality-control approval or rejection of release for distribution under Part 111, Subpart F.

FDA’s dietary-supplement CGMP compliance guide explains that rejected material should be identified, held and controlled under quarantine pending appropriate disposition. It also explains that reserve samples exist for appropriate investigations and should be identified by lot. FDA dietary-supplement CGMP compliance guide.

What FDA’s drug OOS guidance adds—and what it does not

FDA’s guidance titled Investigating Out-of-Specification Test Results for Pharmaceutical Production applies to drug testing under Parts 210 and 211. It is not a dietary-supplement regulation and not a kratom-specific mandate. FDA OOS guidance.

Its technical concepts are still useful because they address common analytical problems:

• Prompt assessment of accuracy and validity;
• Preservation of original observations and data;
• A defined laboratory phase and broader manufacturing phase;
• Scientifically justified retesting plans;
• Careful evaluation of averaging and resampling;
• Consideration of all results in batch disposition; and
• Follow-up corrective action.

FDA has also pointed dietary-supplement manufacturers to this drug guidance when discussing failing, OOS, OOT and unexpected results in warning letters. That does not transform every passage into a Part 111 requirement. It demonstrates that the investigation principles are relevant quality-system references. One example is FDA’s November 12, 2025 warning letter to DeVere Manufacturing. FDA warning letter.

Immediate actions after a suspect result

1. Place or keep the lot under control

If the lot has not been released, maintain quarantine. If it has been released, identify where it went and assess whether distribution should be paused or other action is warranted. Do not allow a result under investigation to be bypassed by changing status in a second system.

The hold should extend to related material when evidence suggests shared risk—for example, lots made from the same component, a common blend or the same labeling run.

2. Preserve the original record

Preserve:

• Original instrument files;
• Chromatograms, spectra, plates or images;
• Sample and standard preparations;
• Worksheets and calculations;
• Sequence and audit-trail information;
• Remaining sample quantity;
• Retain or reserve samples; and
• Communications relevant to the test and decision.

Do not overwrite an integration, delete a failed run or replace the original COA without a controlled history. FDA’s drug data-integrity guidance states that even a legitimately invalidated result should remain in the full record with the investigation that justifies invalidation. This is drug guidance used here as a technical recordkeeping principle. FDA data-integrity guidance.

3. Notify qualified reviewers

The analyst should promptly notify the laboratory supervisor or designated technical reviewer and the responsible quality unit. The commercial team should not control the investigation or decide what result will be reported.

4. Record the event without deciding the cause

Open a uniquely numbered record that identifies:

• Product and lot;
• Sample and accession;
• Test and specification;
• Original result;
• Date and analyst;
• Current material status; and
• Immediate containment.

The initial description should be factual. “Lead result 1.24 mg/kg against maximum 1.00 mg/kg” is stronger than “lab probably contaminated sample.”

Phase I: the laboratory assessment

The laboratory phase looks for evidence that the analytical process failed. It should be prompt because preparations, instrument conditions and analyst memory can change.

Review the method and actual execution

Compare the approved procedure with what occurred:

• Was the correct method and version used?
• Was it appropriate for leaf powder, capsules, extract or liquid matrix?
• Were required preparation and extraction steps followed?
• Were actual weights, volumes and dilutions recorded?
• Were reagents and standards valid?
• Were hold times and storage conditions met?
• Did system-suitability and quality-control samples pass?
• Did the instrument operate within requirements?
• Were calculations and unit conversions correct?

Interview the analyst

An analyst interview should reconstruct the work, not coach a desired answer. Useful questions include:

• Did anything unusual occur during weighing, transfer, filtration or injection?
• Was material spilled, lost or visibly inhomogeneous?
• Were any vials, plates or worksheets relabeled?
• Did the instrument alarm or stop?
• Was any method setting changed?
• Were calculations copied manually?
• Is remaining preparation available for examination?

The interview should be documented contemporaneously.

Examine raw data and metadata

The reviewer should inspect more than the final PDF. For chromatographic data, this can include:

• Calibration and continuing checks;
• Blanks and controls;
• Retention or identification criteria;
• Integration and reprocessing history;
• Dilution range;
• Carryover;
• Sequence order; and
• Audit trails.

For microbiology, review sample portion, media and controls, enrichment, incubation, colony interpretation and confirmation. For ICP-MS metals, review digestion, blanks, internal standards, calibration and interference checks.

