Thread Content
Typical laboratory problems 1: Question: Are reports required for the experiments conducted as part of the internal quality control plan? Things like personnel comparison and instrument comparison – how should the two sets of data obtained from testing be evaluated? Reference discussion: A report needs to be prepared. There is the tolerance method; use that. If it’s not applicable, you can use the En-value method and the t-test method. Question 2: Quality control requires that laboratories obtaining recognition for the first time participate in proficiency testing prior to such recognition. Then, do CMA laboratories also need to participate in proficiency testing before they are first recognized? No basis for CMA could be found. Reference discussion: Many of the requirements stipulated in CMA are conventional practices, with no specific basis to back them up. This suggestion should still be implemented; to verify our capabilities, we can use external methods to determine them. If the laboratory does not have the capacity to do everything, it should still try to carry out as much as possible. It is possible to participate in proficiency testing as well as inter-laboratory comparisons. At the very least, typical projects in each field should be carried out. Question 3: Regarding the report, what should be done if the test results fall between the detection limit and the quantitation limit? A value greater than the detection limit indicates that something has been detected; conversely, a value below the quantification limit means that the quantification is inaccurate. In such cases, is it acceptable to report the data based on the calculated actual values? Reference discussion: If the standard specifies the reporting method, follow that. If not specified otherwise, report the actually measured data, and then note what the limit of detection and the quantitation limit are. Also, explain what the limit of detection and quantitation limit mean in relation to the data results, providing a brief description of the overall results. Question 4: Regarding reports for sample testing, that is, when no sampling is done and the samples are provided by the customer, are there any particular considerations for such reports? To give a simple example: a hospital sends over a bottle of water for analysis of a certain indicator. Should the report include the information noted by the client at the time of sample submission—such as the sampling location, the exact time of sampling, and whether a preservative was added? Additionally, are there any requirements regarding whether the submitting entity has the necessary qualifications and capabilities to perform this type of testing? Reference discussion: We should note that we’re only responsible for sample receipt. We are not responsible for verifying the accuracy of the information provided by customers in the statement. That’s it. It must be made clear in both the commission contract and the report. Additionally, the client’s capabilities meet our requirements; therefore, there is no need for us to verify them. 5 Question: When samples are transferred within the laboratory, can the client’s name be indicated on the sample labels? Reference discussion: It is best not to reveal this information in order to ensure fairness; the experimenters should not know where the samples come from or whose they belong to, so as to avoid any differential treatment. 6. Should the original records document the results by batch, or should there be a separate original record for each order? Reference discussion: It is recommended to keep a separate original record for each commission order; if a customer comes to view the records, they shouldn’t be able to see others’ records. It can be done in batches, but when organizing the records, it is still best to keep the original records separate. Question 7: Can the technical supervisor serve as a tester? Reference discussion: Yes, it’s necessary. The person in charge of technology shouldn’t stay in the office all day long; it’s difficult to maintain one’s technical skills over the long term without working in the laboratory. Question 8: For detecting the concentration of the heavy metal Cd, equipment comparison methods are required, and t-tests are used to analyze the results. During sample preparation, should one sample be processed multiple times to obtain measurements, or should N separate samples be processed, each time to be measured? Reference discussion: It must involve processing N samples and making N measurements. Question 9: In some experiments that generate electronic spectra, the sample number was entered incorrectly. Since the samples cannot be tested again, is it acceptable to manually write down the correct number on the spectrum and note the reason in the remarks? Reference discussion: It’s okay, but it’s necessary to note who wrote it when and why it was written that way. Question 10: Can the report reviewer review the experiment records they themselves conducted? Reference discussion: First of all, if there are no specific requirements in the relevant field that mandate a third-level review, then a review can be carried out. However, this is not recommended, as the purpose of the third-level review is to identify issues that were missed during the initial reviews; if one reviews themselves, then that purpose cannot be achieved. 11 Question: When retaining samples, should samples whose shelf life has expired but still fall within the retention period stipulated by the company be kept? Reference discussion: You can check how it is specified in your internal documents; generally, such regulations take into account the shelf life of the sample itself. There’s no point in keeping it past the expiration date. If you feel that what’s written in the document isn’t appropriate, you can revise it. Also, the shelf life of this sample must be specified in the commission contract. We need to meet the retention period agreed upon with the client.