Out of specification (OOS) FAQs

These FAQs and interactive presentation provide guidance on what to do if an out of specification result is identified during laboratory analysis.

Raised hand 
1. Has the MHRA produced any guidance?
2. Why is there a need to conduct an investigation of an OOS test result if the decision has been taken to reject the batch?
3. Who should investigate OOS?
4. How is an out of trend result handled?
5. Is it acceptable for a contract laboratory (contract acceptor) to use the contract givers’ procedure when handling OOS results?
6. How is a meaningful OOS investigation conducted?
7. At what point should a manufacturing investigation be initiated?
8. What should be done if unexpected results are obtained and there is no obvious explanation?
9. Under what circumstances could test results become invalid?
10. What should be done in the case where part way through testing the analyst realises there is an error?
11. When should the analyst inform the supervisor that they have an OOS results?
12. What should be done when the phase 1 investigation does not reveal an assignable cause or evidence of error remains unclear?
13. How many repeat tests should be conducted?
14. What should be done if after retesting there is a combination of OOS results and pass results?
15. What should happen if the OOS investigations are inconclusive?
16. When is it acceptable to average test results?
17. When is it not acceptable to average test results?
18. At what stage should retesting occur?
19. At what stage should re-sampling occur?
20. When is it appropriate to use outlier tests?


1. Has the MHRA produced any guidance?
Out of specification investigationsPDF file (opens in new window) (290Kb)

<back to top>


2. Why is there a need to conduct an investigation of an OOS test result if the decision has been taken to reject the batch?
A phase 1 investigation should always be conducted in order to try and establish an assignable cause and determine whether any other batches may be affected. In determining the assignable and root cause of the problem appropriate corrective and preventative actions can be undertaken.
<back to top>
3. Who should investigate OOS?
Both the manufacturers and the laboratories should be involved in the investigation.
<back to top>
4. How is an out of trend result handled?
Results that are out-of-trend (OOT) should be handled similarly to OOS investigations.
<back to top>
5. Is it acceptable for a contract laboratory (contract acceptor) to use the contract givers’ procedure when handling OOS results?
There is an expectation that contract acceptors should follow their own procedures and that these should be flexible enough to accommodate the needs of the contract giver.

It is assumed that the contract giver has assessed the contract acceptor’s procedure for handling of out of specification results and has agreed it as being suitable for their intended purpose. Any issues should have been discussed prior to conducting any analysis.|
<back to top>


6. How is a meaningful OOS investigation conducted?
A meaningful OOS investigation should be thorough, timely, unbiased, well-documented, and scientifically defensible.
<back to top>
7. At what point should a manufacturing investigation be initiated?
This should be initiated as part of the phase II investigation and as a result of the phase 1 investigation not revealing a conclusive laboratory error or the error remains unclear with no assignable cause.
<back to top>
8. What should be done if unexpected results are obtained and there is no obvious explanation?
These are also referred to as aberrant/anomalous. Preliminary laboratory investigation should occur and they should be handled similarly to OOS investigations.
<back to top>
9. Under what circumstances could test results become invalid?
If there is clear evidence of a determinant error. Or where the system suitability/method validity checks fail.
<back to top>
10. What should be done in the case where part way through testing the analyst realises there is an error?
If there is clear evidence of the error and it can be corrected without compromising the results or the validity of the method; for example a dilution error 20 ml volumetric flask used instead of a 25 ml volumetric then it should be handled as a deviation and the results are still valid. If there is any doubt as to the impact of the error which could mean the results may not be accurate, for example sample spillage then the testing should be stopped and the issue handled as a deviation to explain what happened.
<back to top>
11. When should the analyst inform the supervisor that they have an OOS results?
In the first instance, the analyst will be responsible for the preliminary laboratory investigation. This will involve them checking their work and confirming that there is no obvious error prior to informing their supervisor and initiating a phase 1 investigation. This should be done within a timely manner, preferably on the day of generating the results.
<back to top>
12. What should be done when the phase 1 investigation does not reveal an assignable cause or evidence of error remains unclear?
A phase II investigation is initiated, which will involve communication between the laboratory and the manufacturer/contract giver. The decision to undertake any further testing should be agreed and approved within a pre defined testing plan.
<back to top>
13. How many repeat tests should be conducted?
The minimum number of retests should be documented within the procedure and be based upon scientifically sound principles. Any statistical review with regards to %RSD and repeatability should relate to the values obtained during method validation, ie accuracy, precision and intermediate precision. The number of retests should be statistically valid.
<back to top>
14. What should be done if after retesting there is a combination of OOS results and pass results?
All results should be reported unless there is clear evidence of a determinant error or an assignable cause that could invalidate any of the results.
<back to top>
15. What should happen if the OOS investigations are inconclusive?
The certifying qualified person should fully consider all of the information prior to making any decisions as to the final disposition of the batch. Any decision to release a batch where OOS results have not been invalidated should come only after a full investigation has shown that the OOS result does not reflect the quality of the batch. In making such a decision quality assurance and the Qualified Person should always err on the side of caution.
<back to top>
16. When is it acceptable to average test results?
Where averaging of separate tests is appropriately specified by the test method, a single averaged result can be reported as the final test result. The validity of averaging depends upon the sample and its purpose. Using averages in the case of microbiological assay can provide more accurate results because of the innate variability of the microbiological test system. For example the kinetic scan of individual wells or endotoxin data from a number of consecutive measurements or with HPLC consecutive replicate injections from the same preparation where the determination is considered one test one result.
<back to top>
17. When is it not acceptable to average test results?
Averaging cannot be used in cases when testing is intended to measure variability within the product, such as powder blend/mixture uniformity or dosage form content uniformity. In the context of additional testing performed during an OOS investigation, averaging the result(s) of the original test that prompted the investigation and additional retest or resample results obtained during the OOS investigation is not appropriate because it hides variability among the individual results.
<back to top>
18. At what stage should retesting occur?
Retesting occurs at phase II of the investigation. The initial hypothesis testing can involve re-measurement of the original preparation or working solutions, however retesting is when the original sample or composite sample is used to perform analysis. Hypothesis testing and retesting are part of the phase II investigation. Only if the original sample is depleted or compromised should a new sample be used.
<back to top>
19. At what stage should re-sampling occur?
Re-sampling at phase II of the investigation should only occur if the original sample is depleted or compromised and the same method should be used. If the investigation determines that there were errors with the initial sampling method only then should a new accurate sampling method be developed, qualified and documented.
<back to top>
20. When is it appropriate to use outlier tests?
Statistical analysis for Outlier test results can be as part of the investigation and analysis. However for validated chemical tests with relatively small variance and that the sample was considered homogeneous it cannot be used to justify the rejection of data.
<back to top>


Page last modified: 28 August 2013