Check calculations independently

Recalculate from source values. Verify:

• Decimal placement;
• Dilution factor;
• Purity correction;
• Sample mass and final volume;
• Moisture basis;
• Unit conversion;
• Serving calculation;
• Significant figures; and
• Rounding at the decision boundary.

A correct instrument response can produce a wrong COA result through one calculation error.

Can Phase I invalidate the result?

Only a demonstrated cause should invalidate the original result. Examples might include:

• A balance ticket proves the wrong mass was entered;
• The analyst used an expired or incorrectly prepared standard;
• A vial position was conclusively assigned to the wrong sample;
• Instrument failure corrupted acquisition;
• The formula omitted a documented dilution; or
• A sample preparation was visibly lost during transfer and the loss was documented.

Vague possibilities are not enough. “Maybe pipetting error” is a hypothesis, not proof.

Phase II: the full material and manufacturing investigation

If Phase I does not conclusively demonstrate laboratory error, the investigation should evaluate whether the result reflects the material or process.

Review the component history

For botanical leaf or extract, consider:

• Supplier and origin records;
• Component lot and receipt condition;
• Supplier COA and Kiody verification data;
• Identity evidence;
• Moisture and storage history;
• Sampling plan;
• Previous lots from the supplier; and
• Any change in season, plant material, milling or extraction.

Do not assume a supplier’s historically acceptable performance makes a current failure impossible.

Review manufacturing and packaging

Examine the master manufacturing record and actual batch record:

• Components and quantities;
• Equipment and cleaning status;
• Sifting, blending or encapsulation steps;
• In-process results;
• Yields and reconciliations;
• Hold times and environmental conditions;
• Deviations and maintenance;
• Label and package version;
• Other products on the same equipment; and
• Personnel observations.

An elevated 7-OH result in supposed pure leaf may prompt review of supplier material, segregation, formulation, changeover and possible mix-up. A microbiology failure may prompt review of component contamination, handling, environment, cleaning, moisture and package integrity.

Review related lots

The investigation should not stop at the tested lot when a common cause could affect others. Consider:

• Other lots from the same component;
• Batches made before and after the event;
• Products sharing equipment or processing conditions;
• Lots tested in the same questionable laboratory sequence;
• Lots released using the same calculation template; and
• Distributed product connected to the same source.

The scope should be evidence-based and documented.

Retesting: a planned tool, not a search for a passing number

Retesting can provide useful evidence if its purpose, number of tests, sample source and interpretation are defined before results are known.

A retest plan should state:

• The hypothesis being evaluated;
• Whether the original preparation, a new preparation or both will be tested;
• Who will perform the work;
• How many determinations will be made;
• Which method and version apply;
• What controls are required;
• How all results will be evaluated; and
• What conditions would stop or expand the investigation.

Retesting the original preparation

Reinjecting an existing solution may help evaluate instrument behavior or a transient injection problem. It does not test whether the original weighing, extraction or dilution was correct.

Preparing a new test portion

A new preparation examines repeatability from a new portion of the same laboratory sample. It can help identify preparation variability but does not automatically represent the entire commercial lot.

Testing by a second analyst or laboratory

Independent work may add information, especially when a method, instrument or analyst issue is suspected. Different results must be reconciled through method comparability, sampling and data review; selecting the more favorable laboratory is not a resolution.

When retesting becomes “testing into compliance”

Warning signs include:

• No predefined number of retests;
• Repeating until one value passes;
• Reporting only passing determinations;
• Changing the method after each failure without controlled justification;
• Using commercial product as informal system-suitability material; or
• Discarding the original result solely because a later value is lower.

FDA’s drug data-integrity guidance specifically warns against using actual samples in test, preparation or equilibration runs as a way to disguise testing into compliance. FDA data-integrity guidance.

Resampling: when the question is the sample, not the test

Resampling means collecting another portion from the batch or lot. It may be appropriate when the original sampling plan was not followed, the submitted sample was compromised or evidence suggests the lot is nonuniform and more spatial information is needed.

Resampling should not be used merely because retesting the original sample continued to fail.

A resampling plan should identify:

• Why the first sample may not represent the lot;
• Locations or units to be sampled;
• Random or stratified selection method;
• Number and size of increments;
• Composite or individual-testing design;
• Chain of custody; and
• How the original and new results will be combined in the decision.

If a lot is heterogeneous, a second composite can hide the pattern. Individual or stratified samples may be more informative.

Averaging: useful calculation or misleading eraser?

Averaging can be appropriate when the test method and approved sampling design define replicate determinations as one reportable result. It is not appropriate to average away a genuine failure without scientific justification.

