It is common that at times you will get results on samples of known materials that will lie outside the specified range of values. This problem is common in quality control laboratories where almost all analysis is carried out on known samples and you have to communicate your pass / fail verdict. The analysis report should be withheld and the suggested course of action completed before you communicate the final decision.
What is out of specification result?
Out of specification implies test results which fall outside specifications or acceptance criteria established by the product manufacturer or the laboratory.
Investigation of out of specification results
Out of specification results can result from either of the two situations
Laboratory testing errors
Product manufacturing errors
In this article the focus will be on laboratory testing errors. Before the investigation is started it is advisable to get a confirmation on the following points:
- Tests were performed under prescribed environmental conditions
- Tests were conducted by a validated analyst
- Whether validated test methods were adopted and SOP is followed without deviations
- Equipment and glassware used were calibrated
- Reagents used were within their validity
- All results were recorded honestly and simultaneously to the analysis
After completing the initial investigations it is essential that documentation be completed for future reference.
Out of specification investigation procedure
On coming to know about an OOS result the laboratory in- charge should first of all jointly conduct the initial investigations as detailed above and make a written report of the findings. This investigation should be completed before test solutions are discarded. Re-testing should be done from original sample unless it has been consumed or spilled. The same analyst should retest in duplicate from the original sample
In case the original sample is not available re-sampling should be done and record of re-sampling should be made. If duplicate analysis of the sample passes the test the original OOS report can be discarded. On the other hand if duplicate tests fail the OOS is valid and the sample should be declared as failed. In the event of not being able to pinpoint the failure the sample should be sent to another accredited laboratory.
If error can be pin pointed different analyst should conduct the analysis in duplicate and if sample passes the test the OOS report will become invalid and sample will be passed.
All the investigations need to be documented. A corrective and preventive action plan be prepared for record purpose. OOS investigations deserve priority handling and should be completed within maximum three working days.