这两天，金毅老师分享给业内一个SQA邮件群的邮件内容，是关于试验记录中能否通过勾选代替具体数据录入的，现将文件内容分享给大家，并且进行了翻译和简单解读，与大家共同学习。
有位GLP人员在邮件中提出了一个他们在接受监管机构检查中遇到的问题：
As we all know, TK sample analysis is doing under the guidance of effective analytical method after method validation. So the STD\QC preparation and sample processing steps are fixed according to the method. In order to conveniently record and ensure the data accuracy, therefore the lab adopted following measures to optimize their raw data recording way:
- Design and print the record form in advance to having the fixed method information as the expected content, such as Standard\QC sample preparation volume, concentration and sample processing steps. Using the checkmark box to indicate the actual content, also indicating the checkmark in the record form means confirming the actual content is consistent with the expected content. The form information will be QCed according to the method.
- Lab technician will use the above record form template after application and release, due to the form in the lab is controlled which will be stamped after printed. During the TK sample processing, technician only will confirm the actual content is consistent with the expected content, then checkmark to record the raw data.
Surely, this recording way is convenient and accurate. But we worry about if this is deviating from GLP principle about raw data record.
Actually, we have one remote GLP audit last year, the auditor has reported "the record of particular volume was a checkmark in a column, as opposed to entering the exact amount that was added… it is a minor finding in the audit report, which against the regulation §58.3(k) more detailed documentation is need to assist with study completeness for possible reconstruction of the study. "
众所周知，TK样品的分析是在经过方法验证的有效分析方法的指导下进行的。因此，根据该方法确定了STD\QC制备和样品加工的步骤。为了便于记录，保证数据的准确性，本实验室采取了以下措施来优化原始数据的记录方式:
- 事先设计并打印记录表，以标准\QC样品制备量、浓度、样品处理步骤等固定方法信息为预期内容。使用复选框来指示实际内容，同时也指示记录表单中的复选框意味着确认实际内容与预期内容是一致的。表格信息将根据方法进行QC。
- 由于上述表格在实验室是受控的，在打印后需要盖章，所以在申请和发放后，实验室技术人员将使用上述表格模板。在TK样品处理过程中，技术人员只需要确认实际内容与预期内容一致，然后打勾记录原始数据。
当然，这种记录方式既方便又准确。但我们担心这是否偏离了GLP原则的原始数据记录要求。
实际上，我们去年有一次远程的GLP审计，审计人员报告称“特定数量的记录在一栏中打了个勾，而不是输入添加的确切数量……这在审计报告中是一个很小的发现，这违反了§58.3(k)原始记录定义的规定，需要更详细的文件来帮助研究的完整性，以便可能的研究重建。”
对于这个问题，某位专家在邮件中回复到：
In my experience with regulators a checkbox alone for finite values such you describe has been considered insufficient.
In a past life, we had a similar situation where a very detailed form was designed for the lab's convenience. Every step was written out and documentation was via a checkbox and initial/date. During an FDA inspection this was questioned and strongly recommended we modify the form so the technician was forced to record the actual volume. The inspector felt checkboxes were OK as long as finiter values such as theactual volumes, temperatures, etc. were also recorded. We did not get written up on a 483 nor did it show up in the EIR, but we had changed the form before the inspection concluded.
And yes, they did ask about it at the next inspection.
在我与监管机构的经验中，您所描述的有限值的复选框被认为是不够的。
在过去，我们有过类似的情况，为了实验室的方便，我们设计了一个非常详细的表格。每个步骤都是通过复选框和初始日期写出来的。在FDA的一次检查中，有人质疑这一点，并强烈建议我们修改表格，这样技术人员就被迫记录实际的体积。检查员认为，只要实际体积、温度等有限值也被记录下来，复选框就可以。我们没有被写在483上，也没有出现在EIR中，但我们在检查结束前更改了表格。
是的，他们在下次检查时确实又检查了该问题。
由上可知：
试验中的具体数据不能以勾选的方式来代替，如不能通过勾选设定好的符合要求的范围或数值，来代替将具体数值填写在记录中。
FDA检查中会很关注此项内容。
另外，从文件中我们还可以获知记录受控分发的大体流程，即可以是在电子表格上先编辑然后经过QC，打印、盖章，通过申请后发放。具体步骤可以各机构各异，总之是受控管理。
阅读延伸