Consider two examples:

• A validated method specifies duplicate readings whose mean is the result. Averaging is part of the method.
• One test gives 1.30 mg/g against a 1.00 mg/g limit, and a repeat gives 0.70 mg/g. Reporting the average as 1.00 mg/g can conceal important variability and does not explain either result.

Individual results, variability and the reason for differences matter. Averages should follow a pre-established method—not an after-the-fact rescue rule.

Near-limit results and measurement uncertainty

A result can be within the written limit yet too close for a comfortable decision.

Suppose a specification is no more than 0.500 mg/g and the laboratory reports 0.498 mg/g. Questions include:

• What are the unrounded digits?
• What rounding rule applies?
• What is the method’s measurement uncertainty near this level?
• Is the result within the calibrated and validated range?
• Does the sample represent the finished product and serving?
• Is the limit actually expressed in mg/g, dry weight, per serving or percentage of total alkaloids?

A legal threshold is not automatically an internal release target. A responsible specification may include a guard band or decision rule that accounts for method performance and normal lot variation.

For 7-OH, denominator mistakes are particularly serious. A result expressed as percent by product weight cannot be directly compared with a state rule expressed as percent of total alkaloids. Per-serving and per-unit triggers require additional product and label information.

Five worked kratom investigations

Scenario 1: Salmonella detected in a leaf-powder sample

A laboratory reports Salmonella detected in the specified analytical portion of a finished leaf-powder lot.

Immediate action: Keep the lot quarantined, preserve the enrichment and confirmation record, and notify quality control.

Laboratory assessment: Review sample identity, portion size, media and control performance, incubation, confirmation method, cross-contamination controls and all observations.

Full investigation: Review component lots, supplier history, cleaning, environment, handling, moisture, packaging and other batches using the same source material.

Unsound response: Submit a new scoop from the lot repeatedly until one tests negative and ignore the original detection.

Sound decision: Unless the original result is conclusively invalidated, treat the pathogen evidence as material. A negative retest does not prove the organism is absent from every unit in a heterogeneous lot.

Scenario 2: Lead result exceeds an internal specification

An ICP-MS report shows 1.24 mg/kg lead against an approved maximum of 1.00 mg/kg.

Laboratory assessment: Review digestion weight and volume, blanks, calibration, internal standards, interference checks, dilution and calculation.

Full investigation: Examine supplier lot, origin, other elemental results, sampling representativeness and related batches.

Possible valid analytical cause: A documented transcription error entered a tenfold dilution rather than the actual hundredfold dilution. Recalculation from preserved source records is independently verified.

Disposition principle: Corrected math may invalidate the reported failure if the evidence is conclusive. The original report, correction and impact assessment remain controlled records.

Scenario 3: 7-OH result sits on opposite sides of a threshold

The first preparation reports 0.052% by dry weight; a second reports 0.048%. A proposed federal threshold is 0.050% by dry weight.

Important status: The federal 7-OH action is proposed as of August 30, 2026, not final. The example illustrates decision complexity rather than declaring the article controlled.

Investigation: Review sampling, sample homogeneity, moisture correction, calibration, integration, uncertainty, preparation precision, unrounded values and any predefined decision rule.

Unsound response: Publish only 0.048% and call the product legally compliant.

Sound response: Preserve and evaluate both results. Determine whether the difference is expected analytical variation, sample heterogeneity or an assignable error. Obtain product-specific legal review because a boundary result cannot be resolved through favorable rounding.

Scenario 4: Pure-leaf capsule assay varies unexpectedly

Kiody capsules are described as containing approximately 500 mg of pure leaf powder per capsule. A composite assay meets the alkaloid specification, but individual capsule weights show a new, wide spread.

Classification: The composite chemistry result may pass while the fill-weight trend is atypical or fails an in-process specification.

Investigation: Review encapsulator settings, start-middle-end checks, powder flow, equipment adjustments, operator records and whether the composite masks unit variation.

Disposition: Decide against approved fill and label specifications, not the favorable composite alone. Expand the review to units made during the affected equipment window.

Scenario 5: Two laboratories disagree on mitragynine

The supplier laboratory reports 1.45% mitragynine; the receiving laboratory reports 1.08% for the same stated component lot.

Not automatically OOS: Whether either result fails depends on the approved specification.

Investigation: Confirm the samples came from the same lot and sampling plan; compare as-received versus dry-weight basis, methods, reference standards, extraction efficiency, calibration, reporting uncertainty and sample homogeneity.

