Independent Testing Starts with the Sample That Was Sent
Ask independent in what sense
“Independently tested” can describe work performed by an external laboratory, but the phrase leaves several questions open. Who chose the sample? Who paid for the work? Who supplied the sample description? Was the testing performed without the laboratory knowing the expected result? Identify which of these features actually applies.
Payment by a supplier does not, by itself, establish invalid data. Equally, using an external laboratory does not mean that sample selection was independent. Describe the arrangement accurately rather than treating independence as a single all-or-nothing property.
Identify the material selected
A sample may be drawn from bulk material, a finished vial, a retained container or stock bought through ordinary distribution. These routes create different links to what another researcher receives. Ask for the submission and batch references needed to understand that link.
The Eurachem guide on sampling uncertainty explains why sampling belongs within the measurement process. Before extending one result to a batch, establish the sampling target and how the selected portion was intended to represent it. A high-quality analytical technique cannot answer an unspecified sampling question.
Check the journey to the laboratory
Record who packed and dispatched the sample, which identifiers accompanied it and what the laboratory acknowledged at receipt. If the sample was divided, relabelled or prepared beforehand, identify those steps. A report may accurately reproduce a client-supplied description without independently verifying the origin behind it.
The traceability and laboratory-record sections of ICH Q7 provide relevant pharmaceutical context. They should not be used to imply that an external research test certifies a supplier’s manufacturing system or all products offered under the same name.
State the reach of the result
A useful review separates three conclusions: what the laboratory measured, which submitted item it measured and how that item relates to the stock of interest. Evidence may be strong at one level and incomplete at another. Record the boundary rather than converting it into a blanket assurance.
Use the COA library to examine available reports, then ask targeted questions about selection and submission where needed. Keep the answers with the specific report rather than applying them automatically to later documents.
For a UK research project commissioning its own testing, define the selection plan before dispatch and agree the necessary reporting details with the laboratory. The result becomes more useful when its sampling context is established in advance and remains available to whoever interprets the study later.
Related research reading
Explore the research blog and original batch-document library for further context.