Unsound response: Select the higher result because it supports a “stronger” marketing claim.

Sound response: Reconcile the methods and sample histories. If the discrepancy remains unexplained and materially affects release, maintain control and widen the investigation.

Root cause, contributing factors and “no cause found”

A root cause is the most fundamental controllable reason the event occurred. Investigations often confuse a symptom with a cause.

• “OOS lead result” is the event.
• “Analyst error” is usually too vague.
• “The analyst used calculation template version 3, whose locked dilution cell was incorrectly set to 10 instead of 100 after an uncontrolled template change” is a specific, testable cause.

There can be contributing factors, such as incomplete training or inadequate review. Corrective action should address the system that allowed the event, not only the person closest to it.

When no cause is found, say so. Do not invent a root cause to close the record. An inconclusive investigation can require a conservative disposition and increased monitoring.

Disposition options

Quality control should make and document the disposition. Depending on evidence and applicable requirements, options may include:

• Reject and destroy or otherwise appropriately dispose of material;
• Return a rejected component to the supplier;
• Approve a permitted treatment, in-process adjustment or reprocessing;
• Release only when the specification is met and the investigation supports the decision;
• Expand testing or investigation under an approved plan;
• Hold related lots;
• Assess distributed product and potential market action; or
• Refer the issue for regulatory or legal evaluation.

Reprocessing is not a label used to excuse a failure. Part 111 places limits and quality-control oversight on treatment, adjustment and reprocessing, and the corrected material must meet applicable specifications.

Corrective and preventive action

An investigation is incomplete if it identifies the cause but does nothing to prevent recurrence.

Possible actions include:

• Correct and validate a calculation template;
• Revise a sampling plan;
• Repair or recalibrate equipment;
• Improve cleaning or segregation;
• Retrain personnel on a revised procedure;
• Strengthen second-person data review;
• Requalify or disqualify a supplier;
• Add a packaging or moisture control;
• Change an internal guard band; or
• Increase targeted monitoring for a defined period.

Effectiveness checks should be specific. “No complaints received” may not test whether a laboratory calculation problem was fixed. A better check could verify the next ten reviewed calculations against an independent worksheet with zero discrepancies.

Twenty investigation warning signs

1. The lot is released while the investigation is still open.
2. The original result is deleted or overwritten.
3. A repeat is ordered before the first data review.
4. Retesting has no predefined number or plan.
5. Testing stops as soon as one passing value appears.
6. Only passing results appear in the report package.
7. “Analyst error” is assigned without evidence.
8. A possible cause is treated as a proven cause.
9. Different methods are used without comparability review.
10. A new sample is collected because the first sample failed, not because sampling error was demonstrated.
11. Results are averaged after the fact to obtain a pass.
12. Units or denominators change between reports.
13. Rounded numbers determine a near-threshold decision without an approved rule.
14. The supplier’s result is accepted solely because it is more favorable.
15. A laboratory-only review closes an issue that may involve manufacturing.
16. Related lots or distributed product are not considered.
17. The disposition was decided by sales or purchasing rather than quality control.
18. A corrected COA silently replaces the original.
19. Corrective action is only “retrain analyst” with no system review.
20. The investigation closes with a root cause that cannot be verified.

A proposed 34-field Kiody investigation record

Field	Purpose
1. Investigation ID	Creates unique control
2. Open date and time	Establishes chronology
3. Product and SKU	Identifies the affected item
4. Product form	Leaf, capsules, extract, liquid or other
5. Lot or batch	Connects records and inventory
6. Component lots	Supports supplier and cross-lot review
7. Sample and accession	Connects laboratory evidence
8. Test or examination	Defines the affected attribute
9. Method and version	Identifies the approved procedure
10. Specification and version	Shows the pre-existing acceptance rule
11. Original result	Preserves the event
12. Units and basis	Prevents denominator confusion
13. Classification	OOS, OOT, atypical or other
14. Reporter	Establishes attribution
15. Immediate containment	Records hold or distribution action
16. Raw-data location	Protects source evidence
17. Analyst interview	Reconstructs actual work
18. Method-execution review	Compares procedure with performance
19. Standards and reagents	Checks validity and traceability
20. Instrument and calibration	Checks system status
21. Controls and suitability	Evaluates run validity
22. Calculation verification	Confirms formula and inputs
23. Phase I conclusion	States whether lab cause is proven
24. Manufacturing review	Assesses batch history
25. Supplier review	Assesses component history
26. Related-lot scope	Documents breadth of review
27. Retest protocol	Defines authorized additional work
28. Resampling protocol	Defines new collection if justified
29. All additional results	Prevents selective reporting
30. Root cause and evidence	Links cause to facts
31. Disposition	Records reject, reprocess, hold or release
32. CAPA and owner	Assigns corrective work
33. Effectiveness check	Verifies the correction worked
34. Quality approval and closure	Documents final authority and date

This is a proposed editorial review tool, not a regulation or a substitute for a qualified quality system.

Current federal 7-OH and derivative status

As of August 30, 2026, two federal actions must remain separate:

• DEA’s proposed temporary scheduling of 7-OH above a specified threshold remains proposed, and the comment period runs through September 10, 2026. Federal Register comment extension.
• Mitragynine pseudoindoxyl, MGM-15 and MGM-16 entered federal Schedule I on August 26, 2026 under a separate final temporary order. DEA temporary scheduling order.

An investigation cannot convert an intentionally added or controlled compound into compliant botanical leaf through averaging, renaming or selective reporting. Kiody does not sell concentrated 7-OH.

Frequently asked questions

1. Does one failed test automatically condemn the lot?

It triggers control and investigation. The outcome depends on whether the result is valid, the specification, the material and the full evidence. A failure cannot simply be ignored.

2. Can a passing retest replace the original failure?

Not by itself. Both results must be preserved and evaluated. The original should be invalidated only when evidence demonstrates an assignable analytical cause.

3. How many retests are allowed?

There is no universal kratom number. The protocol should define a scientifically justified number before results are known.

4. Is OOT the same as OOS?

No. OOS is outside a specification. OOT is unusual relative to history or expected behavior and may still meet the specification.

5. Can botanical variation explain an unusual result?

It may contribute, but “natural variation” should not be used as an unsupported conclusion. Sampling, origin, season, processing and method performance should be evaluated.

6. Can two laboratories legitimately report different values?

Yes. Differences can arise from sampling, preparation, method, reference standards, moisture basis, uncertainty and calculation. Material differences require reconciliation.

7. Is averaging replicate results acceptable?

It can be when the approved method defines the average as the reportable result. It should not be invented after a failure to obtain a pass.

8. What is testing into compliance?

It is repeated or selectively reported testing designed to obtain a preferred passing result rather than investigate the original evidence.

9. When is resampling appropriate?

When evidence supports a sampling problem or when additional lot locations are necessary to understand heterogeneity—not merely because the first sample failed.

10. Who decides whether the lot is released?

Qualified quality-control personnel should make and document the disposition under the applicable quality system.

11. What happens if no root cause is found?

The investigation should be labeled inconclusive. The unresolved uncertainty should be considered in the disposition, scope and follow-up monitoring.

12. Does a COA marked “pass” prove an investigation was adequate?

No. The final report may not show the original result, retests, root cause or disposition record.

13. Should near-limit results be rounded?

Reports can use controlled rounding, but a legal or quality decision should follow a pre-established rule and consider the unrounded result and method performance.

14. Does ND mean absolute absence?

No. It means not detected under the method’s conditions. The detection or reporting limit is necessary for interpretation.

15. Can a supplier COA close an OOS investigation?

No. It is one piece of evidence. The submitted samples, methods and lot connection must be compared.

16. Does an OOS investigation prove a product is safe or effective?

No. It evaluates a quality event against defined evidence and specifications. It does not establish medical safety, efficacy or universal legality.

17. Does Kiody sell concentrated 7-OH?

No. Kiody is 21+ and does not sell concentrated 7-OH.

Primary sources and editorial references

1. 21 CFR §111.113 — material review and disposition requirements.
2. 21 CFR §111.70 — required specification categories.
3. 21 CFR §111.75 — determining whether specifications are met and scientifically valid methods.
4. 21 CFR Part 111, Subpart J — laboratory controls, methods and records.
5. FDA dietary-supplement CGMP compliance guide — current Part 111 explanations.
6. FDA OOS guidance — drug-specific, nonbinding technical framework.
7. FDA data-integrity guidance — drug-specific technical framework for preserving original data and avoiding testing into compliance.
8. FDA DeVere Manufacturing warning letter — dietary-supplement example pointing to OOS principles.
9. FDA and Kratom — FDA’s current federal kratom position.
10. Federal 7-OH comment extension — proposed threshold action.
11. Federal MP/MGM-15/MGM-16 order — effective August 26, 2026.